Space, Place and Identity: Wodaabe of Niger in the 21st Century 9781789206371

Citation preview

Space, Place and Identity

Integration and Conflict Studies

Published in association with the Max Planck Institute for Social Anthropology, Halle/Saale Series Editor: Günther Schlee, Director of the Department at the Max Planck Institute for Social Anthropology Editorial Board: Brian Donahoe (Max Planck Institute for Social Anthropology), John Eidson (Max Planck Institute for Social Anthropology), Peter Finke (University of Zurich), Jacqueline Knörr (Max Planck Institute for Social Anthropology), Bettina Mann (Max Planck Institute for Social Anthropology), Ursula Rao (Leipzig University), Stephen Reyna (University of Manchester), Olaf Zenker (Martin Luther University) Assisted by: Viktoria Giehler-Zeng (Max Planck Institute for Social Anthropology) The objective of the Max Planck Institute for Social Anthropology is to advance anthropological fieldwork and enhance theory building. ‘Integration’ and ‘conflict’, the central themes of this series, are major concerns of the contemporary social sciences and of significant interest to the general public. They have also been among the main research areas of the institute since its foundation. Bringing together international experts, Integration and Conflict Studies includes both monographs and edited volumes, and offers a forum for studies that contribute to a better understanding of processes of identification and intergroup relations. Recent volumes: Volume 21 Space, Place and Identity: Woɗaaɓe of Niger in the 21st Century Florian Köhler Volume 20 Mobile Urbanity: Somali Presence in Urban East Africa Edited by Neil Carrier and Tabea Scharrer

Volume 16 Difference and Sameness as Modes of Integration: Anthropological Perspectives on Ethnicity and Religion Edited by Günther Schlee and Alexander Horstmann Volume 15 On Retaliation: Towards an Interdisciplinary Understanding of a Basic Human Condition Edited by Bertram Turner and Günther Schlee

Volume 19 Playing the Marginality Game: Identity Politics in West Africa Anita Schroven

Volume 14 ‘City of the Future’: Built Space, Modernity and Urban Change in Astana Mateusz Laszczkowski

Volume 18 The Wheel of Autonomy: Rhetoric and Ethnicity in the Omo Valley Felix Girke

Volume 13 Staying at Home: Identities, Memories and Social Networks of Kazakhstani Germans Rita Sanders

Volume 17 Bishkek Boys: Neighbourhood Youth and Urban Change in Kyrgyzstan’s Capital Philipp Schröder

Volume 12 The Upper Guinea Coast in Global Perspective Edited by Jacqueline Knörr and Christoph Kohl

For a full volume listing, please see the series page on our website: http://www.berghahnbooks.com/series/integration-and-conflict-studies

Space, Place and Identity Woɗaaɓe of Niger in the 21st Century Florian Köhler

berghahn NEW YORK • OXFORD www.berghahnbooks.com

First published in 2020 by Berghahn Books www.berghahnbooks.com © 2020 Florian Köhler

All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without written permission of the publisher. Library of Congress Cataloging-in-Publication Data Names: Köhler, Florian, author.  Title: Space, place and identity : Woɗaaɓe of Niger in the 21st century /    Florian Köhler.   Other titles: Integration and conflict studies ; v. 21.  Description: New York : Berghahn Books, 2020. | Series: Integration and    conflict studies ; v. 21 | Includes bibliographical references and    index.  Identifiers: LCCN 2019051315 (print) | LCCN 2019051316 (ebook) | ISBN    9781789206364 (hardback) | ISBN 9781789206371 (ebook)  Subjects: LCSH: Woɗaaɓe (African people)--Niger--Social conditions--21st    century. | Group identity--Niger. | Rural-urban migration--Niger. |    Nomads--Sedentarization--Niger. Classification: LCC DT547.45.B67 K64 2020  (print) | LCC DT547.45.B67     (ebook) | DDC 305.89632206626--dc23 LC record available at https://lccn.loc.gov/2019051315 LC ebook record available at https://lccn.loc.gov/2019051316 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library

ISBN 978-1-78920-636-4 hardback ISBN 978-1-78920-637-1 ebook

Contents

Contents

List of Illustrations

vii

Acknowledgementsix Note on Language and Transcriptions

xi

Introduction1 Part I.  Taariihi: Mobility and Group Formation in Historical Perspective   Chapter 1.  The Woɗaaɓe in Niger: Structure as Historical Process

21

  Chapter 2. A History of Migrations: Placemaking Processes in Diachronic Perspective

28

Part II.  Duuniyaaru: Spaces of Social Interaction   Chapter 3. Inter-ethnic Relations: The Balance of Integration and Conflict55   Chapter 4. A Meta-ethnic Social Space: The Continuum of Identity and Difference

63

Part III.  Ladde: Transformations in the Pastoral Realm   Chapter 5. From Nomadic Pastoralism to Sedentarization and Economic Diversification

77

  Chapter 6.  Consequences of the New Spatial Strategies

99

vi  Contents

Part IV.  Si’ire: Appropriating the City   Chapter 7.  New Resources in the Urban Space

123

  Chapter 8.  Social Interaction in the City

138

  Chapter 9.  The Translocal Dimension of Urban Migration

149

Part V.  Gassungol Woɗaaɓe: The Translocal Network of the Ethnic Group Chapter 10.  The Translocal Community and Social Reproduction

173

Chapter 11.  Cultural Change and the Reproduction of Difference

190

Conclusion203 References209 Index229

Illustrations

Maps   0.1   2.1   5.1   5.2

The study region. Migrations of Ɗawra Egoyi and his ancestors. The main field sites and the northern limit of farming. Seasonal transhumance movements of the Gojanko’en in the Koutous region.   8.1 Urban clustering of Woɗaaɓe Gojanko’en in Zinder.

4 30 78 82 140

Figures  1.1 Woɗaaɓe clan structure, current leaders within the study group and sites of attachment. 22   2.1 The internal structure of the Woɗaaɓe Gojanko’en. 33   2.2 The kinship relationship of Atiiku and Giiye. 34   2.3 Giiye and his descendants. 37   2.4 Marriages of Hamma Beleti. 46  3.1 Woɗaaɓe in the court of their hosts’ house in the market village of Gueza.59   4.1 Young Boɗaaɗo performing a mock-Fulɓe dance in front of Fulɓe Ndoovi’en.68   5.1 The two leaders of the Gojanko’en community in Ganatcha. 80   5.2 The genealogical relation of Ɗawra and Umaru. 89   5.3 Schoolchildren in Ganatcha. 95   7.1 Urban dwelling in a construction site. 129

viii  Illustrations

  9.1 10.1 10.2 10.3

Urban connectors. 162 Choosing ritual at the climax of a geerewol-dance.177 Preparing the presentation of the ngaari ngaanka.182 A string of bark fibre is fixed to the nares of the ngaari ngaanka.183

Tables   4.1 Identity categories and the socio-cultural dimensions of identification.72   6.1 The leaders of the study group and their politico-administrative affiliation.108 10.1 The role of the two main forms of marriage for social cohesion. 175

Acknowledgements

I owe gratitude to the following persons and institutions who have contributed in some way or the other to the work that has resulted in this book. First, I want to thank my numerous Woɗaaɓe-hosts in Niger for endless hours of conversation during which they shared their knowledge and their experiences, their stories and histories, their joys and sorrows, their milk and their tea with me. Many other people in Niger were also very welcoming and helpful in providing emotional, intellectual, logistic, administrative or other support during my fieldwork. In particular, I want to thank Amadou Siddo, Moussa Baouada, Abdou Issa, Mamane Ousseini, Anna Coendet, Mirco Göpfert, Eric and Halima van Sprundel, KarlHeinz and Heidi Siekmann. I am grateful to the Max-Planck-Society for the financial and institutional support that made this work possible, and to the supporting members of the Max-Planck-Society for an additional language learning grant. I greatly profited from the generous working environment of the Max Planck Institute for Social Anthropology in Halle (Saale), Germany that provided ideal working conditions with an excellent library service, cartographers and scientific assistants. My particular gratitude goes to Günther Schlee, the director of this institute’s department for ‘Integration and Conflict’, who trusted in my ability to do this scientific work after almost a decade outside academia, and to Nikolaus Schareika, who had helped me during that decade to hold on to the aim of developing an anthropological research project on the Woɗaaɓe and to develop ideas for it. Numerous colleagues have commented on draft versions of chapters and contributed with their suggestions and critical remarks: Al-Amin Abu-Manga, Lucie Buffavand, James Carrier, Solange Guo-Chatelard, John Eidson, Joachim Görlich, Stephen P. Reyna, Tabea Scharrer and Timm Sureau and in particular Martine Guichard from whose complementary perspective

x  Acknowledgements

I profited a lot for putting many phenomena into the wider context of Fulɓe studies. The greatest thanks, however, I owe to my family, both in Niger and in Germany, for their patience, trust, support and love. I therefore dedicate this work to them.

Note on Language and Transcriptions

Interviews and narrating sessions took place mainly in Fulfulde and Hausa, the principle being to allow the interlocutors the choice of language. Most interlocutors spoke Hausa with more or less ease and especially in the initial phase of fieldwork, my Hausa was more developed than my Fulfulde. In the transcriptions, Fulfulde texts are not marked while Hausa texts are indicated by a ‘(H)’ behind the text to facilitate differentiation for the non-familiar reader. The Fulfulde transcriptions follow the orthographic conventions agreed upon at the 1966 conference of Bamako, with exceptions due to regional specificities. For example, the phoneme ‘ʧ’ in Fulfulde, generally written as ‘c’ in standard transcriptions, is often rather pronounced ‘s’ by Woɗaaɓe speakers in the study region and has been transcribed accordingly. Another peculiarity is that long vowels are often split and pronounced as two short vowels separated by a glottal stop, hence for example ‘o’o’ instead of ‘oo’. The Hausa transcriptions equally render the language actually spoken rather than applying a standardized orthography that would level out regional and idiosyncratic peculiarities. Vowel length and tonality have not been marked with diacritics. The Hausa spoken in Niger – and in particular by non-native speakers such as the Woɗaaɓe – tends to lose much of its tonality so that an exact linguistic transcription would be as awkward for Hausa linguists as would a transcription using the standard tonality falsify the impression of the language spoken by the Woɗaaɓe. Since the text is not primarily addressed to linguists, I have opted for relatively free translations of the original accounts, focussing on semantic content rather than labouring to stay true to structural peculiarities of the spoken language where these might have obscured the meaning and made the reading difficult.

Space, Place and Identity

Introduction

Living in Niger over several years in the 2000s, I was faced with an intriguing puzzle: while the sedentarization of nomadic pastoralists and a decrease in pastoralist mobility was a current issue, the society as a whole was becoming increasingly mobile. This book asks how this second, more general, trend is linked to the first: how do sedentarization processes among nomads translate in terms of actual mobility? What are the new forms of mobility that emerge in the context of alternative livelihood activities such as work migration, which is a significant phenomenon in many pastoral societies today, and how are they linked to the rapidly developing facilities of public transport? What does such change in mobility patterns imply for the relations of pastoralists to space and place? What are its wider consequences in terms of social group formation and collective identification, for questions of integration into the wider society and the structures of the modern nation state, and with regard to social and cultural reproduction? These questions are pursued by analysing the case of the Woɗaaɓe, a group of Fulɓe pastoralists who had once been an almost paradigmatic case of highly mobile cattle nomads, but who are today characterized by sedentarization and livelihood diversification, based mainly on agro-pastoralism and urban work migration. It was rather per chance that, immediately after my first arrival in Niger in 2004, I became closely acquainted with an urban-based Woɗaaɓe family. Taafa Buuyo, the head of this family, had worked as a watchman for the previous tenant of my house in Diffa, a provincial capital in the extreme east of the country, where I would work on a project for conflict prevention and resource management. In Niger, as in other African countries, it is customary for Western expatriates to employ watchmen at private domiciles. At the time, there was no obvious need to have them, since the security situation in Diffa was comparatively relaxed then,

2  Space, Place and Identity

but the employment of watchmen was an established custom and had become an important sector of local employment. Taafa and his family were living in a small house on my compound, hoping for a new work contract upon my arrival. We thus lived in close proximity and I soon spent most of my spare time with this family, sharing meals, tea and company. Over time, our relation developed into friendship. What drew us close to each other – despite our cultural and economic differences – might also have been the fact that we were both work migrants far from home: Taafa originated from the distant Zinder province and, apart from his immediate family, did not have any relatives in Diffa. By and by, I became familiar with Taafa’s life story. As with many other Woɗaaɓe, he had abandoned the pastoralist livelihood as a young man to work as a watchman in different cities across Niger and Nigeria. Although far removed from his wider family, he was in no way isolated from them, but imbedded in a tight lineage network. I understood that his urban activities were part of a joint strategy to sustain the extended family’s pastoral economy. Taafa regularly sent parts of his earnings home to his younger brother who invested them in animals, and despite the distance, Taafa regularly returned to his pastoral home camp for visits. Over the years, he had worked for different European expatriates, which had enabled him to establish a social network that constituted an important economic backdrop. Later, I realized that Taafa’s story combined elements that were recurring themes in contemporary Woɗaaɓe biographies and that similar life stories as his could be found in greater number among his lineage mates. His case exemplified the significance of both translocal kinship-networks and networks with Westerners, which were to become an important focus of my research – and of which I had myself become a part. After three years in Diffa, I left Niger, but came back a year later to work on a similar project, this time in the city of Zinder. Taafa still had employment in Diffa, but since many members of his extended family lived in Zinder as migrant workers or as stranded drop-outs of the pastoral economy, I was soon addressed by Taafa’s half-brother (FS), Baji, who asked for work as a watchman. His family thus moved to my compound and I learned that the situation in Zinder was quite different from what I had experienced in Diffa. Not only was there an urban migrant community of Baji’s lineage living in Zinder, the city was also a waypoint between the pastoral areas of the Damergou and Koutous regions, where the families of most Woɗaaɓe migrants in Zinder live, and the city of Kano, in northern Nigeria, which also attracts significant numbers of Woɗaaɓe migrant workers from Niger. As a result, many friends and relatives of Baji soon regularly spent their days and sometimes longer periods on my compound – migrants in transit to or from Kano, or lineage mates who had come to town and were looking for work but had momentarily no place to stay. I thus quickly learned more about their lives – their experiences, aspirations and challenges – in which I became more and more interested.

Introduction  3

The wish to record the history and to document the contemporary condition of this particular group of Woɗaaɓe thus developed gradually over years during the course of my close acquaintance with those who were my principal company when I first tried to make a home in Niger, and who were to become my principal interlocutors in the context of anthropological research in the proper sense, on which I finally embarked in November 2010.

The Woɗaaɓe of Niger Niger is a landlocked state in the West African Sahel. The young and fast-growing population of about 17 million, according to a census from 2012 (Republic of the Niger 2013), is concentrated in the relatively fertile south-west, while the vast arid north is characterized by the Sahara desert that covers over 80 per cent of the country’s land area. Niger is regularly assessed as one of the poorest countries in the world: in the 2013 UN Human Development Index, it was ranked last of 187 countries (UNDP 2014). Apart from the extraction and exportation of raw materials, notably uranium and, since recently, petroleum, the economy of Niger is still dominated by subsistence agriculture and livestock rearing, the latter notably in the form of pastoralism of a varying degree of mobility. The Woɗaaɓe (sg. Boɗaaɗo) are part of the large group of Fulɓe who are today dispersed in wide parts of West-Africa, and to a lesser degree across the whole continent, from Senegal (Dupire 1970) to Ethiopia and Sudan (Braukämper 1992; Delmet 1994, 2000; Feyissa and Schlee 2009; Schlee 1997, 2000b, 2011, 2012, 2013a), and Central Africa (Boutrais 1990). The cultural differentiation between different Fulɓe groups has historically been very pronounced, ranging from nomadic pastoralist and sedentary agro-pastoralist groups, to warrior aristocrats, Islamic scholars and founders of emirates. Within the total population of the Fulɓe, which, in more recent writing, has been estimated at about sixteen million (Diallo 2008: 7), the Woɗaaɓe represent only a small minority. According to estimates from the 1980s and 1990s, their number in Niger then amounted to approximately 100,000 (Bonfiglioli 1988: 12; Paris 1997).1 While smaller populations can be found in Cameroon, Nigeria, Chad and the Central African Republic, but even as far east as Sudan (Mohamadou 1969: 72; Osman 2013), and as far west as Senegal (Faliu 1980; Kane 2004), the majority of Woɗaaɓe live in the Republic of Niger, predominantly as mobile pastoralists specialized in the breeding of zebu cattle. In the course of their migration to increasingly northern regions of the Nigerien Sahel, and in reaction to historic processes, notably the Fulɓe Jihad in the nineteenth century and later the French colonization and the gradual extension of agricultural lands, the Woɗaaɓe have, since the end of the nineteenth century, developed the high degree of nomadic mobility which has since then remained characteristic (Braukämper 1971; Bonfiglioli 1988; Boesen 2004a: 212, 2007a: 31, 2007b: 209f.).

4  Space, Place and Identity

Map 0.1  The study region.

The pastoral Woɗaaɓe are organized in loose migration groups, formed by a varying number of households. These kinship-based groups of patrilineal descent conceive of themselves as segments of more complex lineages and clans (Dupire 1970: 303). The different clans, and in many cases even the segments of one clan do not live in spatial proximity, but are dispersed over important distances and sometimes isolated one from another. A rather vague notion of the unity of the group is being maintained through a network of relations between regional clan segments, but institutions or occasions that would unite the Woɗaaɓe as a whole

Introduction  5

across clans do not exist (Dupire 1962: 319; 1970: 300f.). In the absence of any central political institution, the most important political function is that of the arɗo (pl.: arɗuɓe). Originally a pastoral leader and a political and moral authority on the level of a clan segment, he is nowadays attributed administrative functions by the state. The authority of an arɗo is based solely on his personal qualities as a leader, and any family head can at any moment withdraw his allegiance to follow another arɗo. Decisions concerning interior affairs of the clan or between clans are taken communally by a council of elders. The most important institution for maintaining inter-clan relations are ceremonial meetings (ngaanka), which take place at the end of the rainy season on the basis of reciprocal visits between two clans or their regional segments. By fostering exogamous inter-clan marriages (te’egal) and thus translating inter-clan relations into kinship ties, these meetings are an important tool for strengthening the cohesion of the otherwise fragmented ethnic group. Paradoxically, this is achieved by what are perceived as acts of aggression, since the clan-exogamous te’egal marriages are by principle arranged with women who, in their own clan, are already married, generally by clan-endogamous betrothal from early childhood (kooɓgal-marriage). The ngaanka ceremonies are thus an arena in which two clans ritually approve of mutual te’egal elopement marriage and lay the basis for it by exposing married women and men to each other during male dance contests, for which the Woɗaaɓe are probably best known in the West (Dupire 1970: 67; Paris 1997; Boesen 2008a). Although forms of elopement marriage are known among other groups of pastoral Fulɓe as well (e.g. Bocquené 1986: 247ff.; Burnham 1996: 111f.; Reed 1932: 433), the particularity of Woɗaaɓe te’egal marriage is that it occurs within a formalized regulatory framework based on inter-clan agreements that, generally speaking, sanction the practice between clans and ban it from within one clan. These inter-clan agreements are the principal issue at stake in ngaanka, and ultimately, the participation of a group in the network of ceremonial and marital relations is what defines Woɗaaɓe ethnic identity and group membership. Since the 1970s and especially since the 1980s, as a result of animal losses after recurring droughts in the Sahel region, many Woɗaaɓe have taken up work migration to regional urban centres (Maliki et al. 1984; Loftsdóttir 2000, 2002a, 2004; Boesen 2004a, 2007a). Income from migrant work has become a significant economic factor and in many cases the pastoral economy has today been transformed into a mixed system subsidized by urban revenues (Boesen 2007b: 210). Social networks being of major importance for finding paid labour, different professional specializations among migrant workers can be more or less associated with clan groups (Loftsdóttir 2000: 249f.; Boesen 2004a: 215). The Woɗaaɓe Gojanko’en in the Zinder province, on whom my research focused, have for a long time been particularly well positioned in the job market for watchmen at expatriates’ homes and offices in the provincial

6  Space, Place and Identity

capital Zinder. The second important city to which they migrate is Kano in northern Nigeria. Most major works on the Woɗaaɓe have put a strong emphasis on aspects of ‘traditional’ pastoral nomadic culture and livelihood. This holds true not only for the classic monographs by Stenning (1959) and Dupire (1962), and for Bonfiglioli’s seminal Duɗal (Bonfiglioli 1988), but also for more recent contributions such as those by Schareika (2003a) on environmental knowledge, Krätli (2007) on resource management in cattle breeding, or Loncke (2015) on song and dance. Although anthropologists, in particular Boesen (2004a, 2007a, 2007b, 2010) and Loftsdóttir (2000, 2001a, 2002a, 2004), have also addressed complementary economic strategies outside pastoralism and the new forms of mobility which they entail, little effort has been made to analyse them in a comprehensive way, i.e. to explore the systemic logic in the interplay of different economic activities both in the rural and the urban sphere, and of the different actors involved in them. One aim of this book is to fill this gap, based on the premise that the (agro-)pastoral and the urban context are today complementary spheres of Woɗaaɓe social and economic activities, closely linked by multiple ties of mobility. I argue that for understanding the dynamics characterizing this society today, it is indispensable to analyse not only these two contexts, but in particular the complex and multiple translocal relations between them.

The Study Group and Central Research Questions As the perspective of conducting anthropological research in this context became more concrete and I pursued my investigations, I soon learned that the migration of this group of Woɗaaɓe into the Zinder province occurred rather recently, in the early 1970s. Also, they constituted only a relatively small section of their clan, the Gojanko’en, the majority of whom today live in the Ader region of central Niger. Within this small group, economic diversification covered a wide spectrum from pastoralism of different degrees of mobility to agro-pastoralism and urban work migration. Individuals moved flexibly between the different economic models and between the urban and the rural realm. Ultimately, this diversity of different livelihood models that coexist today within a relatively small faction of a Woɗaaɓe clan seemed to offer a reasonable framework for research. I thus made the Gojanko’en in the province of Zinder my principal group of investigation (henceforth called the study group). I put a focus on both rural areas (Damergou and Koutous regions) and urban locations (Zinder), and I used existing contacts to Woɗaaɓe from other clans and regions, in Diffa and Tahoua provinces, to collect comparative data. I learned that the study group’s migration to the Damergou region was only one move in a long history of migrations – and hardly the end-point, as the most recent migration of one small faction from the Damergou to the Koutous region

Introduction  7

indicates. Mobility was a central issue in more than one regard: in the context of pastoral migrations and pastoral day-to-day mobility, in the form of a more recent rural–urban mobility, and in the form of the socio-economic mobility, or flexibility, that it takes for switching between these different spheres and between different livelihood strategies. However, I also learned that in parallel to this continuing thread of mobility and migration, there was a more recent and seemingly rather opposite tendency of sedentarization of pastoral Woɗaaɓe in the vicinity of wells, which has led, in recent years, to an increasing differentiation of the regional clan community along different points of local attachment. This trend towards territorial fixation can be understood as a strategy for securing legal rights over resources, and as an attempt at selective integration into state structures in order to be included in processes of resource distribution by governmental and non-governmental development programmes. It must be seen in the light of contemporary developments within Nigerien society as a whole – particularly urbanization and the periodically massive presence of international aid-organizations – which have had an enormous impact on the society. These facts opened the perspective on two major issues that were to become central points of reference for my research: (1) the aspects of mobility and migration (both in the pastoral and in the urban context), and the processes of placemaking and local attachment which they entail; (2) processes of social group formation and collective identification. Among the Woɗaaɓe, mobility and migration go together with processes of group formation. Social groups form in a continuous process of reconfiguration, following a double and inverse pattern of fission or disjunction of descent groups, on the one hand, and affiliation or fusion of local groups after periods of co-residence or coordinated mobility within a particular area, on the other (Dupire 1962; Bonfiglioli 1988). This pattern of group formation induces processes of collective identification based on differentiation and redefinitions of internal boundaries, and others, based on the constant renegotiation of external boundaries that the migration experience, both in the pastoral and in the modern urban context entails. In his classic work Ethnic Groups and Boundaries, Fredrik Barth (1969) showed how group identities are shaped not in isolation of one group, but in confrontation with others along boundaries that are constructed in this process. Although Barth’s use of ‘boundary’, i.e. a spatial metaphor, for what is in fact an ideological social construction has been much criticized (Cohen 2000; Wood 2009), the concept can nonetheless fruitfully be applied today if the boundary is not regarded as a fixed spatial barrier, but as a dynamic and situational interface with the other (Burnham 1996: 161f.). Crucial group interfaces that define and shape the possible identification categories of a Boɗaaɗo have to be analysed both on the external level (interfaces between ethnic groups) and on the internal level (interfaces between the segments of Woɗaaɓe society on different levels).

8  Space, Place and Identity

One central research question thus concerns identification processes in a context of mobility and migration, and the roles that space and place, or locality, play in this process. Different places provide different constellations of neighbourhood and interfaces with other groups and hence, potentially produce different identities. This is doubly relevant in the context of this study: First, the Fulɓe are a paradigmatic case of a group that, over the course of its history, was characterized by high mobility and continent-wide migration, and had to redefine its identities along ever new constellations of neighbourhood with various cultural others (Diallo and Schlee 2000; Schlee 2011). Second, the context of urban migration is of particular interest for an analysis of identification processes because it offers, in a nutshell, permanent interface situations in the sense of Barth’s group boundaries along which processes of identity formation and (re-) negotiation unfold. The multi-ethnic context of the city works as a catalyst for identification processes, because the close neighbourhood with ethnic others entails a constant contestation of group identities and thus requires a constant re-negotiation of identity and difference, of belonging and otherness (Schlee 2013b). Another question concerns the impact that the exposure to a modern urban lifestyle and the resulting culture change in the contemporary condition have on questions of cultural reproduction and cultural continuity. The issue of spatiality is also of significance for the reproduction of social groups in a highly mobile society: how is a social group maintained as a community over spatial distance, and what is the role of places and placemaking in this context? The issue of connectedness over space in a dispersed community introduces several concepts that need to be defined, notably ‘community’ and ‘locality’, and ‘translocality’.

The Translocal Production of Community The terms ‘translocality’ and ‘translocalism’ have variously been used since Arjun Appadurai (1995: 216) proposed a better understanding of the processes of ‘production and reproduction of locality’ in a context of increasing global mobility with the term ‘translocalities’. More recently, ‘translocality’ has become a catchword for researchers from various disciplines concerned with the phenomena of migration, mobility, transfer and spatial interconnectedness (Greiner and Sakdapolrak 2013: 373), yet the use of the term is often characterized by a relative vagueness (Ben Arrous 2004; Greiner and Sakdapolrak 2013). While, generally speaking, the concept refers to processes and relations that span different locales, owing to the recent career of the term in social sciences, the different nuances of meaning that it carries merit a closer look. On the one hand, using the concept of translocality presupposes an understanding of the notion of ‘locality’. In a first, more elementary sense, the term refers to a spatial form, a place in the geographic space. A second, more sociological use

Introduction  9

of the term applies it to a form of sociality. Appadurai (1995: 222) has argued in this sense that locality is negotiated and socially ‘produced’, and thus ‘primarily relational and contextual, rather than scalar or spatial’ (ibid.: 204). Appadurai admits that ‘[t]here is no ideal way to designate “localities” as actual social forms’ (ibid.: 222), and he proposes the term ‘neighbourhood’ which, however, does evoke a specific spatial location. The term ‘community’, by contrast, has the advantage of not being locally but rather interactionally defined (De Jong 1999). It has been argued that ‘the apparently immediate experience of community is in fact inevitably constituted by a wider set of social and spatial relations’ (Gupta and Ferguson 1997b: 7) and that ‘communities can exist without being in the same place’ (Massey 1994: 153), i.e. as translocal communities. This has become an important position in social sciences due to the fact that today ‘the geography of social relations is changing’, in many cases being ‘increasingly stretched out over space’ (ibid.: 154). On the other hand, translocality must be understood in close relation with the better-established concept of transnationalism (Freitag and von Oppen 2010: 5). The debate on transnationalism in migration studies has made an important contribution to shifting the focus from viewing migrants as uprooted and disconnected from their home communities to stressing their connectedness with their countries and regions of origin (Basch et al. 1995; Glick Schiller et al. 1995; Hahn and Klute 2007). They continue to engage with the places they have left behind (Hedberg and do Carmo 2012: 3) and they are often simultaneously situated across different locales (Brickell and Datta 2011: 4), thus forging identities of multiple belonging (Greiner 2010: 135). However, while transnationalist approaches, as the term implies, have put a strong focus on nation-states and the borders between them, the latter can be rather arbitrary and in some contexts highly irrelevant. It has also been pointed out that transnational practices ‘are actually embodied in relations which are situated in specific local contexts’ (Brickell and Datta 2011: 9). This led to a new focus in transnational studies on the situatedness of migrants and on local–local relations, and eventually to a development towards a more grounded transnationalism, or, ‘transnationalism from below’ (Smith and Guarnizo 1998). Authors began to stress the simultaneous mobility and locatedness of migrants, and put a greater emphasis on migrants’ agency (Brickell and Datta 2011: 9). Gradually, the concepts of ‘translocality’ or ‘translocalism’ emerge from these discussions; the discourse on ‘transnational migration’ is increasingly complemented by the terms ‘translocal migration’ and ‘transmigration’ (Glick Schiller, Basch and Szanton Blanc 1995). With this shift of focus, the translocality approach widens the scope to include aspects of internal or regional migration which had been neglected by transnational approaches and migration studies (Greiner and Sakdapolrak 2013: 374), although they are of growing importance, notably in Africa (e.g. Adepoju 1995; Aina 1995; Salih 1995; Rain 1999; de Bruijn et al. 2001; Marfaing 2014a). The

10  Space, Place and Identity

translocality approach thus builds up on the insights of transnational migration studies and widens the concept rather than rejecting it (Freitag and von Oppen 2010: 13; Hedberg and do Carmo 2012; Greiner and Sakdapolrak 2013: 374f.). While the concept of translocality is generally used in contemporary migration contexts, it can arguably also be relevant for an analysis of mobile pastoralist societies such as the Woɗaaɓe, e.g. in terms of social networks expanding across places, or processes of exchange and transfers (Freitag and von Oppen 2005: 2). Translocal orientation is not per se a modern phenomenon, but merely more accentuated in the contemporary condition of increasing global exchanges and flows. In a similar vein, de Bruijn and Brinkman (2011: 51) have called nomadic societies ‘[m]obile communities avant la lettre’: ‘The Fulani do not define their community in geographical space but in social relations that expand to cover large geographical areas, i.e. strings of people’ (ibid.: 52). Similar arguments have been made with regard to mobile communities in other geographical contexts.2 In the Sahel, translocality has even been suggested to be a more general feature that concerns not only mobile pastoralists but to some extent all parts of the population (Retaillé and Walther 2012: 19). Although not explicitly under the label of translocality, the debates on ‘cultures of migration’ (Hahn and Klute 2007) or ‘cultures of mobility’ (Boesen 2004b: 107, 2007b: 213f.; Boesen and Marfaing 2007) in the Sahara-Sahel region point into a similar direction: Mobility and translocal livelihood patterns are a common part of Sahelian identities (Rain 1999; de Bruijn et al. 2001: 69; de Bruijn and van Dijk 2003; de Bruijn 2007; Boesen 2007b: 213f.; Retaillé 2010, 2013), or of a ‘condition [Saharo-] Sahélienne’ (Gallais 1975; Boesen and Marfaing 2006). Hence, while recent migration research has been particularly interested in understanding how questions of identity and difference are spatialized in new ways in a time of increasing deterritorialization of cultural difference (Gupta and Ferguson 1997b), in this book I examine these questions in a Sahelian group of mobile pastoralists for whom deterritorialization and translocality are not new phenomena but rather a cultural constant. I argue that in contemporary Woɗaaɓe society, translocal orientation has several dimensions. On the one hand, translocal aspects are readily visible in the context of urban work migration that has become a significant phenomenon since the major droughts of the 1980s. Migrants generally maintain close contacts with their home communities, which remain a strong focus of social orientation, and translocal kinship-­networks connect actors belonging to the same home community who are dispersed across different sites (Chapter 9). In the specific case of the study group, a further level of translocal orientation is of importance: It emerges from networks with Western expatriates that originate in the urban migrant context yet also have a strong impact on migrants’ pastoral home communities. As I will show in Chapter 7, these networks are economically significant as they involve transfers of material and immaterial goods. On the other hand, however, long before

Introduction  11

urban migrant work became a relevant phenomenon, translocal community structures emerged from the patterns of pastoral mobility that, in the scarce environment of the Sahel, demand a relative dispersal of social groups for a substantial part of the year cycle. In addition, a network of translocal relations between clan groups is the very basis for the constitution of the Woɗaaɓe as an ethnic group (see Chapter 10).

Social Space and Place Although the Woɗaaɓe pastoral strategy of constantly reaching out for better pastures in order to optimize herd fertility and performance (Maliki 1981; Schareika 2003a, 2003b; Krätli 2007, 2008) entails a tendency of constant migration, this does not mean that at the same time there is not an attachment to specific places. As social beings, people are entangled in both translocal kinship networks and local neighbourhood ties. In fact, it is precisely this interplay of transcending places and becoming attached to new ones that makes the translocal dimension palpable. If translocality is concerned with both mobility and places, while putting a focus neither on the one nor on the other, but rather on the relation between the two (Freitag and von Oppen 2005: 3, 2010: 4; Greiner and Sakdapolrak 2013: 380), we have to define the concepts of place (and space) in a context of mobility. Several more recent conceptualizations of space and place, both in the social sciences and human geography, as (1.) relational and socially constructed and (2.) mobile and dynamic rather than fixed seem relevant and helpful as part of an approach to studying translocal phenomena and placemaking processes in a highly mobile society. Important references for most theoreticians of space as socially constructed are the works of Lefebvre and de Certeau. Lefebvre (1991 [1974]: 101) distinguishes between natural space, which merely juxtaposes things and living beings, and social space, which is the product of encounters, simultaneity and assembly at or around particular places. Similarly, for de Certeau, space is a ‘practiced place’ (1984: 117). It is produced by human social practice. While, in this fundamentally social understanding of space and space-production, place is still conceptualized as fixed and spatially defined, more recent approaches also regard place itself as socially and relationally constituted. Ingold’s (2007, 2009) concept of placemaking by ‘wayfaring’ is of particular interest here. For Ingold, wayfaring is particularly well exemplified in the mobility of nomads and hunter-gatherers, but ultimately it is a fundamental characteristic of human existence. Human movement is constant and leaves a ‘trail’ (2009: 33). Places are constituted by the convergence of human movement. Where people meet, their entangled trails form ‘knots’ (ibid.). The process of wayfaring, i.e. the progressive course of human existence, is thus characterized as fundamentally ‘place-making’ (2007: 101)

12  Space, Place and Identity

Massey (1994, 2005) also conceptualizes places not as localized entities, but as relationally defined. They are points of intersection and encounter, ‘articulated moments in networks of social relations’ (1994: 154). Massey puts a particular emphasis on the network of social relations and interconnections that stretch beyond the boundaries of a physical place itself, which she defines as relational or social space (2005: 184). A similar idea is expressed by Olwig, for whom ‘social practices, processes and interaction can expand beyond single places, thus manifesting a more encompassing place that can be of regional or even global dimensions’ (1997: 4). In the same vein, Sheller and Urry underline the ‘complex relationality of places and persons connected through performances’ (2006: 214). Places are defined by the ‘copresence’ (ibid.) of people who are simultaneously in the same place and engaged in interaction. Here, time becomes a central variable for the definition of place, which is regarded as temporary and ephemeral. This is also the position of French geographer Retaillé, whose ‘mobile space’ model is of particular relevance also because it was developed on the basis of field work in the same regional context as ours, i.e. the Koutous region of Niger (Retaillé 1984, 2013). Based on his observation that in the highly mobile Sahelian societies a place – be it a nomadic camp or a village – can ‘move within a given spatial structure and yet keep its intrinsic properties’ (Retaillé and Walther 2011: 89), Retaillé asks what defines place in a mobile social space. He concludes that a place itself can move (Retaillé 2013: 63) and, not unlike Ingold, argues that movement is even fundamental to place. Places are junctions where people and things that are constantly in motion meet (ibid.: 55). Since movement is constant, however, ‘spatial differentiations between places are constantly at work’ (Retaillé and Walther 2012: 18), and place itself becomes a mobile conception. Such non-static and processual concepts of place and space seem particularly useful for understanding the translocal social networks of Woɗaaɓe both in the pastoral and the urban migrant context, and their realizations through mobility and movement. Placemaking can thus be understood as the social process by which (geographical) space is transformed into place (Gupta and Ferguson 1997c: 36). Appadurai’s (1995) notion of locality production equally expresses this social dimension of placemaking. Next to the social, however, placemaking also has an intrinsic political dimension (Gupta and Ferguson 1997b: 6), which is defined by the aspect of resource appropriation that the appropriation of space comprises. Placemaking can therefore be regarded as a power struggle (Turton 2011) and is thus always a matter of integration and conflict. These aspects are of particular relevance with regard to the contemporary tendencies of sedentarization and territorialization, but no less in the context of urban migration. Placemaking opposes groups and involves a construction of difference and identity (Gupta and Ferguson 1997b: 13). It is thus intrinsically intertwined

Introduction  13

with ‘people making’ (ibid.: 4), i.e. ethnicity, processes of identification and the construction of social group boundaries.

Mobile and Multi-sited Fieldwork An approach of studying not only the relations that Woɗaaɓe establish with their neighbours in a multitude of different environments, but also the translocal connections that they maintain across the different, seemingly disparate spheres in which they live today makes research inevitably multi-sited and mobile. In a globalizing world, characterized by increasing flows of goods, ideas, technologies and people, static approaches to the anthropological field have increasingly been questioned and theorists have proposed to focus on shifting locations rather than bounded fields (Gupta and Ferguson 1997a: 38). The insight that the objects of research were increasingly mobile and could no longer be confined to specific locales (Sökefeld 2000: 51), fuelled the career of the methodological paradigm of multi-sited fieldwork for the study of mobile and multiply situated groups (Marcus 1986, 1995). This approach owed a lot to the field of migration studies, in which multi-sited fieldwork was already an established practice when the concept gained wider recognition in anthropology (Hannerz 2003: 202). The multi-site approach emphasizes the translocal linkages, and the interconnections between field sites (ibid.: 206). As these links are materialized in relations between people, it is crucial to analyse the social practices across sites on which they rely. Arguably, research on mobile groups inevitably has to be mobile, yet whether this mobility makes it automatically multi-sited, is a different question. One might argue that the ‘field’ itself can be mobile and move through space with the people studied (Boesen 2010: 31f.). In the case of the pastoral Woɗaaɓe, however, the composition of residential units varies a lot over the year cycle and over time, due to the pastoral mode of production which demands a relative fragmentation into small units during a significant part of the year. Social groups split and regroup in shifting agglomerations yet remain closely connected. In this context of translocality, the multi-site aspect becomes obvious. Urban migration adds a further dimension to the dispersal of social groups. Even a rough micro census of pastoral residence units reveals that many members of nuclear families are absent for a greater part of the year and might only come on and off for short visits. People switch between the pastoral and the urban realm, and they move within both. The situation is characterized by constant and multiple rural–urban, rural–rural and urban–urban flows of people. The potential ‘fields’ are thus dynamic and not bounded. My research was in this sense inevitably multi-sited, not only because I focused on a variety of different actors who were located in different sites, but also because many of

14  Space, Place and Identity

these actors were highly mobile across sites, in their daily lives and in the pursuit of their economic and social activities. The study context cannot be understood by analysing the pastoral and the urban realm as separate field sites, but only by focussing on the connections that exist between the two spheres. The ‘field’, in this translocal perspective, is less defined by physical places than by social relations that are reproduced and that manifest themselves in variable forms over time and space. The translocal field of my research was thus defined by relational spaces, or, ‘spaces of social relating’ (de Bruijn and Brinkman 2012: 47), spanning different locations and spheres. In such a situation of multiply connected and shifting potential ‘fields’, the problem of delimitation, the question of which sites to include into the scope of the work and which others not, was not without difficulty. I approached the problem by putting an initial focus both on specific sites and, within or across them, on a number of interlocutors, while being open to widening or changing the scope as my research progressed. For instance, I had first been most closely in contact with the Gojanko’en community in the Koutous region, but, given that their migration had taken place only recently and mutual visits to and from the Damergou region occurred frequently, it was evident that it would be important to include the latter region into the scope of the study and to explore the prevailing connections. It was equally significant to follow the links that both pastoral communities entertained with their urban-based members who stayed as migrant workers in Zinder, and who play a crucial role in keeping their spatially separated home communities connected. My research was thus characterized by an emergent object of study whose contours, sites and relationships materialized only gradually (Marcus 1995: 102). Faced with a multitude of interconnected sites, however, it was even more important to delimit at least a number of localities on which to focus the analysis. I decided to put an emphasis on one urban (Zinder) and two rural sites (Ganatcha in the Koutous region; Salaga in the Damergou region), while remaining open to following links to other, connected sites and using these data for comparison and to enrich the analysis. For research in a community as tightly connected by mobility as it is dispersed in space, the method of ‘following the people’ (Marcus 1995: 106) seemed particularly well adapted. Building up on, and going beyond it, my own approach was often not so much to follow the people but to travel directly with them, i.e. to conduct research while accompanying them during their mobility (Sheller and Urry 2006: 217f.). This method proved to be especially fruitful in the cases of urban migrants visiting their families in the pastoral context. On such occasions, news are generally exchanged, recent developments discussed and relevant information passed on, which offered welcome opportunities to collect data and gain insights en passant. Mobility thus became a central principle of my own research.

Introduction  15

Collective Social History Revealed through Autobiographical Narratives While literature on pastoral mobility has often put a considerable focus on the agency of pastoralists, emphasizing the option of choice and the constant need for decisionmaking (e.g. Gulliver 1975: 371), in the contemporary condition, in which nomads and other indigenous minority groups are faced with radical and sometimes existential change, forced to adapt and find economic alternatives, they are often depicted one-sidedly as victims in a global power play (Gupta and Ferguson 1997c: 44; for the Woɗaaɓe, see Loftsdóttir 2000). However, it is not simply the conditions that bring about change in a society, but ultimately the people, who react to changing conditions, challenges and opportunities with conscious action that can also challenge or subvert existing orders (Gupta and Ferguson 1997c: 47). Although my interlocutors also sometimes depicted themselves as victims – as a minority-group excluded from political participation and processes of resource distribution – I perceived them just as much as rational agents, thinking, planning and deciding deliberately and strategically. My aim was therefore to put a focus on agency and to document strategies (as well as their limits, and their failures) to integrate into the wider society and state structures in order to participate at processes of resource distribution, but also the opposite tendency of the reproduction of difference and withdrawal from the state in order to subvert imposed structures. I followed this aim by focussing on a number of key interlocutors, whose cases and life stories allowed, in complementary perspective, to grasp variations of a theme, and thus to arrive at a comprehensive view of the contemporary condition of this group. My principle of ‘following the life or biography of interlocutors’ (Marcus 1995: 109) added a historical dimension to the continuum of connected field-sites: as the accounts that I collected opened up a view on the itineraries of the past, I took up this thread, mapping not only contemporary movements between physical places, but also historical ones, thus tracing a spatio-historical continuum of mobility. My approach thus merges a diachronic perspective with an analysis of the contemporary condition. Principally based on extensive participant observation, my research methods involved mainly semi-structured and open interviews encouraging ­thematically-focussed autobiographical narration. Near the end of the twentieth century, biographical approaches increasingly made their way into the social sciences and it has since been widely recognized that biographical material can show how social group processes translate on an individual level and how abstract historical processes of cultural change actually take place (Bertaux and Delcroix 2000: 73; Fischer-Rosenthal 2000). Bonfiglioli’s (1988) Duɗal is a good example of how oral accounts can shed light on individual agency in the context of wider historical processes.

16  Space, Place and Identity

In order to make visible the variety of socioeconomic transformations that can be found in the study group, I recorded a number of autobiographical accounts that reveal how recent change is experienced and coped with by the people, what strategies they pursue, what answers they give to the new questions and challenges, and what obstacles they face. Instead of recording complete life histories of a great number of individuals – an approach that, if taken seriously, would demand an extremely high degree of depth and complexity with questionable usefulness for the envisaged results – I favoured a focus on chapters or ‘portions’ of life histories (Vansina 1980: 266), i.e. biographically based yet thematically-focussed accounts of specific events. The members of the study group are all linked by ties of kinship. The kinship group offers a system of interconnected actors, embedded in nets of reciprocal commitments, sharing common goals, strategies and values or being in conflict over them (Bertaux and Delcroix 2000: 73f.). In juxtaposition, the cases complement each other, as an example might illustrate: The life stories of the brothers Nano, Taafa and Maalam Buuyo (Chapter 9), if regarded in isolation, seem to present almost antithetic schemes: while the youngest brother has never turned away from pastoralism, the other two have spent a considerable part of their lives as migrant workers in different urban centres. What might seem disparate at first reveals itself to be two aspects of a joint project in which different roles are attributed to the different actors. The brothers follow a common strategy in which two of them seize urban job opportunities to subsidize the pastoral family economy, while the herds are managed by the third. The principal interest in the migrations of the first two is rooted in the local concerns which all three of them share. Unfortunately, the format of the monograph limits the possibilities of including ethnographic data of any greater length. Although I quote from the autobiographic narratives or summarize parts of them as case examples throughout the book to substantiate generalizations, it was not possible here to give the voices of those studied the room they would have deserved. To compensate for this limitation and to give the interested reader a possibility to look into the primary sources on which the findings of this book are based, selected accounts have been made available in full length in a separate publication. This publication is available online with open access (Köhler 2017a, http://www.eth.mpg.de/ pubs/series_fieldnotes/vol0020.html). References to texts featured in this volume are made by indication of the numbers of the respective text and paragraph(s).3

How This Book Is Organized This book is organized into five parts, each carrying a Fulfulde title that summarizes the main aspect that is investigated. The first part is titled Taariihi, a term derived from the Arabic tarikh (‫)تاريخ‬, meaning history, story or chronicle. The

Introduction  17

two chapters of this part deal with the history of the study group and of the Woɗaaɓe more generally, analysing how migration processes are connected to processes of group formation. Placemaking processes are examined here in historical perspective. First, in Chapter 1, by analysing how the structure of Woɗaaɓe society has historically developed in processes of fission and fusion, of disjunction and re-aggregation of social groups in a context of migration and mobility, and then, in Chapter 2, by reconstructing such processes using the concrete example of the historical itinerary of mobility of the study group. In a complementary view, these two chapters show how social groups are constructed and reconstructed in a continuous process, based on the principles of mobility and placemaking. The second part, Duuniyaaru, examines inter-group relations and the question of the study group’s integration in the wider poly-ethnic society of the study region. Duuniyaaru is another Arabic loanword, derived from the term dunia (‫)دُنيا‬, meaning the world. In Fulfulde, the term can designate either the world at large, the world out there, or ‘the people’, other people. The Woɗaaɓe are part of a wider social, economic and political landscape, which they share with other ethnic groups and which I refer to as a meta-ethnic social space. Just as between individuals within a group, social space is produced by interaction between and across groups. Social space on a meta-ethnic systemic level (Schlee 2001b: 19) is thus the product of social interaction and exchange between the different groups that constitute the study region’s poly-ethnic society. Chapter 3 examines how this meta-ethnic social space has been constituted historically by long co-­existence and by the development of particular social institutions that favour exchange and communication between groups and thus have an integrating function. Chapter 4 looks more concretely at the mechanisms along which the integration of the different heterogeneous groups in a meta-ethnic social space works. Although the ethnic groups in the study region distinguish each other along clear boundaries, there are also levels of shared identification along different dimensions of identity (e.g. language, socio-economic profile, religion). Identification can change situationally, expressing changing perceptions of sameness and difference that are re-interpreted and re-negotiated depending on the context of interaction. The third part, Ladde, refers to the pastoral space, the ‘bush’, the open rangeland. In chapter 5, I analyse more closely the recent transformations in spatial strategies and in relations to space and place, and I analyse the reasons for these changes. In chapter 6, their effects on the social, political and environmental level are assessed. Part IV focusses on the urban space. The Fulfulde term siʾire can refer to a town or city, but also to a small village. Generally considered to be the sphere of the non Fulɓe others, the Haaɓe, it is the space that is opposed to ladde, the sphere of pastoralism and of the Woɗaaɓe. Nevertheless, as I argue in chapter 7,

18  Space, Place and Identity

for many Woɗaaɓe the city has developed into an important complementary sphere of social and economic activities, since urban work migration was first taken up as a coping strategy during the major drought of the 1980s. Chapter 8 deals with social interaction in the city and describes how, similar to the rural poly-ethnic setting described in Chapter 4, the urban realm also allows for multiple possibilities of identification along different dimensions and categories for the formation of more or less inclusive urban communities. Chapter 9 argues that translocal social relations are crucial for identity construction among urban migrants. This point again relates to concepts of social space as manifested in the networks of relationships that an individual or a social group maintains (Ingold 2007, 2009; Massey 1994, 2005): The spaces that connect the urban and the pastoral spheres between which many Woɗaaɓe today move are socially constructed by networks of people who are translocally linked. Part V, Gassungol Woɗaaɓe, puts the focus on questions of the translocal reproduction of the ethnic group and on the challenges for cultural continuity that evolve from the contemporary situation of urban migration. Gassungol Woɗaaɓe is used as a metaphoric designation for the ethnic group. Gassungol is a special kind of rope that is used to tie up the household loads on a pack animal when camp is moved. The ethnic group is thus conceived as an ensemble of clans that are tied together by the translocal politico-ceremonial alliances of ngaanka like a net, or a network. In this last part, I examine more closely how this network of translocal relations is established and maintained. Intra-ethnic processes of differentiation (between different clans) are at least as relevant for identity construction among the Woɗaaɓe as differentiation from external others (Boesen 2004b: 120; 2008a). While Chapter 4 has put a focus on processes of identification and differentiation along interface situations with other ethnic groups, in Chapter 10, the analysis is moved to internal processes of identification, i.e. to interface situations between clans and their role for social group formation and reproduction. This last part rounds out the argument of translocality as a fundamental principle of social group formation and community production in Woɗaaɓe society. In chapter 10, I argue that the continuity of the ethnic group relies on the maintenance of a network of ceremonial and marital inter-clan alliances across an extended social space; in chapter 11, I show how, in the contemporary condition, the exposure, particularly of urban migrants, to alternative value systems represents a challenge to the translocal reproduction of the ethnic group.

Notes 1. Bovin (1998: 94), while also referring to Paris, gives a figure of around 125,000. 2. E.g. north western Namibia (Greiner 2010). 3. References to other interviews or field diary entries are made by indication of the name of the interlocutor and the date.

Part I

Taariihi Mobility and Group Formation in Historical Perspective

Chapter 1

The Woɗaaɓe in Niger Structure as Historical Process

The Woɗaaɓe in Niger today comprise fifteen clans that belong to either of two opposing clan clusters, Degereewol and Alijam.1 Although research was conducted among different Woɗaaɓe communities in Niger,2 the main focus of this study was on a relatively small regional faction of a clan segment. This group is part of the Kuskudu maximal lineage, a branch of the Gawanko’en (or Gojanko’en Gawo), who are themselves a segment of the Gojanko’en clan, which is part of the Degereewol clan cluster (see Figure 1.1). More precisely, the study group has been delimited and defined as those members of the Kuskudu maximal lineage who live today in the Damergou and the Koutous regions, both situated in the Zinder province of east-central Niger. Their migration into the region was preceded by increasingly difficult climatic conditions towards the end of the 1960s, culminating in the catastrophic drought seasons of 1973–4. At that time, larger numbers of Gojanko’en (and other groups of Woɗaaɓe) migrated to the Damergou from the Ader and Agadez regions, but a vast majority of them returned later, when conditions became better again. Only relatively few, roughly numbering 100 households, remained in the Zinder province and live there today in the Damergou and Koutous regions, engaged in a process of partial sedentarization around 5 pastoral wells that individuals have acquired or for which they have obtained management rights (see Figure 1.1 and Chapter 5). Those Gojanko’en who stayed all belong to the Gawanko’en clan segment, among them one leader of the Mbuuldi lineage, Arɗo Kiiro, with his followers (around 40 households), and a number of households of the Kuskudu lineage, which are the main object of my study and which I will discuss in more detail later. Kuskudu and Mbuuldi are the names of the founding ancestors of two maximal lineages that together constitute the Gojanko’en

22  Space, Place and Identity

Figure 1.1  Woɗaaɓe clan structure, current leaders within the study group and sites of attachment (diagram by F. Köhler).

Gawo, or Gawanko’en. The two are said to have been brothers, Mbuuldi being the senior one. Most elders can trace their lines of descent four generations back to these ancestors.3 Figure 1.1 shows the genealogical relations of some of the key figures of this study with the lineage founder, Kuskudu.4 In order better to understand the position of the study group in the framework of the Woɗaaɓe clan structure, it seems helpful to have a closer look at how this structure has formed historically.

The Woɗaaɓe in Niger  23

Woɗaaɓe society is composed of complementary segments of different orders, and has thus often been identified as a segmentary lineage society (e.g. Loftsdóttir 2001c; Boesen 2008a). Dupire was rather reluctant to apply this label. Although she roughly qualified the Woɗaaɓe as a lineage society (1970, 1975), she argued that one would hesitate to recognize the existence of lineages, were it not for the Woɗaaɓe’s own presentation of a system that so neatly matched with the classic scheme (Dupire 1970: 303). In practice, the Woɗaaɓe lineage system is only to a limited degree segmentary. In principle, solidarity between equal segments when it comes to defending common interests against external actors is recognizable even at the inter-clan level, as Schareika’s (2007, 2010a) analysis of political decision-making processes has demonstrated. However, the geographical dispersal of the segments of the same level and their lack of territorial attachment sets clear limits to the functionality of the principle (Dupire 1975: 337). Neither the clan clusters nor the clans ever act as corporate groups due to their fragmentation into regional factions.5 Dupire (1975: 322f.) argued that the Woɗaaɓe lineage model is not merely a construction of anthropologists, but used by actors themselves as a framework for explaining their social structure – which is not unusual for segmentary lineage societies either (Holy 1996: 80f.). Even Evans-Pritchard (1940: 143) claimed that the logic of segmentary opposition between lineages was a ‘folk-model’ held by the Nuer themselves.6 In the case of the Woɗaaɓe as in classic segmentary lineage societies, the lineage model is an ideological and ideal model of society rather than an operational model for political action (Holy 1996: 85; Salzman 1978). Schareika has correctly objected that groups among the Woɗaaɓe do not simply form naturally according to the logics of Evans-Pritchard’s classic model: the interaction of individuals and the social ties that bind them together are not merely the result of their being part of an autonomously existing lineage (2007: 163). Rather, the segmentary patrilineage constitutes a normative framework of rules and values that influence people’s orientations and offer a reference for their interpretations and judgements of social relations and appropriate interactional behaviour. In practice, individuals may manipulate the model according to their needs and interests (ibid.: 164), and other factors come into play as well. There are diverging lines of solidarity that interfere with those imposed by the logic of the lineage system. The seemingly fundamental dichotomy of the two opposing clan clusters, especially, plays a relatively minor role in daily practice. Other relations of solidarity, which can transcend the division of the clan clusters, are of greater importance for social practice. For example, some clans are closely allied for different historic reasons, often across clan clusters (Paris 1997: 77). While the emic interpretation of group relations is that of a (relatively) coherent structure of patrilateral descent between segments of all levels, historical data collected by Dupire, and later Bonfiglioli (1988), have shown that the postulated genealogic relations between segments are in many cases rather artificially

24  Space, Place and Identity

constructed. At least since Sahlins’ Islands of History, it is an established position in anthropology that social structure is not a monolithic fact, but a result of historical production (1985: vii). The clan structure of the Woɗaaɓe is a case in point, and it is noteworthy that not only is the structure here pragmatically adapted to the facts, but historic facts are also adapted to fit into a once established structural scheme. The present structure of clans and clan-clusters is the result of a complex combination of processes of segmentation of patrilineal descent groups and simultaneous counter-processes of constant re-affiliation and fusion of groups – a pattern characteristic of the Fulɓe in general (Dupire 1970; Riesman 1974; Lombard 1981), and even more generally of many pastoralist societies (Fabietti and Salzmann 1996). On the one hand, Fulɓe communities show a tendency of splitting apart at an early stage of complexity. Ecological factors play a determinant role here: the scarcity of resources in the arid climates of the main region of distribution of the Fulɓe demands a more or less pronounced degree of mobility and does not permit a too great concentration of herds and households (Dupire 1970: 85; Lombard 1981: 188). Equally significant are economic factors linked to pastoralism. From a certain herd size, to split up the herd and the social group can be strategically advantageous (Schareika 2007: 118). Finally, sociological factors favouring individualism also contribute to the segmentation of patrilineages, in particular the system of pre-mortem inheritance that allows young men to establish independent households quite early (Riesman 1974: 45; Dupire 1970: 94; Lombard 1981: 187f.). On the other hand, local proximity and co-residence of migratory groups often leads to alliances on the basis of common pastoral interests. Such clusters of local groups that have fused on the basis of economic and political considerations can over time develop a common identity that, through marital exchange, finally materializes in real kinship-relations and becomes rationalized ex post in the assumption of a coherent, but putative, genealogical structure (Dupire 1962: 319ff.; 1970: 296, 303f; 1975; Bonte 1979: 216; Bonfiglioli 1988: 107; Maliki 1988: 23). Dupire (1962: 320) convincingly argues that what matters is not the presence or absence of actual genetic relations, but the perception of a cultural homogeneity that manifests itself in the affirmation of a common origin. The manipulation of genealogies can thus be regarded as a means of integration of heterogeneous groups (Bonte 1979: 222; Holy 1996: 82). Fusions between Fulɓe clans of Woɗaaɓe and non-Woɗaaɓe origin are well-documented. Particularly during the nineteenth century, due to the societal transformations brought about by the Jihad of Usman Dan Fodio and the formation of the Sokoto empire, significant cultural reorientations took place within Fulɓe society at an accelerated pace (Bonfiglioli 1988: 63, see also Chapter 4). Entire clans or lineages split away from the Woɗaaɓe and affiliated with other Fulɓe groups, and groups of other Fulɓe sections assimilated to the

The Woɗaaɓe in Niger  25

Woɗaaɓe and were over time incorporated along the above-outlined pattern. Against this background, the Woɗaaɓe of Niger must be regarded as a cluster of Fulɓe lineages and clans of heterogeneous origin that have regrouped in the present form and over time forged a common identity. These historic processes of group reconfiguration are apt also to explain why in different regional contexts clan patronyms can be found in different, sometimes contrarious affiliations. For instance, according to Reed, in Borno there were two distinct groups of Jaɓto’en (Njapto’en), one belonging to the Woɗaaɓe, the other to the Fulɓe Waila, both claiming not to have anything in common but their names (1932: 423, 443ff.; see also Dupire 1994: 266f.). As for the two clan clusters (Degereewol and Alijam), their alleged genealogical relation seems no less a historic construct than those between their constituting clans (Dupire 1962: 306f.; 1972: 27). This becomes evident from a look at Stenning’s work on the Woɗaaɓe of Borno. Here, the Alijam appear not as an overarching category, but – just as the Degereeji (Degereewol) – as a clan among others, more precisely as a section of the Ɓii’Eggirga’en clan of the Woɗaaɓe (1959: 196).7 Obviously, the agnatic relations between clans claiming a common origin were no less putative in Borno than they are today in Niger. In the words of Stenning, ‘the Woɗaaɓe overcame the defects of memory by treating as agnatic the relationship of groups in the same cluster, and employing fictions to do so’ (1959: 54). From such information one can conclude that the Alijam and Degereewol existed as clans before the formation of the contemporary Woɗaaɓe clan structure in Niger (Dupire 1962: 306; Loncke 2015: 126). It can be assumed that under the umbrella of each of the two clans that were to give their names to the contemporary clan clusters, a number of clans or lineages united and developed a common identity (Dupire 1972: 27). Due to their pronounced mobility over the course of the nineteenth and twentieth century, the Woɗaaɓe are today dispersed over a wide region across different nation states. It is therefore difficult to speak of ‘the Woɗaaɓe’ or even ‘the Woɗaaɓe of Niger’ as a clearly delimited group. Dupire (1975: 323f.) has aptly described the Woɗaaɓe of Niger as a rather loose entity with fluid limits. In the same vein, Stenning has called the ‘tribe’ a ‘vague cultural entity’ (1957: 58). The ethnic group as an entity remains hypothetical, a merely ‘imagined community’ (Anderson 1991). It has neither a single representative nor any level of interaction that coordinates all of its constitutive segments. Above the level of the primary lineage, or rather its regional segments, ‘society’ does not have any concrete manifestation apart from the periodical meetings in the course of ngaanka inter-clan ceremonies, which primarily serve to reconfirm a bond and an understanding of unity, which, however, generally concerns only the regional clan segments involved. Following Reed (1932), Dupire (1962) considered the Woɗaaɓe of eastern Niger as distinct from the groups of the Ader and Damergou regions. Reed

26  Space, Place and Identity

had maintained that although they comprised groups carrying the same clan patronyms as some Woɗaaɓe clans in Central Niger, they did not recognize any kinship relationships with those groups (1932: 425). For the contemporary groups of Woɗaaɓe in eastern and east-central Niger this cannot globally be confirmed. Although, owing to geographical distance, ceremonial relations between these groups are not maintained, the sense of historic connectedness is in some cases strong and contacts with lineage members across regions are sometimes maintained. Notably, interlocutors from a group of Ɓii Ute’en from the Manga region in eastern Niger confirmed that they are still in good contact with their relatives in the Damergou region whom they left behind upon their migration in the 1970s.8 The same consciousness of being part of a widely dispersed group was expressed by Njapto’en from the Tchintabaraden area, a faction of whose lineage once moved to Chad and live there today, even though they have not remained in contact.9 Similarly, a Gojanko’en interlocutor from the Koutous region made reference to a clan faction that moved to Chad around 1980, where they now live in an area south-east of Ndjamena (Köhler 2017a: text 3,10). A consciousness of one’s own kinship group being dispersed across different areas is pronounced, and the sentiment of being related often remains strong. Arguably, the degree of connectedness depends on the historical distance of the separation. Unfortunately, Reed does not give any hints as to the period of separation of the groups that he mentions. With regard to the study group, as I will discuss in Chapter 2, it is striking that numerous Kuskudu of the Damergou region today entertain vivid contacts with their former home range in the Agadez region, from which they migrated around 1970, whereas their contacts with lineage members in the region of Tahoua, from which their fathers had first migrated to the Agadez region around 1940, are much lesser. The sense of historic relatedness, however, remains undiminished. The Woɗaaɓe can probably best be considered, much like the Fulɓe of whom they are part, as a continuum with fluid limits, both spatially and in terms of group boundaries. In the course of their history, the high degree of mobility has led to a great dispersal of the Woɗaaɓe and to hybrid forms which have developed by close contact with neighbouring groups and through integration of other Fulɓe groups into Woɗaaɓe society. In such a setting, the ethnic group can probably best be defined as the largest unit of social and ritual interaction among clans or their regional segments that share an understanding of being part of a wider whole. The pronounced spatial mobility of the Woɗaaɓe in the process of pastoral migrations makes them a highly dynamic case in terms of social group formation and reconfiguration and ultimately, identity formation. Historically, ethnic group boundaries have always been in a constant process of renegotiation, characterized by a high permeability and fluidity. The close link between mechanisms of group formation and migration means that social and spatial processes have to be regarded in close association.

The Woɗaaɓe in Niger  27

Woɗaaɓe, as with other pastoral Fulɓe, live in small scattered groups. These groups, however, are not isolated units, like the islands of an archipelago, but are rather closely and multiply interconnected, and they have a strong consciousness of their connectedness over time and space. Co-residence in a loose mobile unit defines the basic social group, but for the functioning of community and society on a larger scale, translocal social relations and networks play a crucial role. In the following chapter, I will demonstrate, through the example of the study group’s migration history, how the processes of socio-spatial differentiation and group (re-)formation and the maintenance of translocal links between the dispersed parts over time look in practice. The actual mechanisms and strategies for maintaining social contacts and in fact a notion of community within a group widely dispersed in space makes the meaning of concepts such as ‘translocal society’ and ‘translocal social space’ palpable.

Notes 1. These fifteen clans are: Degereeji (also called Shahidooji), Gojanko’en, Jiijiiru, Njapto’en, Suudu Suka’el, Kasawsawa, Baagel’en and Hadaali (all from the Degereewol cluster); Ɓiɓɓe Denke, Ɓii Ute’en, Ɓii Korony’en, Ɓii Nga’en, Yaamanko’en, Ɓii Hamma’en (also called Kabaawa) and Alamoƴo (all from the Alijam cluster).   In the terminology here adopted, ‘clan’ and ‘clan cluster’ correspond to Dupire’s (1962) terms ‘lignage primaire’ and ‘lignage maximal’, respectively. I reserve the term ‘maximal lineage’ to refer to the highest level of segmentation that can be considered as a lineage in the proper sense. On the level of the clans, descent can generally not be traced in an uninterrupted line (see Holy 1996: 75). The term ‘clan cluster’ further emphasizes the fact that, as I will show, the claim for common descent to explain group relations on this level is rather a social construction than a historic fact. 2. Apart from the Gojanko’en in the Damergou and Koutous regions, these notably comprise a group of Yaamanko’en in the Koutous region, groups of Ɓii Ute’en and Suudu Suka’el in the Diffa region (eastern Niger), and, to a limited extent, a group of Njapto’en in the area of Tchintabaraden (central Niger). 3. For the Mbuuldi branch, my data is further confirmed by Maliki (1982: 36f.). 4. A more detailed genealogical chart featuring all the lineage members mentioned in this book can be found in Köhler (2017a:139). 5. Critical reconsiderations of the segmentary lineage model (Peters 1967; Holy 1979, 1996; Kuper 1982) have revealed similar discrepancies between the model of segmentary opposition and empirical reality in many classic cases of segmentary lineage societies. In the case of lineages of a higher order of segmentation, corporateness has often been assumed rather than empirically observed (Holy 1996: 89). 6. See, however, Kuper (1982: 84), who rejects the contention that the emic conception of the Nuer resembled the theoretical anthropological model. 7. Reed (1932: 443) identifies both the Alijam and the Ɓii’Eggirga’en in Borno as subgroups of the Woɗaaɓe, while the Degereeji are identified, in accordance with Stenning, as a sub-group of the Ɓii’Eggirga’en. 8. Nyeriiɗo Tambaya, Ɓii Ute’en Diffa, December 2011. 9. Muusa Tchawɗo, Njapto’en Tchintabaraden, April 2011.

Chapter 2

A History of Migrations Placemaking Processes in Diachronic Perspective

The Woɗaaɓe are characterized by a history of mobility and migration. Contemporary spatial strategies and patterns of mobility thus have direct historical antecedents, and their historical embedding can illuminate the ways in which Woɗaaɓe communities had been translocal before the contemporary phenomena of migrant work and rural–urban networks became relevant. In this chapter, I will give a sketch of the historical migrations of the study group, based both on the accounts of elders who have kept the orally transmitted memories of past events, and on former research on Woɗaaɓe migration history, notably the works of Dupire (1962, 1970), Maliki (1982, 1988), Bonfiglioli (1988) and Paris (1990). More general studies of Nigerien history (Baier 1980; Fuglestad 1983) put these sources into a wider historical context. It is a common characteristic of oral history that the period for which detailed accounts can be given moves on with time (Cooper 2005: 192). Since the Woɗaaɓe do not have specialists for, nor an institutionalized way of remembering history, the time span that can be covered with decently reliable information often does not exceed more than two or three generations back from the oldest living generation. Nevertheless, it is remarkable that in many cases, the oral information has meticulously been transmitted over the generations. My principal interlocutor with regard to historical issues was Ɗawra Egoyi, an elder from the faction that is today based in the area of Ganatcha in the Koutous region. My historical overview will be strongly based on his perspective, because his account of the group’s migrations was the richest and the most consistent.1 Details such as the name of a cow that was presented as an honorary gift to a colonial officer were remembered just as the name of a slave who was charged with transmitting such gifts (Köhler 2017a: text 2,05). References to memorable

A History of Migrations  29

occurrences like the appearance of a comet, the succession of political leaders or major droughts and famines often permitted to locate events in time. While at other times, the selectiveness of oral transmission and simply oblivion may have led to losses in historical detail, the general picture has become quite consistent by putting together my own data with those supplied by the existing sources.

The Historic Migrations of the Study Group The migrations of the study group can be traced back relatively well at least to the mid nineteenth century, when their ancestors lived in Sokoto in what is today northern Nigeria. However, this period is hardly remembered with more than some vague references to Usman Dan Fodio (1754–1817). This religious leader and founder of the important Islamic Fulɓe Empire of Sokoto plays a significant role in the collective memory of the Woɗaaɓe and he is repeatedly mentioned with clearly positive connotations as a quasi-mythical hero and identification figure (see also Paris 1990: 196f.; Loncke 2015: 28; Köhler 2017a: text 1). He is said to have defended, and indeed encouraged, the mobile way of life of the pastoral Fulɓe by personally advising them to lead their animals to the rich pastures of the north (Bonfiglioli 1988: 191f.; Paris 1990: 200). In such quasi-mythical explanations of the important northward movement of the Woɗaaɓe in the nineteenth century, elements that can be historically traced become mingled with others that remain doubtful, but are emphasized because they serve as legitimizations of a mobile pastoralist identity. An overall positive image of Usman dan Fodio is prevalent also among contemporary Woɗaaɓe. According to Ɗawra, it was in reaction to Usman’s foundation of the Caliphate in 1809 that many Woɗaaɓe migrated to Sokoto: ‘It was because Usman Dan Fodio had taken the power, many Muslims [including many Woɗaaɓe] went there’ (‘Sabo da Sheefu Ɗan Foodye ya samu, Musulmi sun koma can’–H).2 This statement is interesting as it explicitly claims an Islamic identity for the Woɗaaɓe. But Usman is also significant in Ɗawra’s account, because he – or his successors, yet this differentiation tends to get lost (see Bonfiglioli 1988: 192) – appointed Woɗaaɓe leaders to political offices and conferred titles (giifol)3 to them, which still remain an important source of legitimization of political leadership. Ɗawra proudly underlines that his ancestor received his title from Usman (Köhler 2017a: text 2,01). The role of the Woɗaaɓe and other pastoral Fulɓe in the jihad was ambivalent. On the one hand, pastoral Fulɓe have undoubtedly been an important factor in jihad warfare (Braukämper 1971: 70; Boutrais 1995: 34f.). Many, however, seem to have participated less out of religious conviction than with rather profane motivations, like the hope for better conditions under the new rulers. This hope was not implausible given that the latter were Fulɓe like ­themselves – unlike the Hausa rulers, under whom pastoralists had suffered from high taxation and

30  Space, Place and Identity

arbitrary treatment (Braukämper 1971: 67; Maliki 1982: 16). Yet more opportunistic motives, like gaining wealth and slaves from their military engagement, might have been considerations as well (Stenning 1966: 389f.; Braukämper 1971: 70; Maliki 1982: 17). According to Dupire (1962: 25f.), the Gojanko’en were among the clans who engaged most actively in the jihad. Other groups participated rather grudgingly and only for fear of losing their liberty if they would not. The seasonality of warfare with its focus on the dry season – due to terrain conditions and logistic considerations, and to facilitate recruitment of peasant populations (Smaldone 1977: 73f.) – was not easily compatible with the pastoralists’ increased labour-demand during this time of the year (Bonfiglioli 1988: 67; Boutrais 1995: 34). Overall, it can be assumed that only a minority of the pastoral Fulɓe was able to profit considerably from the jihad and that the majority rather tried to keep at a distance from the warfare activities. The new Fulɓe reign had a significant impact on the development of pastoral mobility in the Sokoto Empire. The ‘pax fulani’ enforced by the new polity offered increased security to pastoralists, who began to leave the fortified villages,

Map 2.1  Migrations of Ɗawra Egoyi and his ancestors.

A History of Migrations  31

in which they had previously lived, and to disperse in small groups (Braukämper 1971: 94). This process was additionally reinforced towards the end of the nineteenth century by the colonial conquest (ibid.; Stenning 1959: 5; Hopen 1958: 49). The high nomadic mobility of pastoral Fulɓe in the region is thus a relatively recent phenomenon. While by 1890 the majority of the groups of Woɗaaɓe lived north of Sokoto, in the region of Wurno (Bonfiglioli 1988: 83), the infiltration into the territory of what is today the Republic of Niger began gradually towards the end of the nineteenth century. Two major factors explain the fragmentation of Woɗaaɓe society that occurred between 1890 and 1920: the first is the Rinderpest epizootic which had severely reduced the Woɗaaɓe’s herds during the early 1890s; the second is the colonial conquest that began shortly afterwards.4 These factors caused northward movements of many splinter groups and their subsequent reformation, which laid the basis for the contemporary clan structure of the Woɗaaɓe of Niger (Bonfiglioli 1988: 107). Around the end of the nineteenth century, the Gojanko’en, just as other Woɗaaɓe clans, had to recover from severe animal losses. In the aftermath of the epizootic, many pastoralists were forced to settle down temporarily in the outskirts of villages and practice agriculture (Baier 1980: 134; Bonfiglioli 1988: 93f.; Braukämper 1971: 71; Maliki 1982: 25; de St.Croix 1972: 13). It took them different periods of time to reconstitute a stock and return to a pastoral way of life, which in part explains the dispersal of the groups, during several waves of migration, into different regions (Paris 1990: 199). Another cause of the fragmentation was that during the epizootic, many pastoralists, for fear that their animals might be infected if they stayed too close to other herds, had reacted with increased mobility and dispersal (Maliki 1982: 25). Around the same period the colonial advance caused significant insecurity. In fact, the earliest event in Ɗawra Egoyi’s account that can be precisely dated, is the destruction of Birni N’Konni by the French Voulet-Chanoine mission in May 1899 (see Comte 1988: 163–78; Fuglestad 1983: 55ff.). As Ɗawra reports, his great-grandfathers had their camps not far away in the surrounding bush at the time and saw the city burning in the distance (Köhler 2017a: text 2, 02).5 Between 1897 and 1908, the colonial conquest was marked by military repression and massacres, before entering into a more quiet, formative period, during which the French established the structures of the colonial state (Bonfiglioli 1988: 87). Lasting until 1922, this period was less marked by violence, yet instances of massacres still occurred, intended to suppress localized efforts of resistance (Fuglestad 1983: 99; Paris 1990: 197f.). The introduction of taxes, first in the form of naturals, later monetized, accompanied the colonial conquest from early on (Bonfiglioli 1988: 91ff.; Fuglestad 1983: 87f.). As far as taxation is concerned, many sources contend that the Woɗaaɓe were rather successful in escaping the attention of the colonial

32  Space, Place and Identity

administration, due to their fragmentation into small and dispersed family units (Bonfiglioli 1988: 105; Schareika 2003a: 137). Henceforth, they further increased their mobility, preferably moving into remote areas that were less successfully controlled by the colonial power, to avoid conflicts or simply too much attention (Paris 1990: 198; Maliki 1982: 26). In other regards as well, small mobile groups like the Woɗaaɓe were better able to cope with the effects of the colonial conquest than the sedentary populations. From the Woɗaaɓe’s perspective, the euphemistic dictum of the colonial ‘pacification’, however problematic it might otherwise be, was rather applicable in the sense that they suffered less from violent raiding by the Tuareg (Braukämper 1971: 72; Boutrais 1994: 141f.). The colonial advance against the Tuareg, who were considered by the French as the major source of resistance, thus allowed the Woɗaaɓe access to new pastoral spaces in the north, which the Tuareg had denied them in pre-colonial times (Baier 1980: 133; White 1987: 240). In this sense the colonial conquest paved the way for the migration movements that led the Woɗaaɓe to the northern fringes of the Sahel, which they inhabit today. At the same time, the necessity to orient themselves towards more northern pastures increased at the turn of the twentieth century because colonial fiscal and agricultural politics caused a steadily increasing density of fields in the south (Fuglestad 1983: 87; Bonfiglioli 1988: 91ff.). Each year during the agricultural period, the nomadic pastoralists moved north to avoid conflicts with farmers, and as the agricultural front moved northward, the itineraries of the pastoralists’ herds also gradually expanded further north. Like most Woɗaaɓe (see Bonfiglioli 1988: 104), in the first decade of the twentieth century the ancestors of my interlocutors from the study group stayed in the area of Birni N’Konni, while spending the rainy season around Madaoua. Between 1915 and 1920, the Gojanko’en migrated into the Ader region, more precisely to the area of Chadawanka, under the leadership of Atiiku from the Gawanko’en clan segment. The office of an administrative chief (arɗo) had been passed on to Atiiku by his father Mbuuldi, to whom it had purportedly been transferred by Mbuuwa, arɗo of the senior clan segment, the Gojanko’en Goje (see Figure 2.1).6 Ɗawra Egoyi also confirmed that Atiiku from the Mbuuldi maximal lineage was the arɗo who led the group into the Ader region. Shortly after this migration, the first recorded encounter with the French colonialists took place, which for Ɗawra (and thus from the perspective of the Kuskudu maximal lineage) marks the point where the Kuskudu established their supremacy over the Mbuuldi lineage within the Gojanko’en Gawo. The civil administration of the colonial state had been established by 1922 and the narrated events are likely to have taken place shortly after. According to Ɗawra, it was his grandfather Giiye who first met with the colonizers, while they were in the area of Bouza, west of Dakoro. Probably this

A History of Migrations  33

Figure 2.1  The internal structure of the Woɗaaɓe Gojanko’en (diagram by F. Köhler).

was in the person of a colonial officer who had come with a troop of goumiers, indigenous soldiers in the service of the French. The administrator asked Giiye to assemble the Woɗaaɓe – possibly to assess them for taxation – but Giiye told him that the Woɗaaɓe had dispersed and would not be back around Bouza in larger numbers until after the end of the rainy season. Back in the pastoral camp, Giiye told Atiiku, the son of Mbuuldi and arɗo of the Gojanko’en Gawo, of his encounter, reporting that the administrator had asked him to present the Woɗaaɓe family heads at the colonial post upon their return to the area. Atiiku, however, asked to be left out of this, because ‘he was afraid of the French’. Instead, Giiye was accompanied by one of Atiiku’s slaves, who carried a big calabash of butter to the French. Giiye himself took a young bull from his herd and brought it, together with the butter, to the French administration, as a customary act of honouring and welcoming an authority. In the cold dry season, when the bulk of the Woɗaaɓe homesteads were back in the region, Giiye took the men to Bouza and was awarded with a political office by the French colonial administration, whereas his great-uncle (FFBS) Atiiku was too afraid of the French to go himself. This anecdote is very rich in terms of its interpretive potential. Regardless of the question of the objective correctness of the historical facts, the rhetorical strategy that Ɗawra Egoyi pursues clearly shows his intention of claiming and justifying the supremacy of the Kuskudu lineage over the Mbuuldi lineage within the clan segment of the Gawanko’en. Ɗawra is a direct patrilineal descendant of Giiye and thus of Kuskudu. Atiiku and Giiye’s father Riiga were

34  Space, Place and Identity

Figure 2.2  The kinship relationship of Atiiku and Giiye (diagram by F. Köhler).

patrilateral parallel cousins (FBS – ɓiɓɓe baaba’en), since their fathers, Mbuuldi and Kuskudu are said to have been elder brother and younger brother (see Figure 2.2). Thus, in terms of seniority, the Mbuuldi are the branch with the older founding ancestor and would thus enjoy higher status owing to their seniority. But it is exactly this status which is put into question by Ɗawra’s account in which Atiiku is mockingly described as being too frightened to go and meet the colonial officers. Giiye took his chance and was rewarded with the office of a chief. The story thus serves to justify not only the supremacy of the Kuskudu over the Mbuuldi maximal lineage, but also the legitimacy of Ɗawra’s own claim to the office of a chief, which is here traced back to his grandfather, Giiye. Upon the establishment of their colonial administrative system, the French appointed indigenous chiefs at the lower levels of the territorial subdivisions as a link between the colonial administration and the people (Spittler 1981; Maliki 1982: 28; Fuglestad 1983: 66ff.). The principal levels at which these administrative chiefs were put in place in Niger were the village and the district (canton). For the predominantly pastoral populations, two corresponding offices were introduced, the chef de tribu and the chef de groupement, called arɗo, and laamiiɗo in local Fulfulde.7 The chiefs function as arbiters in cases of civil jurisdiction, and as functionaries who collect the taxes of their tributaries (talaka’en) to remit them to the state administration. This system of ‘traditional’ chiefs with its double division into two hierarchical levels and two categories – ‘sedentary’ and ‘nomadic’ – exists to the present day in Niger. Although called chefferie

A History of Migrations  35

traditionnelle, the ‘tradition’ of this institution, which was largely modelled by the colonial administration on partly real, partly imagined pre-colonial structures, can be questioned (Spittler 1981: 74ff.; Fuglestad 1983: 66ff., 80; for Benin, see Bierschenk 1993, 1997). From a colonial history perspective, the appointment of Giiye to the office of a chief can be read as an example of how the colonial power arbitrarily rewarded those willing to collaborate with titles and offices, thus breaking up existing hierarchies and political power structures (Fuglestad 1983: 84–9; Abba 1990: 51ff.; Bako-Arifari 1998; Lund 1998: 66f.). The French strategy of appointing chiefs did not only cause considerable upheaval in the local political order, but also, in the case of the Woɗaaɓe, had a direct impact on the fission of clans. Later on, especially since the 1940s and the 1950s, this impact became stronger, as the colonial administration appointed greater numbers of chiefs in order better to control certain areas (Bonfiglioli 1988: 126). The multiplication of chiefs caused a fragmentation not only of local political power, but also of social structures (ibid.). Giiye and Atiiku went on to stay together in the Ader region, probably between Chadawanka and Bouza. While the majority of the Gojanko’en stayed in this area and still live there today (Maliki 1982: 37), others moved further north to the Agadez region, especially in the 1940s and 1950s (Paris 1990: 200). Although this new wave of migrations might again be interpreted as part of the aforementioned defensive strategy of avoiding both the expanding agricultural areas and the colonial administration, in this case it was also a deliberate response to the new opportunities opening up after the colonial advance in the north: The French encouraged Fulɓe presence in the region of Agadez, in order to undermine the local supremacy of the Tuareg, whose warrior aristocracy was still regarded as having a dangerous potential for resistance (Baier 1980: 118, 133). From the 1930s, as a part of this strategy, the French attributed a number of newly constructed, cemented pastoral wells in the region to Woɗaaɓe groups (Maliki 1982: 41; Paris 1990: 200). The Woɗaaɓe herds were also welcome in the north for providing meat to the garrisons in Ingal and Agadez (Paris 1990: 200), and northward movements of nomadic Fulɓe were therefore further encouraged by reduced taxes in the northern region (Baier 1980: 133; Maliki 1982: 41; White 1987: 240; Paris 1990: 200). While these conditions alone would have made the Agadez region attractive, the climatic conditions of the period were also exceptionally favourable for pastoralism (Bonfiglioli 1988: 121). Among the Gojanko’en who migrated to the Agadez region were Ɗawra’s father Egoyi and several of his brothers with their families. According to Ɗawra, his paternal uncle (FB) Jaataw was the first of the family to migrate to this region, more precisely to an area south of Ingal:

36  Space, Place and Identity

My father’s elder brother, Jaataw, moved to the Agadez region. . . . He was the first Boɗaaɗo there. There were no other Woɗaaɓe beside him, only Tuareg. . . . They stayed in the Agadez region, south of Ingal. This is where they constructed a pastoral well of the name of Alala. It was the first well in the region. My father and his brothers made this well accessible to all the other Woɗaaɓe. (Ɗawra Egoyi, January 2011) Later, Jaataw and his brothers constructed a well at Alala which is still in use and still in the hands of the family. It was the first well of solid materials that the group ever owned. The exact time of arrival in the area is difficult to reconstruct, but probably it was not earlier than in the late 1940s or around 1950, since Ɗawra (born around 1945) and even his younger brother Araba had been born in the Ader region. Both remember having been small boys when the migration took place. This probable period of the migration to Alala makes Ɗawra’s claim that Jaataw was the first Boɗaaɗo in the Agadez region rather doubtful. The first Gojanko’en had already been registered in the area of Ingal around 1920 (Paris 1990: 191). Araba acknowledged that while there might already have been other Woɗaaɓe, notably from the Ɓii Korony’en and Ɓii Nga’en clans, Jaataw was at least the first from the Gojanko’en clan to arrive in the region.8 Before leaving to Ingal, Jaataw transferred his office of chef de groupement to his brother Kardaw, but later on, Kardaw himself followed his brother to Ingal: Kardaw continued to stay in the Ader region. Jaataw was the one who went to the Agadez region. Later, Kardaw would follow Jaataw. At the time there were excellent pastures and the herds were abundant. Everybody enjoyed pastoralism. Before moving, he [Kardaw] took a young man of the name of Laayaaru aside. He said: ‘Alright, you will stay here’. He caught him where he was with his cattle and told him: ‘We will not give up our office! You will stay here in the Ader region. As for us, we will move on [to Ingal]’. So, while their paternal cousins continued to carry the title of a chef de groupement in the Ader region, our fathers went to the Agadez region where [Jaataw] finally acquired a title of chef de groupement. (Ɗawra Egoyi, July 2011) Jaataw thus established himself as a chef de groupement in Alala. He passed his office on to Kardaw, after whose death his son Ndonya and finally Ndonya’s son Njaange continued to carry it. Today it is in the hands of Njaange’s son Baawa. Other members of the Kuskudu lineage, however, migrated further on to the Damergou region, yet others even as far east as Tesker, where the conditions during the 1940s and 1950s were similarly favourable for pastoralism (Retaillé 1984: 191).

A History of Migrations  37

Figure 2.3  Giiye and his descendants (the diagram includes only those mentioned in the text) (diagram by F. Köhler).

Ɗawra also migrated to the Damergou region, yet much later and in response

to a series of difficult years, characterized by little rainfall and bad pastures. This period began towards the end of the 1960s and culminated in the catastrophic drought of 1973–74. Ɗawra’s father Egoyi (the younger brother of Kardaw and Jaataw), as well as some other members of the Kuskudu lineage, anticipated the worst of these years and moved away in time to avoid the severe losses that their relatives in the Agadez region had to face. The year of the first arrival in the Damergou region was 1970, and the group at first stayed in the valley of Eliki. It was not yet a season of general famine but of bad pastures that pushed Ɗawra’s group to migrate (Hitaande nde gawri e woodi, geene ngalaa – ‘That year, there was millet, but there was no grass’). At the time, many Gojanko’en migrated to the Damergou region. Most of them, however, were from the Mbuuldi lineage and the majority later returned to the Ader and Agadez regions.9 After exploratory missions, Egoyi and his sons finally also decided to move there, together with a small number of other Kuskudu. They stayed for two or three years in the area north of Tanout before moving on to Tesker, where they joined another segment of the Kuskudu lineage. This was a migration group led by Siddi, a second cousin (FFBSS) of Egoyi who had migrated much earlier to the Damergou region and further to Tesker (Köhler 2017a: text 3). Shortly after, the lack of rainfall and pastures during the drought season of 1973–74 (locally called gandaw) forced Ɗawra and his family to leave Tesker again and to move south, to Matsena, just across the Nigerian border. After the drought, they moved back to the Damergou

38  Space, Place and Identity

region, which they still considered as their home range (Köhler 2017a: text 2,17). Like most of the other Gojanko’en who moved to the Damergou region, Ɗawra affiliated himself to Laamiɗo Riiya from the Woɗaaɓe Ɓii Ute’en in Gourbobo.10 They now remained permanently in the Damergou region, yet for lack of their own pastoral well they remained rather mobile, driving their cattle to the areas where the pastures were best and where access to water was the least difficult. They habitually watered their animals at a cemented public well near Gourbobo or at a smaller one in a place called Tumulle. During the 1970s, the pastoral well of Tumulle was widely frequented by the Gojanko’en, but later it attracted Kanuri settlers to the point that Tumulle developed into a sedentary village and the mobile pastoralists even tended to be excluded from access to water there. Later in the 1970s, Buuyo, a son of Egoyi’s brother Hamma, was the first from the Kuskudu in the Damergou region who finally acquired a well of his own, in a place called Salaga, near Farak. Access to water being a question of major importance for pastoralists, acquiring or constructing wells became an important strategy of securing the access to pastoral resources and has remained so to date. Ɗawra’s patrilateral cross cousin (FZS) Usman followed the same strategy and shortly after Buuyo, still in the 1970s, also constructed his own well, situated in Ngel Tireeji11 near the village of Eliki Njaptooji. This well is today in the hands of Usman’s son Laɓɗo and remains the point of attachment for this lineage segment. The account that Laɓɗo gives of the construction of the well hints at the rivalry over resources that already existed at the time even between different groups of Woɗaaɓe. It also shows how significant social networks beyond one’s own habitual home range had already been for the successful establishment in a new area: The Ɓii Ute’en did not want us to dig a well because they said it was too close to their own well, called ‘Also’. So, Usman went to see Laamiɗo Riiya, and Riiya said: ‘No, all the Woɗaaɓe are one. A Boɗaaɗo shall not hinder another to build a well. Besides, Usman’s father Mo’aadu had come to the Damergou region regularly since a long time ago. . . . Usman shall go and build his well’. And so the Ɓii Ute’en accepted. (Laɓɗo Usman, June 2011) Egoyi, who did not yet possess a well of his own, continued to be very mobile throughout the 1970s, watering his herds either at the cemented, ­government-constructed public well of Kekeni, near Gourbobo, or north of Salaga at a well called Kafarnda, among others. Being on good terms with his nephew (BS) Buuyo, he also regularly used the well in Salaga. In the meantime, Egoyi’s eldest son Ɗawra had established himself as an independent head of his own herd and family. In the second half of the 1970s,

A History of Migrations  39

he spent about four years between the areas of Damagaram Takkaya and Tesker, again in close contact with Siddi’s group and especially with Siddi’s son Umaru. In 1983–85, another severe drought swept the wider region and this time became a major threat for the Woɗaaɓe in the Damergou region as well. This major drought is locally called banga-banga, or, when in reference more specifically to the worst year, 1984, al buhari. The latter name is derived from the name of the then president of Nigeria, Muhammadu Buhari, whose government closed the borders between Nigeria and its neighbouring countries in 1984 (Spittler 1993: 123). Although not in direct connection with the critical situation in the northern Sahel, this decision had severe consequences for the pastoralists attempting to flee the drought. 1984 is thus remembered by the Woɗaaɓe as the year when they were hindered by the Nigerian state from crossing the border, or even expelled. While some did manage to move their herds to the south, for many an evasion of the drought by increased mobility was not feasible this time and many herds were reduced to a critical point, forcing their owners to settle down at the margins of larger towns. In the case of the Damergou groups this was Zinder, where many destitute Woɗaaɓe moved to benefit from the distribution of relief food (taimako). Ɗawra, like many other members of the Kuskudu lineage, stayed in Zinder for about two years. While he was able to return to the pastoral economy in 1986 with a few animals, many others remained in town. The division of family groups between the pastoral economy and urban migrant work that is characteristic of the Woɗaaɓe today was henceforth established. Both Ɗawra’s father Egoyi and Ɗawra’s paternal cousin (FBS) Buuyo died towards the end of the drought period, and when Ɗawra moved back to the Damergou region, he took care of Buuyo’s adolescent sons (Taafa, Nano and Maalam) who continued to live with Ɗawra’s family. Ɗawra returned to Salaga, where he took care of the pastoral well formerly owned by Buuyo until the latter’s sons were old enough to manage the inherited well themselves. In subsequent years, Ɗawra constructed his own well, Yalema, only about four kilometres from Salaga, in a place called Intrika. Today this well belongs to Ɗawra’s youngest brother Boyi, to whom he sold it before migrating to the Koutous region with his sons. This most recent migration of Ɗawra’s minimal lineage was prepared by the construction of a new pastoral well near the Kanuri village of Ganatcha, beginning in 2003. Upon completion of the well, Ɗawra and his sons successively moved to the Koutous region with their families between 2005 and 2006.12 The background and consequences of this latest migration are the object of a detailed analysis in chapters 5 and 6.

Integration and Withdrawal Nomads are often accused by sedentary state institutions of deliberately using mobility as a means of evading control by the state, primarily with the

40  Space, Place and Identity

aim of escaping taxation and other measures interfering with their interests. Consequently, states have been notorious for attempting to settle mobile populations. Their ability to exploit rather marginal ecological niches has indeed allowed the Woɗaaɓe and other pastoralists to keep at a distance from the centres of political power. The identification of the open pastoral rangeland (ladde) as an asset granting a relative political autonomy has been pointed out with regard to different Fulɓe groups across regions (Schareika 2007: 117; Ciavolella 2010: 75f.). But generalizing statements about the state-avoiding attitudes of pastoralists have to be seen in the light of a cost–benefit ratio and against the alternative of integration, because the central question is what the state has to offer. Withdrawal occurs mainly when the state is perceived more as a nuisance, while integration can be considered attractive if it is seen as a structure that provides resources (Azarya 1996: 1). Ethnographic works on the Woɗaaɓe (e.g. Stenning 1957: 71) confirm that their strategy of withdrawal in certain historic situations was motivated by the application of this rationale. In Fulɓe studies, the phenomenon of withdrawal has been variously referred to as an ‘exit’ option (Azarya 1988; Bierschenk 1997; Dafinger and Pelican 2002; Osman 2009). The concept is borrowed from Hirschman (1970), who distinguishes between ‘exit’ and ‘voice’ as two principal ways of reacting towards dissatisfaction with superordinate structures. ‘Exit’ is basically economically motivated, e.g. in reaction to dissatisfaction with the offers or performance of the state, and translates into circumvention, evasion or avoidance, while ‘raising one’s voice’ is a political reaction of active engagement or participation. Hirschman’s idea of ‘exit’ mainly as withdrawal from participation applies well to the case of the Woɗaaɓe. Their strategies of state evasion, although based on high mobility, did not necessarily imply flight from the territory of a given polity, but often rather withdrawal into less controlled marginal areas, in order to escape the reach of authorities (e.g. Bonfiglioli 1988: 105; Paris 1990: 198; Schareika 2004: 177). This tactical principle of pastoralists has also been grasped by the concept of interstitiality (Stenning 1966: 389; Diallo 2008). Diallo’s definition of interstitial spaces (ibid.: 7) makes explicit mention of frontier regions and thus recalls Kopytoff’s (1987) concept of the interstitial frontier, the no-man’s land between established polities characterized by a sparse population and a weak presence of the political institutions. With some qualifications, this concept also applies to the contemporary situation in the rural zones of the study area. Although there is no longer any de jure political no-man’s land, either in Niger or elsewhere (see Schlee 1984: 141; Schlee 1990a: 1), an effective control of rural areas is given only to a limited extent. Owing to the vastness of the territory, state control is here merely theoretical or partial. Strategies of exit into such interstitial areas have been documented for the Woɗaaɓe at least since the later period of the Sokoto Empire (Bonfiglioli 1988) when the initial euphoria about the new Fulɓe rulers increasingly gave way to

A History of Migrations  41

disenchantment. Confronted with religious ideologies discouraging their central cultural practices and values (Paris 1990: 196f.), or with state policies plaguing them with taxation and urging them to settle down (Adamu 1986: 60; Maliki 1982: 23), the Woɗaaɓe reacted with strategic retreat. In accounts of the colonial and post-colonial period, the characterization of the Woɗaaɓe as following evasive strategies continues to be prominent. Many Woɗaaɓe regarded the newly introduced school education with suspicion as they feared it would alienate the young generation and extract essential workforce from the pastoral economy. Again, the answer was exit. Against the background of this attitude, the marginality of the Woɗaaɓe in the contemporary Nigerien state and society has been characterized as the result of a once self-chosen condition (Boesen 2009: 70). In contemporary Niger, however, a less self-determined and thus more perturbing implication of the term ‘interstitial populations’ has become prevalent: mobile pastoralists today often simply do not have access to other than quite literally interstitial spaces; they have to graze their herds in the spaces in-between, in the narrow enclaves of uncultivated land between the fields of the agricultural populations. Due to the dramatic population growth, areas that used to be unclaimed rangeland and offered important pasture resources are increasingly transformed into farmland. The sedentary populations often own considerable numbers of cattle themselves, and the pasture resources that used to be free of access for the mobile pastoralists tend more and more to become the object of competitive strategies of appropriation. The mobile populations are often the losers in this race for resources and increasingly, it is less in the sense of a strategy that they make use of the interstitial spaces, than for lack of better options and as a result of increasing exclusion. Looked at from this angle, the assumption of the deliberate marginality of the Woɗaaɓe can be maintained only in a more nuanced way. But even beyond these considerations, the assumption of strategic withdrawal cannot be generalized with regard to the Woɗaaɓe. A more common strategy than strict evasion and avoidance has been to keep a low profile towards authorities, and at times, the approach comprised obedience and cooperation with the state to avoid conflicts (Paris 1990: 198; Schareika 2004: 181). Within the study group, although examples of state evasion have been reported (e.g. Köhler 2017a: text 4,15), they are not a systematic tendency. From the very first encounters with the colonizers one can rather recognize an oscillation between withdrawal and integration and in fact controversial discussions about the q­ uestion of which attitude was to be favoured. The episode in Ɗawra’s account that depicts the relation between the ancestors of the Mbuuldi and Kuskudu maximal lineages is representative of this controversy: according to Ɗawra, Atiiku from the Mbuuldi branch was ‘afraid’ to meet the colonial officers, while Giiye from the Kuskudu was willing to interact and cooperate with them, and was rewarded with a political office. The historical

42  Space, Place and Identity

anecdote shows from an emic perspective how the early encounters between the Woɗaaɓe and the colonizers were negotiated, and how the question of integration or withdrawal was unresolved, yet fervently discussed and answered differently from the very beginning. One might characterize Giiye’s attitude as opportunistic, or even see in it an act of collaboration with the colonial power. For Ɗawra, such doubts do not seem to be of much concern. The colonizer is pragmatically regarded as a new player on the political scene, relations with whom can be used to profit on the internal political level, e.g. by the legitimization of political titles and thus the consolidation of power. The question of evading the state or, on the contrary, even seeking a rather close contact with representatives of the colonial and postcolonial state is a recurring theme in the history of the study group and has remained so to this day. One example is Ɗawra’s acquaintance with the first president of independent Niger, Diori Hamani (see Köhler 2017a: text 5), another, more recent one is the active participation in the election campaigns of a presidential candidate with whose daughter some of the women from the study group were friends. In turn, this friendship facilitated the application process for having a rural community school accredited. In the political reality of Niger, private contacts to authorities can be used to facilitate the achievement of ends.

Historical Perspectives on Mobility, Space and Place Stenning (1957) differentiates between three major forms of movement among pastoral Fulɓe: transhumance, migratory drift and migration. Transhumance refers to movements between different ecological zones in order to adapt to seasonal variation and for the herds to profit from different seasonal pasture resources.13 Migratory drift means that the recurring seasonal transhumance movements are always slightly variable, which can lead to gradual, yet over time considerable, shifts of habitual grazing areas. Migration, finally, designates a movement into a new area, out of the framework of the habitual pastoral itineraries. These three forms of pastoral movement cannot always clearly be separated but are tightly interrelated (Diallo 2001: 154). While migratory drift occurs as a result of adaptations of transhumance itineraries to ecological variations, even long-range migrations are in practice realized by a sequence of short-range movements that each have the character of routine camp relocations (gonsol). Diallo (2001: 155) differentiates between direct and indirect migration: while the latter term expresses a similar idea to Stenning’s ‘migratory drift’, the former stresses the rationale of migration as a conscious decision. The major migrations movements outlined above were clearly also of this conscious, direct nature and were accordingly referred to as perol. In the literature, this term has been translated differently, either as ‘flight from intolerable conditions of a political or ideological nature’ (Stenning 1957: 70), or more

A History of Migrations  43

broadly as major migration movements due to ecological, social or political reasons (Bonfiglioli 1998). The latter definition fits better in the study context as it considers push and pull factor alike. The study group’s migration movements cannot monocausally be interpreted in terms of flight from a politically repressive situation or by deteriorating or even catastrophic ecological conditions in the area left behind. The migration from the Ader to the Agadez region in particular was not due to political pressure or drought, but occurred in response to the favourable pastoral conditions in the target area of migration. This corresponds perfectly to the picture drawn by Maliki (1982) and Paris (1990). Dupire also mentions rich pastures and numerous newly constructed pastoral wells as a decisive factor for this wave of northward migrations (1962: 79). The process of day to day camp relocations (gonsol) can be understood in the terms of Ingold’s ‘wayfaring’ approach in which places are, on an elementary level, regarded as ‘moment[s] of rest along a path of movement’ (2007: 96). A¹

→

A²

→

A³

→

etc.

If we thus think of a place (A) in Retaillé’s (2013) sense as a temporary manifestation of social space and as mobile itself, A¹ and A² can be understood as different manifestations of the same place. Placemaking, in this perspective, is a result of human movement and resting. Movement produces space; resting materializes places. This wayfaring pattern characteristic of the Woɗaaɓe is closely linked to pastoral necessities and the specific responses to them. Beyond this fundamental aspect of places as resting places on a trail of movement, however, the social aspect of place also comes into play here. Again with Ingold, we can regard the places created by Woɗaaɓe mobility as the entanglements of the lines of human movement in ‘knots’ (2009: 33).14 What seems important to note is that, despite a high degree of mobility, the different factions of the Kuskudu maximal lineage have always had a geographic attachment. Individual migratory groups are never just moving about, but they are always attached to a home range, and often more concretely to a core area within this range, or even to a specific site, typically a pastoral well. Especially in more recent decades, the latter has become a key element in the process of geographic attachment. In this context, places are quite concretely made: they are physically materialized by constructing wells (and increasingly, as I will show later on, also schools and private houses). The construction of a well literally creates a place where there had not been one before, because the location of the infrastructure defines the place where the itineraries of humans will henceforth intersect, just as it informs people’s perceptions and definitions of their home ranges. The well thus becomes a site in which social space ‘takes place’ (Retaillé 2013).

44  Space, Place and Identity

However, this geographic attachment can be changed. The historic migrations of the study group show a pattern of placemaking by repeated establishment of new home ranges through migration into new pastoral areas. The reasons for migrations (perol) range from pastoral opportunities or constraints to political motives. A succession of migrations leads to a sequence of areas and sites of historic attachment. However, since lineages generally do not migrate as complete entities, but rather in varying agglomerations of families or small migration groups, migration typically goes together with the separation and reconfiguration of social groups. With regard to the aspect of placemaking, this pattern is not one of a simple relocation (A¹ → A² → A³), but of a replication of places: A

→

B (A’)

→

C (A’’)

→

etc.

From the historical accounts, this close link between placemaking processes and processes of group-(re-)production – ‘the intertwined processes of place making and people making’ (Gupta and Ferguson 1997b: 4) – becomes apparent. Another important principle of this replication process is that the different places that it produces remain, to a varying degree, inter-connected over time and space. Sites or areas of a former attachment of a faction that has migrated thus often remain sites of a current attachment for other factions that have chosen not to migrate. For instance, at the time of the northward migration into the Ader region, parts of the Gojanko’en remained back in Sokoto. The same applies for the migration from the Ader to the Agadez region. The spatial separation, although it often goes together with a politico-administrative split, most often does not entail a complete rupture of social contacts. The accounts of my Kuskudu interlocutors confirm that social contacts with members of those parts of the group that had stayed behind while others had moved on to establish a new home range continue to be maintained. When the minimal lineages of Ɗawra, Usman, Buuyo and others moved from the Agadez to the Damergou region, other parts of the Kuskudu remained back in Alala or returned there after the drought years and are still living there today. The ties with this part of the community have been maintained, just as the relations between the groups that have migrated to the Koutous and those that have stayed behind in the Damergou region. Hence, from a diachronic perspective, we can recognize sequences of ‘lines’ (itineraries of movement) and ‘knots’ (points of convergence and concentration where these itineraries intersect) (Ingold 2007, 2009; Sheller and Urry 2006: 214; Retaillé 2013), i.e. chains of temporary manifestations of place, created by a specific pattern of human movement. From a synchronic perspective, we can recognize a network (or ‘meshwork’, Lefebvre 1991 [1974]) with ‘knots’, i.e. a web of places that, to a varying degree, remain connected by human movement.

A History of Migrations  45

As historic sites of past attachments, these chains of manifestations of place become a means of referring to the past (as mental maps of the group’s migration history), but to a certain extent they also remain a significant potential resource in the present (as mental maps of social contacts). Social networks are actively maintained over time to the degree that the general conditions permit. In some cases, the patterns of seasonal transhumance movements might permit visiting acquaintances or relatives. Inter-clan ceremonies can bring together people from sometimes distant regions and are important occasions for renewing social bonds (see Chapter 10). But even extended travelling that primarily serves the purpose of keeping in touch and maintaining social relations is very common. While the vastness of Niger and the difficulties of public transport in the more remote areas limit the possibilities of regularly staying in touch with spatially more distant relations, regular contact is no prerequisite for the maintenance of a social network over time. Social relations can remain potentially functional even if they have been inactive for long periods, and if the situation requires, they can quickly be reactivated. A good example is that of Ɗawra’s younger brother Araba: after having lost most of his small stock during the drought season of 2010, he travelled from the Damergou region to Alala to visit some of his lineage-mates whom he had not seen for many years. He returned home with ten loan animals provided by his relatives, which allowed him to reconstitute a herd of smallstock.15 It is obvious that the geographical distance of a migration plays a central role for the question of whether, and to what degree, social ties continue to be maintained. Furthermore, not all individuals continue to maintain active contacts across sites. With regard to the study group, connections are strong particularly between the older lineage members in the Damergou region and their age-mates in the Agadez region, with whom they grew up. Social contacts are maintained by mutual visits, and they are manifest in animal loans between individuals across regions. Younger men and women, who were not yet born when the most recent migration took place, are hardly attached to the former home range of their parents or grandfathers, since they generally do not know any of their relatives of their own age in the concerned region. The historical distance of a migration movement is thus another determining factor for the question of whether, and to what extent, contacts are maintained across space. The links to lineage members from the Ader region, for instance, are much weaker for reason of the historically more distant migration.

46  Space, Place and Identity

Figure 2.4  Marriages of Hamma Beleti (diagram by F. Köhler).

The anecdote on how Kardaw ensured that a trusted member of the family took over his office as a chef de groupement before he migrated to Ingal shows how translocal ties are consciously constructed and strategically maintained. Contacts to a former site are not easily abandoned when moving on, as they constitute a valuable fallback resource. They grant social security, leave an option for return in difficult years, and – as the following example illustrates – can even widen marriage options. Hamma Beleti, a man of about 35 years of age, grew up with his mother in the Damergou region after his parents separated, whereas his father lived in the Agadez region (see Figure 2.4). A kooɓgal betrothal marriage had not yet been arranged for Hamma by his father. Instead, his mother now arranged a marriage for him with the daughter of one of his maternal uncles (MBD). When this marriage failed, however, marital options seemed limited in his mother’s lineage, and during the time of my fieldwork, Hamma returned to the Agadez region to negotiate a new marriage with a daughter of a paternal uncle (FBD). Although Hamma’s immediate family was split across two regions, both local groups potentially offered marital options for him. Also, these options were not diminished by the fact that he had not been in contact with his paternal kin for many years. Relational networks across places, i.e. translocal networks, can thus provide access to a variety of resources (water, pastures, animals, marriage alliances). Their maintenance can be regarded as an efficient risk-management strategy (Dafinger and Pelican 2002: 16) and more generally as an important principle of mobile economic strategies. If one might characterize the just described maintenance of social relations with lineage members in former home ranges as ‘retrospective’, they can be regarded as ‘prospective’, from the point of view of those staying behind. Actively maintaining relations with lineage members who have migrated to a different region is equally advantageous as it gives the option to join them and profit from their support in case migration into the respective region should be considered

A History of Migrations  47

attractive or necessary at a later point of time. For the same reason, contacts in different areas are also actively established at the occasions of travelling, exploratory missions and more episodic transhumance movements beyond the habitual itineraries. Access to new pastoral areas is not necessarily conditioned but much facilitated by previously established social relations in the target area. The significance of such prospective networking becomes apparent, e.g. from the example of Usman’s establishment in the Damergou region. Obtaining the permission to construct his well in Ngel Tireeji was made possible by the fact that Usman’s father Mo’aadu had established friendly relations with the local chief of the Woɗaaɓe Ɓii Ute’en during repeated transhumance movements into the region. Similarly, in the case of Ɗawra’s migration from the Damergou region to Ganatcha, Ɗawra had already been staying close to Umaru before migrating to the area, and the old network ties facilitated both the obtaining of permission to construct a well and the access to farm land (see Chapter 5).

Claims on Supremacy: The Importance of Being the First to Arrive Social networks are relevant as insurance and Woɗaaɓe can generally rely upon their lineage-mates for help if they are in difficulties. However, it is preferable to be in a position oneself to share one’s own wealth for the benefit of the lineage-community. Moving on and making new resources accessible, such as by constructing wells and thus opening up new pastoral spaces, is a cultural ideal and allows one to gain considerable prestige and status (Schareika 2003a: 91ff.). A pioneer has the potential to become a leader, and in fact, the term arɗo that today also designates a political and administrative chief (chef de tribu), originally refers to leaders – in particular leaders of pastoral migrations – and etymologically means ‘he who walks ahead’ (Kintz 1985: 98). The legitimacy and renown of political leaders owes a lot to their agnatic descent from such a pioneer of pastoral migration (ibid.: 103). The one who stays behind (ngonooɗo), on the contrary, is generally rather pityingly regarded by those who have moved on. Hence also the importance of being the first to arrive. Kopytoff (1987: 148) recognizes ‘the principle of firstcomer authority’ as being widespread in African politics. Among the Woɗaaɓe, the supremacy of the firstcomer is visible in the social sphere, namely in the hierarchy between brothers according to their age, or that between co-wives according to the seniority of their marriage. It is of particular importance also in the political sphere, such as with regard to the order of arrival of different groups in a given area and the resulting legitimacy of their claims to local resources. Although Woɗaaɓe society can be considered as egalitarian in the political sense that it lacks a central authority and that its different segments are autonomous, the social structure is characterized by hierarchical principles in which firstcomer status plays a crucial role.

48  Space, Place and Identity

Dupire (1962: 307) has demonstrated that the hierarchy between clans, although defined in terms of descent-based seniority, depends much on their order of arrival in a given area. Relative seniority is thus the idiom in which hierarchies are expressed, and the group that first arrived in a given area (or the one that can successfully maintain such a claim) is generally considered as senior by more recently arrived groups and enjoys certain priority rights, e.g. concerning the order of watering at a given well (Schareika 2007: 207). As the seniority status of a clan is not absolute, but depends on the regionally differing, relative order of arrival of clan sections, in some cases contradictory assumptions about the seniority of clans can be found across regions (Dupire 1962, 1970; Schareika 2007). For example, in the Damergou region, the Jiijiiru maintain a position of higher seniority status than the Suudu Suka’el, whereas in Eastern Niger, the reverse applies (Schareika 2007: 305, 348). Furthermore, the hierarchies are also determined by power relations and are thus frequently contested (Paris 1997; Loncke 2015) or a matter of debate, based on diverging claims. The lack of a central institution that could enforce a valid version of the hierarchy among clans gives the rhetorical management of versions of history a particular relevance. Historical accounts are thus often highly ideological and politicized among the Woɗaaɓe (Loftsdóttir 2002b: 300) as in other contexts in Africa and elsewhere (e.g. Murphy and Bledsoe 1987). At the same time, accounts of oral history are a part of social action (Tonkin 1998: 96). History and, in consequence, the rights deriving from its interpretation, is a matter of negotiation and a question of persuasiveness and the capacity to convince a majority (Schareika 2007). In such a context, the manipulation of historical facts can be used as part of a political strategy (Murphy and Bledsoe 1987; Packard 1987). Different versions of mythico-historical narratives of the relations between clans show how such rhetoric strategies have been applied in the struggle to define the meaning of the past (Cooper 2005: 198; see also Köhler 2017a: text 6). What holds true for the hierarchy between clans is equally relevant at the level of individual status and renown, and, as noted above, for legitimizing individual claims to leadership. The rhetoric used in the historical narratives highlights this aspect: Firstcomer claims are a constantly recurring theme, and they are often emphatically staked. Ɗawra stresses that his father’s brothers were the first to construct a well in the Agadez region (‘ka’iru ariti ɓulli fu’ – ‘it was the first of all wells’).16 But a second aspect, that of making this resource available to others, is just as important: ‘ɓe kolli boɗaaɗo fuu woindu’ – literally: ‘they showed it to all the Woɗaaɓe’). Egalitarianism does not imply the absence of material or economic differences between individuals (Salzman 2004: 71), nor the absence of differences in individual status and prestige. Although Woɗaaɓe society is characterized by some institutionalized mechanisms that contribute to levelling out extreme

A History of Migrations  49

material inequalities – such as the institution of animal loans or the communal sharing, instead of individual monopolization, of resources –, these mechanisms at the same time have an effect on individual status differences. The productive property circulates and is invested in social networks, but to a certain extent it is also invested in prestige. An individual who is in a position to share his resources to the benefit of others (be it in the form of animal loans or in the form of a well that is put to the profit of the wider community) can earn considerable renown and status. Such increased status, however, does not lead to a stronger stratification of the society, since economic differences are likely not to be stable over time, due to the high risks related to pastoralism (Salzman 2004: 74). Rather, the possibility to gain prestige and status by sharing has the effect that the mechanisms of internal redistribution of resources function in the sense of a social insurance. As I will show in Chapter 7, accessing and granting others access to new resources such as urban jobs can be a similar source of prestige, and the same firstcomer rhetoric is therefore prevalent also in contemporary narratives. The significance of such claims for first comer status, and the priority rights deriving from this status, have to be regarded in close relation with the fact that, among the Woɗaaɓe, placemaking processes in the political sense of an appropriation of space historically did not go together with territorial claims. Their form of space appropriation can be characterized as a combination of zeroand one-dimensional representations, i.e. points and lines, or places and paths, as opposed to the two-dimensional conception of space as bounded territories, prevalent in farming societies (Schlee 1990b: 18f.). The Woɗaaɓe did not occupy the land in the sense of exclusive territories with property claims, but rather, land was ‘something that is moved through’ (Watson 2010: 207). In such a conception, the rights concerning the use of the shared resources were therefore a matter of constant renegotiation, and the above-described translocal social networks of pastoral Woɗaaɓe can be understood as a relational and networked, in contrast to a territorial mode of space organization (Retaillé 2013: 67; Retaillé and Walther 2012: 2f.). This practice clashed in many cases with the practice of new arrivals from a sedentary farming background, who, pushed by growing population density in the south, ventured into the pastoral zone with a strategy of resource-appropriation which was, in contrast to that of the mobile pastoralists, occupational and territorially oriented. In the long run, such processes of territorial occupation have often led to an exclusion of the formerly established pastoral populations from access to crucial resources by latecomers who followed the territorializing strategies also favoured by the state legislation. One example would be the aforementioned case of the village of Tumulle (see also Chapter 5), another the settlement history of Tesker, as related to me by one of my interlocutors (Köhler 2017a: text 3,01ff.): at first a place inhabited by Tubu hunters and herders and by Woɗaaɓe pastoralists, Tesker was later artificially transformed into a town,

50  Space, Place and Identity

through the establishment of an administrative and military outpost. Here, again, a placemaking process was initiated by groups who did not follow strategies of territorial appropriation and who were finally pushed out. The production of space through movement clashed with the control of space through borders (Retaillé and Walther 2012: 4). In Chapter 5, I will show how Woɗaaɓe today adapt their strategies in the light of such experiences by incorporating a more territorially based element into their patterns of appropriating natural resources. These tendencies have an important influence on the nature of place-making processes and, as a consequence, on social relations both internally, in terms of the constitution of social groups, and externally, in terms of inter-group relations. Before proceeding to this analysis, however, I will, in the following chapter, show how the Woɗaaɓe’s non-territorial approach to land use and the resulting need to constantly negotiate access to land resources, has also contributed to integrating them into a complex social space that they share with the other ethnic groups of the region.

Notes  1. For a transcription of the complete account, see Köhler (2017a: text 2).  2. Ɗawra Egoyi, July 2011.  3. Literally: ‘turban’. The turban was a principal insignia of office and remains so today in the case of customary chiefs. The term is thus used in the political context as a synonym for the title of a political office.  4. On the connection between the Rinderpest epizootic and the colonial conquest respectively, see Loimeier (2011).  5. The same historic event is used as a reference point in oral accounts collected by Paris (1990: 197) among Woɗaaɓe from the Agadez region, more than twenty years before my own fieldwork.  6. According to Maliki, Mbuuwa had been appointed by authorities in Sokoto, but he had chosen not to migrate (Maliki 1982: 36f.). Maliki objects that this version, which shows the perspective of the Gojanko’en Gawo, is put into question by the Gojanko’en Ute, another clan segment, who claim that Mbuuwa had only ‘lent’ the title to the Gawanko’en, but that legitimately, it belonged to them (ibid.). Loftsdóttir (2002b: 297) states that according to some versions Mbuuwa was originally from the Ader region, went to Sokoto to pledge alliance to Usman dan Fodio and decided to stay while a part of his family returned. In this version, the lineage’s migration to the Ader region looks like ‘a return to their area of attachment’ (ibid.). Consequently, Loftsdóttir puts into question the dates indicated by most of the existing sources for the first arrival of Woɗaaɓe in the territory of present-day Niger. She admits, however, that historical narratives are ideological and highly political (ibid.: 300). My own interlocutors, without being able to give any precise details, also vaguely mentioned that their ancestors had migrated to Sokoto from the north.  7. Although the term laamiiɗo for the superior chief evokes the great Fulɓe chiefs in Sokoto or Adamawa, his status and function in the context of colonial and post-colonial Niger is not comparable. Bierschenk (1993: 225) has correctly pointed out that the terms arɗo and

A History of Migrations  51

 8.  9. 10.

11.

12. 13. 14. 15.

16.

laamiiɗo should not be regarded as unchanging archetypes, but in their historical context. On the terms laamiɗo and arɗo, see also Bonfiglioli (1988: 126f). Araba Egoyi, December 2011. Araba Egoyi, December 2011. The Ɓii Ute’en were among the first Woɗaaɓe clans to migrate into the Damergou region in the 1920s. One of their leaders was appointed chef de groupement by the colonial administration in 1947 (Sambo 2008: 49; see also Dupire 1972: 25, 28; Bonfiglioli 1988: 126f.). Ngel Tireeji (‘the small [pond] of giraffes’) is a Fulfulde name and indicates not only the presence of giraffes but also of Fulɓe pastoralists in the past. Woɗaaɓe had been present in the area at least since the second decade of the twentieth century and had marked their presence by ascribing place names in Fulfulde (Paris 1990). On the significance of naming places as part of the process of placemaking, see Tilley (1994: 18), Appadurai (1995: 206), Bender (2006: 306), Schareika 2007: (207f). Abdua Ɗawra, December 2010. For other definitions and discussions of the term transhumance, see also Krader (1959), Ingold (1987), Schlee (2001a: 455f), Schlee (2005), Bonte (2006: 19), Homewood (2008). See also the similar conceptions of Massey (1994: 154), Sheller and Urry (2006: 214) and Retaillé (2010, 2013). In Woɗaaɓe animal loans (haɓɓanaaye), rights in female animals are temporarily transferred. The receiver can use the milk and keep the first calf (in the case of cows) or the first two or three lambs (in the case of sheep or goats), before returning the animal to its owner. Multiple loans establish or strengthen social relationships between giver and receiver and thus constitute important social security networks. Animal loans are widespread both within and across clans, and to a lesser degree also with members of other ethnic groups (see Scott and Gormley 1980; Bonfiglioli 1985). For a similar firstcomer rhetoric to claim priority use rights over resources, see Schareika (2007: 207).

Part II

Duuniyaaru Spaces of Social Interaction

Chapter 3

Inter-ethnic Relations The Balance of Integration and Conflict

Located between the Sahara desert in the north and the more fertile coastal regions further south, the semi-arid region of the West-African Sahel is characterized by both ethnic and socio-economic diversity. The region is a zone of cross-cultural contact (Horowitz 1972; Retaillé 1984, 1998), notably between predominantly farming and pastoralist populations, who are both, in the region, characterized by a high spatial and economic mobility (Gallais 1975; Retaillé 1984; Boesen et al. 2014; Marfaing 2014a). In depictions of inter-ethnic relations in Sahelian societies, much attention has been paid to farmer-herder conflicts (e.g. Bernus 1974; Babiker 2001; Moritz 2006), while aspects of their peaceful interaction have often not gained the same attention (see however Guichard 1996; Hagberg 1998; Dafinger and Pelican 2002; Pelican 2006; Dafinger 2013). A potential for such conflicts, which typically evolve around the question of crop damages, clearly exists in the Niger context, and the relations between farmers and mobile pastoralists are seasonally strained (see Chapter 5).1 Nevertheless, peaceful coexistence is rather the rule. Open conflict does occur, yet the time of increased tensions between farmers and pastoralists generally remains limited to a relatively short period during the agricultural season. During the rest of the year, conflict potentials are relatively low. In fact, there are a number of mechanisms and institutions that contribute to balancing the conflict potentials and mutual reluctance, prejudice and latent distrust that exists between pastoralists and farmers, and integrating them in a shared social space.

56  Space, Place and Identity

A History of Coexistence The Koutous region exemplifies on a minor scale the Sahelian characteristic of being a contact zone between different cultures and different socio-economic groups, and it can thus serve to illustrate the history of inter-ethnic coexistence in the wider region. While the Kanuri Dagra constitute the main population of the sedentary villages, many traders in the local market villages and some of the more recent settlers are of Hausa origin. The surrounding pastoral areas are inhabited by different groups of Fulɓe (south), Tuareg (north and north-west) Tubu and Arabs (north-east). Essentially, all the ethnic groups of the Eastern part of Niger are represented in the area: a veritable inter-ethnic crossroad. When the first settlers arrived, they found highly attractive conditions. While the plains were characterized by favourable water and farming conditions, the foot of the mountain plateau offered strategically advantageous defensive positions for villages (Retaillé 1984: 185f.). Kanuri settlers arrived in several waves between the late sixteenth and the beginning of the twentieth century, first from the area of the Kanem Empire and later from north-eastern Nigeria (ibid.). Before their arrival, the region had been inhabited by Tubu, who were hunters and caravan traders dominating the trade routes leading from Bilma, via Termit and Tesker to Boultoum in the Koutous region and further to the south. During the nineteenth century, Kel Aïr Tuareg warriors arrived in the region from the Agadez via the Damergou region and established themselves by forcefully subjugating the Kanuri peasants. Only after the colonial defeat did the Tuareg transform into pastoralists, breeding mainly caprines. Many from their former servant caste (Iklan, locally called Buzu) took up agriculture upon their emancipation by the French, became settled and assimilated to the Kanuri, while others continued to be pastoralists (ibid.: 189). Similarly, as in the Damergou and Agadez regions, Fulɓe pastoralists profited from the defeat of the dominating Tuareg and Tubu by the French and infiltrated into the region throughout the twentieth century. More recently, Arab groups (notably Ouled Sliman and Mohamid) arrived from the east and north-east as camel breeding pastoralists and as traders. Placemaking is a process of spatializing power relations (Gupta and Ferguson 1997c: 36) and involves winners and losers, or ‘colonizers and resisters’ (Turton 2011: 170). It is thus a matter of integration and conflict, but not necessarily violent conflict. The spatialization of groups in interaction can also find a balance based on common interests, complementarity and mutual exchange. Arguably, the history of coexistence in the Koutous region has not always been peaceful. The subjugation of the peasants by the Tuareg warrior aristocracy is the most obvious case in point. However, the groups of the region over time got involved in various forms of exchange. The mutual interest of exchanging animals for farm crops and trade goods soon led to the development of local markets that

Inter-ethnic Relations  57

also became centres of social exchange and integrated the different groups in a meta-ethnic socio-economic system (Retaillé 1984: 188). It has been argued that economic complementarity can favour peaceful coexistence between groups (Schlee 2003b: 91). Although economic specializations in the region were never fixed exclusively along ethnic lines, but have rather remained flexible over time and between groups, the different ethnic groups can roughly be associated with particular economic profiles: while Kanuri and Hausa are predominantly farmers, the Tubu are pastoralists and caravan traders. The Tuareg nobility have transformed from warriors into caprine-breeding pastoralists and their liberated servants took up mixed forms of farming and pastoralism. The Fulɓe, meanwhile, specialize in cattle and the Arabs in camel breeding. Stressing the integrative aspect of economic complementarity does not mean denying the existence of conflict potentials. However, as Dafinger and Pelican (2002: 14) have argued for Burkina Faso, low-level conflicts can also be regarded as ‘part of the process of constructing a common economic, social and political landscape’. One principal conflict potential between farmers and herders results from the fact that the different groups do not have mutually exclusive territories with fixed boundaries but use the resources of the land to different ends and in a system of seasonal alternation. Pastoral and agricultural land use in Niger are not clearly separated, but are characterized by overlapping zones of seasonally changing and multiple use with shifting spatio-temporal boundaries. After the completion of harvest, all farm land theoretically returns to the status of pastoral land with open access to pastoralists, a rule which has been reconfirmed only recently by a new pastoral law (Republic of the Niger 2010: par. 30, 34–5). The law stipulates that a date is annually fixed by the regional authorities (date de libération des champs) after which harvest must be completed and the fields opened for the herds to access the crop residues (ibid.). This particular system of land use, although it bears important conflict potentials, is also a factor of integration. The alternating use of the same spatial resources by different socio-economic groups can be characterized as a sharing system of land use (Dafinger and Pelican 2002; Dafinger 2013: 68ff.), which is characteristic of the West-African Sahel more generally (Gallais 1975; Bonte 1999: 216f.; Retaillé and Walther 2011: 87). The lack of clear-cut boundaries and the fact of a commonly shared territory can be interpreted as elements favouring exchange and interaction. Thus, Watson, (2010) drawing on the ideas of Massey and Ingold, has elaborated on the impact of space-use on inter-group relations. She argues that ‘open’ systems (Massey 2005) where different social groups are spatially ‘entangled’ (Ingold 2007) in their social, political or ritual activities, are potentially characterized by more tolerant attitudes than ‘closed’ systems based on territorial exclusivity. She contends that the former mode of engaging with space is often transformed into the latter through modernization, colonialism and globalization (Watson 2010: 202), or, to cite factors that are

58  Space, Place and Identity

of particular relevance in the study context, by processes of sedentarization, the introduction of boundaries and more exclusive territorial strategies, as they are also favoured by more recent bodies of law in Niger (see Chapter 6). Although the sharing land use system in Niger perpetuates rather than dissolves the structural conflict potentials, it does foster the interaction between groups and their integration in relations of exchange. The fact that group relations remain nevertheless to a certain degree conflictive is not a contradiction, since integration does not exclude conflict. Rather, it can also imply integration into a wider systemic whole characterized by internal conflicts (Schlee 2003b: 78). This wider systemic whole can be characterized as a meta-ethnic social space, which the ethnic groups in the study region share and which they construct through their interaction.

Institutionalized Inter-ethnic Relationships Interaction and exchange between farmers and pastoralists on a regulated and cooperative basis is further supported by certain institutionalized relations. An important example are stranger–host relations, which are widespread across regions and considered to be among the most ancient forms of inter-ethnic cooperation between farmers and herders in west-Africa (Diallo, Guichard and Schlee 2000: 226). As highly mobile populations, pastoralist groups like the Fulɓe rely upon hosts from sedentary groups, e.g. for facilitating their affairs in market villages, or for storing things away during transhumance movements. By their institutionalization across regions, these inter-ethnic relationships have importantly contributed to the integration of mobile pastoralists in newly accessed areas after migration. The term jaatigi, originally derived from the Bambara language (de Bruijn 2000: 24) and used in other Fulɓe contexts to refer to this institution (e.g. de Bruijn 2000; Diallo, Guichard and Schlee 2000: 226ff.; Boesen 2010: 42), is not known in the study group. Rather, a paraphrase such as ‘to jippoto min’ (‘where we stop over’), or the term higo (friend) is used to refer to such hosts. The use of the term ‘friend’ in this context indicates the overlapping of such formalized relations with friendship-like relationships, which in general are often difficult to clearly separate (Grätz et al. 2003; Pelican 2003: 6; Guichard 2014). Although stranger–host relationships are institutionalized and often primarily instrumental (Pelican 2003: 8), they can favour the development of individual friendship or friendship-like relationships (ibid.: Guichard 2014: 10). The monetization of such service relationships is a relatively recent phenomenon. In former times, they were rather based on reciprocal services or the exchange of products. An important form of reciprocal relations were ‘manure contracts’, cooperative arrangements based on the mutual interest of the farmers in the dung provided by the pastoralists’ animals, and of the pastoralists in crop

Inter-ethnic Relations  59

Figure 3.1  Woɗaaɓe in the court of their host’s house in the market village of Gueza (Koutous region), July 2011 (photo: F. Köhler).

residues. Such ‘manure contracts’ typically comprised the reception of pastoralists by farmers in their fields, offering a place to make camp, fodder resources for the animals and often an additional ‘payment’ for the pastoralists in the form of millet or the provision of food and water (de Bruijn 2000). It is often lamented that manure contracts have diminished or ceased to exist as farmers increasingly keep livestock themselves. In the northern part of the study area, manure contracts indeed play a minor role. Farmers sometimes even dislike the presence of herds in their harvested fields here, as they say that too much fertilizer ‘burns’ the soil (see also Sutter 1990: 342). In the south of the study area, however, new forms of such manure contracts have developed over the past decades: farmers invite the pastoralists’ herds into their fields and provide crop residues in a controlled manner and in defined places in order to have specific parts of their fields fertilized (ZFD 2008: 56). In the contemporary condition of livelihood diversification, new forms of mutuality and reciprocity emerge between the socio-economic groups: as noted above, farmers today increasingly invest in livestock. As farming demands a relative local stability during the agricultural period, and as livestock-keeping in the Sahel inversely demands a certain degree of mobility, farmers often entrust their animals to hired herders from the pastoral milieu. The contractual herding of animals that belong to others (jokkere) has thus become an important supplementary livelihood strategy among Woɗaaɓe and other pastoralists since the major Sahel droughts of the 1970s and 1980s (Maliki et al. 1984: 320; Bonfiglioli

60  Space, Place and Identity

1985: 32f.). Inversely, since many Woɗaaɓe today take up farming and become partially settled as agro-pastoralists, yet often lack skills and affinity, they tend to pay neighbouring farmers for doing the agricultural work in their place, if they have the means to do so. These relations of exchange in specialized work across ethnic groups are either based on existing inter-personal relations, or new such relationships can emerge from them as a result. A particular institution that favours communication and interaction between certain ethnic groups in the region are inter-ethnic joking relationships. Here, the Woɗaaɓe (and more generally the Fulɓe) entertain such a relationship in particular with the Kanuri, but also with butchers, i.e. with a socio-professional group. Joking relationships between Fulɓe and their respective neighbours are known from many contexts (Diallo 2006b: 193). In the case of Fulɓe and Kanuri, the institution seems to date back to the time of the Borno Empire, when Fulɓe pastoralists lived in close neighbourhood and exchange with the ancestors of the contemporary Kanuri (Dupire 1962: 324, 1970: 225; Maliki 1982: 12). Inter-group joking relationships have been interpreted in terms of alliances, pacts or contracts (Mauss 2013[1928]; Tamari 2006), aimed at bringing conflicts or hostile relationships to a term and to regulate the conditions of coexistence of different groups who use the same territory (Fay 2006: 758). Apart from joking, the relationships can involve further elements such as generalized inter-group solidarity and mutual help, which gives them a political dimension (Canut and Smith 2006; Tamari 2006). While between Kanuri and Fulɓe this political aspect seems to have been more pronounced in the past (Dupire 1962: 324), the ritualized relationship continues to function as a bond between the groups and thus favours social integration. Between Kanuri and Fulɓe, the institutionalized practice of joking is generalized. This means that any members of the respective groups, even strangers, can exchange provocations and insults with impunity. The victim of a verbal assault is expected to react passively or with a counter-insult, yet the exchange must not escalate into serious hostilities. The relationship is symmetrical in the sense that both sides claim to be the master while treating the other as inferior. Both sides try to impose their rivalling interpretations of the mutual relationship in a humorous verbal combat that produces and reproduces stereotyped ethnic images based on juxtaposition and alterity (Fouéré 2005; Canut and Smith 2006: 703). A recurrent theme are the respective roles of Kanuri as farmers and Fulɓe as pastoralists, the typical conflict potential of crop damages and the mutual blaming for them (see Köhler 2017a: text 8). As such, the joking discourse is part and parcel of a process of ethnicity and ethnic boundary maintenance (Barth 1969). The partners serve each other as constituting other in a process of identification through mutual construction of the other as a polar opposite (Fay 2006: 757). The institution thus contributes to the reproduction of difference and works against assimilation (Fouéré 2005).

Inter-ethnic Relations  61

However, the effect of joking on the relations between Kanuri and Fulɓe is ambiguous: Joking partners refer to each other as ‘joking-cousins’, which points to the similarities that exist between inter-ethnic joking behaviour and that between particular categories of kin – among the Woɗaaɓe notably between grandchildren and grandparents, and between cross-cousins. In some Fulɓe groups, the institution of inter-ethnic joking is even referred to with the same term as the relation of cross-cousins (denɗiraagal or denɗiraaku). Such a terminology characterizes the relationship between the two groups in a kinship idiom which inseparably links them in a positive way. Thus, although ethnicity is used here as a means of social construction of otherness, this otherness is at the same time put into perspective and an emphasis is put on the relation rather than the boundary. However, despite the idiom that characterizes the two ethnic groups as cross-cousins, i.e. as mutually marriageable categories, inter-marriage remains generally dismissed. As joking is rather left aside in negotiations of concrete conflict cases, the institution cannot simply be regarded as a means of conflict resolution or conflict prevention (Diallo 2006a: 177; Fay 2006: 776). However, it can contribute to the avoidance or pre-emption of serious conflict situations by offering an arena for addressing conflicting issues in verbal instead of violent confrontations (Diallo 2006a: 175). The ambivalence of serious discourse behind the shield of the humorous role play, and the controlled framework of prescribed role-­ behaviour, permit the venting of emotions and thus reduce the tensions caused by the latent potential for conflict. In Niger, the capacity of joking relationships to reinforce inter-ethnic cohesion has recently been given official recognition when UNESCO inscribed the social institution in their list of Intangible Cultural Heritage of Humanity.2 While the politicized character of such positive valuations should not be overlooked and the institution’s conflict-reducing potential not be overestimated by overly positivistic interpretations (Fouéré 2005; Diallo 2006a: 178; Canut and Smith 2006: 721f.), inter-ethnic joking definitely has a community building function in the sense that it facilitates communication and exchange between the members of the groups concerned. The ritualized relationship structures the groups’ co-existence by offering prescribed patterns of social interaction. Moreover, the institutionalized relationship on the inter-group level can function as a bridge to inter-ethnic friendship between individuals (Grätz et al. 2003: 5; 13; Guichard 2014: 9). The thesis of integration through difference (Horstmann and Schlee 2001) can thus be confirmed by the case, however in a particular framework of an alliance based on an ironical overemphasis of this difference. The case is an example of integration that does not lead to assimilation of one group into another, but where two groups are integrated in a wider heterogeneous system (Schlee 2003b: 78).

62  Space, Place and Identity

Reflexions on Coexistence Overall, the relation between the different ethnic groups in the study region can best be described as coexistence, defined by Mutie (2013) as characterized by social exchange while keeping up ethnic distinction and social distance, and by latent conflict which, however, remains controlled and on a relatively minor level. The groups live in a common social space that is jointly constructed through their interaction. Whereas rural localities (villages, pastoral camps, semi-sedentary centres) are widely ethnically defined, the spaces of interaction within and across these localities are co-produced by the different ethnic groups who together form a local, and in a wider perspective a regional, society. Although the long history of neighbourhood and of sharing the same socio-economic space favours a feeling of being part of a wider society, relatively rigid group boundaries are maintained. The complementarity of livelihood patterns seems significant for the maintenance of this status (Schlee 2003a: 57ff.). In Chapter 5, I will show how contemporary diversification strategies lead to an increasing approximation of economic profiles, which entails an increase in conflicts, as has also been observed in other regions of West Africa (e.g. Pelican 2006). However, groups continue to be categorized according to economic activities perceived as characteristic (both in the sense of mutual ascription and self-ascription), even where such categorization no longer reflects economic realities (Retaillé 1984; Hagberg 1998; Dafinger and Pelican 2002). Contrary to the findings of other studies (Retaillé 1984), the diminishing complementarity has so far not led to a significant reduction of direct economic exchange between pastoralists and farmers in the Koutous area: the relations between Woɗaaɓe and Kanuri villagers continue to be characterized by mutual economic interest, notably in animal transactions outside the market in order to avoid taxes and transaction fees, but also contract herding or small commerce of milk or butter, which gives Woɗaaɓe women opportunities for supplementary sources of income related to pastoralism. Overall, the meta-ethnic system constituted by the predominantly agrarian and pastoralist societies in the region is still widely characterized by relatively peaceful coexistence rather than strict opposition.

Notes 1. To be precise, the term ‘damage’ (F: wonere; H: bana) also covers injuries inflicted by farmers upon pastoralists’ animals, whom they regard as a constant threat to their crops and often drive away with stick beatings when they approach their fields. The Nigerien law provides penalties for such damages just as for those caused by animals in the fields (République du Niger 1987b, par. 4). 2. http://www.unesco.org/culture/ich/RL/01009, accessed 5 April 2015.

Chapter 4

A Meta-ethnic Social Space The Continuum of Identity and Difference

The Woɗaaɓe are not only an integral part of the region’s poly-ethnic society, but also of a larger macro-ethnic entity. They can be regarded as a sub-ethnicity of the Fulɓe. However, there is no unanimity among scholars regarding the question of ethnic unity across different Fulɓe groups. This is due to the pronounced differences between groups not only over vast distances, but sometimes between groups who live in immediate proximity to one another. The main internal differentiation concerns more sedentary and predominantly agro-pastoral Fulɓe groups, on the one hand, and more mobile and predominantly pastoral groups, often collectively called Mbororo, on the other. Some authors have argued that these internal differences are too pronounced, and indeed are perceived by the diverse groups themselves as too great to justify the view of the Fulɓe as a whole as one ethnic group. For instance, van Santen (2000: 132) contends with regard to northern Cameroon that sedentary and pastoral nomadic groups of Fulɓe do not consider themselves as belonging to the same ethnic group. With regard to the Woɗaaɓe in Niger, there is no agreement about the question of ethnic boundaries either: Loftsdóttir maintains a similarly exclusive view of ethnic boundaries as van Santen, stating that the Woɗaaɓe claim to be a distinct ethnic group and not part of the Fulɓe (2007: 67). Other authors treat them rather implicitly as a distinct ethnic group (Bovin 1985; Schareika 2003a, 2003b, 2007), or, despite recognizing the Woɗaaɓe’s significant emphasis on difference, refer to them as a sub-ethnicity of the Fulɓe (Dupire 1962, 1970; Boesen 2004b).

64  Space, Place and Identity

The Fulɓe and the Question of Ethnic Boundaries Different scholars have suggested putting a greater emphasis on internal differentiation within Fulɓe society instead of artificially creating an imaginary, seemingly homogenous category that does not correspond to any reality. Thus, the concept of pulaaku, a set of moral qualities and cultural values that have often been regarded as the ‘essence’ of Fulɓe culture, have been shown to have very different meanings and connotations in different contexts (Breedveld and de Bruijn 1996).1 Accordingly, these authors have stressed the importance of analysing Fulɓe cultures in their respective geographic, social and historical settings instead of generalizing cultural concepts across regions and groups (Bierschenk 1992, 1995). Amselle (1987, 1998[1990]) has further shown how ‘the Fulɓe’ as an artificially essentialized category were indeed ‘invented’ by colonial administrators and researchers. Later, this gave rise to a new macro-ethnicity, a ‘trans-regional pan-Fulbeism’ (Dafinger 2013: 19), which is today sometimes put forward in local political discourses (Guichard 1996; Bierschenk 1992, 1995). However, what is often represented as a distinct entity is in fact the result of a constant modification of political formations (Amselle 1998[1990]: 49) – in continuous processes of integration and exclusion, of assimilation and rejection, much in the sense of the above described processes of fission and fusion – which, in their sum, make the idea of a ‘Fulɓe essence’ highly questionable (ibid.: 46). Other scholars reject the deconstruction of Fulɓe ethnic identity by arguing that the notion of pulaaku is regarded as essential not merely by outside scholars, but by Fulɓe themselves (Burnham 1996: 163). They maintain the notion of a Fulɓe ethnic group and argue for the utility of sub-ethnic categories in illustrating the variety of identities within a wider whole (Diallo, Guichard and Schlee 2000: 235; Schlee 2000a). Generally, in more recent writing, the plurality of identities is stressed also by those authors who use the concept of Fulɓe sub-ethnicities (e.g. Diallo, Guichard and Schlee 2000; Schlee and Guichard 2007, 2013), and it has been argued that it should be given priority over the question of a dualistic or monistic view of Fulɓe culture (Botte and Schmitz 1994). Hopen has argued that Fulɓe society is constituted by a number of distinct ‘sub-cultures’ and that the way in which the members of any of these see themselves in relation to the members of any other depends on the social or political context (1958: 2). This perspective seems interesting as it leaves room for historical and situational variation. According to Barth (1969), ethnicity becomes manifest when a group constitutes itself along boundaries that define its identity by mechanisms of exclusion and inclusion. For the Fulɓe, the most clear-cut demarcation setting the we-group apart from the ethnic other is the dichotomy of Fulɓe vs. Haaɓe (Sg.: Kaaɗo) – a pejorative term for non-Fulɓe peasants, in the study area particularly Kanuri and Hausa.2 The ethnic boundary between Woɗaaɓe and Haaɓe, is highly

A Meta-ethnic Social Space  65

essentialized. This is quite directly expressed for instance in the way people speak about marriage preferences: Whereas marriage with members of other Woɗaaɓe clans, other Fulɓe groups and even with Tuareg is considered perfectly acceptable, marriage with Haaɓe is generally looked down upon with a considerable disdain. Fulɓe thus identify themselves largely through the construction of a polar opposite embodied in the Haaɓe category and maintained by a discourse of contrast (Riesman 1974: 118f.). In the case of the Woɗaaɓe in Niger, however, a triadic model of contrasted categories has developed as a result of historic dynamics: Woɗaaɓe vs. Haaɓe vs. Ndoovi’en, a tendentially pejorative term which is used by the Woɗaaɓe to designate most other Fulɓe groups (Bonfiglioli 1988: 63).3 Based on Barth’s reflections, Schlee (2000a: 7f.) has argued that an ethnic identity which has been defined in a given context and in a process of differentiation from neighbouring ethnic groups, will be redefined if this context changes, i.e. if a new boundary situation emerges. Such an emergence of new boundaries is not necessarily the result of new constellations of groups, but can also result from a re-interpretation of existing constellations and hence a re-definition of group boundaries, described by Elwert (2002) as the switching of identity discourses. Historical reconstructions show how social transformations in the nineteenth century led to such redefinitions of ethnic boundaries (Bonfiglioli 1988; Paris 1990). These sources present the Woɗaaɓe not just as one group of Fulɓe increasingly differentiating themselves from other Fulɓe groups, but at the same time also as the spearhead of an ideological cultural movement, almost a revitalization movement for pastoral Fulɓe values. Groups that were more involved in processes of acculturation were accused by the Woɗaaɓe of betraying the Fulɓe values of the pulaaku. This led to an increasing self-identification in declared opposition to these groups and eventually to the formation not only of the concept of Ndoovi’en, but also to the proclamation of a Woɗaaɓe-specific form of the concept of pulaaku, called mboɗangaaku4 and to the increasing perception of their own language, Woɗande, as distinct from Fulfulde (Bonfiglioli 1988: 63). These concepts formulated claims of difference and thus introduced new criteria for we-group definition: ethnic categories were redefined in a historical situation of alienation, by a re-interpretation of basic concepts of cultural identity and belonging: language, moral qualities and behavioural rules. This becomes even clearer if one takes into account a second meaning of the term pulaaku, not in the sense of a set of values, but as the community of the Fulɓe (Breedveld and de Bruijn 1996). However, the category Ndoovi’en is rather vague and not as rigid and essentializing as the differentiation between Fulɓe and Haaɓe. The use of the term seems to be based rather vaguely on perceptions of cultural difference, which often let it appear like a general term for the Fulɓe other. When I visited a (French-speaking) Boɗaaɗo in his pastoral camp, we saw a group of donkeys loaded with women, children and household goods passing by at some distance – a mobile Fulɓe

66  Space, Place and Identity

family moving camp. My interlocutor remarked with some disdain: ‘Ce sont des Peuls sédentaires’ (‘these are sedentary Fulɓe’), ignoring the apparent contradiction with their being obviously very mobile. In the course of our discussion it became apparent that ‘Peuls sédentaires’ was his French translation of Ndoovi’en. Although criteria of acculturation, notably tendencies towards sedentarization or the adoption of an agro-pastoral rather than a nomadic pastoral livelihood seem to serve as defining criteria for Ndoovi’en, the term is more of a polarizing ascription than a clearly defined category. The category Ndoovi’en as an appellation for more sedentarized Fulɓe groups recalls its counterpart Mbororo, which is used in many local contexts as an umbrella term for groups of predominantly pastoral Fulɓe.5 In the study context, however, the term Mbororo is not used in this sense. Although Bororo as an exonym is used by the Hausa and Kanuri, it generally refers, however imprecisely, to the Woɗaaɓe rather than to pastoral Fulɓe in general. As a synonym for the Woɗaaɓe, the term Bororo is still widely in use, especially in the francophone literature addressed to a more popular audience (e.g. Paris 1990; Loncke and Durou 2000). Other exonyms that locally serve as umbrella terms for mobile pastoral Fulɓe groups more generally, are Hilanin Daji (Hausa: ‘Bush Fulɓe’) or simply Hilani, or Fulani. Another reason why the category Mbororo is problematic in the context of east-central Niger is that the term Mbororo’en is indeed used by the Woɗaaɓe and other Fulɓe, yet in a quite different sense, as a designation for a specific Fulɓe clan, the Weeweɓɓe, who are recognized as historically part of the Woɗaaɓe, but today considered by the latter as belonging to a different Fulɓe section (see also Boesen 2004b: 118). The Mbororo’en (Weeweɓɓe) are generally highly mobile pastoralists and clearly do not match the category Ndoovi’en. They are respected by the Woɗaaɓe as their seniors (mawɓe amin) and their conformity with the Fulɓe values of pulaaku is undoubted, yet they are said to have ‘followed a different way’ (ɓe tokki ngo’ngol laawol’). I follow Schareika’s (2007: 32) argument that the question of ethnic boundaries should be answered individually for each given historical and local context, because the social significance of the criteria for defining different categories as ‘we’ or as ‘other’ are subject to historical change. Historically, as described in Chapter 1, group boundaries between the different (sub-)ethnic categories within the Fulɓe continuum, although clearly defined, have been all but impermeable and the interfaces between groups could be characterized as ‘gateways’ (Wood 2009) rather than boundaries.6 Woɗaaɓe identity was not essentialized but apparently highly flexible, yet conditioned by rather ideological questions like the adherence to cultural principles such as a specific form of mobile pastoral livelihood. While in northern Cameroon, processes of ‘transfulɓeization’7 – in the sense of nomadic Fulɓe assimilating to politically dominant Fulɓe groups while simultaneously becoming sedentarized – have been documented (Burnham 1996; Schlee, Diallo and Guichard 1999), the case of the Woɗaaɓe shows the

A Meta-ethnic Social Space  67

inverse: a group of Fulɓe who rejected the Fulɓe identity of more sedentarized groups (whom they accused of having assimilated to the culture of the Haaɓe) while simultaneously accentuating their pastoral mobility (Boesen 2004b: 118). In both cases, pulaaku served as an ideological concept for identification, yet was quite different in its meaning. With regard to the question of ethnic entities, the present situation must be viewed in light of the above-outlined historic processes of differentiation, and ultimately, as a still ongoing process of constant renegotiation of identity and difference, based on the recognition of a historically distant, more inclusive level of social segmentation (Burnham 1996: 12), and the since then prevailing perception of mutual otherness, respectively. Today, in the study area, a consciousness among Woɗaaɓe of sharing certain basic cultural features with other Fulɓe groups (i.e. the Fulfulde language, certain qualities and values characteristic of pulaaku) can be recognized, yet remains secondary and does not lead to any more palpable manifestations of social interaction among these groups, nor to a permanent common identification. In day-to-day interaction, ‘Woɗaaɓe’ is clearly the more relevant reference category for self-identification than ‘Fulɓe’. Actors more frequently refer to themselves as Woɗaaɓe than as a part of the wider collective of Fulɓe. The Woɗaaɓe distinguish themselves clearly from other Fulɓe groups and emphasize their difference from them (Bovin 1985; Boesen 2004b): central features of ethnic identification are either markedly different or they are perceived as such (e.g. external ethnic markers like male and female hairstyle, facial marks, dress, etc.; see Bovin 1985). Also, the Woɗaaɓe share no common ceremonies with other Fulɓe groups. The complementary distribution of central cultural practices like ngaanka dance contests (Woɗaaɓe) versus soro flagellation rituals (Ndoovi’en) equally rather accentuates the dichotomy. Even on occasions where joint celebrations across ethnic lines take place between Woɗaaɓe and Ndoovi’en on a neighbourhood basis at the local level, cultural exchange generally remains minimal and the festivities rather become occasions for mutual confirmations of difference. This becomes manifest for instance in dances, which might be performed simultaneously, yet separately and generally with an attitude of indifference for, at times even a certain mockery about the performances of the respective others (see Figure 4.1).8 The question of ethnic boundaries certainly remains debatable. In my own use of the terminology, I base myself on the reflections of Ronald Cohen (1978) who has read Barth’s (1969) definition of ethnic ascriptions against Joan Vincent’s (1974) ideas about broader or narrower definitions of ethnic group boundaries, based on political needs and interests. Barth defined ethnic ascriptions as classifying a person ‘in terms of his basic, most general identity, presumptively determined by . . . origin and background’ (1969: 13). Cohen paraphrases that in this view ethnic groups would then be ‘those widest scaled subjectively utilized modes of identification used in interactions among and between groups’ (1978:

68  Space, Place and Identity

Figure 4.1  Young Boɗaaɗo performing a mock-Fulɓe dance in front of Fulɓe Ndoovi’en – Ganatcha, September 2011 (photo: F. Köhler).

385). With reference to Vincent (1974), however, he objects that ‘[e]thnicity is not a “most general” or widest scaled identity but rather it can be narrowed or broadened in boundary terms in relation to the specific needs of political mobilization’ (Cohen 1978: 386). In fact, Vincent’s (1974: 376) argument that ‘ethnic identifications are broadened when greater political mobilization is required, and narrowed when exclusion is sought’ anticipates an important aspect of Schlee’s more recent theories on strategies of inclusion and exclusion (Schlee 2004, 2008; Schlee and Werner 1996): In social interaction, either sameness or difference can be emphasized, depending on the specific interests in a given situation, as well as broader strategic considerations (Schlee 2008: 43ff).9 This constellation, where actors are simultaneously part of different social categories and refer to them situationally, has been described by Georg Elwert with the terms ‘polytaxis’ or ‘polytactical potential’ (Elwert 2002: 39). This concept seems apt for describing the mechanisms of ethnic identification prevalent in the study context. The relations between neighbouring Fulɓe groups have been described as characterized by the dialectics of sameness and difference (Diallo, Guichard and Schlee 2000: 235), or as oscillating between identification and demarcation (van Santen 2000: 131). In the study area, Fulɓe ethnicity can best be understood as a continuum of possible identity categories, which are per se not mutually exclusive, but subsist in a taxonomic relation of greater or lesser inclusiveness (Donahoe et al. 2009: 16). The categories ‘Woɗaaɓe’ and ‘Fulɓe’ are characterized by different degrees of inclusiveness. Woɗaaɓe might, in one situation, rather refer to Woɗande and the mboɗangaaku to emphasize difference, but in another, refer rather to Fulfulde and the pulaaku. In other words, they can make ‘[a]ppeals to broader or narrower defined identities of the same kind’

A Meta-ethnic Social Space  69

(Schlee 2008: 46). Similar to what has been observed in northern Cameroon (Burnham 1996: 12), sameness among Fulɓe groups is expressed most clearly in contrast to the Haaɓe, while in direct day-to-day interaction, otherness is rather emphasized. Such situational shifting of the emphasis on identity or difference might also be the basis for statements such as the one documented by Loftsdóttir (2007) concerning the ethnic distinctness of Fulɓe and Woɗaaɓe: this extreme position may indeed be maintained by some actors in certain situations, yet rather not categorically. Say someone from the study group identifies themselves primarily as a Boɗaaɗo Gojankeejo (sg. of Woɗaaɓe Gojanko’en). In opposition to the neighbouring Gojanko’en Mbuuldi in the Damergou region they will stress their identity as a Kuskudu. Since all the Gojanko’en in the Damergou region are from the Gawanko’en segment, the distinction from the two other clan segments, the Gojanko’en Ute and Goje does not play a particular role here. The affiliation with the overarching clan cluster is also a relatively minor concern and instead plays a role primarily in inter-clan meetings (ngaanka). Supposing, this individual is a grandchild of Usman, they can define themselves in opposition to the descendants of Buuyo (Ɓiɓɓe Buuyo) or the descendants of Egoyi (Ɓiɓɓe Egoyi) while at the same time stressing that they are all one (minon fuu min go’o). On these different levels within Woɗaaɓe society, the mechanisms of identification roughly follow the logic of segmentary opposition within the lineage system. However, mechanisms of situational identification along categories of levels above the ethnic group, both within and beyond the Fulɓe-category, also exist. Commonalities with other Fulɓe groups are recognized and can situationally be put forward. One of the most impressive examples for a situation in which Woɗaaɓe and other Fulɓe groups in Niger focus on their shared identity in order to achieve common goals is a mechanism called daangol pulaaku. This term literally means ‘the calf rope of pulaaku’ and thus refers to a cultural symbol of the Fulɓe.10 When the state of daangol pulaaku is declared, a community or an institution, e.g. a local livestock-market, becomes the object of a general boycott by the entire local Fulɓe population (PSSP 2009: 54ff.; Schareika 2007: 208). With this tool, the Fulɓe groups make strategic use of their economic weight as livestock producers in order to defend their common interests as pastoralists.11 Identification is situationally shifted from a lower, more exclusive taxonomic level that is generally more relevant in day-to-day interaction, to a more general and more inclusive one. Such collective action however remains temporary and situationally bounded, and does not lead to permanent social integration. The mechanism that allows for collective solidarity across Fulɓe groups under the banner of pulaaku resembles the logic of the segmentary lineage model, which has similarly been defined as ‘a social means of temporary consolidation . . . for concerted external action’ (Sahlins 1961: 342). In both cases, social groups situationally consolidate themselves by referring to a dimension of identification that

70  Space, Place and Identity

is of a sufficient degree of inclusiveness to comprise all of them. The difference to the lineage model is that identification in this case is not based on descent, or descent only in the widest sense, referring to a distant common historical origin. This could be regarded as an extension of lineage logics above the level of lineages in the proper sense; and one might be tempted to regard it as an argument for defining the ethnic group more inclusively as comprising all the Fulɓe groups of the region. Ultimately, however, the functioning of the mechanism on this level is also of little use as a criterion for defining ethnic group boundaries because consolidation for concerted action can also function across ethnic lines above the Fulɓe category.

A Continuum of Identity Categories Potentially, other pastoralist groups in the region, notably the Tuareg, can have similar interests to the Fulɓe with regard to their relations to the farming populations. Pastoralists’ associations in Niger today make attempts to mobilize large assemblies across ethnic lines under the umbrella of their common socio-professional identity as pastoralists. The organization of thematic panels typically includes cultural manifestations of different ethnic groups to celebrate the diversity and yet emphasize commonalities. This does not mean that the identity category of ‘pastoralists’ is artificially constructed. Rather, this category is used by the actors themselves to emphasize their commonalities. An argument such as ‘minon fuu min waynaaɓe’ (we are all pastoralists) is frequently heard, expressing the wish to decrease tensions in an acute conflict. It is more likely for the different Fulɓe groups to consolidate themselves situationally as a solidary group than for Fulɓe and other pastoralist groups. This is due to the shared respect for the values of pulaaku that constitutes a uniting element. In the relations between Fulɓe and other pastoralist groups, such a uniting element in the form of specific, shared cultural values is lacking and it is therefore often more difficult to put the widespread perceptions of cultural distance and mutual reluctance aside. In cases of conflict among different pastoralist groups, however, other dimensions of identity can be referred to in order to decrease tensions – notably religion. As the smallest common denominator, so to speak, Islam can be made appeal to as a common identity reference. A case of conflict resolution that I witnessed in 2006 in the region of Diffa in eastern Niger can serve as an example to illustrate this point. The conflict concerned the question of access to a pastoral well that was managed by a group of Woɗaaɓe Ɓii Ute’en. A group of Mohamid Arabs12 on their passage with a huge herd of camels demanded to water their animals, which put the local Woɗaaɓe into an uncomfortable situation: according to customary law, they could not refuse the strangers access to water (Thébaud 1990: 17), however the sheer number of camels would have threatened to surpass the limited capacities

A Meta-ethnic Social Space  71

of the well. In the heated discussions, a Woɗaaɓe elder repeatedly tried to calm the opponents by stressing that they were all Muslims and as such brothers who should not fight among each other. The conflict was finally resolved peacefully by way of a compromise: the Arabs were not denied access to the well, but they agreed to limit their consumption to a minimum and pass on quickly to the near Komadugu river for proper watering of their herds. The mechanisms of situational identification can thus function in a de-escalating way as a means of conflict prevention and conflict resolution. The overlapping identity references allow for different degrees of mobilization of different social groups in the poly-ethnic society. Islam, as a commonly shared feature of identification, across basically all ethnic boundaries, can potentially even bridge the gap between Fulɓe and Haaɓe. This aspect once more points to the importance of analysing ethnic categories in their historic and geographic context: in Niger, Islam is, although with substantial variations in the degree of observance and practice, a general part of people’s identities across ethnic lines and thus a common denominator. In other regional and historical contexts such as nineteenth-century northern Cameroon, however, Islam has been precisely the opposite – an important criterion of distinction and a marker of political hierarchy and prestige. Here, the Fulɓe were identified as Muslims (juulɓe) and the Haaɓe as ‘pagans’ (Burnham 1996: 48ff.). In Niger, on the contrary, the Islamic identity of the Haaɓe (Hausa, Kanuri) is beyond any doubt. Neither is Islam a distinguishing aspect of Fulɓe identity here, nor is religious difference an intrinsic part of the distinction between Haaɓe and Fulɓe. However, the commonalities concerning religion are situationally put forward rather than generally being perceived as such. Although virtually all Woɗaaɓe would publicly acknowledge being Muslims, many non-Fulɓe and even many sedentary Fulɓe would rather place the Islamic identity of the Woɗaaɓe into question. For the very reason that everybody is nominally Muslim, even small differences in religious practice can be at the basis of stereotypical perceptions of others as different. Not religion as such, but assumptions about the religious practice and observance of respective others, can become criteria for differentiation, distinction and demarcation. In fact, many pastoral Woɗaaɓe, young people in particular, do not pray regularly. Although there is a tendency for both men and women to become more inclined towards religious practice with age, there are also elders who mock their more pious age-mates. A major issue on which the more Islamized populations of the region base their critical assessment of the Woɗaaɓe’s religious status is the institutionalized practice of marriage by elopement of already married women with men from other clans (te’egal), which is in obvious contradiction of Islamic practice, but which most Woɗaaɓe today hold on to as a central cultural element (see Chapter 11). While Islamic religion in Niger is neither a matter of particular prestige, nor associated with particular ethnic or status groups, it is, however, a cultural standard

72  Space, Place and Identity

à  Increasing degree of inclusiveness  à

Table 4.1  Identity categories and the socio-cultural dimensions of identification. Societal level

Identity categories

Dimension of identification

Muslim society

Arab, all Fulɓe groups, Hausa, Kanuri, Tuareg

religion (Islam)

(predominantly) pastoralist groups

Arab, pastoral Fulɓe groups, Tuareg

socio-economic specialization

pan-ethnic group (macro-ethnicity)

all Fulɓe groups

socio-linguistic (Fulfulde) and cultural (the values of pulaaku) criteria

ethnic group: network of inter-clan alliances (ngaanka)

Woɗaaɓe

the values of mboɗangaaku, institutionalization of marriage by elopement (te’egal) between clans

clan cluster

Alijam, Degereewol

socio-political affiliation based on putative descent

clan

Ɓii Ute’en, Gojanko’en, Jiijiiru,

putative descent

maximal lineage

Kuskudu, Mbuuldi

descent

minimal lineage

the descendants of Egoyi, Buuyo, Usman, and others

descent

and others

which is given a considerable importance by the social elites. In ­consequence, the prevailing image of the Woɗaaɓe as not being proper Muslims tends to put them into a position of social marginalization. And yet, despite the potentially stigmatizing aspect of variations in religious practice, they also bear an important potential for integration between the Woɗaaɓe and neighbouring populations: within the study group, there are no religious specialists, and in general, they are rare if not absent among pastoral Woɗaaɓe. However, the involvement of an Islamic specialist (maalamiijo) in domestic rituals like marriages and name-giving ceremonies is today generally considered desirable. Therefore, specialists from neighbouring villages are often invited to lead collective prayers and carry out the prescribed sacrifices. Such a reliance on religious specialists from other ethnic groups favours social exchange and the maintenance of inter-ethnic relations, and can thus be regarded as a factor of integration. As argued above, the Woɗaaɓe are an integrated part of a meta-ethnic social space in the study region. Their position therein can be described as oscillating between integration and participation, on the one hand, and differentiation, on the other, according to situationally varying interests.14 Just as with the level of inter-ethnic group relations, the intra-ethnic level also offers potentials for groups to define themselves as similar to or different from each other on the basis of overlapping categories and dimensions of identification. The identity

A Meta-ethnic Social Space  73

categories within the Woɗaaɓe ethnic category, which are based on kinship, and those above and across ethnic lines, which are based on other dimensions of identification (cultural values, socio-economic specialization, religion), together form a continuum (see Table 4.1). Their level of inclusiveness or exclusiveness is based on perceived similarities or differences, which can, however, situationally be re-interpreted and re-negotiated: social space can be widened or narrowed down. Boundaries are principally maintained between all ethnic groups, yet they are of different degrees of permeability. The continuum is characterized on one side by an increasingly strong identification and on the other by an increasing tendency to essentialize differences and to maintain rigid identity boundaries. The options for identification as either the same or different thus favour integration without leading to a significant degree of assimilation among groups. Despite a long history of co-residence, difference is emphasized, maintained and continuously reproduced.

Notes   1. For discussions of the concept of pulaaku, see Stenning (1959: 55f.); Dupire (1970: 189; 1981: 169); VerEecke (1993: 146f.); Schareika (2007, 2010a).   2. The term literally signifies ‘the bitter ones’ (Boesen 1997: 27; Stenning 1959: 57). In the study area, the concept does not apply to non-Fulɓe in general: Members of other pastoralist groups like Tubu, Tuareg or Arabs do not qualify as Haaɓe.   3. Some authors use Farfaru instead of Ndoovi’en (Dupire 1962; Mohamadou 1969: 61f.; Bovin 1985; Boesen 2004b; Loftsdóttir 2000, 2007). In the study group, the term is not used.   4. The term mboɗangaaku is today notably used by the Woɗaaɓe in central Niger (Dupire 1962: 296; Paris 1990: 196f.). In eastern Niger (Schareika 2007, 2010a) and in the study region, the term pulaaku prevails and it alternates, in some contexts, with the widely congruent term fulfulde, in the sense of adequate conduct. In order to refer to the specificity of the Woɗaaɓe values and qualities as against the more general Fulɓe qualities, mboɗangansi rather than mboɗangaaku is used in the study group (see also Loftsdóttir 2000 for central Niger).   5. While the Mbororo category is often understood as a Fulɓe sub-ethnicity (e.g. Diallo, Guichard and Schlee 2000), Burnham (1996) distinguishes ‘Fulɓe’ and ‘Mbororo(’en)’ as distinct ethnic categories and emphasizes the internal differentiation of the Fulɓe society in northern Cameroon, while using the term ‘Fulani’ in a more inclusive sense to refer to ‘the shared traditions of common cultural and racial origin’ (Burnham 1996: 12). The Woɗaaɓe, who are also part of the ‘Fulani’ society in northern Cameroon, are qualified by Burnham as a sub-ethnicity within the ethnic category ‘Mbororo’ (ibid.: 99). Burnham’s use of the term ‘sub-ethnic groups’ thus aims at highlighting the heterogeneity within the superordinate categories ‘Fulɓe’ and ‘Mbororo’. However, his distinction between ‘ethnic’ and ‘sub-ethnic’ labels is not altogether consistent: while terminologically subdividing the Fulɓe populations of the region into separate ‘ethnic categories’, he later on argues in favour of conceptualizing the Fulɓe as a whole as an ethnic group (ibid.: 158). Similarly, Schlee has used the terms Fulɓe-ethnicities and sub-ethnicities interchangeably (e.g. Schlee 1997: 17; 2011; Schlee and Guichard 2013)

74  Space, Place and Identity   6. On the flexibility of social identifications in African pre-colonial societies, see also Kopytoff (1987), Amselle (1998[1990]), Ranger (1983), Burnham (1996: 159), Hagberg (1998: 72f).   7. Although the term ‘fulɓeization’ suggests processes of assimilation of non-Fulɓe to a Fulɓe group (Schultz 1984; VerEecke 1993), it has equally been used to refer to processes of identity change between different Fulɓe groups (Burnham 1996). This is due to the fact that in northern Cameroon the term Fulɓe is used in two different senses, on the one hand referring in a more inclusive way to the ensemble of Fulfulde speaking groups, on the other hand referring exclusively to the sedentary Islamic Fulɓe elites, while excluding the pastoral Mbororo. In order to describe assimilation processes between Fulɓe groups more aptly, the term ‘transfulɓeization’ has been proposed (Schlee, Diallo and Guichard 1999: 4; Schlee and Guichard 2013: 36). An important reason why processes of assimilation of non-Fulɓe to Fulɓe do not play a role in the study region is that in contrast to Cameroon, where the identity change both comprises a change in religious identity and entails significant changes in status or prestige, in Niger this is not the case. While the Fulɓe were the politically dominant group in Cameroon at the time of the Fulɓe emirate of Adamawa, in Niger they did not play a comparable role but infiltrated gradually as pastoralists and agro-pastoralists.   8. On dance as a means of affirming ethnic boundaries, see Bovin (1974/5: 469).   9. With regard to Fulɓe–Haaɓe relations, see Hagberg (1998: 72f). 10. On the symbolic content of the calf rope, see Stenning (1959: 122f.), Schareika (2003a). For the symbolic content of ropes or cords in connection with the pulaaku concept, see Schareika (2007: 320ff.). 11. On the importance of pastoralist livestock production for the Nigerien economy, see Bayard et al. (2001), ZFD (2008: 11f.). 12. The Mohamid, a group of camel breeding Arabs, have migrated into eastern Niger since the 1980s due to political violence and persecutions in Chad (Anderson and Monimart 2008/9a: 3). As relatively recent immigrants, they have a reputation of disregarding established local rules for the use of common resources and sometimes violently enforcing access to wells. 13. See Chapter 10. 14. See also Boesen (1998: 234) for Fulɓe in Benin.

Part III

Ladde Transformations in the Pastoral Realm

Chapter 5

From Nomadic Pastoralism to Sedentarization and Economic Diversification

Throughout the twentieth century, the Gojanko’en Kuskudu have roughly corresponded to the picture drawn in most texts on the Woɗaaɓe: they were predominantly mobile pastoralists specialized in the breeding of zebu cattle, and only in times of economic crisis did they temporarily take up agriculture as a fallback activity. Since 1984, urban migration has played a far greater role in terms of economic diversification than agriculture, and pure pastoralism has remained the – however seldom realized – economic ideal to this day. In recent years, however, a trend towards stronger local stabilization, and indeed territorial appropriation, has become more manifest.1 Local communities have established a relatively permanent presence in the vicinity of pastoral wells. Within the study group, there are three individually owned wells in the Damergou region around which such local communities have formed: (1) one in Salaga, owned by the sons of Buuyo; (2) one in Yalema Intrika, about 4km south of Salaga, which used to belong to Ɗawra Egoyi and which he sold to his brother Boyi upon leaving for Ganatcha; (3) one in Ngel Tireeji, owned by the sons of Usman. In the Koutous region, near the Kanuri village of Ganatcha, the Gojanko’en form a single community around two wells located in immediate proximity (see Map 5.1). The process of sedentarization is the most pronounced in the Ganatcha community, the group that Ɗawra joined with his sons and their families upon their recent migration from the Damergou to the Koutous region. The first part of this chapter focuses on this community and gives an image of their recent spatio-economic strategies. Although the three wells in the Damergou region similarly serve as points of local fixation, important parts of the respective communities continue to be more mobile and undertake longer pastoral movements.

78  Space, Place and Identity

Map 5.1  The main field sites and the northern limit of farming.

The Local Woɗaaɓe Community in Ganatcha The local Woɗaaɓe community that has formed in the Koutous region near the Kanuri village of Ganatcha basically consists of three sub-groups that live in close co-residence. The first subgroup, in the chronological order of arrival, is a lineage segment of Gojanko’en Kuskudu under the leadership of Arɗo Umaru, a patrilateral parallel third cousin and matrilateral second cousin of Ɗawra (see

Nomadic Pastoralism to Sedentarization  79

Figure 5.1). Umaru grew up in the area of Tesker, about 200km east of Tanout, to where his father Siddi had once migrated. Siddi was also part of the migration group that left the Ader region towards Agadez, yet he did not stay long but continued eastward, via the Damergou region, to Tesker (see Köhler 2017a: text 3). After the severe drought of 1984–5, which he spent, like many other members of the study group, in the western outskirts of the town of Zinder, Umaru did not return to Tesker but went to the area north of Damagaram Takkaya, at the western fringes of the Koutous mountains. He remained rather mobile until the second half of the 1990s, when he finally became more definitively attached to the area between the Kanuri villages of Ganatcha, Gongoboul and Gueza. He obtained land rights for cultivation and settlement from the chef de canton in Moa, the main town of the district. At first, access to water was difficult because the relations with the farmers of Ganatcha were strained and the herds had to be watered at wells in the nearby villages. Just after the year 2000, a cemented pastoral well (Ganatcha I) was built by the government about 5km west of the village of Ganatcha, and Umaru was appointed the local representative in charge. The second sub-group that constitutes the local community of Ganatcha is a lineage segment of Woɗaaɓe Yaamanko’en, an Alijam clan, under the leadership of Arɗo Ali. Previously, his habitual rainy-season pastures had been in the area north of Ganatcha, then he spent many years in northern Nigeria but returned to Ganatcha after the construction of the well. The third sub-group, finally, is that of Ɗawra Egoyi and his three adult sons with their families, who arrived only in 2004, after having constructed, since 2002, their own well (Ganatcha II) in close proximity to the well managed by Umaru. The local community is called wuumre, the same term that designates migratory groups in the mobile pastoral setting. The three sub-groups each have their own, independent duɗal, the unit of humans with their jointly managed herds (Maliki et al 1984: 293; Bonfiglioli 1988). The wuumre is thus generally composed of several duɗe (pl. of duɗal). Ɗawra’s duɗal comprises seven households, together numbering 44 persons and 40 bovines, comprising 7 milk cows, 7 camels and approximately 100 heads of smallstock.2 The whole Woɗaaɓe community of Ganatcha comprised roughly 150 persons when fieldwork was conducted, yet exact numbers are fluctuating due to the still significant mobility of parts of the population. Located at the foot of the Koutous mountains on a gently sloping plain, the pastoral reserve of Ganatcha is a thorn savannah characterized by a relatively dense vegetation of bushes and shrubs, intensifying in the depressions and along the temporary riverbeds that cross the area. Ɗawra’s well is situated in such a densely covered spot. Some large Acacia radiana trees offer generous shadow which serves the men as a meeting place to spend the afternoon hours after the work at the well is done. The mobile homesteads are generally located a bit further uphill to the north, moving about regularly but within distances of

80  Space, Place and Identity

Figure 5.1  The two leaders of the Gojanko’en community in Ganatcha, Arɗo Ɗawra (left) and Arɗo Umaru (right), Ganatcha, 2011 (photo: F. Köhler).

generally less than one kilometre. A more extended household mobility has been given up for different reasons. One was the construction of a community school in 2008, another the fear that extended periods of absence from the well might encourage other groups of pastoralists to settle in the area and appropriate the pastoral resources. At the time of fieldwork, permanently fixed dwellings had not yet been constructed and the only form of habitation in use was a round, tent-like shelter called tukuru. Camp relocations occurred in variable intervals of generally not more than four weeks, less out of pastoral considerations than because of the seasonal advantages and disadvantages that different terrains offer. The relative density of the vegetation on the foot of the slope provides more protection against the cold wind during winter months, while this area could not be used in the rainy season when the riverbeds temporarily transform into streams. In the hot season, by contrast, open spaces offer a cool breeze at night. Another reason why mobile tents are still preferred over permanent houses is hygiene. After some weeks, the surroundings of the homesteads become polluted with dung and faeces, and it is preferable for humans and animals to move on and allow the site some time for regeneration (see also Benfoughal and Boulay 2006: 17; Watson 2010: 207).3 Sedentarization is thus an ongoing process in the Ganatcha community. Arguing against romanticizing views of nomadic mobility, Dupire stressed that the Fulɓe regard it as a necessary submission to environmental conditions

Nomadic Pastoralism to Sedentarization  81

rather than a value in itself (1972: 52). In keeping with the law of least effort, mobility is readily reduced if sufficiently good pasture conditions allow for constant grazing in a limited area around a water point (1962: 79). Sedentarization processes among pastoralists can thus in fact be caused not only by impoverishment but also by great wealth (Salzman 1980; Azarya 1993). Many of the Woɗaaɓe in Ganatcha express their satisfaction with the reduction of mobility. Unsurprisingly, it is especially the women who welcome a more sedentary lifestyle, as they bear most of the hardship of pastoral camp relocations: all activities in relation to the deconstruction, packing up and reconstruction of the mobile houses are exclusively female tasks (see Köhler 2016). The local fixation has gone hand in hand with economic diversification. Cultivation of millet and beans was taken up by many as a supplementary livelihood activity next to pastoralism. The regular labour demand during the farming season is another factor limiting household mobility. Although a common practice, agriculture is not enthusiastically embraced. Rather, it is rationally seen as a necessary subsidiary activity and many, if they have the means, would rather pay a Kanuri farmer to do the farm work for them. However, a generally negative attitude towards farm work, such as has been described by other authors (Bovin 1990 38f.; Loftsdóttir 2004: 56), cannot be confirmed for the study group. Rather, farming is increasingly perceived as a normal livelihood activity. Most farming Woɗaaɓe cover only a small part of their consumption needs in cereals with their own production, while the bulk has to be bought in the market. Farming is only one among other, sometimes more important, subsidiary activities next to pastoralism, which generally remains central. It is arguable whether the term ‘agro-pastoralists’ can reasonably be applied here or whether the amount of crop production is too small.4 Ironic self-mockery about one’s own lack of farming skills is often heard. The pastoral economy has also been diversified. The breeding of small stock, in particular caprines, has been intensified. The pastoral zone of Ganatcha is well suited for the grazing needs of caprines, who feed well on the leaves of bushes. They multiply well and thus contribute significantly to the family economy. This is one reason why their husbandry has been developed here, together with occasional camels that favour a similar diet yet are kept as prestigious riding animals rather than being raised in numbers. For the cattle and sheep that mainly need pastures of grass and herbs, adequate pastures are not sufficient in the immediate surroundings. For this reason, the Woɗaaɓe in Ganatcha have adopted a system of separation of their different herds. While the cattle and sheep are driven to more distant pastures depending on the seasonal conditions, the goats are always kept in the surrounding of the main camps. The herd relocations to the open rangelands of the northern pastoral zone during the rainy season are primarily motivated by nutritional considerations, but

82  Space, Place and Identity

Map 5.2  Seasonal transhumance movements of the Gojanko’en in the Koutous region.

another aspect that makes them imperative is the proximity of the main camps to the farms of the neighbouring Kanuri villages. If the rains start belatedly in the north and the herds are forced to stay near the farms, close supervision to avoid crop damages in the newly sown fields is necessary even at night – a task that is regarded as highly demanding. The goats, by contrast, are tied to ropes and can be kept in kraals at night. Pastoralism has thus changed its character from a nomadic to a semi-­ sedentary mode, characterized by a separation of the herds, or parts of the herds, which remain relatively mobile, from the bulk of the households, which become spatially more stable. A system of associated mobility, in which herds and households move together, has given way to one of dissociated mobility, whereby they are separated (Anderson and Monimart 2008/9b: 65f.). The Fulfulde term for this system is biggal. It is based on the principle that generally only a few herders accompany the cattle, while the households remain either stationary or follow a

Nomadic Pastoralism to Sedentarization  83

different itinerary with other parts of the stock in cases where different species with diverging grazing needs are raised. Herds and households form a unit at the beginning of the rainy season, around June or July, and again around mid-November to December, in order to wait for the harvest activities in the south to be concluded. As soon as fresh pastures are available in the northern pastoral zone between Adjeri and Zabetan, the cattle and sheep are moved there to allow for optimum grazing. Only a few lactating cows are then kept in the main camps with their calves to guarantee a minimum supply of milk for the households. The herds are accompanied by two or three young men (biggankooɓe, sg. biggankeejo) and normally one woman with her household, forming a highly mobile satellite camp. The itinerary and duration of the rainy season transhumance varies from year to year as it is determined by the highly unstable distribution of pastures. The time of fresh pastures is short, typically lasting no longer than three months, but the dried-up herbs remain a good source of nutrition. The herds stay in the north as long as the availability of surface water permits easy watering and then gradually return within the vicinity of the well. They keep up shorter distance movements and return to the well for watering every two days. If the pastures permit, the herds are kept in walking distance from the main camps to allow for a better milk supply of the households. When harvest activities in the area south of Ganatcha are advanced around November, the animals eventually return towards the main camps, waiting for the sorghum and millet fields around Rafa, Moa and Damagaram Takkaya to be opened for grazing. As soon as this is the case, the herds are driven south in order to feed on the crop residues, mainly stalks and leaves that the farmers have left behind. This source of alimentation has for long been an important component of the herds’ nutritional cycle. Even lactating cows are now driven south to make them fit for the dry season. The southward movements after harvest again demand a dissociation of herds and households. Generally, significant herd concentrations are encountered in the area during this period, and a pronounced competition over the best spots occurs, resulting in a temporally extremely high herd mobility. During this period, the biggal-herders are often no longer accompanied even by a single household unit. However, they are equipped with cooking pots and a stock of millet and rice, and they prepare staple food in addition to feeding on the milk of their cattle. The herders keep at least one mounting camel with them and every two or three days, depending on the distance to the main camps, one of them returns home with a supply of milk for the family. Gradually, the herds will return closer to the homesteads and the system of one or two household units accompanying the joint herd is taken up again. A closer contact is again maintained with shortening distances between the main camps and the herds.

84  Space, Place and Identity

To Migrate or Not to Migrate – The Decision-Making Process In May 2011, I visited Yalema Intrika, the well owned by Ɗawra’s younger brother Boyi. Incidentally, two young men from the Ganatcha community were at the same time present for a visit, and I was able to witness a long discussion between Boyi and the two visitors about the potentials of the two sites. Boyi did not know the Koutous region yet was interested to learn more about its potentials and bombarded the young men with questions. It became clear that not only Boyi, but other Kuskudu family heads as well were considering the possibility of joining Ɗawra in Ganatcha. However, there was also some reluctance. This debate inspired me to investigate in more detail the process of decision-making about migration movements. The analysis of the arguments brought forth by different interlocutors for and against a possible migration to Ganatcha readily shows that any monocausal explanation for migrations is problematic. Although pastoral considerations are undeniably crucial for explaining the movements of herds and households, the dynamic and complex patterns of mobility and the social reconfigurations they entail cannot easily be reduced to ecological factors alone, but equally have to take political and social considerations into account (Gulliver 1975; Burnham 1979). Most classic texts on the Woɗaaɓe identify the quality of pasture resources as the most important consideration for decisions about migration (Maliki 1981; Bonfiglioli 1988, 1998; Schareika 2003a, 2003b; Krätli 2007, 2008). This is still valid today. A main argument brought forth by the Woɗaaɓe of the Damergou region against the area of Ganatcha is that the pastures are insufficient for assuring satisfactory cattle fertility. While the vegetal resources in Ganatcha are favourable for breeding goats and camels, the cows and sheep cannot stay in place but seasonally have to move further north where the pastures better suit their needs. Studies of the nutritional value of herbs in different areas of the Sahel have indeed shown that the northern pastures in areas with an average rainfall of less than 300mm are considerably richer in protein than those in the south (Bonfiglioli 1988: 127f.; Paris 1990: 200; Sallah 1999: 265). In this regard, the pastures north of the Koutous are considered to be of equally good quality as those in the Damergou region. The biggal-system is designed to compensate for the insufficient pasture conditions for cattle around the wells of Ganatcha, but the practice is a compromise insofar as it entails an absence of the cattle, including most of the lactating cows and thus of the daily milk supplies, during a significant part of the year. This clearly has an undesirable effect on household nutrition. An argument in favour of Ganatcha is that the vegetation of bushes and trees is relatively rich, whereas around Salaga, the massive cutting of trees by professional fire-wood vendors has significantly reduced the ligneous vegetation over the course of the past decades. The Woɗaaɓe of the Damergou region, conscious

Nomadic Pastoralism to Sedentarization  85

of the problem of erosion caused by deforestation, foresee that in a few years, the area will be transformed by desertification. They say that ‘the bush has died’ (ladde waati). Trees and bushes are essential not only as an additional nutritional resource for livestock (Sallah 1999: 269), but also as protection against sun and wind. Another important aspect is the respective geographic location of the two sites. While Salaga is situated north of the official northern limit of cultivation, Ganatcha is located below this limit. In Niger, the agro-pastoral and the pastoral zone are legally separated by a line that roughly follows the latitude 15°10’ N (Bernus 1974: 141; Bonfiglioli 1988: 117; Retaillé 2006: 220f.; 2013: 56; see Map 5.1). A resolution from 1954, turned into a law in 1961 and confirmed by a new pastoral legislation from 2010, defines the zone north of this line as pastoral land and officially declares the creation of new cultivation-sites above this limit illegal (Republic of the Niger 1954, 1961, 2010). Theoretically, this line fixes a strict northern limit for agriculture. In practice, however, the principle of the pastoral/agro-pastoral divide has never been strictly applied (Boesen 2004b: 105; Mansour and Tan 2008: 4). Due to demographic pressure in the south, farmers have long advanced into the pastoral zone (Arzika et al. 2007: 9f.; ZFD 2011: 30), and over the past decades the agricultural front has considerably moved north (Hammel 2001: 3; Arzika et al. 2007: 10). In Salaga too, there are sedentary villages with farming populations nearby, and some of the Woɗaaɓe have themselves taken up agriculture to a varying degree. With regard to the question of conflicts about crop damages, what difference does the location of a site either in the agro-pastoral or in the pastoral zone make? Theoretically, the legal status of the pastoral zone should exclude the possibility of crop damages here altogether. Where agriculture is prohibited, damages in fields should logically not entitle claims for compensation payments. However, since the agricultural front has de facto long transgressed the legal limit, the same rules are applied in the north as in the south: once transformed by cultivation, the land becomes in practice recognized as farmland and crop damages are handled accordingly. The new pastoral law from 2010 even clearly states that the status quo of the already existing agricultural surfaces in the pastoral zone can be maintained, hence, that they are ex post facto legalized (Republic of the Niger 2010, par. 9). Against this background, the move from Salaga into the agro-pastoral zone of Ganatcha did not make a great difference, since the more recent developments in the area of Salaga had de facto also transformed it into an agro-pastoral area. However, one important difference is that in Salaga open pastoral spaces are situated immediately to the north and the density of fields around the site is lesser. The site of Ganatcha, on the contrary, is more or less surrounded by fields, and crop damages, which are not a particular concern in Salaga, here constitute a major source of conflicts with the neighbouring farmers. During the agricultural

86  Space, Place and Identity

period, the relationship between Kanuri farmers and Woɗaaɓe is thus rather strained. Although the herds are driven to the northern pastures, among other reasons, to avoid crop damages, such damages nevertheless regularly occur. In the surroundings of the main camps, it is not the cattle, but rather the goats and donkeys that cause damages when they are insufficiently attached at night. The southward movements after harvest into the agricultural areas of Moa, Rafa and Damagaram Takkaya regularly cause clashes between the biggal herders and farmers, and increasingly also with armed forces that are mobilized by the district chiefs – officially to protect the harvest activities, yet with often arbitrary and legally questionable measures, restricting the mobility of the pastoralists even when no damages have occurred (see also Arzika et al. 2007: 32). A major problem for pastoralists is the way conflicts about crop damages are handled. State law provides legal procedures for compensating damages, but in practice they are often not applied. Instead of measuring the actual damage, generalized fees are often raised by local authorities according to the numbers of animals that have presumably entered into a field, independent of the damage they have actually caused. The civil jurisdiction is generally in the hands of the local authorities (chefs de village), which means that the farmer in whose fields the damage has occurred is normally represented by his own village head, while mobile pastoralists during transhumance often do not have their representatives nearby and thus often do not enjoy a fair legal assistance. Even clear cases of abuse of power by authorities occur frequently (PSSP 2009: 50). But the closer and more permanent neighbourhood relations that result from the process of local stabilization also facilitate the resolution of conflicts. Interlocutors have pointed out that knowing the field-owners personally and being engaged in day-to-day relations makes negotiations easier than they would be between strangers (Köhler 2017a: text 9; see also Dafinger and Pelican 2002: 13). As the economic use of pasture land depends on access to water (Thébaud 1990; Moutari 1999: 429), the conditions of the latter are another important criterion for evaluating the potential of a site. Although the Woɗaaɓe communities in both Salaga and Ganatcha are in possession of their own pastoral wells, the regular herd relocations necessitate the use of different water points. Pastoralists thus generally rely on a network of wells for always ensuring good access to water (Thébaud 1990: 18). Ɗawra’s brother Boyi in the Damergou region, for instance, said he would join the Koutous faction, if he was able to buy or construct a pastoral well in the area north of Ganatcha, which he considers very favourable. Boyi’s concern is that he does not want to rely on other well owners for watering his herds on the northern pastures. In the Damergou region, conflicts with Tuareg about the access to wells are a recurrent problem. As explained in Chapter 3, Tuareg pastoralists had inhabited the region before the Fulɓe, who profited from the defeat of the

Nomadic Pastoralism to Sedentarization  87

former by the French and infiltrated the region in several waves during the course of the twentieth century. In contrast to the Woɗaaɓe, the Tuareg had since long followed a more territorial strategy of space appropriation with a more exclusive understanding of use rights over wells and pastures (Sallah 1999: 267f.). Since the colonial transition, the struggle over the control of wells and other water points has been a way of acting out power relations. Although disrupted by the colonial forces and later by state legislation determining the status of public wells, firstcomer claims of the Tuareg seem to play a central role in these conflicts over the access to wells and pasture, which have been a recurrent issue between the Tuareg and Fulɓe since colonial times, occasionally leading to violent clashes (Bernus 1975: 70; White 1987: 242). Conflicts over the access to wells and to pasture land are intrinsically linked, because denial of access to water means de facto exclusion from the surrounding pastures (Hagberg 1998: 60). For Ɗawra and his sons in Ganatcha, the question of water does not pose a particular problem. Their close ally Arɗo Ali from the Yaamanko’en clan possesses a well near Guirdigiskil in the northern pastoral zone, which can be jointly used during the rainy season, when the cattle of both groups move to the northern pastures. Similarly, Ɗawra’s well in Ganatcha is jointly used by both groups when their herds are in the area of the main camps. Hence, these two groups have resolved the question of access to water across two zones by way of a cooperative arrangement. The fact that Ɗawra possesses his own well constitutes a significant advantage. The public well managed by Umaru is often highly frequented, which can entail extended periods of waiting for the herds. Water shortages usually do not occur, but water is pulled on up to four ropes simultaneously, which increases the danger of accidents. The water quality is also better at Ɗawra’s private well as it is only used by his own family and the Yaamanko’en households. High frequentation plays against water hygiene because particles of animal dung fall down each time the bucket-bags are plunged into the well. Although agriculture remains only of secondary importance as an economic activity, the question of access to farmland is also an aspect that can make a site either attractive or problematic. Farming land is customarily allocated by the local ‘traditional’ authorities, generally against small amounts of money and customary gifts of animals. In the case of Ganatcha, Umaru has thus obtained a large strip of farmland from the district chief in Moa, smaller portions of which he later allocated to other members of the local group on the basis of informal grants (aro) (see Köhler 2017a: text 3,16f.). Ɗawra and his sons, although all involved to some extent in agricultural activities, have so far also just borrowed land from Umaru, but never approached the authorities to acquire farmland themselves. In the case of Salaga, Nano Buuyo at first tried to acquire farmland for his community from the village chief of Tumulle, but it became obvious that

88  Space, Place and Identity

the village population wanted to exclude the Woɗaaɓe from cultivation. Nano thus addressed the district chief with whom he was on good terms and obtained farmland near Tumulle from him, which the village population could no longer refuse after the intervention of the superior authority.5 Other problems that the Woɗaaɓe in the Damergou complain about are frequent animal theft and more generally a situation of growing insecurity. The latter is closely associated with Tuareg ex-rebels who are still well armed and have drifted into armed banditry since the end of the second Tuareg rebellion in 2009. Bush taxis connecting local markets with the town of Tanout have thus been attacked by armed bandits and the passengers robbed of their valuables. The Damergou region also has a bad reputation with regard to animal theft. Racing camels in particular run the risk of being stolen as they represent a major value, and occasionally, even entire herds of sheep disappear. In the Koutous region, the problems of theft and insecurity play a minor role. The area was not particularly affected by security threats during the last Tuareg rebellion from 2007 to 2009 and overall enjoys a good reputation among the Woɗaaɓe for its peaceful conditions. Finally, social factors also played a certain role in Ɗawra’s decision to settle in Ganatcha. When he left his brothers, Araba and Boyi, behind in the Damergou region he joined his cousin Umaru with whom he had already stayed closely together in Tesker in the 1980s. Additionally, the existing kinship bond had been reinforced at that time by a marriage alliance when Ɗawra took a sister of Umaru as his second wife (see Figure 5.2).6 The fact that Umaru had meanwhile established himself as a chief and as the local representative of a public pastoral well was a good starting point. Umaru offered Ɗawra land for cultivation and for definitive settlement, and with his good relations to local authorities he helped Ɗawra get the permission to construct his own pastoral well (see Köhler 2017a: text 3,24f.). However, it seems that social factors have influenced Ɗawra’s decision to migrate to Ganatcha in yet another sense. In particular, a lineage-internal conflict between Ɗawra and his sons, on the one hand, and the sons of Buuyo, on the other, is often cited by lineage members as one reason why Ɗawra and his sons left the Damergou region.7 In fact, while farmers often perceive pastoral mobility as threatening and as a source of conflicts, a different capacity of mobility comes into play here: it can also function as a means of avoiding confrontation or resolving conflicts (Stenning 1966: 388; Gulliver 1975; Bonfiglioli 1988: 107; Wilson-Fall 2000; Schareika 2004: 177), both in relations with other groups – such as in the case of flight migration in times of political repression – and on the intra-group level. The day-to-day nomadic movements of household units that constantly bring about new social configurations can express the distance or closeness of relationships (Woodburn 1972: 204f.; Gulliver 1975: 379; Ingold 1987: 177). Mobility permits approaching or avoiding others in a diplomatic

Nomadic Pastoralism to Sedentarization  89

Figure 5.2  The genealogical relation of Ɗawra and Umaru (diagram by F. Köhler).

way without being rude: by always leaving a door open and even more so as this door is immaterial (Bernus 1995: 95). Ɗawra himself has never openly acknowledged the lineage-internal conflict as providing the reason for his decision to migrate. While sensitive aspects touching on interpersonal conflicts are not easily addressed when discussing the motives for household relocations (see Ingold 1987: 178), the interpretive ambiguity of Ɗawra’s move arguably also implies strategic advantages. In fact, the interpretation of his migration as a defensive move to avoid conflict is by no means imperative. It can also be understood as a pioneering move to open up new pasture resources in a new area offering more favourable conditions – a privately owned well, a less conflict-dominated environment, in short, a better place to make a living. In this sense, Ɗawra’s migration appears less passive and more like an example for the use of mobility as a political tool. Ɗawra has left the immaterial door behind himself wide open and if others decided to follow and join him as their arɗo they would be more than welcome, as he never ceases to underline.

90  Space, Place and Identity

Grazing the Meadows of International Aid – Villageization, Schooling and the Notion of the Terroir d’Attache As argued above, while nomadic mobility has often been interpreted as a strategy of state evasion, in the case of the Woɗaaɓe such interpretations have to be applied with some caution and in a more differentiated way. While historically, self-marginalization through conscious withdrawal was to a certain extent indeed deliberately chosen (Boesen 2009), at least since the droughts of the 1970s and especially of the 1980s, one-sidedly evasive strategies would lead to a clearly undesirable state of exclusion from political and economic participation. Henceforth, the state, hand in hand with international aid and development organizations, has more or less regularly acted as a distributor of aid and development resources. Participation in this process of distribution constitutes an obvious interest at least in a minimum of integration into administrative structures. The question today is less how to evade the state than how to integrate those aspects of the state that seem rewarding: how to access the resources and services that the state has to offer – schools, development projects, infrastructure, food supplies and aid. The Woɗaaɓe try to selectively appropriate specific aspects of what is locally often quite undifferentiatedly referred to as ‘the state’ (gomnati), while they reject others and try to keep at distance from them. Such an attitude towards the institutions of the state – trying to profit from state services wherever possible, while at the same time keeping up a maximum of autonomy – may look opportunistic. However, there are many constraints that make this ideal difficult to achieve, one of the most important probably being the contemporary marginalization resulting from mobility: for a long time efficient as a mechanism of escaping environmental and political constraints, mobility has today in many cases turned into a political handicap (Botte and Schmitz 1994 : 17; Boesen 2009: 71, 77). An important reason is that the lack of a permanent physical base materialized in a concrete place leads to exclusion from many offers of the state and development programmes that are generally designed with a sedentary logic in mind and thereby neglect the specific needs of mobile populations (Schlee 2005: 30; Grémont 2011: 183, 2014: 34). This is one reason why today many Woɗaaɓe have opted for the model of the locally fixed centre. The current tendencies towards reduced mobility and sedentarization in the vicinity of a pastoral well can in this sense also be interpreted as a strategy of resource appropriation, the principal interest being participation in processes of distribution of resources by state and development actors. Not only the state’s institutions, but also non-governmental development programmes have difficulties with adapting to the mobility of pastoralists. As a result, their projects are often conceived with a sedentary bias and in consequence often play hand in hand with the state’s efforts to settle pastoralists. Most authorities and development organizations, in order to distribute relief food or other aid,

Nomadic Pastoralism to Sedentarization  91

ask mobile groups to indicate a permanent place at which the population can be reached and make this a pre-condition for their interventions (Greenough 2003: 105f.). The essential requirements the communities are generally asked for are a locality name, a head of community and a permanent local presence. Although the latter is hardly compatible with the necessities of pastoralism, many Woɗaaɓe, due to their interest in the resources at stake, make concessions and organize themselves, at least partially or temporarily, in so-called centres – semi-sedentary camps that are still moved, yet with minimum mobility around a fixed point in space. The term is derived from the French development jargon in which ‘centre de regroupement’ designates a place of regular gathering of pastoral populations. The term thus indicates that the creation of centres is directly linked to the intervention of development and humanitarian programmes and at the outset a reaction to their requirements.8 The following quote also confirms this link: Whenever we assembled in a centre, it was at times when aid was distributed to the pastoralists because their animals were dying. This is what is behind it: development agents and state administrators. They came and asked for our centre where they could help us. So we established our centre around our well. (Laɓɗo Usman, June 2011) The first time this mechanism was put to work was during the drought of 1984–5 (banga-banga) that forced many to rely on distributions of relief food by the government and international organizations. Although during the drought of 1973–4 many pastoralists had already been forced to move to humanitarian camps in Niamey and elsewhere, for the Woɗaaɓe Gojanko’en in the Zinder province, the first occasion on which they had to move to towns to obtain aid was the crisis of 1984–5. The following account of the situation by Umaru Siddi makes it clear that the distribution of aid was even more centralized then than it is today: had the nomadic populations not moved towards towns, they would not have been reached by the distributions at all. At that time, many Woɗaaɓe went to Zinder because of the famine. We stayed close to the town, because that is where relief food was distributed. They did not make an effort to bring it to the pastoral zone. (Umaru Siddi, July 2011) Apart from aid distribution, development organizations have over the years executed diverse projects aimed at sustainably improving the economic situation of the population. Yet these interventions generally also work with the premise of sedentary beneficiaries and require a permanent presence. The households group together at the site of the centre and manifest their presence if an organization announces an activity that seems economically rewarding, but for most, access

92  Space, Place and Identity

to the project’s resources is the only reason for staying in place. Generally, the greater part of the group quickly disperses as soon as the external actors have left the site, or even before, if the requirements of their herds necessitate new movements. The request of state and development actors for a local presence was thus often met with a minimal adaptation. It has not, from the outset, led to sedentarity or even to a significant and sustained decrease in mobility, because the interventions have only occurred occasionally. The creation of the centres has thus entailed new cycles of concentration and dispersal, occasioned by the presence and by the activities of development projects. The degree of mobility remains generally high and the process of sedentarization concerns only particular parts of the society. The artificial creation of the centre and its highly fluctuating population make the notions of ‘community’ and ‘locality’ with regard to it rather difficult to grasp. It is probably best to understand the community of the centres as temporal and partial manifestations of a community that is constituted on a wider scale, on a translocal level (Gupta and Ferguson 1997b: 7). One example for project interventions that are embraced with keen interest by the Gojanko’en Kuskudu today are school projects. School education is also an excellent example of how the attitudes among Woɗaaɓe have changed from state evasion towards a greater concern for integration into state structures. For a long time, there has been considerable reluctance against government efforts to put their children to school. The model of the so called ‘écoles nomades’ was first introduced in Niger in 1944 (Clauzel 1992: 109). These ‘nomadic schools’ were in fact locally fixed and for nomads, rather than being conceived as mobile units (ibid.). However, they provided canteens, which allowed the children to stay in place and attend school even if their parents had to move on with their herds. Although such offers are considered very attractive today, at the time they were not. One reason was that the children’s workforce was needed in the pastoral economy, yet a more general scepticism against modern education was also prevalent (ibid.). The time of the introduction of these schools coincided with a period of relative wealth for the Woɗaaɓe and the benefits of modern education, which would eventually lead the children to drop out of the pastoral economy, was not as evident as it seems today. The colonial policy towards schooling was one of strict enforcement and many elders recall how children were seized and put to school by force, even against the will of their families. Many Woɗaaɓe responded with evasion, based on their mobility. From a contemporary perspective this resistance seems almost paradoxical, since the current reality looks quite different: as with mobile pastoralists in other regional contexts (Krätli and Dyer 2009: 46; Siele et al. 2013), most Woɗaaɓe are today well aware of the benefits of modern education. Statements such as the following are frequently heard:

Nomadic Pastoralism to Sedentarization  93

Whether in town or in the bush, we are way behind. What has retarded us is that we have never sent our children to school. The way I see it, this is the reason for our problems. Before, there was more rain and the bush was abundant. Schools were not of our concern. Today the rains have diminished, the trees have been reduced and desertification is accelerating. This is why today we pastoralists are seeking to obtain farmland and to send our children to school. (Laɓɗo Usman, June 2011) The statement indirectly confirms the thesis that the contemporary marginality of the Woɗaaɓe is closely linked to the long-time refusal of modern education (Boesen 2009). It also shows that one reason for the growing interest in school education is to open up economic alternatives to the young generation in a time when, for many, pastoralism alone is not a realistic future option. Being able to handle administrative matters directly without relying upon outsiders is another motivation. Many parents would like to enrol their children, but the access to schools in the rural areas outside villages remains difficult. If a certain reluctance regarding school enrolment is still prevalent, this is mainly linked to the fact that the existing rural schools are poorly adapted to the needs of pastoralists, which is also a more widespread problem in other African contexts (e.g. Schlee 2005: 30; Siele et al. 2013). One problem is that the teachers are generally outsiders to the pastoral milieu and do not speak Fulfulde, while the pupils in their first years generally do not have sufficient Hausa skills to allow even for a basic communication, let alone for following lessons which are, as in all Nigerien schools, in French. Although efforts are being made to employ teachers who have themselves a pastoralist background (Galy 2010: 152), the success is limited simply for a lack of candidates who dispose of the required training. Schools that literally move with the mobile households, as desirable and as reasonable as such a solution might be in other nomadic contexts (Schlee 2010: 11), are clearly dismissed by the Woɗaaɓe in Salaga as not feasible in their context: the seasonally pronounced dispersal of mobile household units due to the scarcity of resources would make it difficult to maintain classes of sufficient size in a single – even mobile – place. Modern solutions that are presently being experimented with, both in the East-African and West-African contexts, are distance learning based on widespread media like the radio and mobile phones (Aderinoye et al. 2007; Krätli and Dyer 2006, 2009; Siele et al. 2013). For the moment however, these are no practicable alternatives in Niger either. The only realistic access to schooling for nomadic Woɗaaɓe is either to settle down in or near a village where the children can be enrolled at an existing primary school, or, again, to establish a centre. If a sufficient number of children at the age of being enrolled can be confirmed, a school will officially be created by the administration. The community will then be provided with a teacher and the

94  Space, Place and Identity

necessary equipment. In return, the community generally provides accommodation for the teacher and a shelter that serves as a class room. The example of Salaga attests to the importance that is accorded to school enrolment today. Here, the process of getting a school was difficult and took some time. Before the administration finally agreed to put a headmaster in place, the community took the initiative to support the salary of a teacher by their own means between March and June 2011. Schools with canteens are of particular interest. A government programme for rural canteens exists, yet covers only a small percentage of rural schools (Galy 2010: 152). But an increasing number of projects with international funding also provide meals at selected rural schools. From the perspective of development actors, canteens are a means of assuring the effectiveness of teaching by providing a balanced alimentation for the pupils. In the case of pastoralists in particular, they allow pupils to stay in place and follow their lessons even if their parents have to move away seasonally with the herds. From the perspective of many Woɗaaɓe parents, however, school canteens are also a means of relieving the family economy, which makes it sometimes difficult to discern to which degree educational projects are regarded as a value in themselves and in how far they are also seen as an economic asset (Greenough 2003: 110). Today, in all the three centres of the Kuskudu in the Zinder province (Ganatcha, Salaga and Ngel Tireeji) school projects have been initiated and have advanced to various degrees. The first local Woɗaaɓe community to have its school accredited was Ganatcha in 2008. The school officially has 37 pupils.9 In Salaga, the school initiative started in 2011. Its 126 pupils come from three different local communities: the initiating Woɗaaɓe Gojanko’en Kuskudu, the Gojanko’en Mbuuldi and a neighbouring Tuareg community. At the third site, Ngel Tireeji, the number of pupils is the smallest with only 29. In all the three sites, the school projects have comprised the construction of clay houses that serve as domiciles for the teachers, and others for stocking the supplies for the canteens. Also, in Ganatcha and Ngel Tireeji, classrooms have been constructed with an external funding. In Salaga, the construction activities have further entailed private initiatives of house construction, thus establishing the nucleus of a sedentary village. In contrast to the distribution of relief food, which is occasional and irregular, the regular distribution of school meals in community run canteens has a stronger effect on permanent local residence, if not for complete families, then at least for the core households. Hence, schools, just as pastoral wells, are crystallization points for a more permanent materialization of localities and reflect the direct impact of governmental and non-governmental development programmes on the current processes of local fixation. But accessing new resources offered by state and development actors is not the only motivation for Woɗaaɓe to settle in centres. Another major concern is

Nomadic Pastoralism to Sedentarization  95

Figure 5.3  Schoolchildren in Ganatcha, February 2014 (photo: F. Köhler).

their prevailing lack of legal security for access to the resources on which they depend for their existence and which are increasingly competed. The problem is partly based on the fact that since colonial times, the status of pastoral land has been defined only negatively as vacant, unclaimed land (‘terres vacantes et sans maître’), which has not been transformed by cultivation and which belongs to the state (Hagberg 1998; Bonte 1999: 217f.; Hammel 2001: 5). The dichotomy of personalized land rights for farmers and mere use rights for pastoralists finds its expression in the fact that the ‘sedentary’ populations are represented by chefs de village and chefs de canton who administer clearly delimited territorial units and decide about the allocation of land, while the ‘nomadic’ chefs de tribu and chefs de groupement do not have any formal rights over the distribution of land, their competences being limited to administrating their tributaries (Hammel 2001: 2f., 14f.; see also Fuglestad 1983: 80; Moutari 1999: 428; Boesen 2009: 77). This administrative division also had the effect of artificially essentializing and dichotomizing the categories of ‘sedentary’ (farmers) and ‘nomadic’ (pastoralists) along ethnic lines, although in practice they are rather poles of a continuum on which groups and individual actors move with considerable flexibility (Boesen 2004b). Even seemingly sedentary farmers are often highly mobile and sometimes ‘more nomadic than the nomads’ (Gallais 1975: 180). Although the dichotomy of ‘sedentary’ and ‘nomadic’ is thus highly problematic in the context of the Sahel (Retaillé 1984, 1998; 2006, 2013; Retaillé and Walther 2011; Boesen et al. 2014; Marfaing 2014a), the categorization continues to have significant legal consequences as far as access to land resources is concerned.

96  Space, Place and Identity

More recent Nigerien bodies of law, notably the land tenure law Code Rural (Republic of the Niger 1993) and an ordinance specifically concerning pastoralism, often referred to as Code Pastoral (Republic of the Niger 2010), translate efforts to more explicitly codify pastoralists’ options for legally claiming rights on the land resources on which they economically depend. When it was introduced in 1993, the Code Rural was progressive in that it represented an effort to integrate customary rules for land use into a modern body of law (IIED 2006; Dyer 2008). Thus, the Code Rural explicitly recognizes customary law as a valid legal basis for claiming property rights in land (Republic of the Niger 1993, par. 5), yet at the same time demands inscription in the land tenure register (dossier rural) in order to ensure full property rights (ibid., par. 8–9). Similarly, the Code Rural and a more specific complementary text (Republic of the Niger 1997a) grant official recognition of priority use rights to pastoralists in their terroir d’attache. The latter is defined as a territorial unit in which a group of pastoralists habitually lives for a significant part of the year and to which it retains a link during times of longer-range pastoral migrations into other areas (Republic of the Niger 1997a, par. 2; Lund 1998; Hammel 2001; Dyer 2008).10 This legal definition introduces a relative permanence of a link to an area as one criterion of recognition. Other criteria are the historical depth of a link to an area as well as the investments made in terms of infrastructure, notably the construction of wells in the pastoral zone, that allow to make economic use of the area’s resources. The first of these criteria remains rather vague. Although the notion of terroir d’attache carries strong connotations of historic and habitual rights (Hammel 2001: 5), a geographical attachment, in order to be recognized, need not be based on a real historic link, but can in fact be quite recent. More important in current practice is therefore often the second criterion, i.e. the creation or possession of infrastructure (Republic of the Niger 1997c). The owner of a well in the pastoral zone enjoys de facto priority use rights over the surrounding rangeland (Thébaud 1990). This general practice, together with a steadily increasing competition, has led to an increasing race for resources and many pastoralists have begun to participate in this struggle, trying to obtain or construct pastoral wells of their own. A central aspect of the recognition of a terroir d’attache is that it does not grant property rights but merely privileged use rights (Republic of the Niger 1997a, par. 4; Hammel 2001: 5). The land remains part of the public domain of the state (Dyer 2008: 17). The local community enjoys privileged access to the pasture lands and water points, but it cannot deny other users access to these resources. Although the recognition of priority use rights is an important step towards a legal protection of the right of access to pastures, the new laws reconfirm the existing bias between farming and pastoral populations with regard to land-­ ownership. While the texts allow farmers to transfer their inherited claims to

Nomadic Pastoralism to Sedentarization  97

land into written land titles officially granting them property rights, the same is not easily possible for pastoralists concerning the land resources of their habitual grazing areas. The texts do acknowledge the possibility to obtain property rights on pastoral resources, yet only on condition of a permanent local fixation (Republic of the Niger 1993, par. 28., 1997a, par. 8). This indirectly excludes mobile pastoralists from individual property rights on pastoral lands and thus potentially discourages mobility. Another point is that priority use rights are granted to collectivities rather than individual pastoralists.11 The dichotomy between individual property in land, among farming populations, and common property and joint resource use, among pastoral populations, is thus not resolved by the new pastoral and land tenure laws. The consciousness for the problem of land rights among Woɗaaɓe and the strategy of assuring claims on pastoral land by constructing or obtaining pastoral wells have been documented notably for Central Niger (Loftsdóttir 2001c; Boesen 2007b, 2009a; Loncke 2015: 182ff.). The recent migration of Ɗawra and his sons from the Damergou region to Ganatcha must also be seen in the light of this strategy. The Woɗaaɓe, as with other groups of pastoralists in Niger and elsewhere, are today aware of their possibilities and constraints in securing land use rights or claiming land titles. If the potential for translating legal options into definitively recognized rights is not yet consequently used, this is generally due to a lack of knowledge about the legal procedures: for official recognition of terroirs d’attache, inscription in the land tenure register, held by the land tenure commissions (commissions foncières), is necessary (Republic of the Niger 1997b).12 Although the localized groups of Woɗaaɓe Kuskudu in the Damergou and Koutous regions consider their respective sites as ‘our land’ (lehidi amin), much in the sense of the terroir d’attache as an habitual grazing area, none of them has so far taken measures to legally assure their claims by obtaining written titles as prescribed by the legal texts. Many pastoralists today are well aware of the necessity of a relatively permanent presence for claiming priority use rights not merely on the pastoral resources in the surrounding of their wells, but also on these wells themselves (Hammel 2001: 13; Boesen 2009: 78). In the northern Damergou region, numerous wells originally constructed by pastoralists later attracted farming settlers who established villages in their vicinity. The already mentioned village of Tumulle near Salaga is a case in point: while it was Woɗaaɓe pastoralists who first established a well there, a farming village finally emerged and today the Woɗaaɓe even tend to be denied access to water. As the main reason for their exclusion, interlocutors identified their relative mobility or, more precisely, their lack of a permanent presence and of territorial appropriation. In a time of increasing pressure on resources, an effective claim for rights over them can only be defended by a relative permanence. A prolonged absence gives other groups the opportunity to settle in the area and claim the resources

98  Space, Place and Identity

for themselves. Even very mobile groups of Woɗaaɓe therefore generally make an effort to come back to their home range regularly and at least for a short period, even if the pastures are not very abundant (Schareika 2007: 207). This becomes even more important today, as the continuity in the use of local resources or infrastructure is recognized as an important argument for legal claims. Against this background, the current trend towards local stabilization in the vicinity of wells must also be interpreted as a rational and strategic effort to secure control over pastoral resources (Anderson and Monimart 2008/9b; Boesen 2009). The phenomenon of the emerging local centres, although it might seem to reflect reactionary or even opportunistic and primarily economically-motivated attitudes at first sight, also has a more strategic and political side.

Notes  1. For western Niger, see Boesen (2009); for eastern Niger Anderson and Monimart (2008/9a, 2008/9b).  2. Human micro census data from December 2010, animal census June 2011.  3. This regeneration becomes increasingly problematic as basically all modern waste articles for which no system of disposal is provided, are left behind in the abandoned sites.  4. See also Müller-Dempf (2014: 20) for the Turkana in Kenya.  5. Nano and Maalam Buuyo, December 2011.  6. On the role of marriage alliances as strategies for securing access to pastoral resources in the study region, see Retaillé (1998: 76), PSSP (2009: 52f.). For northern Nigeria, see Hickey (1978).  7. Taafa Buuyo, May 2011.  8. See also Boesen (2009: 75) and Lassibille (2009: 318, 329) for the Woɗaaɓe in central Niger, and Grémont (2011; 2014) for the Tuareg in Northern Niger and Mali. On the sedentary logic of development programmes in the Sahel and their impact on spatial structures, see also Retaillé and Walther (2011).  9. Numbers refer to 2013. 10. The French term terroir does not have a direct English equivalent. For the history of the concept and its shifting meanings and connotations in the West African context, see Bassett et al. (2007). 11. The texts are not altogether consistent on this matter: Although the possibility of individual use rights by pastoralists is mentioned (Republic of the Niger 1997a, par. 10), ‘pastoralists’ are in an introductory article defined as ‘groups’ (ibid., par.2) without any specification of the criteria (ethnicity, kinship, political affiliation) that define such groups. 12. The general task of the land tenure commissions is to implement the Code Rural. In the course of the political decentralization process in Niger, the commissions have gradually been installed, over the past decade, at all administrative levels. They have considerable planning and decision-making power, especially concerning the registration of land titles and the settlement of land tenure disputes (Mansour and Tan 2008: 10). An important principle of the commissions is that they include representatives of all the different socio-economic groups represented in the area of their competence (see also Köhler 2017a: text 3,26).

Chapter 6

Consequences of the New Spatial Strategies

The previous chapter drew a picture of a society involved in a process of profound change. This change has many facets: a transformation of the pastoral economy and a diversification of livelihood patterns, the embracing (although hesitantly) of modern education or, more generally, a stronger integration into the structures and institutions of the state, and a more pronounced local attachment that has led to new patterns of mobility. It seems difficult to determine a directionality in these different aspects, or to single out one reason to be at the basis of a causal conjunction. Is local stabilization a result of the increasing farming activities, or has agriculture been intensified because the sedentarization allowed for it? Has the creation of schools been possible as a result of sedentarization or was it rather its very reason, as some interlocutors claimed? Even with regard to the construction of wells, the direction of the causality is not always clear: although a well is basically a prerequisite for sedentarization, it has been pointed out that obtaining funds for its construction is much easier once a sedentary base has been established (Loncke 2015: 252). While I have so far concentrated on reasons and conditions for the transformation processes, in this chapter I will go on to assess their impact on the groups concerned.

Spatial Fixation, Placemaking and Territorialization The appropriation of land by pastoralists has been characterized as less direct and less marked in comparison to agricultural land use (Hagberg 1998, 2000: 172; Kintz 1982: 217; Schlee 1990b: 1). This characteristic is closely related to another fundamental difference between farmers and pastoralists: among

100  Space, Place and Identity

pastoralists it is not spatially fixed resources, but movable productive capital, i.e. animals, that are appropriated and transmitted from one generation to the other (Ingold 1987: 168). Territory in a mobile context is not to be understood in the sense of exclusive collective appropriation of land, but as the optional range of movement within which natural resources are appropriated by the community (Bonte 1979: 215; Retaillé 2006: 238). In this regard, the Woɗaaɓe in Niger are a paradigmatic case. Historically, as shown in Chapter 2, they did not own land. Placemaking was a constant process resulting from the strategy of reaching out for ever new and better pastoral resources. It was characterized not by occupation, but rather by non-territorial ‘habitation’ of space (Ingold 2009: 37). The tendencies towards stronger geographical attachment have entailed new patterns of placemaking and new modes of territorial representation. They are place-making in a quite literal sense as they lead to the emergence of new, and increasingly permanent, places. In a first step, these places materialize in the sense of Ingold (2007, 2009) as the intersections of the traces of movement. Cattle paths leading from all directions to pastoral wells are a readily visible example of how a pattern left by the traces of human and animal movement is literally inscribed onto the surface of the earth and thus makes place visibly a result of the social relations and contacts of people coming together, meeting and again separating, and defining a particular place by their interaction (Lefebvre 1991: 118; Massey 1994: 154; Sheller and Urry 2006: 214; Ingold 2007: 80, 2009; Retaillé 2010, 2013). The conceptions of place as socially produced and interactional, developed by authors like Lefebvre, Ingold, Massey or Retaillé, are thus apt for describing the initial phase of the process behind the emergence of new locally defined centres among the Woɗaaɓe. In a second step, the construction of schools, cereal banks and private houses adds to the ‘socialization of space’ (Appadurai 1995: 206), as it increases the intensity and permanency of the placemaking process. A third step, finally, is the increasing involvement in agricultural activities and the more territorially defined ­appropriation of space that, in the long run, goes together with it. As explained in Chapter 3, the relations between mobile pastoralists and predominantly farming populations in Niger were, despite a constant conflict-potential about crop damages, in the past characterized more by complementarity than by competition. The same spatial resources were interchangeably used among groups according to a pattern of seasonal alternation that favoured social exchange and integration. Over the last decades, however, the increasing pressure on land has led to a situation in which the former complementarity of pastoral and agricultural production has made way to an accelerated competition over the same resource (land) for different means (farmland vs. grazing land). Simultaneously, the increasing investment of farmers in livestock leads to more competitive relations between themselves and mobile pastoralists about fodder resources. This growing competition among different socio-economic groups

Consequences of the New Spatial Strategies  101

about the same resources increasingly leads to practices that play against the pastoralists. While the use of crop residues by the pastoralists was in the past facilitated by the already mentioned ‘manure contracts’, today, many farmers tend to either collect the residues from their harvested fields as they own animals themselves, or they make access for the pastoralists conditioned by payments (ZFD 2008: 25; Randall 1994: 330). Although access to pastures, including harvested fields, is in principle free, the legislation leaves farmers considerable options for monetizing the pasture resources on their land. For example, the stocking of crop residues in the fields has become a current practice among farmers. If a labour input has been made to bundle the stalks and to collect them in one place, they are legally considered as private property. Farmers can thus sell them to the pastoralists who depend on them, or claim compensation money if ‘damages’ on these bundled residues occur, even after the official date for opening the fields. The increasing trend of cutting and collecting grass in pastoral enclaves for putting it on stock and selling it to pastoralists at high prices when they are in need of it is another highly conflictual issue (ZFD 2008: 37). The pastoralists are generally in a weaker position as they depend on these resources, yet do not control the territories on which they are available. In this situation, the sedentarization of Woɗaaɓe and other pastoralists in Niger, and their increasing engagement in agriculture, accelerates the prevalent resource competition with established farming populations. Where place­ making takes the character of a strategy for securing access to spatial resources by staking territorial claims over them, the formerly predominantly zero-and one-­dimensional mode of space appropriation characteristic of Woɗaaɓe migration patterns increasingly makes way to a two-dimensional mode (Schlee 1990b). The sharing system of an ‘open, ongoing production of space’ (Massey 2005) is increasingly becoming ‘closed’. Relations to space become more fixed and more exclusive, and, as a result, inter-group relations tend to become more competitive (Watson 2010: 211). The stronger geographical attachment and ultimately territorial approach to land has the effect that placemaking increasingly becomes an exercise of power (Gupta and Ferguson 1997c; Turton 2011). With regard to the study group, this becomes manifest, as mentioned above, notably in Salaga, in conflicts between Woɗaaɓe and Kanuri about access to farmland and water points. This is not to say that in the past, power relations were not relevant. As mentioned above, the hierarchical order among Woɗaaɓe clans has historically determined the distribution of grazing areas and the ‘pecking order’ among clans at wells. Although expressed in the idiom of seniority, this hierarchy can also change over time according to numerical strength and political influence (Loncke 2008: 160ff.; Köhler 2017a: text 6). Similarly, among different groups of pastoralists, access to spatial resources is also often a matter of power relations. As mentioned in Chapter 5, this makes the relations notably with Tuareg in the

102  Space, Place and Identity

north of the study region often conflictive. While the use of the spatial resources by the different groups of pastoralists was in the past regulated on the basis of a relative complementarity of spatio-pastoral strategies (Moutari 1999: 427f.), today tensions arise increasingly from the more exclusive claims over pastoral resources that leave less room for negotiations and social interaction. The territorial strategies currently pursued by the Woɗaaɓe and other pastoralist groups in the region thus have an effect on the balance of integration and conflict in the sense that they accentuate existing potentials for conflict (see also Boesen 2009: 75).

The Strategic Appropriation of Boundaries As observed above, pastoral and agricultural land use in Niger have historically not been spatially separated by fixed boundaries. In principle, the new legal texts confirm this principle: farmland still becomes a pastoral resource after the completion of harvest and the terroir d’attache is a rather vaguely defined and not clearly demarcated area. There are, however, instances where borders become increasingly important. The Code Rural stipulates the elaboration of land use plans on a provincial level that specify and delineate the modes of economic use of the land (Republic of the Niger 1993, par. 127ff.; Younfa et al. 2010). Especially in the more densely populated southern agro-pastoral zone, this has become an increasingly relevant issue for the reason of accelerated resource conflicts due to the pronounced demographic growth. As stipulated by the legal texts, pastoral reserves and habitual cattle corridors have in recent years been identified and legally protected by inscription in the land tenure register. Officially part of the public domain, agricultural activity within their borders is formally prohibited (Republic of the Niger 1987b, par. 4; 2010, par. 54). The pastoral reserve of Ganatcha is a case in point: it was officially delimited in 2001 and is marked with cement posts. This introduction of borders has been welcomed by the local Woɗaaɓe community as a means of legal protection of their pastures. In many other local contexts where the status of pastoral enclaves within the agro-pastoral zone is not yet legally secured, the pressure on the pastures by gradually expanding farms is a common problem (Hammel 2001: 15). The phenomenon of millet or sorghum fields anarchically being sown into pastoral enclaves and cattle corridors is widespread in contemporary Niger (PSSP 2009: 49; ZFD 2011: 9). In Ganatcha, in contrast, the protection of the pastoral resources is already legally established. The clearly demarcated status of the pastoral reserve facilitates the negotiation with farmers who do not respect the limits, and it is thus a further aspect that plays in favour of Ganatcha as a site, despite being located in the agro-pastoral zone and surrounded by farms. At the same time, access to farmland is still relatively unproblematic and with a few exceptions, fields do not yet touch the limits of the grazing area. The

Consequences of the New Spatial Strategies  103

declared aim of the Woɗaaɓe agro-pastoralists has therefore been to obtain plots of farmland which are located directly on the limits of the grazing area.1 Apart from practical considerations of proximity, this strategy serves a double purpose: first, (1) it protects the grazing area against abusive intrusion of fields by local farmers; second, (2) the strategy is apt to reduce conflicts with neighbouring farmers, since crop damages caused by the Woɗaaɓe’s animals will thus predominantly concern their own fields, the latter functioning as a buffer range between the pastoral zone and the fields owned by farmers from the neighbouring Kanuri villages. Where, once, agricultural and pastoral land use were not spatially separate, today the Woɗaaɓe of Ganatcha appropriate new state-introduced and clearly demarcated borders in order to assure future access to resources and to minimize the conflict potentials inherent in the immediate spatial proximity of pastoral and agricultural land use.

Territorialization, Placemaking and Lineage Segmentation Doing research in central Niger in the 1950s, Dupire observed the principle that the ownership of pastoral wells supported the social cohesion of the agnatic group: they were meeting points and thus important centres for the social life of the clan segment during the dry season (Dupire 1970: 289f.). While this was a rather seasonal phenomenon at Dupire’s time, it is of more permanent relevance today as locally defined communities emerge in the vicinity of wells. This also becomes obvious from statements such as the following: Since we have got our own well, we began to stay constantly in its surroundings. We do not go anywhere [else]. Only if there is a lack of pasture do we move away with our animals. [Otherwise] we always stay near our well, in the rainy season as well as in the dry season. (Laɓɗo Usman, June 2011) Wells are thus crystallizing points for the formation of localized communities and for the emergence of new places: meeting places in the sense of Dupire (but also Lefebvre or Ingold) at first, and permanent localities, or centres, in a second step. Although the positive effect on social cohesion described by Dupire is thus in principle still valid, a second phenomenon that is prevalent today rather has an inverse effect: the multiplication of private wells also leads to new splits within social groups. The decision to construct a new well often triggers a process of group fission and subsequent re-formation of a new localized group around the new well. The fact that the Gojanko’en Kuskudu are today attached to five pastoral wells in three rather distant areas in the Zinder province has led to a pronounced geographical dispersion and to a relative separation, because in each of the sites the same tendency of sedentarization can be observed.

104  Space, Place and Identity

The pattern of placemaking in the contemporary context of the multiplication of wells and increasing fragmentation by differentiation according to localized sites leads to an accelerated multiplication of such sites, hence: A

→

A'

→

A

→

A" →

→

etc.

A"1, A"2 →

etc.

A'1, A'2

The trend towards obtaining pastoral wells has thus become a factor that also accentuates the fragmentation of the agnatic group rather than only favouring social cohesion. As shown above, the process of lineage segmentation is a gradual and ongoing process. Today, however, it is reinforced by the increasing race for resources. The possibility of creating centres and thus opening access to new resources has led to a certain trend among Woɗaaɓe to follow this model. More and more individuals strive to establish their own well and those who own wells establish themselves as independent leaders of centres that form around them. The currently observable spread of the model of the centre, including schools with external support and other projects that grant access to resources, is additionally fuelled by a competition among clans and lineages, and even within lineages, to be more successful in attracting projects and funds than others.2 The accelerated fragmentation in combination with the emergence of more spatially fixed places in the form of centres has the effect that lineage segments are increasingly locally attached. The local stability of a nucleus of the community leads to a gradual shift in the focus of social interaction from the translocal lineage community to the local community. As discussed above, close co-­residence over time entails a process of convergence with other social groups and eventually leads to the forging of new identities (Bonfiglioli 1988; Dupire 1963, 1970). Although this is a long and gradual process, the first signs of it can be recognized in the study group. The localized communities remain largely homogenous in ethnic terms, but the example of Ganatcha, where the community is constituted by Gojanko’en and Yaamanko’en, indicates that the co-residence of Woɗaaɓe from different clans can be intense to the point of forming a tightly knit community. Such close co-residence of Woɗaaɓe from these two clans is facilitated by the fact that their mutual relations are of a particular nature: in contrast to the majority of other clans, they have prohibited inter-marriage by elopement (te’egal) in an ancient ritual ban, which is still widely functional today (Paris 1997; Köhler 2017a: text 7). This has the effect that one major source of inter-clan conflicts, i.e. te’egal, was in principle dispelled from the beginning of their co-habitation.

Consequences of the New Spatial Strategies  105

In the mixed community in Ganatcha, Gojanko’en and Yaamanko’en share important ritual and social institutions. Not only do they jointly celebrate important community constituting rituals like marriages and name-giving ceremonies, they also share a joint age-group for young men and girls, called daɗɗo. One important function of the daɗɗo is to approach young women and men and to instruct them in matters of dance and of courtship. Both boys and girls enter the age-group at puberty. Sexual contacts are not only encouraged, but rather enforced: girls who refuse to go to the daɗɗo are disciplined by the male leaders of the age-group.3 The daɗɗo is thus a place of control of the sexuality especially of the young women, but more generally of the young (see Schareika 2007: 209ff.; Loncke 2015: 258ff.). While girls attend the nocturnal daɗɗo meetings only until they join their betrothed husbands, young men continue to do so even after marriage. Although this shared social institution can be considered as an indicator of the degree to which the two clan groups in Ganatcha form a joint community, there are also limits to the process of convergence. As boys and girls from the two clans grow up and attend the age-group together, elopement marriage (te’egal) with partners from the adverse clan, although condemned by the elders, does in practice occur. Although the alternative of arranged inter-clan marriages (kooɓgal) with the consent of both families would be considered acceptable by some, for the time being, the elders still tend to generally refuse inter-marriage. The limits of convergence of the different clan communities also becomes obvious on the political level. Within the relatively small Woɗaaɓe community in Ganatcha, there are three leaders, none of whom functions as a community chief. In politico-administrative terms, the community is not more than a loose association of its constituent sub-groups. Within each sub-group the respective arɗo is internally the first authority, and on occasions that concern the community as a whole, the three appear on equal terms. Only when it comes to outside representation of the community does Arɗo Umaru enjoy a privileged position, due to the fact that he was officially entitled with management functions for the local pastoral well. This function makes him the principal interlocutor for development actors and state officials in all matters concerning the representation of the local pastoralists. Arguably, in the case of the Woɗaaɓe community in Ganatcha, locality has become an important new parameter of group definition and identification, in addition to patrilineal descent (clan and maximal lineage). This becomes manifest, for instance, during inter-clan meetings (ngaanka), when young dancers weave improvised lines of praise for their ‘wuumre Ganatcha’ into their chants.4 As noted above, the term wuumre designates both the pastoral migratory group and the localized, semi-sedentary group of co-residence. The young dancers’ identification with their localized group of co-residence rather than with their

106  Space, Place and Identity

clan or maximal lineage can be regarded as an indicator of a process towards stronger locally-defined identities. Paradoxically, the trend towards localized communities does not globally imply that the process of segmentation becomes less fluid. For the owner of a well, the possession of the latter is a factor that restricts his mobility, because claims over the local resources can be effectively secured only by assuring a permanent presence. Together with a group of lineage mates, ideally at least with his minimal lineage, he therefore generally forms the nucleus of the localized community. For the wider lineage community, by contrast, a new well marks a new place, a new potential resource, and thus constitutes an additional option for pastoral mobility. Therefore, while a core group consisting of the community leaders and their minimal lineages forms a relatively stable residential unit, there continues to be a relative fluctuation of group members. This results from absences of individuals leaving for urban migration, but also from entire families joining the group or leaving it to join other segments and localized groups in other regions, not very different than in the nomadic context of mobile migratory groups. After a catastrophic season in 2010, the conditions were overall mediocre in 2011, yet highly varying across regions. While the situation in the Damergou region was difficult, the pastures in the Koutous region were comparatively good. During the cold dry season, towards the end of 2011, many pastoralists from the Damergou region decided to move there at least temporarily. In mid-November Boyi Egoyi visited his brother Ɗawra in order to explore the area of Ganatcha, and he was struck by the good pastures. In February 2012, Boyi moved his herds and households to Ganatcha. However, this was not a definitive migration: in the rainy season of 2012 he moved back to his home range in the Damergou region. The regional lineage segment, although split into different localized groups, is in this way still tied together by a flow of people across these localized groups and characterized by a continuing process of internal differentiation. The fluid movements and the group dynamics that they entail depend both on individual decisions and on roughly prescribed lines of solidarity within the lineage (see Gulliver 1975: 373f.). The resulting social configurations are reversible. A split is not necessarily definitive, but rather, at least at an early stage, it can open up new options for the mobility of individuals. Migration adds new places to existing webs of places across which there is a constant flow of lineage members according to considerations of either seasonally changing pastoral conditions or more long-term developments. As I have shown in Chapter 2, translocal links and social networks across places have for long been actively maintained in order to level out environmental risks by multiplying options for pastoral movements. Today the situation is characterized by localized groups with spatially relatively stabilized nuclei and among them a certain fluctuation of personnel according to environmental conditions.

Consequences of the New Spatial Strategies  107

The process of lineage segmentation does not necessarily become less fluid in the contemporary context. The apparent split in the lineage segment that Dawra’s migration to Ganatcha has caused might just as well turn out to be merely temporary. If the conditions in the Damergou region continue to deteriorate, Ganatcha might well become a pole of attraction for more Gojanko’en. At the time being, other lineage members also make up their minds as to whether or not to join Ɗawra’s group, and it is still an open question as to which parts of the lineage the process initiated by the migration of Ɗawra will, in the long run, lead to definitive separation, and which other parts will re-unite with Ɗawra’s group in Ganatcha. This fluid dynamic clearly gives mobility and migration political relevance.

Shifting Political Affiliations and the Politics of Mobility Analyses of the political organization of the Woɗaaɓe have emphasized the ease with which individuals or groups can change their political affiliation (Dupire 1970: 295, 1994: 266; Bonfiglioli 1988: 51). It has been pointed out that a chief’s personality is crucial for his capacity to attract followers (talaka’en), and that politico-administrative affiliations are often subjected to highly strategic and opportunistic considerations. In practice, changes in affiliation are often linked to processes of group fission and to mobility and migration. When they first migrated to the Damergou region, all the Kuskudu pledged allegiance to a laamiiɗo from the Ɓii Ute’en clan, who, at the time, was the only chef de groupement of the Woɗaaɓe in the region. Later, when a Jiijiiru leader was appointed chef de groupement in Njaptooji, not far from Usman’s well, this occasioned a change of allegiance on the part of Usman and his followers, as his son Laɓɗo reports: Today, we are affiliated to Kirta Girma, chef de groupement of the Jiijiiru in Njaptooji. Before, we were affiliated to the Ɓii Ute’en chief in Gourbobo, like everybody else here, including Kirta Girma. When he became chef de groupement himself, we affiliated ourselves with him. You know, the Gojanko’en and Jiijiiru are very close. (Laɓɗo Usman, June 2011) The statement makes clear that the choice was motivated by considerations of socio-cultural closeness: the two Degereewol clans Jiijiiru and Gojanko’en consider each other as closely related, with their clan founders sometimes conceptualized as twins. In practice, this translates into strong ties of solidarity. The Ɓii Ute’en are also Woɗaaɓe, yet not from the Degereewol but from the opposing Alijam clan cluster. However, even ethnic identity is not a decisive criterion for politico-­ administrative affiliation in the region (Retaillé 1984: 199). Bonfiglioli observed

108  Space, Place and Identity Table 6.1  The leaders of the study group and their politico-administrative affiliation. Name

Local attachment

Politico-administrative affiliation

Ethnic category

Ɗawra Egoyi Umaru Siddi Laɓɗo Usman Boyi Egoyi Nano Buuyo

Ganatcha Ganatcha Ngel Tireeji Yalema (Intrika) Salaga

chef de canton, Moa chef de canton, Moa chef de groupement, Njaptooji chef de groupement, Gourbobo chef de groupement, Tenhya

Kanuri Kanuri Woɗaaɓe Jiijiiru Woɗaaɓe Ɓii Ute’en Tuareg

in the 1980s that although a number of Woɗaaɓe chefs de tribu were generally grouped together under the authority of one chef de groupement, there were also exceptions to this rule. He cites the case of a leader who was affiliated with a Hausa chef de canton (1988: 127). While an affiliation with a chief from a different ethnic group seems to have been rather exceptional at the time, it is quite common today in the region (PSSP 2009: 59f.). With regard to the study group, out of the five Kuskudu arɗuɓe in the Zinder province only two are affiliated with Woɗaaɓe chiefs (see Table 6.1). These are Laɓɗo Usman and Boyi Egoyi, who follow the chiefs in Njaptooji (Jiijiiru) and Gourbobo (Ɓii Ute’en), respectively. Nano Buuyo is affiliated with a Tuareg chef de groupement in Tenhya, although he is ‘indépendant’ insofar as he remits the taxes for his group directly to the municipality. The remaining two leaders (Ɗawra Egoyi and Umaru Siddi) are affiliated with the Kanuri chef de canton in Moa. This might seem unusual at first sight, but appears not implausible when regarded more closely. Given that agriculture is increasingly significant as a supplementary economic activity, an affiliation with a chef de canton can be advantageous insofar as he is in charge of the allocation of land. An affiliation with a chef de groupement does not necessarily facilitate access to farmland, since, as outlined in Chapter 5, only district- and village chiefs (chefs de canton/chefs de village) have land tenure competences while pastoral chiefs do not. Commenting on ‘dissidents’ who register with Tuareg or Arab chiefs while continuing to live with their respective lineages and migratory groups, Bonfiglioli (1988: 126) has pointed out a relative separation between administrative and social matters. Indeed, the political splits within the study group generally do not go together with fissions of social groups. The affiliations with Kanuri or Tuareg chiefs seem pragmatic more than anything else, the most obvious rationale being that of local proximity. A chef de groupement should serve the interests of the chefs de tribu under his authority and, more generally, those of their followers. An ally nearby, however, is more helpful in this regard than one far away. It is noteworthy that Ɗawra and Umaru followed the same rationale of local proximity in order to enjoy administrative advantages when they affiliated themselves to the chef de canton of Moa, the main town of the municipality. An administrative affiliation nearby is also encouraged by the state, in line with the process of political

Consequences of the New Spatial Strategies  109

decentralization that began to become palpable in more remote areas since the first municipal elections took place in 2004.5 Umaru Siddi, when explaining his affiliation with the chef de canton, makes this connection quite clear: When the rural municipalities were established it was said that everybody had to register and pay their taxes in their municipality. As for those who possessed a well, they were supposed to pay their taxes in the municipality where the well was located. This is why I affiliated myself here [i.e. with the Kanuri chef de canton]. (Umaru Ɓii Siddi, July 2011) With regard to the pastoral populations in particular, this requirement can be understood as a means of facilitating an effective control by the decentralized administration. In the present context, however, such a localized affiliation is not conflicting but rather perfectly in line with the territorial strategies of the Woɗaaɓe. Increasingly, there are incentives for the pastoralists to comply with this request. The municipalities have become indispensable as they provide essential services. Schools and political elections are cases in point, as are aid-­ distributions, when the municipalities are involved alongside with the customary chiefs. While an affiliation with a chief in a faraway municipality would thus easily entail exclusion from these assets and resources, an affiliation at local proximity facilitates integration into state structures and ensures consideration in resource distribution.

The Political Impact of New Resources Since the decentralized administration is implicated in the distribution of the resources of humanitarian aid, good relations with an administrative chief are definitely advantageous. In this context, the mechanism of establishing and maintaining institutionalized relationships of mutuality on the basis of animal loans (haɓɓanaaye) is used, among others, to affirm political affiliation with and assure consideration by customary chiefs and elected representatives alike. Laɓɗo Usman, for instance, has given animal loans to the prefect of Tanout, which can be interpreted as networking aimed at more easily obtaining funds for projects that have to be approved by the authorities, and of ascertaining to be duly considered in aid distributions. Governmental programmes for relief food distribution follow a top-down process. This means that ever smaller portions are handed down from one hierarchical level to the next in order to be redistributed, and the individual beneficiary generally receives his share at the end of a chain of instances that have all had their hands on the resources. Regularly, food supplies do not arrive at their destination as a result of misappropriation during this distribution process. Therefore, it is desirable either to entertain good relations with an administrative chief or, preferably, to

110  Space, Place and Identity

achieve oneself a position that allows one to actively participate in the distribution process. This is one reason why many Woɗaaɓe today aspire for political leadership by obtaining titles of chefs de tribu, which contributes to a political fragmentation and a multiplication of offices. When the Kuskudu first migrated to the Damergou region, Ɗawra Egoyi was the first to establish himself as an administrative chief. Usman obtained his title shortly after, and for more than 30 years the two remained the only Kuskudu chiefs in the region. Today, there are five. The phenomenon has occurred since at least the mid-twentieth century (Bonfiglioli 1988: 127; Boesen 2009: 77). Dupire (1970: 329) observed in the 1950s that the pace of fragmentation was accelerating as a result of a growing political competition. Today, however, the above-described new spatial strategies also add to the political fragmentation of lineage groups. When Nano Buuyo and Boyi Egoyi only recently established themselves as administrative chiefs, for instance, this was directly linked to the establishment of their centres. Also, just as the creation of centres is influenced by development projects and relief programmes, the same is also true for the processes of political fragmentation. According to Laɓɗo Usman, the main reason why Nano Buuyo aspired to the office of chef de tribu was that when he acquired project funds for rehabilitating the well inherited from his father, the donor agency made the signature of an administrative chief a requirement in the application procedure, as is generally the case with infrastructure projects: In the past, Nano was affiliated with [my father, Arɗo Usman] . . . When he wanted to repair his well, a Christian association agreed to finance the work. But they requested that a chef de tribu sign the papers. Nano refused any chief to sign concerning his well. He wanted to sign himself. So, he decided to establish himself as a chief. He looked for followers and he found enough of them, so he established himself as a chief. (Laɓɗo Usman, June 2011)6 From this perspective, Nano’s split from Arɗo Usman and his decision to establish himself as a chef de tribu looks pragmatic rather than motivated by aspirations for power. The requests of donor organizations for administrative representatives at the local level produce such representatives just as they produce the centres, thus also contributing to a politico-administrative fragmentation of the lineage. Although the current tendencies of fragmentation might primarily be caused by external factors, conflicts or rivalries within the community can further reinforce them. And the episode also reveals a certain lack of internal cohesion: apparently afraid that a signature from a third party might entail claims of rights over the well, Nano was reluctant to have anybody else – even a close relative – sign in his place.

Consequences of the New Spatial Strategies  111

The longer established chiefs regularly put into question the authority of the new leaders of the emerging centres, and indeed, their status is ambiguous. Upon establishment of a centre, its leader is at first just a representative without any official administrative status, a mere spokesman who assures the contact with state officials and development agents. If the centre develops into a permanent settlement it can in principle over time acquire the status of an official village and the leader will then be raised to the status of an official administrative chief. Often, the Woɗaaɓe refer to their new category of leaders with the Hausa term mai gari (village chief) instead of the Fulfulde term arɗo. On the one hand, this term, which reflects the jargon and the sedentary logic of development agents, might express an understanding of a process of villageization (Boesen 2009: 81). On the other hand, the careful distinction between the new leaders and the established chiefs might express the latter’s interest in emphasizing the different status of the new category. Ɗawra, for instance, insists that a ‘real’ arɗo must be invested by the chef de groupement and state representatives, thus dismissing the legitimacy of his younger brother Boyi’s claim to the status of a chief: Boyi is not an arɗo . . . In order to become an arɗo, the chef de groupement has to come to your house and gather all the people. The prefect has to come and everybody else. You have to pay some cows and then you will be invested as a chef de tribu. . . . Only because of the aid-distributions did Boyi quit [and establish himself as a chief], but he is not an arɗo. (Dawra Egoyi, July 2011) The statement shows how the multiplication of offices under the influence of development and relief programmes goes together with an internal differentiation of notions of power. The request of development actors for representatives of local centres has led to the emergence of a new category of leaders in the political landscape of the Woɗaaɓe – the mai gari. Inversely, access to this new ‘office’ depends on the creation of a locality, the leader of which one can then become. As shown above, this is achieved through possession of locally fixed infrastructure, notably wells. The possession of wells has thus become a new form of access to power. The new option for acquiring an office has led to new strategies for political careers, and these new strategies have a direct effect on the fragmentation of the lineage and on patterns of spatiality and territoriality. But the introduction of this new category of leaders has also entailed an internal debate about the status of such new leaders in relation to the established chiefs (arɗuɓe), and it has thus given a new dimension to internal political dynamics. Where new offices emerge, the internal struggle for power focalizes on the nuances in status and meaning of the different classes of titles. In small scale, primarily descent-based societies like the communities of the Woɗaaɓe, where kinship ties link practically everybody among each other, one

112  Space, Place and Identity

might expect a tendency for strong solidarity. Although the maximal lineage does play a certain role as a solidary community, in practice units of solidarity are often much narrower, and the stronger bonds on the level of the minimal lineage and the potential opposition between different minimal lineages often inhibit the consolidation on the higher levels (see also Boesen 2009: 85). An example from the Damergou region shows in a telling way how internal competition hinders the achievement of common goals.7 Several years ago, when Ɗawra was still based in the Damergou region as a chef de tribu, an educational programme was ready to finance a school for the Gojanko’en community. The three local chiefs at the time, Arɗo Ɗawra, Arɗo Usman (both from the Kuskudu maximal lineage) and Arɗo Kiiro (from the Mbuuldi maximal lineage) were asked to designate a location for the school to be built. Each of the arɗuɓe proposed his own well and they were not able to find an agreement. The people of Arɗo Ɗawra said it should be at Salaga, and Arɗo Kiiro said, no, it should be where they were staying. Everybody wanted the school close to the place they were staying. . . . Later they did not come back to the matter of the school. That was the end of it all. (Laɓɗo Usman, May 2011) Finally, due to internal fragmentation and the inability to consolidate for a common cause, the community did not manage to obtain the school that they had wished for and that had already been offered to them.

Internal Competition and Fragmentation The reputation of Fulɓe and other pastoralists of not being capable of collective action has been pointed out by researchers (e.g. Bierschenk 1995: 462) and is also frequently brought forth particularly against the Woɗaaɓe by both local populations and development agents (Loncke 2015: 251). In fact, political and pastoral associations are today highly ethnicized and their difficulties with organizing themselves for concerted action show the same oscillation between fusion and fission, the same struggle of centripetal and centrifugal forces at work that is visible in the merging and splitting of segments of Woɗaaɓe lineages (Boesen 2009: 86; Lassibille 2009: 311). According to the logics of situational identification and interest-group formation discussed in Chapter 4, consolidation is temporarily possible if common interests, notably in opposition to an external other, are strong and commonly shared values make a situational coalition seem plausible. The power of the political leaders to unite larger groups in order to pursue common goals, however, is often insufficient and particular interests and power struggles easily interfere with efforts of forming more stable interest groups (see also Schareika 2010a).

Consequences of the New Spatial Strategies  113

Hence, the limitations in collective mobilization and concerted action seem to be conditioned by structural aspects of the socio-political organization: the relative independence of individual family heads to decide on political matters, have the effect that political projects are difficult to realize on the higher societal levels of the clan or the regional clan section, and even on the level of the maximal lineage. Schareika (ibid.: 221) has called this a ‘paralyzing balance of forces’. Individualist tendencies and the refusal to submit to the authority of others might have prevented the emergence of a central political power (Loncke 2015: 226f.), but they have also limited the power of political leaders to enforce decisions. Members of the study group speak quite openly and in a self-critical way of this problem and recognize its negative effects on their political and economic advancement. They say for instance that if the Tuareg are in a better economic and political position today, this is due to their ability to act collectively for their interests: ‘Mu kaɗai, ba mu iya haɗin kai’ (H) – ‘we are the only ones unable to act collectively’.8 Although the matter is rather controversially discussed (see Schareika 2007: 225ff.), it is recognized by many that the fragmentation of power has the effect of weakening the community: Today, the chiefs have multiplied. Nano has become an arɗo. Boyi has become an arɗo. They have become many. As far as I am concerned, I do not like the fact that we are all separated. I would prefer all of us, arɗuɓe of the Gojanko’en, to be together. (Laɓɗo Usman, June 2011) The lack of unity and the dominance of individual ambitions over collective interests also has a negative impact on the political participation of the Woɗaaɓe. Their prevailing absence from the new political offices created in the context of political decentralization, even at the municipality level, is not only due to their own reluctance and conscious abstinence, nor merely to marginalization resulting from long-prevailing evasive strategies (Boesen 2009). Several individuals from the study group have attempted without success to run for the 2011 elections to the municipal councils. In order to obtain a seat, a valid strategy would be for a whole community to corporately support one political party and make this support conditioned by the inclusion of one candidate from the community on the list of the party. In practice, however, this strategy of corporate political action often does not work out. Instead, individuals who aspire to political office often compete against one another by joining different parties, but then fail because of insufficient numbers of supporters.9 Such internal competition is common not only between individuals from different clans or maximal lineages, but also between individuals from the same patrilineage. This problem seems to be a more widespread phenomenon and not limited to the study group, as several similar cases from other groups of Woɗaaɓe and

114  Space, Place and Identity

other semi-sedentarized Fulɓe attest. In January 2011, I heard of a case from the Suudu Suka’el in Diffa, where candidatures for the municipal elections, and more generally the power play of party politics, were apparently used by different Woɗaaɓe leaders from one maximal lineage for fighting out an internal political power struggle, and where the occasion not so much provoked as revealed a cleavage within the lineage.10 VerEecke has observed a similar phenomenon in the Adamawa region of Nigeria, where internal rivalries and animosities led to the defeat of Fulɓe candidates in local government elections as many Fulɓe voted even against their immediate kin. To explain this, the author refers to a concept called ngayndi (translated as ‘wickedness’ or ‘being mischievous’), which can originally be understood as ‘a playful sort of competition, an attempt to thwart the success of another’ (1993: 156) and which today often manifests itself in ‘deliberate attempts to prevent the advancement of others’ (ibid.). According to VerEecke, this practice, although considered as ‘the opposite of pulaaku’ (ibid.) has entered the political arena in Nigeria as a political statement to self-seeking and corrupt candidates and thus paradoxically as a ‘call for a reversion to traditional Fulɓe ideals’ (ibid.: 157). While the concept is known in the study group11 and a similar attitude can be observed here in the political sphere as well, the interpretation that this attitude expresses a growing awareness of the corruption of political leaders who seek power for its own sake (ibid.: 156) does not seem to be applicable here. Rather than by altruistic concerns for moral control, the internal rivalries within the study group are motivated by goal-oriented considerations of new opportunities for resource appropriation that are associated with political offices. It seems, however, that the efforts for sustained political consolidation as larger units with overall representatives are not only inhibited by an individualistic strive for success and renown but also by the internal mechanisms and dynamics of social group formation that are based on opposition and that favour fragmentation and obstruct joint action. In more general terms, this means that the failure of many attempts at integration is not only due to the marginalization of an ethnic minority group, but also structurally determined. Although it is often recognized that consolidation is necessary for successful involvement in local politics and would thus be the better strategy to achieve political goals, this awareness does not seem to suffice to counter the tendencies of fragmentation. One effect of the aforementioned structural factors is that the original logic of development projects, as in the case example of the school, is bound to be disregarded and development offers regarded primarily in terms of the resources that they represent.12 The acquisition of projects is used as a political asset to consolidate one’s own position against others (see Schareika 2007: 233; 2011). As a result, instead of serving the group as a whole, the implementation of projects and the formation of associations that often go together with them (e.g. steering committees or cooperatives) often rather accentuate existing rivalries between

Consequences of the New Spatial Strategies  115

different political camps within groups and become a new arena in which these are acted out. While the emergence of associations has at the outset often been directly linked to the interventions of development projects that made them a requirement for the creation of infrastructure, today, such associations begin to be created by the Woɗaaɓe themselves in order to obtain project funds more easily. These structures are often referred to as civil society organizations (e.g. Boesen 2009). It is debatable whether this term is apt to describe their nature: although ‘civil society’ is generally conceptualized as autonomous of state structures, it is the product of an engagement that is explicitly political in the sense of Hirschman’s (1970) voice option (Azarya 1994: 97). In many cases, the interest of Woɗaaɓe in creating associations seems to be primarily economically motivated and can be interpreted as an opportunistic strategy to access the new resources of the development sector, rather than being politically motivated with a reflected aim of structurally enhancing the level of inclusion, integration and political participation. Among the Woɗaaɓe in central Niger, success in the creation of such associations has in certain contexts today become a new criterion for socio-­political status (Lassibille 2009: 319). The different Woɗaaɓe clans and sometimes clan-­ segments have organized themselves in a significant number of associations and even in an umbrella organization (Collectif des Associations des Eleveurs Nomades du Niger – Djingo) (Boesen 2009; Lassibille 2009). Within the study group, the necessary knowledge and skills to establish NGOs and associations have not yet been fully developed. Numerous attempts at associative self-­organization have been made, but so far with very limited success. While the necessary administrative procedures are certainly one obstacle (Boesen 2009: 79), the study context shows that the problems are also of a structural nature.

The Politics of Place At the beginning of the rainy season of 2011, which was a critical time because the hot dry season of 2010 had been catastrophic and everybody was eagerly awaiting signs of relief, I was in Zinder at the house of Baji Buuyo, when Alaji, Ɗawra’s eldest son, called him on the mobile phone from Ganatcha. Alaji reported an abundant rain shower and gave an enthusiastic report of the pastoral conditions in the area. When I discussed this with Baji later, he said that Alaji was only exaggerating in order to give a positive image of Ganatcha, because Ɗawra had an interest in others joining him. If more members of the Damergou faction decided to definitively move to Ganatcha, this would shift the power relations between the different local leaders within the lineage segment again to the advantage of Ɗawra. The discussions within the study group about the advantages and disadvantages of the respective sites, which I have documented in Chapter 5 in terms of

116  Space, Place and Identity

pastoral, economic and social aspects, are also of a particular political relevance. The political power relations between the leaders of localized lineage sections depend on human flows that are significantly conditioned by seasonal variation in local conditions. Nevertheless, efforts at influencing them through rhetorical strategies are common. The different local community leaders compete with each other about followers – a competition that is largely acted out with verbal means, by maintaining and spreading a positive discourse about one’s own and a rather negative one about other sites. The leaders pursue such inclusive strategies for different reasons. On the one hand, the administrative chiefs are paid for their service as tax collectors with a compensation that depends on the number of their tributaries. On the other hand, a greater number of followers increases the internal political influence of a chief (Schareika 2007: 190). In the case of newly-created centres, inclusive strategies are particularly relevant because the leaders depend on a substantial number of followers to consolidate their administrative status and thus substantiate their claim for leadership towards the state. Finally, a larger community also means a greater weight in the race for the new resources of the state and the development sector, e.g. when it comes to questions such as aid distribution, or the number of children necessary to justify the creation of a school. The competition about political followers is thus closely linked to concerns about resource appropriation. Although a larger group size may be advantageous for accessing the new resources of the aid and development sector, the pastoral resources, on the contrary, are very limited and have to be shared by more individuals if group size increases. The inclusive strategies that can be observed today are thus potentially problematic in the sense that they accelerate the pressure on the local pastoral resources and are thus questionable in terms of environmental sustainability.

Environmental Consequences of Sedentarization Systems of mobile pastoralism in arid regions have long been assumed to have a degrading effect on the environment by accelerating erosion and desertification (Aubréville 1949; Peyre de Fabrègues 1987; Arzika et al. 2007: 71; PSSP 2009: 15). One argument often brought forth against mobile pastoralists is that since they are not permanently attached to the land they use, they do not care to use its resources in a sustainable manner but tend towards abusive resource exploitation and overgrazing. Such assumptions were also expressed in Hardin’s (1968) classic problem of the ‘tragedy of the commons’. More recent research has shown, however, that local populations have generally been able to develop mechanisms to regulate the problems of communal resource use, and that the discourse on pastoralism as degrading the environment was in many cases rather politically motivated and aimed at promoting the sedentarization of mobile pastoralists (Homewood 2008: 79ff.).

Consequences of the New Spatial Strategies  117

In Niger, pastoralists are regularly accused of accelerating the prevailing deforestation by abusively cutting branches off trees to feed their animals with the leaves (fourrage aérien). More recent studies have shown, however, that the cutting of branches, if rationally executed, can even be of benefit for the trees, and that pastoralists are indeed often conscious of a sustainable and rational use of the resources to which they habitually return (ZFD 2008: 52). More generally, it has been demonstrated that livestock keeping is the less environmentally degrading, the more mobile it is (Dodd 1994; Bonfiglioli 1998; Schwartz 2005; Hiernaux and Maidaji 2006; Arzika et al. 2007; ZFD 2008). Similarly, pastoralist mobility has often been regarded as irrational and incoherent. Nomadism was seen as a ‘traditional’ way of life that followed an emotional ideal (‘élevage contemplatif’),13 but was economically inefficient (Peyre de Fabrègues 1987: 310). The solution generally proposed was a sedentarization of the herds. More recent work has shown that the movements of the herds are not random and irrational, but based on well adapted strategies (Maliki 1981: 100ff.; Galaty and Bonte 1991; Bonfiglioli 1998; Niamir-Fuller 1999; Khazanov and Schlee 2012; Catley et al. 2013; Schlee 2013c). The Woɗaaɓe’s pastoral strategy, which builds on the constant adaptation of the numbers of cattle to the available resources and on the use of mobility to level out seasonal shortages (Sandford 1983: 38), makes them less vulnerable to the endemic drought in the Sahel region and is thus also an important risk-reduction strategy (Bovin 1990). The more mobile the herd, the better it can generally cope with localized pasture shortages. For this reason, mobility also supports herd reproduction. In Niger, mobile pastoralism is thus significantly more productive than stationary livestock keeping (Colin de Verdière 1998: 147), and overall of remarkable significance for the national economy: as the open range system allows for low-cost production, regional exportation to the coastal countries is highly profitable and makes livestock the second important product of exportation after uranium (Bayard et al. 2001; Mansour and Tan 2008: 8). Numerous livestock projects and pastoralists’ associations have taken up this line of argumentation and contributed to rehabilitating mobility as a rational mode of pastoral resource use (Maliki 1981; Anderson 2007; ZFD 2008; PSSP 2009). One consequence of these efforts was the integration of two important principal statements about pastoral mobility into a new pastoral law adopted by the Nigerien government in 2010: (I) mobility constitutes a fundamental right for pastoralists, and (II) mobility is recognized as a rational and sustainable mode of using the pastoral resources (Republic of the Niger 2010, par. 3). In the light of such discussions, the development towards spatial fixation that is characteristic of many Woɗaaɓe today might look rather questionable, both economically and in environmental terms. However, such a conclusion needs some qualification. As pointed out above, the higher degree of residential stability does not mean that the mobility of the herds is given up. Rather, the

118  Space, Place and Identity

separation of household and herds is an indicator of the impossibility of giving up mobility altogether. The biggal system is an adaptive strategy to avoid disadvantages that would result from the reduced mobility. Nevertheless, the sedentarization process has induced important changes in the pastoral economy. The formerly highly specialized system of cattle breeding has, especially in Ganatcha, made way for a more diversified system with a strong component of caprines. The fact that the stock of goats is permanently kept in the vicinity of the main camps indeed constitutes a serious risk of degradation of the pastures over time. This is further accentuated by the prevailing inclusive strategy aimed at attracting more followers. The risk of resource overexploitation, a consciousness of which would pave the way for more exclusive strategies to secure a maximum of the scarce resources for oneself (Schlee 2004; 2008), is not yet a matter of discussion. For the pastoralists in the research area, resource protection implies first of all securing spaces for pastoral vocation and securing their own access to them. The concept of terroir d’attache provides a legal framework for such a rationale. However, the decree on terroirs d’attache transfers not only priority use rights, but also clearly defined responsibilities to the users. While it confers responsibility on a particular community, withdrawal of the priority rights in cases of abuse or neglect of management duties and protective responsibilities are equally mentioned (Republic of the Niger 1997a, par. 12). In general, granting priority rights to particular communities seems likely to decrease the risk of a commons tragedy as it creates perspectives of long-term view and thus gives incentives to actors to act responsibly (see Ostrom 1999: 495). With regard to the Woɗaaɓe in Ganatcha, it can be argued whether their spatial fixation has led to an increased sensibility to the question of environmental resource protection. It is interesting to observe that individuals from the community begin to adopt a position of accusing pastoralists on their passage of abusively cutting wood in the pastoral reserve. The process of local stabilization has here led to a seemingly paradoxical adoption of a critique as it had earlier often been voiced against themselves by the environmental authorities. Although this could be interpreted in terms of a rising ecological consciousness of the need to protect resources, it might also be a reflexive defence of the local resources, which are increasingly perceived as the property of the localized community, against strangers. So far, such new sensibilities in Ganatcha have not led to the exclusion of other pastoralists from access to the resources, as has been reported with relation to other areas in Niger (Anderson and Monimart 2008/9a: 77).

Sedentarization and Culture Change Ingold (1987: 169) distinguishes between two qualitatively different forms of sedentarization processes in nomadic societies, the first of which is characterized

Consequences of the New Spatial Strategies  119

by a reduced mobility and a reduction of animal husbandry in favour of cultivation. This form of process is generally reversible since it implies mainly a shift on a continuum. The second form, in contrast, comprises a profound structural transformation. Ingold’s differentiation is based on his definition of nomads as characterized by their social relations being based on the possession and the distribution of movable property, i.e. animals, in contrast to farming populations, where they are based on the possession and the distribution of land (ibid.: 170). In sedentarization processes of the second kind, this basic structural feature tends to change, which, according to Ingold, makes processes of this kind less likely to be reversible, since they imply a change not of degree but of kind (ibid.). As described in Chapter 2, Woɗaaɓe pastoralism has been mobile to highly varying degrees over the course of history. This capacity to flexibly move from mobility to sedentarity and from economic specialization to diversification is a typical feature of many Fulɓe groups (Braukämper 1971: 62; Kintz 1985: 98) and pastoralists more generally (Bonte and Galaty 1991: 6), and contemporary Woɗaaɓe point it out as a virtue (Loftsdóttir 2002a: 12f.). Sedentarization processes among the Woɗaaɓe have generally been reversible, and the oscillation between mobility and sedentarity can aptly be described as a ‘pendular phenomenon’ (Dupire 1962: 39). The question of mobility or sedentarity cannot be answered definitively but only in a diachronic perspective. For the time being, most of the land that the Woɗaaɓe use is not fully appropriated in the sense of an acquisition of property titles. However, the trend towards investing in pastoral wells as a strategy of assuring use rights over the surrounding land resources seems significant with regard to the question of the reversibility or irreversibility of sedentarization, i.e. the question of a transition in degree or kind. Pastoral wells are either individually or collectively owned, and they can be handed down from one generation to the next, together with the use rights to which their possession entitles. They thus represent a significant family capital and link their owners to a place. The sedentarization processes are too recent to observe mechanisms of establishing social relations by passing on land or spatially fixed property on a significant scale. So far, the distribution and passing on of movable property (i.e. animals) continues to be a principal means for the constitution of social relations, not only through the aforementioned animal loans (haɓɓanaaye) but more generally through the circulation of animals as a source of cohesion between social groups (see Chapter 10). First signs of a gradual change with regard to these matters might, however, be recognized in Ganatcha: Arɗo Umaru, upon his settlement in Ganatcha, obtained a large surface of farmland from the chef de canton. Both Ɗawra and Ali and their respective followers, by contrast, have obtained their farmland not directly from the chef de canton but via informal redistribution by Umaru, who apparently uses this privileged position to consolidate leadership claims. If the sedentarization process continues in the sense of

120  Space, Place and Identity

a more definitive fixation, including land appropriation with written legal titles and the construction of permanent houses, a systemic change in Ingold’s sense, not of degree but of kind, does not seem improbable. Salzman’s (2004) distinction between ‘tribal’ and ‘peasant’ pastoralists might also be of interest for an evaluation of the transformation process in which the Woɗaaɓe in the study region are involved. The former are organized as politically independent entities; the latter are politically and socio-culturally incorporated into a state-dominated society (2004: 106ff.). While, over most of the twentieth century, the pastoral Woɗaaɓe could qualify in this sense as ‘tribal pastoralists’ (see Schareika 2007: 117), one might argue whether the recent processes of sedentarization and increasing incorporation into state structures can be interpreted as a step towards their becoming more ‘peasant’. However, although the creation of centres was, significantly, a response to requirements by state and development actors, this response remains in most cases selective and is widely motivated by opportunistic motives of resource appropriation. One might therefore argue that the pattern underlying the spatial fixation and the creation of centres shows a continuity with the rationale of mobility insofar as it reflects the nomadic ability to react flexibly to the highly variable distribution and availability of vital resources (see Grémont 2011: 181, 2014: 33). Today, such vital resources for pastoralists are not only water and pasture, but also education and the offers of development programmes.

Notes   1. Alaji Ɗawra, June 2011.   2. For a similar situation among the Tuareg in Northern Mali and Niger, see Grémont (2011: 183), and among the Woɗaaɓe in central Niger, see Boesen (2009: 78).   3. Tuuɗi Abdulaye, December 2010.   4. Witnessed in Abdenaser, Damergou region, October 2011.   5. On the history of decentralization in Niger, see Salifou (2008).   6. Information confirmed by Nano Buuyo (December 2011).   7. For a transcription of the complete account, see Köhler (2017a: text 11).   8. Guntu Ɗawra, January 2014.   9. Amadou Siddo, personal communication, February 2011. 10. Gorjo Riima, Woɗaaɓe Suudu Suka’el Diffa, personal communication, January 2011. 11. In the Fulfulde of the study group, ‘ngandi’ translates as ‘hatred’ rather than as ‘wickedness’. The attitude of trying to prevent the advancement of others would in general rather be grasped with the concepts nawta or kiisi (‘envy’, ‘jealousy’). 12. In this regard, the ideologies of development, when offered to the Woɗaaɓe, often suffer the same fate as religious ideologies. On the opportunistic embracing of Christian missionary efforts comprising a relief component by Woɗaaɓe in the 1970s, see Paris (1981). 13. See Herskovits’ (1926) ‘cattle complex theory’ and its discussions in the literature on pastoralism (e.g. Bonte and Galaty 1991: 9f.).

Part IV

Si’ire Appropriating the City

Chapter 7

New Resources in the Urban Space

Urban work migration first became important among Woɗaaɓe during and after the severe drought years of the early 1970s (Maliki et al. 1984; White 1987). Since then it has developed first into a seasonal phenomenon of cyclic recurrence and over time in many cases into a permanent condition. But even before, Woɗaaɓe women could generate revenues in times of economic hardship through undertaking service work for villagers or urbanites, e.g. by mending calabashes, plaiting hair or pounding millet (Dupire 1962: 127). Urban wage-labour thus did not mean a categorical break, but rather a gradual development in response to the new possibilities for economic diversification and supplementary activities outside pastoralism that the city offered. As pointed out above, the flexible response to ecological hazards by economic diversification has been characteristic of the Woɗaaɓe throughout their history. Although such flexibility is often particularly associated with pastoralists, it is also a general phenomenon in the region: not only Niger but the Sahel more generally is characterized by ‘cultures of migration’ (Hahn and Klute 2007) and Woɗaaɓe work migration must be understood in this wider context of rural–urban migration in the region, both on an internal and on a cross-border level (Rain 1999; Oumarou 2012: 77f.; Boesen et al. 2014; Marfaing 2014a). Within the study group, urban work migration first became significant after the 1980s droughts, and more particularly since 1984. Many Woɗaaɓe were severely affected and moved to the provincial capital of Zinder, where they received relief food and started to engage in wage-labour.1 Today, Zinder remains one of the most important targets for migration, together with Kano, Abuja and other cities in neighbouring Nigeria. Maliki et al. identify work migration as the most important coping strategy of Woɗaaɓe in the face of animal loss (1984: 481ff.). About 65 per cent of the

124  Space, Place and Identity

households surveyed by this study in Central Niger had members engaged in migrant work in the early 1980s. Even in the absence of statistical data it is clear that the prevalence of migrant work is even greater today (Boesen 2010: 37). Within the study group, it is a ubiquitous phenomenon and concerns practically all households to some degree. In numerous cases, the pastoral economy has developed into a subsidized system in which remittances from urban migrants play a significant role.2 Although conceived as a strategy of herd reconstitution, some authors have argued that in practice, work migration in the 1970s and 1980s was in many cases rather counter-productive because the workforce drain had a negative effect on herd management and thus in the long run on household viability (Maliki et al. 1984; White 1987; see also Hampshire and Randall 1999: 368). A major constraint was indeed that in contrast to farming systems in the region, which have extended slack periods that allow for the absence of workforce without negative consequences, mobile pastoralism in the Sahel is work-intensive throughout the year (Maliki et al. 1984; White 1987). Both among pastoralists and other Sahelian societies (Rain 1999), the major period for seasonal work migration was the hot dry season, when food shortages were most felt. For pastoralists, however, this coincides with the most work-intensive time of the year, because surface water is lacking, and water has to be drawn from deep wells. From this perspective, migrant work appears to have led impoverished Woɗaaɓe almost inevitably into a vicious circle. Although well aware of the negative effects of migrant work, however, they simply lacked better alternatives (Maliki et al. 1984: 481). While such an assessment of work migration among pastoralists, as being economically unviable and destitution driven, might be valid for acute situations of hardship, such as the major Sahel droughts of the 1970s and 1980s, it seems insufficient for an understanding of its contemporary role in the study context. The problem of a workforce drain caused by migration might still be relevant in some households. In most cases, however, the problem is rather the inverse: workforce is often abundant today. Due to generally high numbers of children per family in combination with reduced herd sizes, sons often do not inherit significant numbers of livestock from their fathers that would allow them to constitute their own herds and to establish independent households. As a consequence, adult sons today often stay in their parents’ households longer than in the past. In this situation, grown up sons are often even encouraged to leave for migrant work and thus contribute to the family economy – or at least relieve it through their absence.3 Also, it is not always the poorest who migrate, but sometimes rather those who can afford the absence of their workforce because they can rely on social networks to compensate for it (Hampshire and Randall 1999, 2005; Boesen et al. 2014: 3). Under such conditions, work migration can be a viable strategy of risk reduction by economic diversification. With regard to the study group,

New Resources in the Urban Space  125

individual economic situations of migrants vary greatly. Kinship-based social networks and the complementary division of economic tasks within minimal lineages are often important elements of economic strategies that include work migration as one component. For those who can rely on such networks, work migration is not merely destitution driven, but it can be an important source of complementary income. Most importantly, however, the dominant patterns of work migration have changed from seasonal migration, which was characteristic of the first generation of migrant workers (Maliki et al. 1984; White 1987), to more permanent urban employment on a contractual basis. Some migrants from the study group have been staying in town for as long as twenty years. They might return seasonally to their pastoral home camps, though usually only for short visits. While individual migrants are almost exclusively male, many migrate with their nuclear families. Women generally do not migrate on their own, but once they arrive in the city with their husbands, they often pursue their own activities to gain additional income. Moving to town with entire families has been interpreted as a strategy to relieve the pastoral economy at home (Boesen 2004a: 214). With regard to the study group, a further reason are longer-term sojourns in town, which make migrants more inclined to migrate with their families than they would be in the case of seasonal short-term migration. In fact, virtually all Gojanko’en migrants who have stayed in Zinder for a more substantial period live there with their spouses and children, if they have any. In these cases, migration has led to a relatively permanent relocation of their centre of everyday life. An urban sub-community has emerged; children are born, and successive generations remain attached to the city.

Urban Activities The Woɗaaɓe have two different terms for ‘work’: kuugal and aikiwal. The Fulfulde term kuugal applies to all sorts of day-to-day tasks, e.g. female household tasks or the male activities of watering the herds. The other term, aikiwal, is a derivation from the Hausa term aiki (work).4 Although the semantic fields of the two terms overlap to a certain extent, the second is particularly used for wage labour. It is plausible that a loanword is used to designate work that conceptually differs from the kinds of activities designated by the emic term. For my interlocutors, urban work was typically characterized by a fixed salary and regular working hours, a definition that owes a lot to the fact that the most common urban activity within the study group is the work as watchmen. Kuugal is therefore rather associated with hard physical labour, while for aikiwal this is often not the case. It has been pointed out that most of the urban activities Woɗaaɓe migrants typically engage in do not imply hard physical labour and can rather be

126  Space, Place and Identity

characterized as non-productive work, and in particular as service tasks. Some of the most often cited migrant activities like repairing calabashes and braiding hair (Dupire 1962: 127; Maliki et al. 1984: 490), selling tea (Loftsdóttir 2001c: 9) or traditional medicines (Boesen 2004a, 2007a, 2007b, 2008a, 2010), or guarding houses are cases in point and seem to confirm an image of the Fulɓe as characterized by their disdain for any kind of physical labour not directly related to pastoralism (Boesen 1996: 193; 2004a: 214; 2010: 35; Bierschenk 1997: 69f.). In some regional contexts, this attitude seems to have generated a veritable discourse of weakness, i.e. the strategic cultivation of a self-image as incapable of hard physical labour (e.g. Bierschenk 1997: 70; Diallo, Guichard and Schlee 2000: 233f.; Schlee 2011: 173). Within the study group this is not the case and cultivating such an image would in fact be counter-productive as a strategy for getting the highly aspired jobs as watchmen, which rely rather upon a reputation of fierceness and bravery. Physical labour is not a principally avoided, nor is it per se regarded as shameful. Apart from taking up agriculture, members of the study group have been employed as labourers in the construction of roads and pastoral wells, and urban-based women are engaged in all kinds of domestic service tasks. That hard physical labour is preferentially avoided if attractive alternatives are available, however, is hardly surprising and not particular to the Woɗaaɓe or Fulɓe. One particular kind of service activity that Woɗaaɓe engage in is the selling of traditional medicines, herbal potions and magical charms – an activity that leads to important seasonal migration movements among the Woɗaaɓe in central Niger (Boesen 2004a, 2007a, 2007b, 2008a, 2010). While it also plays a certain role in the study group (Köhler 2017a: text 15), it is of less statistical importance here. Boesen (2010: 37) mentions the case of a group in which more than 50 per cent of the adults regularly engaged in trade journeys of several months. The trips of Woɗaaɓe from the Zinder region, typically to the areas of Kano or Abuja in Nigeria, rarely exceed a few weeks. A pronounced specialization of women in these activities (ibid.: 36), cannot be observed either. With regard to trade in medicine, the Woɗaaɓe, generally looked down upon by urban populations as ‘savages’ (Bovin 1985), seem to profit from the prevailing stereotypes: due to their life in the bush they are associated with an esoteric knowledge of medicinal plants (Boesen 2004: 214, 218). Medicine trade is sometimes combined with other kinds of migrant work. Migrants may profit from social contacts in towns to sell their products and some can make considerable additional earnings.5 Rather than being primarily destitution driven, medicine trade is thus an example of a kind of migrant activity that is an opportunity for those who can afford a temporary absence from the family herd or who are anyway staying in town. The example of Boyi Egoyi (see Köhler 2017a: text 15) is a case in point. Although he is comparatively well off and highly respected as an owner of considerable herds, he regularly embarks

New Resources in the Urban Space  127

on trading journeys of several weeks to Nigeria. He reinvests the output into his herds, which prosper, and the work migration thus adds to his wealth and social status rather than being a coping strategy in the face of crisis. The work as watchmen (gaadi)6 is a niche of urban employment that has typically been occupied by Woɗaaɓe – and originally to an even greater extent by Tuareg (Waibel 1998) – since the 1970s, when primarily young men went to the urban centres in the south of Niger and in Nigeria to work for individuals or companies as guards. Within the study group, this is clearly the most widespread form of migration work. The particular success of Woɗaaɓe and Tuareg in getting jobs as watchmen has been explained by the ‘ferocious’ image associated with them due to their turbans and swords (Maliki et al. 1984: 489), or with qualities such as attentiveness and bravery associated with a life in the wilderness (Boesen 2004a: 214). The fact that Woɗaaɓe migrants predominantly work as watchmen, without access to better qualified positions, has also been considered as evidence for their marginalization and lack of education due to their long-time strategy of state evasion and their refusal to integrate into modern state structures. The work as watchmen has thus for the most part been described as a badly paid menial activity (Maliki et al 1984: 488) and even as ‘psychologically killing’ (Bovin 1990: 38). This picture, however, seems one-sided. First, even badly paid employment can, as a relatively reliable source of income, significantly help to level out seasonal shortages in the pastoral economy (Hampshire and Randall 1999: 378). Second, the work as watchmen, which has never been analysed in terms of its economic rationale, cannot generally be considered as ‘badly paid’ when compared to average salaries of Nigerien middle-class workers. Although the labour input expected from watchmen in Niger can differ to some degree, on average their job consists mostly of being present during working hours. Receiving friends or relatives for company is generally admitted. In most cases, the tasks do not involve physical labour or only minor tasks, such as small errands or routine gardening tasks. Of course, night-watchmen are expected to stay awake during working hours, but in practice many private employers tolerate their sleeping at the door, on the premise that they stay alert and wake up at the slightest noise. Companies and international organizations apply stricter rules and principally prohibit sleeping during working hours, which Woɗaaɓe often consider as hardship (Köhler 2017a: texts13,7 and 14,31). In order to realistically assess the economic potential of a position as a watchman, the necessary work input must be put into relation to the salary a watchman is able to earn. The latter can indeed vary considerably, and generalizations are thus problematic. At the lowest level are the watchmen of market stalls, who are often paid on an unofficial and irregular basis. The salaries of watchmen who work at local private households, e.g. better off Hausa traders, varies between approximately 10,000 to 15,000 FCFA (ca. 15–25€) per month,

128  Space, Place and Identity

which is a rather modest income. Besides, the working conditions are in many cases informal and precarious. Theoretically, employers have to pay taxes and social security contributions for their employees, who in turn profit from family allowances, a partial restitution of their health expenses, and an indemnity at the end of their work-contract. In practice however, many workers are not officially registered. While employers often seek to minimize costs by avoiding registering their personnel, Woɗaaɓe employees often complain that the social security agency refuses to disburse the indemnities to which they would be entitled. Their distrust, fuelled by negative experiences due to their widespread illiteracy and lack of understanding of administrative procedures, makes Woɗaaɓe migrants prone to accept unofficial work contracts on the promise of a slightly higher salary. The most attractive positions as watchmen are with European expatriates, international organizations and larger scale private companies, where monthly salaries can amount to 35,000–45,000 FCFA (ca. 50–70€). Expatriates and international organizations are generally well reputed for providing fair work conditions on the basis of a fixed contract and for paying salaries regularly and correctly. In addition to the salary, private employers often provide free accommodation. While the form of the latter can range from rudimentary shelters to rather elaborate concrete houses, the watchmen, and sometimes their entire families, benefit from free electricity and access to a water tap on the compound. Although this might seem basic, it easily surpasses the standard of a majority of Nigeriens. The full economic potential of the work as a watchman becomes transparent only if one considers the possibilities for accumulation of additional income-­ generating activities that the nature of the work allows for. This is best demonstrated by an example: Taafa Buuyo has been working as a watchman in various cities – Kano, Niamey, Zinder and Diffa – for many years (see Köhler 2017a: text 13). In 2011 he worked as a watchman for a private school in the daytime – a job that he owed to his long-term relationship with French expatriates. His salary was 35,000 FCFA. The minimal work input that this work required allowed him to pursue a second contractual activity – as a night watchman for a German development worker – with an additional salary of 45,000 FCFA. In addition to this, his wife had employment in the pre-school section of the school as a nursery maid where she earned another 25,000 FCFA. In total, this provided the family with a regular monthly income of more than 100,000 FCFA (ca.150€), which exceeds the average salaries of a nurse or a secondary school teacher. Taafa’s leisurely working hours still left time for doing textile embroidery that he sold on and off via a network of expatriate acquaintances. Accommodation, electricity and water were provided for free on the school compound, the space of which was generous enough for Taafa to keep five goats. During the three months of school vacation, which coincide roughly with the agricultural period, Taafa was allowed to use the schoolyard (basically a large strip of sand) for subsistence farming of millet and beans.

New Resources in the Urban Space  129

Figure 7.1  Urban dwelling in a construction site – Zinder, 2011 (photo: F. Köhler).

Although Taafa’s example is not generalizable, it is not a singular case within the study group either. When I lived in Niger between 2004 and 2010, it was not unusual that a watchman who installed himself with his family on the compound of his employer was doubly employed as a watchman for night and day, thus also doubling his salary. The example shows that the work as watchmen for expatriates can, under certain conditions, come close to optimizing the input/outcome-ratio of wage labour and can make work migration a sometimes very attractive option. One important condition that made this activity so economically profitable was the massive presence of international development organizations in Niger during the past few decades, and the fact that they have created salary conditions for watchmen that are largely disconnected from the conditions for jobs of comparable responsibility and labour demand on the local employment market. In Zinder, the Gojanko’en Kuskudu have been especially well-positioned, over roughly three decades, in the job market for watchmen at expatriates’ homes. Their working conditions are, however, not representative of all the Woɗaaɓe migrants in Zinder. Members of other Woɗaaɓe clans, notably the Jiijiiru and Suudu Suka’el, who outnumber the Gojanko’en in Zinder, also hold positions as watchmen, yet typically rather in the market or in downtown shops, where they have smaller salaries and face more difficult working conditions. Also, their jobs do not generally provide them at the same time with comfortable dwellings and free access to water and electricity. Some just make camp in unclaimed open spaces at the periphery of the city or in construction sites, where a plot owner might tolerate their presence as it reduces the risk of construction materials being stolen. Such informal agreements can sometimes give these urban dwellers an ambiguous status of quasi-watchmen, but their situation is characterized by

130  Space, Place and Identity

precariousness and informality. As merely tolerated dwellers, they depend on the goodwill of their landlords (Köhler 2016). However, even those with a work contract that grants them access to housing can be forced to move out if their contract finishes and the new tenant chooses to replace them. Many of the Woɗaaɓe migrants who have been staying in Zinder for any length of time look back on a history of frequent relocations from one construction site or empty building to another, living at one time in an old garage or being lucky enough to find employment that could provide them with decent accommodation for some time. One migrant, for example, lived in seven different dwellings in Zinder between 2005 and 2007, each time making a new home in uninhabited houses and unclaimed sites within the urban space. One might be tempted to interpret the urban placemaking patterns as a continuation of pastoral mobility patterns, arguing that even in town, the Woɗaaɓe remain highly mobile in order to be flexible enough to use the resources that the unoccupied niches of the urban sphere offer to their best advantage. Here, again, the characterization of Fulɓe as interstitial populations comes to mind (Stenning 1966; Diallo 2008; see also Köhler 2016: 107ff.). However, the intra-urban mobility is more forced than chosen, and the concept of interstitiality applies to Woɗaaɓe migrants rather in the less deliberate sense of marginality and exclusion. The placemaking patterns in the city resemble the non-territorial patterns of resource appropriation in the pastoral context insofar as in both cases the marginal position of the Woɗaaɓe limit the latter’s possibilities for a sustained access to spatial resources. Due to their financial constraints, urban Woɗaaɓe generally do not have the means to get access to proper housing by either obtaining, constructing or renting it. In the end, in both contexts the only options they are left with are on marginal, interstitial resources. In construction sites, public spaces at the urban margins, temporarily unclaimed plots or objects of land speculation, Woɗaaɓe migrants live rent-free but at the price of high uncertainty. Mechanisms for coping with the uncertainty resulting from the current situation of increased competition are also similar in both contexts. As described above in regard to the pastoral realm, assuring a permanent presence is aimed at securing claims for priority use rights over an area with good pasture resources. In the city space as well, a permanent presence is essential for securing claims: one migrant told me that while residing in an empty house in Zinder on the basis of an unofficial agreement with the owner, he once left for just a few days to accompany his family on a visit back home to the Damergou region. On his return, the house had been occupied by somebody else and it was lost for good.

Experts for Expatriates Arguably, the fact that the Gojanko’en Kuskudu in Zinder have been particularly successful in establishing themselves as watchmen for expatriates is largely a result

New Resources in the Urban Space  131

of historical coincidence. In other urban contexts the distribution of activities across clans is quite different. Gojanko’en from the Ader region, for instance, migrate to Niamey rather than to Zinder and have specialized in handicraft and souvenir-trade (Loftsdóttir 2000: 249f., 2001d; Boesen 2004a: 215). Once the first steps were made, however, the Gojanko’en in Zinder were able to develop social networks which helped to ensure a long-term access to this occupational niche and economic resource. The professional specialization and differentiation in the city context according to lineage or clan affiliation indicates the importance of kinship-based networks for getting employment. This also becomes evident from accounts of how people got their jobs. Often, newcomers rely upon friends and relatives who have already been working in town and who have thus been able to negotiate a position for them or introduce them to a potential employer whom they might know via their own employers. The patterns confirm observations from other contexts about the crucial importance of social networks for new migrants to get established and find work, accommodation and support in town (Hampshire and Randall 1999: 377; Rain 1999: 148ff.; Marfaing 2014a). This is one reason why kinship – next to friendship – plays a central role in the decision as to whether to migrate and where to migrate (Guichard 2014: 10). Many of the accounts that I collected among Gojanko’en confirm this typical pattern. Taafa Buuyo, for instance, owing to his brother’s connections, was twice offered a job as a watchman without even having applied for it (Köhler 2017a: text 13,13 and 13,20). The same, seemingly paradoxical, situation of being solicited instead of seeking out for employment was reported by Laɓɗo Usman: [A cousin of mine] asked me whether I was interested in a job, there was an Italian looking for a Boɗaaɗo to work for him. Originally, I was not looking for work, it was rather jokingly that I said yes. But he accepted me and I ended up staying with him for two years. (Laɓɗo Usman, October 2011) Such cases highlight the role of kinship networks for finding employment. From Taafa’s example, the translocal dimension of these networks also becomes evident: while Taafa was living as a pastoralist in the area of Salaga in 2003, his Zinder-based brother was able to organize employment for him in Diffa with the help of his relations to expatriates. The two examples also show once more that the economic role of urban work migration can differ significantly: initially a coping strategy in the face of crisis, it has over the years developed into more diversified forms in which both push and pull factors play a role. In many cases – especially for those who can rely on network ties that give them access to profitable jobs – it is an active response to economic opportunities (Hampshire and Randall 1999, 2005; Klute and Hahn 2007: 11).

132  Space, Place and Identity

Not only are social networks crucial for gaining access to urban employment, but new social networks can also develop out of the employment situation. In the case of the Woɗaaɓe in Zinder this is most notable in the form of their relationships with their employers, who are often Western expatriates (Nasaranko’en).7 The expatriates’ often positive perception of their relationships with their Woɗaaɓe watchmen made it possible for the latter in many cases to give other lineage mates access to jobs as well. Ultimately, good employer–employee relations established by kin are even more relevant for getting jobs than the kinship network alone, since decisions about affectation are not taken by the Woɗaaɓe, but by their employers. The fluctuation of expatriates who generally stay only for a few years continually offers new, albeit temporally limited, job opportunities. Upon the recommendations of their predecessors or other already established expatriates, who expressed their satisfaction with their Woɗaaɓe watchmen, newcomers in many cases explicitly asked for Woɗaaɓe who might want to work for them (see Köhler 2017a: text 12,07). At the outset, the relations between expatriates and their Woɗaaɓe watchmen often have the character of patron–client relations, in which the expatriates, with their wealth and social prestige, tend to patronize their Woɗaaɓe employees. In return for material resources, the latter might take ‘their’ expatriates home to experience a night in a pastoral camp or to see one of the spectacular dance performances. The account of Laɓɗo Usman (Köhler 2017a: text 12,06) can once more serve as an example: he was regularly invited by his employer to join him for meals; they travelled together when the employer had to travel; and the employer was interested in experiencing the cultural life of Laɓɗo’s ethnic community. In this particular case, the distribution of roles of employer and employee eventually became blurred: the relationship was hardly perceived as work by Laɓɗo, but rather like sharing a part of the employer’s life as an equal: ‘I was not [like] a watchman, but rather like the landlord’ (‘Ni ba mai gadi ba ne. Kamar mai gida ne.’ – H).8 This, admittedly ambiguous, employer–employee relationship lasted for six years, but even at the time of my fieldwork, more than ten years after the employer’s definite return to Europe, he still paid the school fees for Laɓɗo’s eldest son and regularly sent significant remittances to support the family. This shows that relations with expatriates can, even after the latter’s return to their home countries, continue to function in a resource-providing sense as translocal security networks, and even be extended to the next generation. The networks thus have a potential for expanding some Woɗaaɓe’s social space to the transnational and global level. Although particularly pronounced in the case of Laɓɗo, the relationships between Woɗaaɓe watchmen and their expatriate employers often develop beyond the basic elements of patronage and reciprocal exchange towards a more or less strong component of friendship, which, as recent research has demonstrated, often cannot clearly be separated from the former (Guichard et al. 2014). It has

New Resources in the Urban Space  133

been pointed out that a certain degree of instrumentality does not contradict, but can rather be one of the basic aspects of friendship (Guichard 2014; Reyna 2014). The Woɗaaɓe actively invest in relations with their employers and thus build up ‘social capital’ (Bourdieu 1986). In a similar sense, their culture functions as capital for the Woɗaaɓe – a resource that can be mobilized to acquire other desired resources (see Marfaing 2014a: 43). By sharing it with expatriates, this cultural resource can be exchanged for material resources on a basis of generalized reciprocity that can partially level out the asymmetry of the relationship. This allows for a win-win situation that makes the association attractive for both sides: the Woɗaaɓe watchmen benefit from fair working conditions, financial generosity beyond the salary and eventually friendship; the employers benefit from social contacts beyond their often narrow expatriate communities, participation in a culture that is regarded as exotic and fascinating, and eventually also friendship. The emotional aspect of such friendship is also particularly relevant in the migration context, which, for the expatriates and their Woɗaaɓe watchmen alike, is characterized by insecurity, change and transition (Grätz et al. 2003: 17). But why should the Woɗaaɓe be particularly well situated to entering into privileged relationships with expatriates? In this context, it should be noted that they are not unique in this regard: the Tuareg are in a widely comparable position towards expatriates. I have already explained how, as medicine traders and as watchmen, the Woɗaaɓe are able to turn ascribed stereotype images of their culture into an asset: an assumed knowledge of the bush and its plants in the first case and the assumed bravery and alertness of herders in the latter. The Woɗaaɓe’s particular relationship to expatriates similarly seems to be based on a cultural stereotype, in this case a mutual attitude of positive racism. While their image as peculiar and ‘different’ is rather negatively perceived by many Nigerien peasants as ‘savage’ (Bovin 1985), it is rather admired by tourists and expatriates (Bovin 1998; Boesen 2004a: 218f.). Cultural characteristics can be valued very differently: what is marginalizing in one context may lead to success in another (Brickell and Datta 2011: 12). The peculiarity of the Woɗaaɓe seems to be especially appealing to, and compatible with, western expectations of ‘exotic’ culture and evokes an image of ‘civilized savages’ (Boesen 2004a: 218), very similar to that often maintained of the Tuareg. It has been remarked that being considered different and strange was not new to the Woɗaaɓe when Westerners first regarded them thus, but it was new to them that they could turn this perception into profit (ibid.: 219; Boesen 2009: 82). Many Woɗaaɓe have learned to capitalize on the resource that their cultural identity constitutes, and to react strategically to the expectations of Westerners. They have become experts for expatriates. Inversely, many Woɗaaɓe consider Westerners as similar to themselves. In contrast to the local peasants (Haaɓe), they are not categorized as ‘other’ in the sense of a dichotomous opposite. A self-perception of Woɗaaɓe as ‘light-skinned’

134  Space, Place and Identity

and ‘beautiful’ in contrast to the ‘dark-skinned’ Haaɓe is prevalent, and just like the Tuareg or Arabs, Westerners qualify as similar for reason of their light skin colour (Dupire 1962: 322; Loftsdóttir 2002c: 311). Objectively speaking, many Woɗaaɓe have a rather dark complexion and sometimes acknowledge this, albeit with some regret. Most, however, would hardly admit to being darker skinned than any particular Kaaɗo. Stereotypes are used to maintain boundaries, and emphasize, sometimes artificially, difference. Earlier traditions of Fulɓe research were characterized by similarly racialist projections: the Fulɓe were imagined to be of a ‘nobler race’, and as originally ‘white’ (Amselle 1998[1990]; Bierschenk 1997: 8–9). Hence, Fulɓe self-­perception corresponded to outside perceptions, which is the case also in contemporary encounters between Woɗaaɓe and Europeans in Niger (Boesen 2004a: 219). While, as outlined in Chapter 4, the category Haaɓe is ‘othered’ along a highly essentialized boundary, the continuum of identity categories allows for the integration of the category of Nasaranko’en, based on a postulated similarity and the ideological construction of an extremely far reaching ­identification (Schlee 2008). The phenomenon of close relationships to expatriates is not unique to the study group but can be found elsewhere as well (Boesen 2009). While contacts with resident expatriates are dominant in the study region, in central Niger, where tourism, especially during the early 2000s, played a much greater role, contacts with visiting tourists led to similarly long-lasting relationships with a significant economic potential (Lassibille 2009). Relationships with expatriates can propel Woɗaaɓe from the margins of Nigerien society right into its centre. It often gives them access to the wider social networks of expatriates and facilitates the access to other economically significant actors such as development organizations and other donors or decision makers who would otherwise be difficult to reach for many Woɗaaɓe. Networks with expatriates have thus become an important means to get access to infrastructure and development funds and some Woɗaaɓe have indirectly become ‘development experts’ (Boesen 2009: 89f.). In all the three rural sites of the Kuskudu in the study region – Salaga, Ganatcha and Ngel Tireeji – such networks have been at the basis of various projects. Either, these projects have been directly financed by Western friends and acquaintances, or the latter have helped by elaborating project proposals and thus convincing external donors to consider the sites of the Woɗaaɓe in their intervention programmes. Relations with expatriates can thus have an impact well beyond the inter-personal level, ultimately benefitting the immediate or extended family or even the wider lineage community. For assessing the significance of the particular relation between Woɗaaɓe and expatriates, Granovetter’s (1973) concept of the ‘strength of weak ties’ can be helpful. The ties within social groups or networks are often multiple and interconnected, and thus ‘strong’, yet rather limited in their reach. Those that bridge

New Resources in the Urban Space  135

different social networks might look ‘weak’ at first sight, but their strength lies precisely in the fact that they connect different groups, even across social classes. Werbner (1999: 27) has pointed out the importance of such ties in migration contexts. Applied to the case of Woɗaaɓe urban migrants, their ties to expatriates are weak in the sense that their social interaction is initially generally restricted to a limited context, e.g. their relation of employer and employee. The social networks of both are otherwise largely separate and the relations are initially highly asymmetrical. On another level, however, these ties are strong because based on the mutual affinity between the two groups, they often bridge this initial gap and lead to relations that last beyond the duration of the employment and involve significant flows of material and immaterial values. The case demonstrates the potential of inter-ethnic friendship for enhancing the social power of actors by facilitating their access to resource-providing institutions (Reyna 2014).

Expatriates as a Resource ‘When we started to work in Zinder, I was the first. From all of us who went to town, I was the first to work for a European’.9 With this claim, Laɓɗo Usman begins his half autobiographical, half historical account of the beginning of migrant work among the study group.10 Whether he was really the first or not – what seems essential is that his story is plausible in the historical context. During the 1980s drought, many Woɗaaɓe Kuskudu moved to Zinder and, among other forms of migrant work, began working as watchmen for expatriates. What makes the quote interesting, however, is that it is another example of firstcomer rhetoric to claim status superiority. Like Ɗawra, who claimed that his fathers were the first Woɗaaɓe to build a well in the Agadez region and thus made the region’s resources accessible to other Woɗaaɓe, Laɓɗo similarly emphasizes that he was able to find jobs with expatriates for several of his relatives: I went and asked [my expatriate employer for jobs]. My younger brother Haarika got a job, and after him Atta, the younger brother of my father. . . . Let me count: one, two, three . . . eight. I put eight people into jobs with expatriates [at that time]. (Laɓɗo Usman, October 2011) In other words, Laɓɗo also claims that he made the new urban resource – work for expatriates – accessible to other lineage members. The case is another example of how status and prestige can be gained through sharing resources, and of how claims for such status are verbally maintained with rhetorical means. The significance of this pattern becomes even more obvious from another example: in July 2011, I followed a discussion in Ganatcha between three sons of Ɗawra, on the one hand, and some of his adult grandsons, on the other. The discussion evolved from the reproach made to the young adult men that they did

136  Space, Place and Identity

not contribute to the family economy by engaging in work migration. Upon this accusation, one of them, a son of Ɗawra’s eldest son Alaji responded that their fathers themselves mostly depended on subsidies that they regularly received only thanks to the close relationship that one of their Niamey-based sisters entertained with a European woman who had, more than ten years ago stayed in Niger. The community indeed received significant remittances from this source each year, which were used to buy millet that was shared among the households of Ɗawra and his children. The funds for Ɗawra’s well in Ganatcha have also been raised with the help of this friend. Interestingly, the heated discussion that followed immediately focused not on the reproach that the fathers themselves depended on subsidies rather than their own productive work, but on the contention that these subsidies were the result of connections that a sister of them had established. This perspective was quickly and fervently corrected by Alaji’s brothers: Alaji had worked as a driver for a project that the expatriate woman was involved with. When she visited the project area, Alaji served as a driver and field guide and organized an overnight visit to the family’s pastoral camp where the friendship originated. Only later did Alaji’s sister develop the relationship further. Alaji’s brothers thus defended Aliji’s firstcomer claim by re-establishing a version of orally transmitted, recent family history that had been put into question by one of his sons. The rivalry that can arise even within the immediate family about such claims of firstcomer status might appear to be in contradiction with the often-maintained image of the ‘communal ethos of the lineage’ (Boesen 2010: 32). Despite the principles of solidarity and amity within the lineage – in the sense of Fortes’ (1983 [1969]) ‘prescriptive altruism’ among kin – competition is just as prevalent. Lineage mates in need are generously supported by those who have the means, and to be in a position to give such support confers status. The internal competition is less about resources as such than about the status that the sharing of resources confers. Therefore, even if the resources that an expatriate can provide are put to the benefit of the community, the question of who first initiated the link to the respective expatriate can be the object of heated debates. Such claims are not without social relevance: just as status is achieved by pioneers and migration leaders accessing new pastoral resources for the benefit of their larger lineage groups, it can today also be achieved and consolidated over the question of how new resources, such as relations to expatriates and the material benefits deriving from them, are accessed and made ‘exploitable’ to other lineage members. In central Niger, it has been observed that relatively young Woɗaaɓe who have the necessary skills for networking with Westerners are today also politically influential and constitute a new category of leader personalities (Boesen 2009; Lassibille 2009). To a certain extent, the successful political career of, e.g. Nano Buuyo as the leader of a rural centre has also been fuelled by his extensive networking with expatriates and his capacity to translate

New Resources in the Urban Space  137

these relations into economic output. The not merely economic but potentially also political significance of personal networks with Westerners is also the reason why information about them is sometimes kept confidential and cautiously controlled (Lassibille 2009). Woɗaaɓe who are involved in networking with expatriates and successfully take positions of brokers for their communities can thus considerably increase their social status. Networking with Westerners has become a basis for social mobility.

Notes   1. See Köhler (2017a: texts 3,13 and 12,3).   2. See also Boesen (2007b: 210) for central Niger and Waibel (1998) for the comparable case of the Tuareg.   3. See also Waibel (1998: 110) for the Tuareg in Niger.   4. Often aiki is also simply used as a loanword.   5. In Boesen’s (2010: 40) sample group, individual gains from a two month journey were between 100,000FCFA (ca. 150€) and 750,000FCFA (ca. 1150€). In the study area, gains are generally less important (see Köhler 2017a: text 15,12)   6. The loanword has entered the vocabulary of Woɗaaɓe via Hausa and is originally derived from the English term ‘guard’.   7. The term Nasarankeejo (pl. Nasaranko’en) is derived from the Arabic term nasara (‫)نصارى‬ meaning ‘Nazarene’, i.e. Christian. In the study region it generally denotes Westerners. An often heard alternative term is the Hausa loanword Nasara.   8. Laɓɗo Usman, October 2011.   9. Laɓɗo Usman, October 2011. 10. For a transcription of the full account, see Köhler (2017a: text 12).

Chapter 8

Social Interaction in the City

In migration studies, it has been observed that cultural exchange resulting from the spatial mobility in the migration context does not only lead to the transcending of boundaries, but can also lead to an intensification of existing boundaries (Freitag and von Oppen 2005: 3). Migration often entails a reproduction of cultural difference in new locales (Sökefeld 2000: 52). However, the ethnic identities along which these boundaries are constructed, can be the object of considerable transformations. Migration demands adaptation to a new physical environment, which involves new kinds of behaviour and interaction, new modes of movements, and new kinds of corporeal experiences (Brickell and Datta 2011: 6). This ‘readjustment of migrants’ habitus’ (Friedmann 2002: 302) ultimately implies new configurations of identity. All this holds true for Woɗaaɓe urban migrants. Moving to town, they have to situate themselves in a new social environment and to construct new social relations and networks. They have to build up new neighbourhood relations and to form new communities, in other words, they have to engage in new processes of locality production (Appadurai 1995). Based on Barth’s (1969) theory of ethnicity, Schlee has argued that urban migration contexts are also predestined to generate new ethnicities: since social identities and differences are constructed in contradistinction to other identities, since they articulate at the ‘boundary’ where self meets other’, it is evident that migrants must develop new forms of ethnicity. ‘They live in new boundary situations along with new others, and if the boundary is where ethnicities are articulated, new boundaries must lead to new ethnicities’ (Schlee 2013b: 7, see also Schlee 2000a). On the one hand, this principle does apply to the poly-ethnic Nigerien context: urban migration involves new contexts in which members of different ethnic groups interact,

Social Interaction in the City  139

resulting in new boundary situations and thus renegotiations of ethnicity. On the other hand, since the principal form of work migration in the study context is regional rather than transnational migration, the principal groups who are thus confronted with each other in a new context are no strangers to each other’s social universes. This applies not only to migration to Zinder, the most important target for migration, but even for the case of cross-border migration to Kano and other cities in northern Nigeria. Although the transnational dimension does come into play here, culturally and in terms of ethnic composition the differences between the regions are only gradual. Ethnic identities are therefore not reinvented in the urban context, but rather, their re-negotiation is oriented at established patterns of inter-ethnic interaction. The interactional modes characteristic of the Fulɓe–Haaɓe dichotomy or the inter-ethnic joking relationship between Kanuri and Fulɓe, for example, serve as moulds for social interaction in the urban context. Ethnic images and self-representations are re-negotiated, yet modelled on pre-existing patterns. In this chapter, I will elaborate on the interaction of Woɗaaɓe migrants with members of different urban communities and on the new group relations that life in the culturally diverse urban milieu brings about. In town, just as in the pastoral context, other ethnic groups are not the only relevant others: social interaction with Woɗaaɓe from different clans plays a role in the urban setting as well and merits particular attention. In principle, an analysis of the interaction of urban-based Woɗaaɓe from the same clan but different lineages would be just as relevant for the study of inter-group relations. In Zinder, the urban locality principally studied, however, virtually all Gojanko’en migrants are from the Kuskudu maximal lineage. This means that the analysis on the intra-clan level can take into account only the intra-lineage perspective.

Urban Intra- and Inter-clan Relations: Spatial Patterns and the Impact of Group Size An analysis of residential patterns of Gojanko’en migrants in Zinder shows a remarkable tendency towards clustering: the vast majority of the migrants’ residences are located in relative proximity, although scattered on separate plots, in residential quarters in the south-western part of the city (see Map 8.1). To a certain extent, this spatial distribution can be interpreted as a result of two aspects that have already been discussed above: (1) the professional specialization of Gojanko’en migrants as watchmen for expatriates and (2) the role of kinship networks for getting access to both accommodation and jobs. On the one hand, already established migrants are more likely able to assist their newly arrived lineage mates in finding accommodation in their own neighbourhood than in any remote quarter of the city. On the other hand, as explained in the previous chapter, notably for watchmen working for expatriates, access to urban housing

140  Space, Place and Identity

Map 8.1  Urban clustering of Woɗaaɓe Gojanko’en in Zinder.

is closely linked to their jobs. The two urban quarters of Zinder with the most significant concentration of Gojanko’en have for decades been preferred residential quarters for expatriates and the location of the offices of numerous development projects. This explains why many of these migrants, who have since the 1980s specialized as watchmen for expatriates, have for at least two decades lived in relative spatial proximity in or in close proximity to these neighbourhoods. In recent years, as the city has immensely grown and new residential quarters offering new job opportunities have emerged, this tendency towards clustering among the Gojanko’en migrants has become slightly less pronounced, yet is still recognizable. Many Woɗaaɓe from other clans, notably Jiijiiru, also live as migrant workers in Zinder, yet few of them in the same residential quarters as the Gojanko’en. Their residences are concentrated further to the east in the commercial centre of Zinder, where many of them work as watchmen at the central market. This suggests that the dependency on kinship networks for access to jobs and the resulting association of specific lineages and clans with specific categories of urban work has an impact not only on the spatial distribution of the urban members of the same lineage or clan community, but also on the relative spatial distribution of different urban lineage or clan communities, in other words, that they lead, as in the case of the Gojanko’en Kuskudu in Zinder, to the urban clustering of lineage mates and to a relative spatial separation of the communities of Woɗaaɓe from different clans represented in the urban space. This impact, however, is only partial. As access to urban accommodation owes a lot not only

Social Interaction in the City  141

to social relations but also to opportunities, the observed tendencies are best considered trends rather than general rules. Besides, it is not only the economic role of kinship networks for providing job opportunities in towns that favours the urban clustering of lineage mates. Social considerations obviously constitute another, at least equally important reason why kinship must be considered as a significant factor in explaining this phenomenon. This becomes obvious from other cases, in which access to urban accommodation does not directly rely on migrants’ activities. In such cases, the tendency to cluster can also be strong and sometimes even more pronounced. In the case of an urban community of Woɗaaɓe Ɓii Ute’en in Diffa, almost the complete urban community lives in a rather dense agglomeration in one single open plot in the vicinity of the urban residence of their leader. The Ɓii Ute’en are not established in the job market for watchmen and thus do not have access to accommodation via their employers. They differ from the Gojanko’en community in Zinder also insofar as the demographic composition of the two communities is not comparable: the Ɓii Ute’en in Diffa are composed to a great degree of elderly people and women with small children, who have been left under the care of their leader while their husbands engaged in work migration notably to Maiduguri in north-eastern Nigeria. In recent years, these transnational migrants increasingly preferred to leave their families behind because of growing insecurity in the area of Maiduguri, yet they continued to seek work in this city as the offers for employment were much more abundant there than in Diffa. In the absence of their husbands, the women and children stay very close to their urban-based leader and to others of their group who are in the same situation as themselves. Here, the clustering is obviously motivated by social and emotional considerations. Patterns of clustering among urban migrants of the same origin or background have been observed in diverse geographical contexts, and their significance has been described as providing ‘affinity environments’ (Friedmann 2002: 309) that offer material and social support and psychologically help coping with the hardship of living in the culturally unfamiliar city context. By clustering, urban migrants actively construct ‘community’ to protect themselves from the rejections of other urbanites (Werbner 1997: 12). While this interpretive framework is widely applicable to the case of the Ɓii Ute’en community in Diffa, it is less so with regard to the Gojanko’en migrants in Zinder, for whom the urban condition is in many cases not characterized by the same existential hardship. Nevertheless, the pattern of clustering observable among the latter also seems to be partly determined by social and emotional considerations and to reflect a principle of preferential social contacts: whenever possible, the migrants tend to seek accommodation in the same city quarters and neighbourhoods as their lineage mates, because social contacts with other members of their own group of origin are preferentially maintained to keep up a notion of the kinship-based home community.

142  Space, Place and Identity

However, the possibility to do so largely depends on the availability of other migrants from the same lineage or clan. In different migration contexts, a community of migrants of the same origin might be of very different dimensions: it might be of a more or less significant numerical strength or it might – in the case of isolated migrants – be altogether inexistent. Group size has been shown to have a significant influence on patterns of social interaction within and across urban migrant communities and with the urban community at large (Schlee 2013b). In the case of the Woɗaaɓe, urban migrant communities can be constituted by members of different factions of a clan or by members of different clans who can develop new forms of community in the city context. Whereas the Gojanko’en form a rather significant urban community in Zinder, in the Diffa region, the clan is not represented and there is only one isolated Gojanko’en family, living in Diffa town. This is the family of Taafa Buuyo. His story (Köhler 2017a: text 13) is interesting insofar as he has lived and worked in different cities and is able to compare and reflect upon these experiences. From his native Damergou region, his work migrations have led him, among others, to Kano, Niamey and Diffa, where he has been living since 2003. [When we were staying in Niamey,] there were a lot of other Woɗaaɓe there, even [Gojanko’en] Kuskudu, but they were all from the west, from the regions of Abalak and Tahoua. There were no other lineage mates of ours from Tanout. But we were staying [in close contact] with them [i.e. the other Gojanko’en migrants], we would regularly visit each other. Whatever activities they were involved in, they would tell me, and we would go and participate together. . . . We were very close. When a Boɗaaɗo is staying in town, this is how it is. If there are no other members of his family, but other Woɗaaɓe, then of course he will stay [in close contact] with them. It is like here in Diffa right now. We are the only Gojanko’en, but there are Ɓii Ute’en and Suudu Suka’el here . . . We are always [in close contact] with them. (Taafa Buuyo, May 2011) Taafa’s statement as well as my observation of other cases confirm what has also been pointed out by other authors: while, in the pastoral context, interaction between Woɗaaɓe belonging to different clans is generally characterized by respectful distance, in the urban context a strong solidarity among Woɗaaɓe usually prevails across clan lines (Loftsdóttir 2000: 226ff., 251). Staying in the foreign city context seems to foster a need for belonging, which can be met by referring to a shared dimension of identity. Uncertainties caused by a strange and potentially hostile environment, as well as by the absence of indigenous structures of protection and decision-making, can be reduced by identifying positively with those ‘others’ who are less ‘other’ than others.1 The new physical environment implies a redefinition of the ‘other’ and new methods of interaction

Social Interaction in the City  143

(Brickell and Datta 2011: 6). This works well within the logics of segmentary opposition: confronted with an external other, e.g. in a foreign context, a level of shared identification, in this case the shared ethnic identity, can situationally be put forward instead of emphasising the diverging clan identities. More generally with regard to migration contexts, it has been observed that the principle of amity or prescriptive altruism, generally characteristic of relations among genealogically traceable kin, can thus be widened and generalized to encompass groups of migrants of the same regional or ethnic background (Fortes 1983[1969]: 154f.; see also Grätz et al. 2003: 12). In the case of Woɗaaɓe urban migrants, this translates into close social contacts across clan lines, regular mutual visits and aid in situations of difficulty. In Niamey, Taafa was not completely isolated from his own clan group: even then, in the 1990s, there was a significant community of Woɗaaɓe migrant workers in the capital of Niger that also comprised considerable numbers of Gojanko’en Kuskudu, although predominantly from the Ader region.2 Given Taafa’s own clan section’s migration history (see Chapter 2), genealogical links to Kuskudu from the Ader region, although they might not readily be known, can easily be reconstructed and it is possible to establish a relationship by referring to a commonly known clan mate or to a common ancestor. In other words, efforts are made to detect (or to construct) commonalities, and to refer to the most readily available shared identity category. This is possible because, as explained in Chapter 4, the relevant categories of identification involved (clan identity, ethnic identity), are in a taxonomic relation of greater or lesser inclusiveness (Donahoe et al. 2009: 16) and actors can make ‘[a]ppeals to broader or narrower defined identities of the same kind’ (Schlee 2008: 46). Generally speaking, the category preferably referred to is the one on the lowest level in the taxonomy that is both available in the given context and allows for a shared identification. In Diffa, where Taafa’s network of clan mates is absent, it is the shared ethnic identity that functions as a common denominator: shared identification and we-group definition as Woɗaaɓe is put forward by Taafa and his family; the isolation from their own clan makes them, in the strange environment, inclined to orient themselves, in their efforts to establish social contacts, to Woɗaaɓe from other clans as the culturally closest categories available. In Zinder, by contrast, there is a numerically rather strong group of Kuskudu migrants who form a relatively self-reliant community of their own. Here, social relations are predominantly focused on intra-group contacts and the resulting social networks are predominantly based on lineage identity. Woɗaaɓe of other clans – notably the Jiijiiru and Suudu Suka’el, who by far outnumber the Gojanko’en – are equally represented in Zinder and form separate communities. These observations confirm the thesis that group size is a key variable for collective identification in cities. When groups are bigger, they are characterized by a higher degree of closure: they tend to form ‘closed spheres of communication

144  Space, Place and Identity

which articulate only at certain points with the wider society’ and become self-sufficient in many ways (endogamy, socializing, mutual support . . .)’ (Schlee 2013b: 1). Small groups, on the contrary, favour widened, more inclusive patterns of identification. Accordingly, relations across clan lines and an emphasis on the shared ethnic identity as Woɗaaɓe play a greater role where only one or a few households from a particular clan live in proximity with other clans in a town setting, as is the case in Diffa. The rather exceptional case of Taafa Buuyo, isolated in a remote region where there are no other migrants from his clan, shows the disadvantages, but also the advantages, of such a condition. Although the isolation from the wider family doubtlessly constitutes a certain hardship, Taafa also seems to enjoy some economic advantages that result from his situation: while in Zinder the kinship network ‘eats up’ (nyaamgo) most of the earnings and it is difficult for an individual to progress economically, in Diffa Taafa has been able to achieve a certain prosperity.3 Relationships with Woɗaaɓe from other clans are close and of emotional importance, yet they do not imply obligations for financial support and sharing to the same degree as would close kinship ties.

Inter-ethnic Relations: Integration and Boundary Maintenance Pastoral Fulɓe have variously been characterized as self-oriented, placing little value on acceptance by others (Schlee 1997: 5; Schlee and Guichard 2013: 31). In a similar vein, it has been maintained with regard to the Woɗaaɓe that their emphasis on a ‘strange’ and ‘exotic’ otherness (Bovin 1998) is a minority group strategy for maintaining boundaries and preserving ethnic identity, or that self-stylization as ‘savages’ (Schareika 2004: 177) can be used as a strategy for interaction with ethnic others, aimed at avoiding ‘intensive social participation and the manifold conflicts related to it’ (ibid.). This strategic dimension of ostentatious otherness, or ‘self-stigmatization’, is also expressed in Boesen’s example of Woɗaaɓe ambulant medicine traders who are said to ‘keep away customs officers, gendarmes, etc. by deliberately neglecting their outward appearance or by feigning the dull’ (2010: 44). The attitude can thus be summarized as a specific form of impression management (Goffman 1959) with the aim of discouraging others from engaging in interaction, in order to remain unbothered. It can thus be regarded as another variant of a mechanism of evasion, the deliberate otherness in this case having a similar function to mobility. According to Boesen the ambulant medicine traders ‘refrain from building up relations with urban inhabitants’ and are characterized by a ‘refusal to engage in interaction’, ‘manifesting deficiency, non-participation and foreignness’ (2010: 50f.). This can be explained by the transitory status of these migrants, whose sojourns in the urban setting are limited to a few weeks or some months. In a context of more permanent neighbourhood, however, it can be assumed that such a behaviour changes

Social Interaction in the City  145

to a certain extent, and it is therefore rewarding to have a closer look at the aspect of inter-ethnic contacts that longer term urban-dwelling Woɗaaɓe migrants such as the Gojanko’en Kuskudu in Zinder or Diffa maintain. Zinder is characterized by multi-ethnic neighbourhoods and permanent inter-ethnic interaction. In general, the degree of inter-ethnic interaction that Woɗaaɓe migrants engage in also strongly depends on the availability or absence of other Woɗaaɓe in the city context. In Zinder’s relatively large migrant community of Gojanko’en Kuskudu, intra-ethnic contacts, and even intra-lineage contacts tend to be the most socially relevant. Nevertheless, inter-ethnic contacts can become important beyond their mere economic significance, even if they often emerge from economic interaction. For men, the work context in particular can be of some social significance. Contacts among colleagues often constitute an entry point to more personal relationships. However, the question of migrants’ social contacts with colleagues greatly depends on the nature of their work. As described in Chapter 7, Woɗaaɓe watchmen often live with their families on the compounds of their landlords. Their work duties often inhibit them from leaving the compound during working hours, while they can generally receive friends or relatives to socialize, share a meal or prepare tea. The urban social life of such watchmen thus often takes place behind the walls of private compounds and contacts with members of other urban groups often remain ephemeral and superficial. As private watchmen generally work alone rather than in teams, this type of work offers fewer opportunities for establishing contacts to colleagues than might be the case in other positions. This is different in the case of watchmen who work for bigger companies or institutions where teams of two or more watchmen per shift are common. The most common basis for inter-ethnic social relationships in the urban environment is spatial proximity, i.e. neighbourhood.4 Neighbourhood ties can develop into intense social interaction, and a general avoidance of social participation beyond ethnic boundaries cannot be observed. However, a gender aspect comes into play here: while men, due to their professional activities, often have more contacts outside the urban lineage community than women, this is to a certain extent balanced by the fact that women are often particularly strong in building up proximity relationships with neighbours. Werthmann (1997) has pointed out the particular significance of neighbourhood relations among Hausa women in northern Nigeria who often live in seclusion. Seclusion is less of concern for women in Zinder and even less so for Woɗaaɓe women, who even in town widely enjoy a liberty of movement outside their homes. Nevertheless, the immediate neighbourhood is the privileged sphere for social contacts. One particularly female way of establishing relationships with neighbours, and at the same time a good indicator for the degree of the social integration of a Woɗaaɓe migrant’s household into a neighbourhood, is the reception and distribution of food, especially on festive occasions. While the ritual sharing of meat,

146  Space, Place and Identity

notably in marriage and name-giving ceremonies, is of particular significance for community-building and -maintenance, food-sharing in more profane contexts also has such a function, and even across ethnic groups in the region. It is very common for a woman who has prepared a dish to send a plate to neighbours’ and nearby friends’ houses, especially on festive occasions or when the recipients have visitors. Sending food to a neighbour’s house can be considered a way of symbolically integrating this person into the social group constituted by the neighbourhood community. Reciprocal consideration of neighbours through the sharing of food and by mutual visits on the occasion of festivities can further consolidate good neighbourly relations and even develop them into friendship relations. Most Woɗaaɓe migrants in Zinder engage in this pattern of social exchange by sharing food across ethnic lines and they are thus both integrated by their neighbours and actively integrating into their local neighbourhoods. Evidently, this way of actively integrating into the neighbourhood community is not so much of an option for young single male migrants. But although they cannot actively participate in such an exchange, they are often integrated by neighbouring families according to this pattern. A comparison between the data of Zinder and Diffa indicates that here again, group size seems to be an important variable. While social contacts in the relatively large urban Gojanko’en community in Zinder are predominantly intra-ethnic, inter-ethnic relationships play a relatively important role in the case of Taafa Buuyo’s isolated family in Diffa, where the urban Woɗaaɓe community, even across clan lines, is far less significant. This might be interpreted as a relative openness to social contacts with ethnic others, due to a lack not only of members of one’s own kinship community but also of one’s ethnic community close by. However, the time-factor seems to be of relevance here, too: when I first met Taafa and his family in 2004, they had only recently arrived and often complained about being rejected by the locals and about a general lack of social contacts apart from those with their Woɗaaɓe neighbours. In 2011, during my fieldwork, the family had many contacts across urban communities and ethnic groups, and they seemed well integrated and overall content with their life in Diffa. Unsurprisingly, the temporal dimension of urban residence apparently plays a crucial role for the question of integration of migrants into the urban society. Integration of any significant degree can only be expected to occur if there is a long-term perspective and a certain identification with one’s own life and position in the city. This has often not been the case for the first generation of urban dwellers, for whom the declared aim was rather to return to the pastoral livelihood as quickly as possible (Loftsdóttir 2000, 2002a). Today, however, long-term migration is an increasingly dominant pattern. Children who attend school in town can be an important factor for the integration of their families. They bring friends back home and the parents thus get acquainted. Hausa being the main language of inter-ethnic communication,

Social Interaction in the City  147

urban Woɗaaɓe children naturally grow up bilingually. Although parents often sharply criticize their children for speaking Hausa at home, language skills are recognized as an important tool for the contemporary economic strategies pursued in town. One aspect of adaptation and partial assimilation that can be observed among school children concerns the gradual abandonment of ethnic markers, notably the traditional male hairstyle of long plaited hair. Under the pressure of their classmates’ mockery, many schoolboys cut their braids, sometimes clandestinely and without the consent of their parents. Schoolgirls similarly tend to abandon the characteristic hair bun, which is also frequently mocked by ethnic others. This assimilation of children to an urban lifestyle is seen rather critically in particular by many older Woɗaaɓe, who often accuse the children of having gone astray (bilki’en bonni). However, male adults suffer from the same comments about their hairstyle as schoolboys do, and more and more of them cut their long hair as well. External ethnic markers thus tend to be partially given up in the urban context, which can clearly be seen as indicating a certain effort at integration. Given the central significance of the canonical male beauty for participation in cultural activities, notably dances, one might assume that the abandonment of the traditional hairstyle would entail a break from these central forms of cultural expression and thus a significant rupture. Often however, this issue is today rather pragmatically dealt with and many men, although they have abandoned the traditional hairstyle, continue to engage in dance contests, wearing fake hairpieces that they produce from their own cut-off braids. For girls it is easier to change their hairstyle and return to the ethnic style when visiting their home communities – an example of situational, context-dependent switching of norms (see Klute and Hahn 2007: 17). Social orientation and identification in town are not one-dimensional. Woɗaaɓe migrants are simultaneously involved with different social relations in different classificatory frameworks. The urban migration context thus represents a special case of the phenomenon of polytaxis (Elwert 2002: 39) and provides options for switching between different frames of identity reference.5 The everyday lives of migrants take place in specific urban contexts that provide different frameworks for the negotiation of identity and belonging and for the production of locality (Brickell and Datta 2011: 17). Migrants thus switch regularly and situationally between multiple identifications along different dimensions of belonging. The neighbourhood, the work context with its relations to both colleagues and employers, for children their relations to schoolmates, and the intra-urban kinship community – all these are potential contexts in which migrants have to situate themselves towards others between the poles of difference and sameness, making appeal to different categories of identification.6 In conclusion, it can be maintained that social orientations in the urban context principally follow the ideal of minimal cultural and social distance. If

148  Space, Place and Identity

a substantial urban lineage community exists, kinship remains the dominant principle of social organization. Where it is absent, kinship-like solidarity can be extended to encompass the clan or the ethnic group at large. This social orientation towards socially and culturally close others does not mean that cross-ethnic social interaction is avoided. It does, however, mean that ethnicity and identity categories below the level of the ethnic group remain the dominant criteria for social orientation. The established we-groups and their boundaries as they had been defined outside the urban context remain primarily intact despite the permanent confrontation with ethnic others in the ethnically mixed neighbourhoods of the city. This becomes evident once more in marriage patterns. Although inter-ethnic marriages are more common in the urban than in the pastoral setting, they are generally restricted to the groups considered culturally close, particularly Tuareg and other Fulɓe groups, while marriage with Hausa or Kanuri (i.e. Haaɓe) remains exceptional also in the urban context.

Notes 1. To twist a quote from Appadurai (1986: 357), based on the twisting of a quote from Orwell’s Animal Farm. 2. Loftsdóttir (2000: 248) gives a rough census of the numbers of Woɗaaɓe migrants in Niamey (as of 1997–8), which vary seasonally between 231 in the rainy season (September 1997) and 876 in the hot dry season (April 1998). The numbers for the Gojanko’en within this group are assessed at 61 and 216, respectively. Together with the Ɓii Korony’en they are the numerically strongest clan represented in Niamey. 3. See Marfaing (2014a: 52) for similar observations among regional migrants in West Africa. 4. See also Pelican (2003: 27) for Fulɓe and Hausa in Cameroon. 5. This point owes inspiration to an unpublished draft of a collective paper on ‘Identification in Social Action’, by Brian Donahoe, John Eidson, Dereje Feyissa, Veronika Fuest, Marcus V. Hoehne, Boris Nieswand, Günther Schlee and Olaf Zenker, where trans-national migration is interpreted as a special case of polytaxis. 6. In this context, see Appadurai’s (2003: 341) notion of ‘disjunct registers of affiliation’ characteristic of increasing numbers of people for whom ’the practicalities of residence and the ideologies of home, soil and roots are . . . disjunct, so that the territorial referents of civic loyalty are increasingly divided . . . among different horizons: work loyalties, residential loyalties, and religious loyalties’.

Chapter 9

The Translocal Dimension of Urban Migration

As argued in Chapter 2, social orientations among pastoral Woɗaaɓe are more translocal than place-bound. In the context of urban migration this applies just as well. But in analysing these translocal ties, the focus must not be narrowed onto the migrants alone because those segments of the population that remain behind can constitute an equally important dimension of connectedness (Brickell and Datta 2011). Although mobility is vital in connecting and transforming places, translocal spaces are co-produced by both mobile and immobile populations (Greiner and Sakdapolrak 2013: 376). This position argues against an often-maintained image of migrants as characterized by a higher degree of mobility in comparison to their societies of origin. Since sedentarity has for long been taken as the norm and mobility and migration as an exception (Gupta and Ferguson 1997b: 6f.; Klute and Hahn 2007: 10f.), those parts of the society who stay behind at home have often been considered as rather immobile (Greiner and Sakdapolrak 2013: 376; Marfaing 2014a: 44). With regard to the Woɗaaɓe, such received notions of stability and mobility often associated with migration have to be critically reassessed: as nomadic pastoralists, they are highly mobile even outside the context of migration. More generally, as noted above, a certain degree of mobility – although greatly varying according to season and circumstance – is characteristic of most socio-economic groups in the Sahel region and has been so for long. The complex patterns of spatial organization can perhaps best be characterized in the terms of Boesen and Marfaing (2006: 3) as defined by a double forking – rural–urban on the one hand, and mobile–sedentary on the other – between which the populations of the Sahara–Sahel region move, depending on their different, and sometimes multiple, activities and forms of production.

150  Space, Place and Identity

In the case of the study group, translocal kinship ties of urban migrants with their pastoral home community generally remain strong and continue – even in the case of long-term migrants – to be crucial in terms of identity construction. The translocal ties are supported by new forms of mobility that migrants develop. However, many rural Woɗaaɓe are no less mobile, not only because they continue to move as pastoralists with their herds, but also because they frequently visit their urban-based relatives, e.g. in order to have access to social services like hospitals, to handle administrative or business matters, or simply to keep up social contacts. This chapter assesses the significance of this translocal mobility for cultural orientations among migrants and for questions of identification and belonging, but also in terms of its economic rationale.

Perceptions of City Life Woɗaaɓe use the terms si’ire (settled areas, ranging from villages to cities) and ladde (the rangeland, and thus, the pastoral environment) in the sense of a clear dichotomy. The two spheres are associated with specific connotations, judgements and values that are squarely opposed. One often stated contrast is the vastness of the open range when set against the town, which is often perceived as restricting and therefore non-adequate for living. Tellingly, a spacious plot (fiiliire) is more important for many urban-based Woɗaaɓe than an elaborate house. While the pastoral rangeland is perceived as pure and clean, the city and even villages are regarded as polluted and contaminated with dangerous diseases (Loftsdóttir 2001b: 284). Villagers or townspeople, in turn, perceive the fact that Woɗaaɓe eat, sleep and live amidst their animals, and thus amidst the latter’s excrements, as rather offensive and questionable in terms of hygiene. Such opposed perceptions can perhaps best be grasped with Mary Douglas’s (1978 [1966]) concept of purity and of dirt as something out of its place. In the same dichotomous sense, Woɗaaɓe associate themselves with the open rangeland, while the town or village is associated with the paradigmatic other, the Haaɓe (Boesen 1998: 222, 236; Loftsdóttir 2000: 146; Schareika 2003a: 12, 83). The dichotomization of ladde and si’ire is thus part of the process of placemaking by producing and spatializing difference (Gupta and Ferguson 1997b, 1997c). Woɗaaɓe apply a marked discourse of contrast to the concepts ladde and si’ire. The general conception of the two spheres as marked opposites with their respective positive and negative connotations informs migrants’ perceptions of the urban environment and makes a positive identification with it difficult. Migrants’ negative view of the urban is reflected in their tendency to say that they were driven to town by economic difficulties rather than desire (Köhler 2017a: text 13,1 and 13,22). But if people’s attitudes towards the city are overwhelmingly negative in their statements, their actual practice is not consistent with this rhetoric. Despite

The Translocal Dimension of Urban Migration  151

the prevalent despise, the city is recognized for its economic potential. Faced with a tempting job-offer, many pastoral Woɗaaɓe would be inclined to leave their animals with a close relative and temporarily move to a city (Köhler 2017a: text 12,07). Today, virtually all Woɗaaɓe have relatives in town and entertain more or less regular relations with them. On the other hand, for most urban migrants the pastoral realm remains the central focus for identification, and the pastoral home community an important focus for social and emotional orientation and communication.1 Whenever possible, they return home for visits and even those who have spent a considerable part of their life in town often return upon retirement or when they can no longer find work that would justify staying in town. The pastoral home camp bears resemblance to what Olwig (1997: 11) has called a ‘cultural site’: it is a reference point for cultural identification and a ‘symbolic anchor’ of community (Gupta and Ferguson 1997c: 39) even for those who live more permanently in urban centres. However, this referential place of cultural identification is not a fixed site but, in most cases, rather a highly mobile place.2 Referring to Mauss (1979 [1904/5]), Boesen (2010) recognizes a seasonal alternation between low points and high points in the social life of Woɗaaɓe who engage in short time migration as ambulant medicine traders. While the trading trips to urban centres, which generally occur during the dry season, are characterized by separation from the community and a less structured social life, the periods in the pastoral home camps, which for most migrants coincide with the late rainy season, are a time of social intensity (Boesen 2010: 52). Boesen compares this with a similar alternation that can be observed in the pastoral year cycle: in the hot dry season, when pastures become scarce around the well, the households disperse. They are still dispersed at the beginning of the rainy season, when individual household units or small groups are extremely mobile to follow the emerging fresh pastures. Finally, the height of the rainy season, when both surface water and pastures are abundant, reunites the community, culminating in the annual gathering of the regional clan segment (worso). The late rainy season is thus characterized by a higher population density and increased social contacts, while the hot dry season entails a relative isolation of small groups.3 Interestingly, a similar observation can be made with regard to Woɗaaɓe migrants who reside more permanently in urban centres than those in Boesen’s example. Similar seasonal cycles can be observed in the Woɗaaɓe community in Zinder, yet the alternation here is exactly the inverse of that in the pastoral context: for those staying in town to work, the rainy season months constitute a time of decreased social contact because during this period significant parts of the urban migrant community visit their relatives in the pastoral camps (see also Loftsdóttir 2001a; Boesen 2007a). School vacations last from July to September, which correlates roughly with the rainy season and thus with the time of the year that is characterized, in the pastoral realm, by a concentration of camps and

152  Space, Place and Identity

by increased communal social activity. Children of urban migrants and their mothers are often sent home during this period in order not to lose touch with the home community (see Case Study 1). While they often spend their entire vacation in the pastoral realm, the men are generally constrained by their work contracts and cannot easily join them for any length of time. For those staying behind this means that they experience the rainy season as a rather lonely period, deprived of the company of their relatives. In half-serious remarks, some even consider giving up their jobs in order to be free to join the rainy season gatherings. This is the season when urban work is considered most difficult. Complaints about the hardship (wahala) of urban work regularly occur during this period of the year. This indicates that the hardship which is associated with urban work does not merely refer to physical hardship, but to an important extent also to the emotional difficulty of being separated from one’s extended family. In a city like Zinder, the relatively large migrant community can to a certain extent level out this emotional drawback, but during the rainy season months, this community decreases significantly in size. Case Study 1: Maintaining Translocal Ties from an Urban Base Toggindo Ngutinga has been staying in Zinder since 2005 when she was about 25 years old and her husband found employment as a watchman there. In order to earn some money, she sometimes carries out domestic tasks for a wealthy Hausa woman who works in the neighbourhood. Toggindo lives in close neighbourhood to her sister, whose husband is also a migrant worker. Her brother lives in Kano as a watchman and whenever he travels home to the Damergou region, he stops by for a night at Toggindo’s place. Toggindo and her children regularly spend the rainy season months, which coincide with the school vacations, in Toggindo’s father’s camp in the Damergou region, where the boys join their cousins in the task of herding goats and thus to a certain extent keep in touch with the pastoral life. Toggindo’s eldest daughter, at the age of about fourteen years, has ceased to spend longer times in the pastoral camps as she is rather kept in Zinder during the vacations to look after the household during her mother’s absence. Via her husband’s work environment, Toggindo came into contact with expatriates and her lineage mates in the Damergou region sometimes give her embroidered wrappers or other traditional craftwork to sell back in Zinder and send them the money.

Rural–Urban Mobility and Translocal Networks The life stories of migrants moving between rural areas of the Zinder province and the city of Zinder show that migration to this urban centre is not incompatible

The Translocal Dimension of Urban Migration  153

with keeping up a rootedness in the pastoral context. Not only are seasonal visits of migrants to their home communities common, but more generally, frequent travelling in order to maintain relationships between urban and rural communities is possible due to the relative proximity of the main centres of migration. In some cases, migrants are virtually simultaneously situated in different localities (Brickell and Datta 2011: 4). Alaji Ɗawra (Case Study 2) for example has two homes – one in Ganatcha, the other in Zinder. Each of his two wives leads a household of her own, and Alaji, leading a highly mobile life, divides his time between the two spheres – between an urban income-generating activity and the pastoral economy. His case might contain rather exceptional details such as his work as a driver and the corresponding, rather abundant salary that even comprises a rent allowance. Concerning the principle of double household structures bridging the urban and the pastoral sphere, however, there are many examples comparable to his. Case Study 2: A Double Household Structure between the Rural and the Urban As a child of about 12 years, Alaji Dawra started to serve as an apprentice to a motor carrier and merchant who connected Zinder and Damagaram Takkaya with a collective taxi. Alaji learned driving from him and finally acquired a driver’s license. Later he found work as a driver in different development projects, first in Tanout, then in Maradi and Niamey before returning to Zinder. Alaji, who has two wives, has rented a house in Zinder, where his first wife resides with her children and where he lives when his work duties keep him in town. Whenever the circumstances permit, however, he goes to Ganatcha, where his second wife lives together with his father and brothers. One of Alaji’s sons with his second wife is staying in Zinder too and attends secondary school. In Ganatcha, it is Alaji’s adult sons and his brothers who take care of the family herd. In addition to pastoralism, Alaji has begun, in recent years, to engage in agriculture in Ganatcha, yet to a rather limited extent. As he cannot take care of the fields himself, but has a fairly good regular income, he generally pays some money to a Kanuri farmer from a neighbouring village for doing the farm work in his place. Good relations with his employer in Zinder allow Alaji to use the project car privately during weekends if he pays for the fuel – an arrangement that permits him to lead a highly mobile life between his two homes and to logistically manage this translocal double household structure. The frequent travelling allows him to buy products where they are the cheapest and supply his households accordingly: while firewood, for instance, is considerably cheaper in the rural markets, many consumer items are cheaper in town. Alaji’s mobility furthermore opens up facilities for other lineage members in Ganatcha

154  Space, Place and Identity and Zinder to maintain translocal ties across the sites, as he can provide them with occasions for free and relatively comfortable travelling. The work as a driver in different development projects has also allowed Alaji to involve his pastoral home community in Ganatcha in certain project activities. The project he most recently worked for is engaged in advocacy for pastoralists and has, over the years, realized several activities in the area of Ganatcha. Although not all of them may have entailed a directly felt positive impact for the community, they certainly helped Alaji to keep up a very steady mobility between Zinder and Ganatcha, in combination with a relatively attractive income.4

Double household structures are also often maintained by administrative chiefs (arɗuɓe) who have generally not only the financial means, but also a particular interest in maintaining an attachment in town, because the position of an administrative chief today demands regular and close contacts with both the pastoral community and administrative institutions and development or relief organizations. Hence, the constant mobility between the rural and the urban sphere is advantageous in itself and an integral part of such spatial strategies, which can be encountered among administrative chiefs across Woɗaaɓe clans and regions, but also among ordinary migrant workers. Attractive, yet sometimes temporally limited, job opportunities in urban centres make the maintenance of double household structures between the urban and the pastoral realm sometimes a feasible option because it allows for a better control of the family herd, which is often entrusted to adult sons (see Marfaing 2014b: 150f.). Observations made in Kenya among Turkana who navigate between the pastoral setting and town in order to exploit the resources of both (MüllerDempf 2014) suggest that similar phenomena are of wider significance among pastoralists across geographical settings. Turkana pastoralists may bring parts of their families to town while others remain in the pastoral camps, and they maintain close connections between the two spheres. Often the activities of such ‘hybrid pastoralists’ (ibid.) can be very diverse and link them to different places across urban and rural/pastoral spaces. An important difference from the Kenyan example is that whereas in the Turkana context such a pronounced rural–urban mobility seems to be possible only for men (ibid.: 20), among the Woɗaaɓe high female mobility is not unusual (Boesen 2010). Although women in the study group generally do not migrate to towns on their own, but rather accompany their husbands, they nonetheless often find opportunities to profit economically from their situation in town (see Case Study 1). The translocal rural–urban ties are characterized by important flows not only of people, but also of material and other resources. Migrants provide financial support with remittances or help their rural kin to get access to job opportunities in town, but they are also involved in networking likely to open up access to new

The Translocal Dimension of Urban Migration  155

resources, and the economic outcomes that might derive from these connections are to an important extent redirected to benefit their home communities. The case of Alaji Ɗawra (Case Study 2) can again serve as an example here. Alaji’s job as a driver for a development project has enabled him to establish advantageous relations with development agents and expatriates. Not only Alaji’s immediate family in Ganatcha, but to a varying degree also the other lineage members and members of the wider local community profit from his networking with expatriates/employers. The most important investment achieved on the basis of these networks was the construction of a well in Ganatcha. But access to food aid, or the involvement of the home community in project activities that generate resources and direct or indirect benefits, are also attractive assets that Alaji was able to achieve for his community. Alaji himself, according to the principles discussed in Chapters 2 and 7, has gained the prestige that comes from making resources available to the wider community. Urban migrants can thus significantly contribute to bridging the gap between the pastoral populations, on the one hand, and the state administration and development actors, on the other.5 They can therefore be regarded as ‘bridgeheads’ to the outside world (Geschiere and Gugler 1998: 310; Marfaing 2014b: 142) for their pastoral home communities. Many urban migrants send money back home and commission relatives to buy animals for them. Although research has shown that migrants’ investments in livestock are often not sufficient to speak of a sustainable strategy of herd reconstruction (White 1987: 99; Swift 1984: 489), they can be interpreted as a way of symbolically and emotionally staying attached to the pastoral community, and thus as an indicator for migrants’ continuing identification with a pastoral life. This implies that the flows are by no means one-way: if material flows go rather from the urban to the rural sphere, there is a return from the rural to the urban members of the translocal community (1) in terms of services (e.g. animal transactions in markets, caring for the animals that the migrants have left behind or, in the agro-pastoral setting, the cultivation of fields), and (2) in terms of emotional resources (e.g. providing a home, a source of identity and belonging). Hence, rather than by an asymmetric dependence, the translocal ties are often characterized by exchange, mutuality and cooperation, or as the case of Nano, Taafa and Maalam Buuyo shows, by a strategic and complementary division of tasks (Case Study 3).6 Such two-way flows of values and services, in which both sides invest into the translocal relationship because both sides have an interest in maintaining it, have been pointed out to be a more general pattern of urban-rural connections in Africa (Geschiere and Gugler 1998: 310). Case Study 3: Complementarity of Brothers Nano, Taafa and Maalam Buuyo follow a complementary strategy of diversified pastoralism. While Nano and Taafa both work as watchmen

156  Space, Place and Identity in Zinder and Diffa, respectively,7 Maalam, as the youngest brother, manages their joint herds in the pastoral zone of the Damergou region and assures a presence at the well of Salaga as well as in the recently created centre not far from it. While Taafa, due to the distance of his place of residence, has entrusted all his animals to Maalam and comes home only about once a year to visit, Nano is regularly travelling to Salaga to look after things. Although Nano has established himself as an ardo in 2010, he continues to work in Zinder, while having delegated his responsibilities and functions in Salaga to his younger brother Maalam, who acts as his official representative. Taafa at times sends surplus money from his urban revenues to Salaga to be invested in animals. Inversely, if Taafa is in need of money, he may ask Maalam to sell an animal from his stock and send the money. Taafa, although spatially detached from his pastoral home community and engaged in urban migration work in Diffa since 2003, nevertheless contributes with significant remittances if investments have to be made, e.g. for maintenance work on the jointly owned well.

Evidently, the division of roles and tasks can only function as an economic strategy if the partners can rely upon each other. In practice, the model does not always work in an ideal way. Some urban migrants complained that their pastoral relatives had sold their animals in their absence, claiming that they were lost due to animal diseases, caught by a hyena, or stolen. One interlocutor claimed that he left 30 head of cattle with a brother when moving to the city in 2003, and that despite the considerable investments he had made, he was left with only 3 animals in 2011. From this perspective, another obvious advantage of short-range migration is that it facilitates the supervision of how livestock investments are managed by relatives in the home community (Marfaing 2014b: 150). The case studies presented in this section show different variations of rural– urban mobility and connectivity. They are merely some out of many similar examples that could be used to illustrate the multiple translocal ties that urban migrants maintain with their home communities. The relative proximity between the different rural and urban sites across which the majority of the study group is dispersed, is used in the sense of a spatial strategy for combining economic and social concerns. The resulting translocal networks can be understood as defining larger economic spaces of mobility in which different members of minimal lineages fulfil defined functions in a division of tasks, and in which they position themselves strategically in different places in order to have access to the economic resources that these places offer. Hence, the urban and the (agro-)pastoral sphere, as much as they are described as distinct and dichotomous by the Woɗaaɓe themselves in their relative qualitative assessment and with regard to their

The Translocal Dimension of Urban Migration  157

relevance for identification, are, in economic terms, rather complementary. They are intrinsically linked and interwoven to form a more encompassing whole – the socio-economic space of contemporary livelihood strategies.

The Infrastructure of Translocal Mobility: Public Transport and Inter-ethnic Social Networks As far as the mobility between the urban and the pastoral sphere is concerned, the case of Alaji Ɗawra is rather exceptional. Most Woɗaaɓe do not have access to cars and cannot drive but rely on the existing infrastructure accessible to the generally highly mobile Nigerien society to facilitate their travelling. Motorbikes begin to a certain extent to replace public transport as a means of travel, yet so far on a rather modest scale. Since the road conditions in rural Niger make owning a motorbike a cost- and repair-intensive enterprise, this remains an option only for a few. The Woɗaaɓe from the Damergou region who work in Zinder profit from comparatively good public transport connections. A trip on a collective taxi from Zinder to Tanout, the principal town in the Damergou region which is frequented on market days by many Woɗaaɓe from Salaga, costs 2000 FCFA (ca. 3 €), and the car takes between two and three hours for the approximately 130 kilometres. On Saturdays – the market day in Tanout – there is often a certain flux of Woɗaaɓe travelling either from Zinder to Tanout in order to visit their home communities, or in the opposite direction, to combine affairs in the city with visiting urban-based relatives. Mobility in rural areas in Niger is closely linked to trade routes. Markets in Niger generally being held weekly, each village has its specific market day (or days). Traders travel from one market to the other and motor carriers have reacted to their requirements by establishing a well-functioning system of transport that connects remote villages to commercial centres and provides commodities and comestibles to rural populations. Travellers can rely on this network for their own needs of transportation. On the respective market day, virtually any market village can thus be reached with public transport. Depending on the conditions of the terrain, mini-vans, Land Rovers, big lorries or off-road trucks can function as rural collective taxis. What makes the connection between Tanout, Zinder and even Kano relatively convenient is also the fact that they all lie on the same ancient trade route that used to connect the coastal regions of the south to the endpoints of the trans-Saharan caravan trade in the north. Today, although their importance for international trade has much diminished, these historic routes remain so important for mobility at the regional level that they even have a noticeable effect on the distribution of settlements: towns and major villages are distributed in chains with a north–south direction (Rain 1999: 145f). Discussions with members of the study group about the advantages and disadvantages of different rural sites

158  Space, Place and Identity

show that the proximity to a connecting road is nowadays also an important consideration for the choice of a site for establishing a centre.8 For the Woɗaaɓe, as for other populations in Niger, the close link between transport utilities and trade networks means that specific days of the week offer specific opportunities for travelling and that rural market villages are important nodal points. Woɗaaɓe who travel frequently therefore entertain a network of acquaintances in the more important market villages on the routes between Zinder and Ganatcha, and between Zinder and Salaga, in order to maximize their options for mobility. The Woɗaaɓe travellers may spend a few hours or, if necessary, a night in the houses or courtyards of their hosts in these villages while waiting for a collective taxi or a lorry. Here, they will be offered some food and decent conditions for spending their waiting time. The case of Abdua Ɗawra (Case Study 4) gives an idea of the conditions of travelling and the variety of means of public transport, but also of the role played by the network of hosts – both intra- and inter-ethnic – for facilitating mobility. Abdua’s case is interesting also insofar as it represents a rather exceptional example not of urban, but rather of rural work migration via urban-based networks. Abdua has been acquainted with a French hydraulic engineer for whom he worked on and off as a labourer and thus acquired, over the years, a considerable knowledge in the construction and maintenance of wells. He thus achieved a certain reputation among both pastoralists and well-construction projects in the region, and he is regularly engaged by diverse clients. In 2011 alone, he was involved with a Zinder-based project in the construction of a series of wells, and engaged on a private basis by three different pastoralists for maintenance work at their private wells. Abdua’s work is thus characterized by a high degree of mobility of a considerable range. Case Study 4: Urban and Rural Hosts Facilitating Rural–Rural Mobility In November 2011, Abdua Dawra embarked from his home in Ganatcha for a trip to the northern Damergou region, where he was engaged by a man from the Suudu Suka’el clan to repair his well. Together with a brother’s son he left his home camp on camel back for the nearby village of Karakaptchi, where a close Kanuri friend and host lives (for the relative location of all the places on this itinerary, see Map 5.1). While his nephew returned home with the camel, Abdua spent the night at the friend’s house, knowing that the latter would go to Damagaram Takkaya early the next morning with his ox cart in order to visit the market. After a three hour ride the two arrived in Damagaram Takkaya, where, given that it was a Sunday, the local market day, it was easy to find a Land Rover collective taxi to Zinder. Abdua arrived in Zinder in the late afternoon and took a motorbike taxi from the central market, where the Landrover

The Translocal Dimension of Urban Migration  159 stopped, to the house of his sister who lives in the city. The next day, he visited friends and relatives and took some rest before continuing on his way to Tanout. Tanout being a bigger town, mini-van taxis are never difficult to find. After a three-hour ride Abdua arrived in Tanout and gave a mobile call to a young man with whom he is acquainted and whom he trusts as a motorbike driver. Abdua negotiated a ride to the remote site north of Tanout and arrived there after about an hour and a half on sand roads. This means of transport is more expensive, but quicker and less exhausting than its alternative, a ride on an overcrowded open off-road truck. On his return after ten days of work, Abdua spent a night at his paternal uncle Boyi’s place near Salaga, before leaving, on a Friday afternoon, for Tanout with some family members on their way to the market, which was to be held the next day. After a three-hour ride on donkeys and camels, the party arrived in a small Kanuri village where Boyi has a regular host. They spent the night and continued their ride to Tanout early the next morning. In the afternoon, Abdua took a collective taxi back to Zinder. This time he stayed for two days, to get some rest and in order to do some shopping before returning home, but also, because Tuesday is a convenient day to return to Ganatcha. Abdua is well acquainted with a baker from the village of Kassamba, between Ganatcha and Zinder, who comes to Zinder each week to buy a load of wheat flour. In addition to the flour, he takes travellers and merchants on their way to the rural market of Gueza. The market day in Gueza is Wednesday. The car leaves Zinder on Tuesday around nightfall, heading to Kassamba, where the flour is unpacked and changed against a load of freshly baked bread to be transported to Gueza. The passengers spend the night in the courtyard of the bakery before the trip continues early the next morning. In Gueza the bread is sold, and the merchants make their business. Abdua thus travelled to Gueza, met his wife and son who had come to visit the market and who had brought an extra donkey for him to ride back home in the evening.

Abdua’s case shows the significance of both social networks and public transport networks for the translocal mobility of migrants. It also points to the importance of familiarity with the logic of the public transport infrastructure that supports rural markets and thus follows their weekly rhythm. The examples of both Abdua’s Kanuri friend in the village of Karakaptchi and the baker, whose role was at the same time that of host and motor carrier, show that institutionalized stranger–host relationships, which have for a long time facilitated pastoral mobility and sojourns in villages, continue to play a central role today for facilitating modern forms of rural–urban mobility and take on various forms on a continuum between interest-oriented and institutionalized formality

160  Space, Place and Identity

and interpersonal friendship. In the cities that are target centres for migration, however, resident relatives from the Woɗaaɓe migrant communities themselves generally take the role of hosts today and the relevance of this institution has thus much diminished here.

The Mobile Phone: A New Dimension of Translocal Connectedness Recent research has stressed the importance of communication for the construction and reproduction of translocal communities: ‘without communication, people lose connection and a community may eventually cease to exist’ (de Bruijn and Brinkman 2011: 55). While such communication over distance has been maintained, among mobile groups like the Fulɓe, by mobility itself, i.e. by travelling, today new communication technologies, in particular the mobile phone, are rapidly gaining importance. However, the mobile phone has not completely revolutionized social relations, but rather logically followed up on more grounded indigenous forms of communication. As de Bruijn and Brinkman (ibid.) have shown for Fulɓe in Cameroon, the new medium is used to support and facilitate such established patterns of social communication and interaction across space. However, the mobile phone has had a significant impact on the scope and on the rhythm of communication. It facilitates not only the exchange and diffusion of information but also the maintenance of social relations at a distance and is thus particularly well adapted to the Woɗaaɓe’s translocal habitus (ibid.: 54). Probably the most important use of the mobile phone among Woɗaaɓe concerns the exchange of information on the situation of pastures in different areas, the conditions of specific water points or market prices for animals or millet. Since the location of good pastureland is highly variable and unreliable on the northern fringes of the Sahel, the success of any pastoral strategy highly depends on quick and reliable access to relevant information (Schareika 2003a:125f.). Although neither direct nor telephone conversation could completely replace exploratory missions, because information concerning the location and quality of pastures is too controversially interpreted for actors to rely only on hearsay in obtaining it, the mobile phone efficiently supports decision-making about pastoral movements and the coordination of herd movements and is today widely embraced by pastoralists (PSSP 2009: 53). It can be argued that the high value placed on an efficient exchange of information relevant for pastoralism was crucial for the way the mobile phone has been embraced by the Woɗaaɓe and other pastoralists in the region. As in other parts of Africa (e.g. de Bruijn et al. 2009), mobile telephony in Niger has developed rapidly. When I first came to Niger in 2004, the first networks were being created, however in regional capitals like Diffa the mobile phone was not yet operative. Five years later it was a widespread means of

The Translocal Dimension of Urban Migration  161

communication through all strata of society, from white collars to cattle nomads, and new networks continued to mushroom, soon covering virtually all the more densely populated south of the country and more and more remote areas. On rural markets, generator-driven battery chargers supply people with the energy necessary to keep connected. At the time of my research, only one of the four rural centres of the Kuskudu in the study area did not have network coverage. This was Ngel Tireeji, where people relied upon visits to the markets of Tanout or Takoukou to keep in touch with other parts of the translocal community. In Salaga, Yalema Intrika and Ganatcha, by contrast, mobile telecommunication was already well established, and ringing phones were a constant phenomenon. When my fieldwork started in 2010, the network coverage in Ganatcha was not yet very developed. Only particular spots offered sufficiently good connectivity and telephones were often hung in the branches of trees in order to improve reception. When, not much later, a new network was established in Gueza, the market village about 10km to the west of the Woɗaaɓe camps, this again enormously facilitated distance communication. The fact that in 2008, when the wife of Ɗawra’s youngest son Mahamadu bore a child, it was nick-named ‘Seluula’ – from ‘cellulaire’, the French term for cell phone – is only one anecdotal indicator of the degree to which the mobile phone has already become an integral part of everyday life within the study group. Seluula got her name because her father, when it came to arranging her naming ceremony a week after her birth, was absent for travel but passed all relevant instructions, including the name that she should bear, via mobile phone. Mahamadu’s in-laws were somewhat indignant and although the child was officially given the name Zaara’u, which her father had chosen for her, the nickname Seluula stuck. Reflecting on this revolutionary development over the past decade, the wife of a migrant worker from the Damergou region in Diffa told me that the new possibilities of the mobile telephone played an important role in her acceptance of a life far away from her home area: When we first came here, if we wanted to send a message to our relatives back home, we first had to ask someone, since we cannot read or write. Then we would give the letter to someone who travelled there. Before we had an answer, it would take some time. Nowadays, living here is not as difficult as that anymore. We can always talk to our relatives on the mobile phone, since there are networks everywhere. (Yooɓe Araba, Diffa, March 2011) Modern mobile communication and the fact that it is affordable even to the less well-off has thus become an important means to cope with the new

162  Space, Place and Identity

dimensions of contemporary mobility. The phone helps migrants far from home to overcome their geographic isolation and maintain a sense of lived community at a distance. Despite the advantages of the mobile phone, its appropriation also has problematic aspects. Its regular use can clearly be qualified as part of the new and costly consumption patterns that often go along with urban life and put the intended economic goals of migrants into question (Loftsdóttir 2001a, 2004: 60). While the phone itself might be relatively affordable, the running costs can be considerable.9 Also, since mobile phones are not only communication devices but also status-symbols (Donner 2006: 44), multi-function phones with integrated photo- and video camera, radio and mp3 player – for the most part Chinese fabrications – become increasingly popular and are often worn visibly on a ribbon around the neck.

Urban Migrants as Connectors Urban-based Woɗaaɓe play a key role for the exchange of information between the different rural sites in the region and can thus contribute significantly to keeping the dispersed lineage group connected. Important news circulates via these urban-based lineage members who are in touch with and regularly receive information from different sides. They facilitate the information flow between lineage factions that have migrated into spatially removed areas and are geographically separated. The significance of migrants as connectors becomes transparent through considering the example of the migrant community in Zinder (Case Study 5). The urban-based Woɗaaɓe from the lineage factions in the Damergou and Koutous regions are in closer contact with one another than their pastoral relatives in these two spatially removed areas. This is due to the fact that (1) Zinder is an important target for migrants from both these regions and (2) both these regions are located in relative proximity to Zinder. In town, the members of these lineage factions

Figure 9.1  Urban connectors (diagram by F. Köhler).

The Translocal Dimension of Urban Migration  163

again form one community, based on their common lineage identity and on social and ritual exchange. Urban migrants can therefore in the contemporary condition play an important role in the translocal integration of the dispersed lineage. More than directly among each other, the members of the different lineage factions stay connected via their urban dwelling relatives. Case Study 5: Urban Connectors Maimuna, Buude and Siibi are all daughters of Dawra Egoyi, who currently lives in Ganatcha with his adult sons and their families. Maimuna is married to Nano Buuyo and lives in Salaga. Buude is married to Labdo in Ngel Tiireeji. Siibi, finally, is divorced and lives in Zinder with her children. Her residence is an important meeting place. When her brothers and sisters or other members of their families come to town, they often stay at her house. She thus maintains an intensive contact with members of the different localized communities of the Kuskudu in Ganatcha, Ngel Tireeji and Salaga and thus represents a significant link for the spatially removed parts of the family – and indirectly for the wider communities that live in the different sites. This is further reinforced by two secondary connectors: Muusa, the son of Buude, and Riskuwa, the son of Maimuna. Both about twenty-two years old at the time of fieldwork, the two are matrilateral parallel cousins (bibbe yaaye’en, MZS). While their immediate families currently live in two different rural sites (Ngel Tireeji and Salaga, respectively), the two young men both stay in Zinder – in Moussa’s case to attend high school and in Riskuwa’s case to work. As age mates (waalde’en) who grew up together in the urban environment, they are emotionally very close and thus, in the urban context, constitute another important link between their families. Moreover, Muusa lives in Zinder at Siibi’s place and is thus regularly in contact with the different rural-based relatives visiting or passing by. Riskuwa lives on the compound of the house where he works as a watchman, yet spends a lot of his spare time with Muusa at Siibi’s place. The close relationship between Muusa and Riskuwa thus additionally fosters links between the communities of Ganatcha, Salaga and Ngel Tireeji.10

Urban migrants can thus be ‘nodal points’ (de Bruijn and Brinkman 2012: 47) for the exchange of information between spatially separated lineage factions. They function as bridges between different rural communities – as ‘linking nodes’ (Hedberg and do Carmo 2012) of translocal interaction and communication. The case confirms that migration can also strengthen social ties and thus favour continuity rather than being inevitably a factor of social disruption (Klute and Hahn 2007: 11). Besides, the example of Siibi and her sisters, who all live in different regions, also shows the important role of women for establishing

164  Space, Place and Identity

and maintaining translocal links between spatially removed communities. While brothers often live in close association, sisters, due to the principles of virilocality and female relocation in the context of marriage, often live spatially removed from one another, yet generally make remarkable efforts to maintain strong relations among each other across space. More recent research on migration and translocality has shown how translocal links between rural and urban places are materialized ‘through repeated communication, flows of knowledge and ideas, and political, cultural and economic activities’ (Hedberg and do Carmo 2012: 3). The mobility of urban migrants is therefore an important stimulus for change in their communities of origin (Marfaing 2014a; Klute and Hahn 2007: 16). Similarly, in the study group, the contemporary forms of rural–urban mobility and the flows that they entail facilitate the diffusion of new concepts and ideas and thus induce significant change in the pastoral realm, e.g. in the fields of habitat, education, or technology. However, in stressing the significance of urban migrants for maintaining translocal contacts across regions, the continuity of direct exchange between rural communities through travel should not be underrated. While there are important urban–rural flows that give impetuses for social change, in a second instance, there are also rural–rural flows both directly across sites and indirectly via urban connectors. An example from which the role of translocal networks and information flows for social change becomes obvious is the creation of rural schools and localized rural centres. As explained above, the materialization of Ganatcha and other sites as proto-villages with wells, schools and several additional buildings, is to an important extent the product of networks with expatriates and development organizations, and ultimately of an interplay between the two. These networks, in turn, are a direct result of the new mobilities of urban migrants and the new connections that they establish across places and regions. After the establishment of the first school in Ganatcha, the idea and the necessary strategies were quickly diffused to the other areas through rural–rural and rural–urban links. As mentioned above, lineage–internal competition also played an important role in accelerating this process of diffusion. The new social contacts that urban work migration brings about – in the study group in particular networks with expatriates – thus have a significant impact not only on social change in the pastoral society but also quite concretely on the production of new places as definitive physically materialized locations, hence at once on mobility patterns and placemaking strategies.11 However, this impact does not work one-sidedly towards sedentarity: the Woɗaaɓe have also learned from Westerners that there are actors today (pastoralist projects, researchers) who defend the cause of mobility. This has encouraged them to develop alternative ways of thinking more positively about mobility and nomadism in the contemporary situation, and to hold on to their mobile way of life with the support of these actors.

The Translocal Dimension of Urban Migration  165

The regular mobility of actors between urban and rural locations, possible on the basis of local proximity in the regional migration context, plays a significant role here. Freitag and von Oppen (2010: 5) have highlighted the significance of circulations and transfers as ‘the outcome of concrete movements of people, goods, ideas and symbols’. The contemporary mobility and the exchange that it entails has an important transformative impact on particular places (Greiner 2010: 137). In this regard, migrants who return to their pastoral communities play a particular role. Urban migration is not an irreversible movement. Individuals who have spent more than ten years in towns can ‘retire’ again to a pastoral life if the motivation for staying in town is no longer given. Most agree that city life is economically viable only as long as one can rely on a steady income. Urban life is regarded as a matter of opportunities, following the rationale of resource appropriation rather than being a value in itself. Young men who do not own a sufficient number of animals to get married and establish an independent household may be urged to set out for work migration. However, a man who has remained in town for some time without finding a job would rather be advised by his lineage mates to return to pastoralism, even if, for lack of animals, he would have to start as a hired herder or by herding the animals of a cousin or brother in the hope of gradually reconstructing a herd. Even long-term migrants generally claim that they eventually want to return to pastoral life (e.g. Köhler 2017a: text 13,22; Loftsdóttir 2002a;), and numerous examples show that an eventual return to pastoral life is not only desirable for most migrants, but also feasible and in fact a recurring pattern. Many biographies contain multiple shifts between periods as pastoralists, followed by others as urban migrants (e.g. Köhler 2017a: texts 13 and 14), and eventually, in cases of older individuals, often a final return back to the pastoral sphere (ibid.: text 12). Many biographies thus reveal patterns of circular or return migration. One case of return migration which exemplifies that the identification with the community of origin often remains vital even in a second generation of urban-born migrants’ children is that of Riskuwa, the eldest son of Nano Buuyo. Riskuwa spent the greater part of his childhood and early adulthood in Zinder, where his father stayed as a migrant worker. He went to school and later received a training in information technology yet did not find other employment than that of a watchman. When a school was established in Salaga, Riskuwa decided to follow a basic instruction as a teacher and he became the first from the study group to work as an instructor in one of the recently established community schools. Riskuwa teaches adult literacy classes and is paid by the internationally funded local NGO that supports the school in Salaga. Riskuwa’s case makes it quite clear that if migrants return home today, this does not necessarily mean that they return to a primarily pastoral activity. Rather, such return migration can fuel transformation processes in the pastoral realm

166  Space, Place and Identity

through the transfer of ideas, knowledge and concepts acquired in town. When migrants return to their home communities, they introduce change by engaging in local processes and by introducing and multiplying new concepts (Freitag and von Oppen 2005: 4). But even migrants who never return to the pastoral setting can nevertheless have an important impact on the shaping of the emerging new localities. The transmission of ideas and concepts, and even material transfers, have today become widely independent of physical contacts due to the new possibilities of connectivity over distance. The strong influence of migrants on their home communities is closely related to the fact that they ‘maintain an identification with their home society while diverging culturally from it’ (Glick-Schiller 1990: 342). This cultural divergence seems significant, as it constitutes the nucleus for change. Hence, although the pastoral home community remains an important centre of identification for urban migrants and a close emotional and social attachment with it is maintained through strong translocal ties, this does not mean that the urban experience is insignificant for identification. Glick-Schiller’s characterization of migrants as ‘fully encapsulated neither in the host society nor in their native [society]’, while they ‘nonetheless remain active participants in the social settings of both locations’ (ibid.: 330) applies well to the case of Woɗaaɓe urban migrants. They construct their identities in relation to both societies and they are limited by the boundaries of neither the one nor the other sphere (ibid.: 342). In short, it is neither the urban nor the pastoral sphere that defines contemporary Woɗaaɓe identities, but rather the mobility between the two. The pastoral home community must therefore not be mistaken as the source of a more authentic or genuine, ‘natural’ identity, while regarding migrant identities as derivations or extensions and thus inauthentic (Gupta and Ferguson 1997b: 7). The home community and the urban migrant community are not to be understood in terms of genuineness and derivation, but as different dimensions, or manifestations, of community. Woɗaaɓe migrants can be regarded as simultaneously situated across different locales (Brickell and Datta 2011: 4) which today increasingly span different spheres, their identities being shaped by the interplay of movement and attachment (Olwig 1997: 19; Fortier 2000: 2). The ‘culture of migration’ (Klute and Hahn 2007) characteristic of Woɗaaɓe migrants, i.e. a dynamic culture, which is ‘contextualized in the societies of origin as well as in the host societies’ (ibid.: 16), allows them to adapt to rural and urban lifestyles and to shift flexibly between them. The Woɗaaɓe case thus confirms the observation that migration is not necessarily perceived as a ‘dangerous undertaking’, exposing migrants to ‘incalculable risks or aggressions’; nor is ‘arriving at a destination in all cases felt as difficult processes of integration and assimilation into strange host societies’ (ibid.: 9). Especially young adults who have grown up in town move in the urban space with as much ease as they do in the pastoral environment. Urban-born children who grow up in a context of translocality can in many cases change

The Translocal Dimension of Urban Migration  167

flexibly between urban and rural identities (see Greiner 2010: 150; Friedmann 2002: 302). Urban migration in this translocal context does not inevitably imply a radical change of lifestyles, and a complete biographical break, as it has long been assumed in a rather generalizing way (Klute and Hahn 2007: 9; Boesen 2010: 28; de Bruijn et al. 2001). The urban experience has been integrated into the contemporary self-conceptions of Woɗaaɓe, i.e. into conceptions of an imagined community that comprises both migrants and non-migrants (Klute and Hahn 2007: 9).

Translocal Identities Just like pastoral migration, urban work migration leads to a dispersal of communities, which, however, remain connected across space as translocal communities. Analysing the contemporary condition of Woɗaaɓe urban migration with a perspective of translocality can shift the focus to connections and processes of economic, political and cultural significance. It reveals patterns of multiple belonging and thus avoids depicting Woɗaaɓe migrants one-sidedly as uprooted victims of drought, lost in an alien and hostile environment. Woɗaaɓe migrants generally do not cut their relationships with their communities of origin but continue to engage with them in processes of exchange and communication, maintaining close links by frequent visits and increasingly with the help of modern communication devices. In the case of Zinder, the proximity of the city to the migrants’ pastoral home areas facilitates the maintenance of translocal community ties. Short-range migration can avoid the disruption of social groups that is characteristic in many other migration contexts and allows for the maintenance of relatively intense and immediate social relations (Marfaing 2014a: 46). The maintenance of these relations is so important that migrants’ choices about their target area for migration are significantly determined by considerations of proximity and the availability of connective infrastructure, notably telecommunication and public transport, which both play significant roles in supporting the translocal maintenance of community. Today, migrants’ children are socialized into this translocal context. They develop multiply-bounded identities which combine rural and urban lifestyles and they can learn to switch between the two. Mobility continues to be a central feature of people’s ways of life. This mobility is often circulatory: Woɗaaɓe migrants in Zinder move back and forth from the city to the pastoral sphere and connect the two through a network of translocal exchange and a circular flow of people, goods and ideas. Although longer periods of urban migration work are a typical feature of contemporary Woɗaaɓe biographies, urban migration is not one-directional, but multiple shifts from pastoral to urban activities and back are a normal part of Woɗaaɓe lives. Next to spatial mobility, socio-economic mobility has become a characteristic feature. The constant exchange between

168  Space, Place and Identity

rural and urban sub-communities is vital for both and has a transforming impact on them. The spaces within which the socio-economic mobility of the Woɗaaɓe takes place can thus be understood as spaces of circulation, or as ‘circulatory territories’ (Tarrius 1994; Marfaing 2014a) – if we understand ‘territory’ in Bonte’s sense as ‘the location of the real appropriation of the totality of natural resources used by the community‘ (1979: 215). Supported by complex social networks, circulatory mobility within these spaces is used in order to appropriate the available resources. In this sense, the social patterns of contemporary mobility show a strong resemblance and continuity with those of pastoral mobility. The city may today constitute a particularly important place of exchange and access to resources (Retaillé and Walther 2011: 94), yet it is only a part of a wider circuit of mobility. The urban space is just one space for resource appropriation among others. Mobility in the Sahel has been identified as a risk management strategy and thus as a general part of people’s identity. This mobility – both in the pastoral and the urban migrant context – functions on the basis of translocal social networks. Here again, then, we have a confirmation of the significance of translocal networks as a risk management and social security strategy (Marfaing 2014a: 54). The contemporary translocal livelihoods across rural and urban spaces that have emerged in the context of urban work migration are in this sense only a new facet of a long-established phenomenon. If relational networks across space have long offered important options for pastoral mobility, the contemporary mobility across rural and urban spaces is no less reliant upon them. The urban and the pastoral components, just as they are today complementary in economic terms, are interdependent for the production of community and as sources of identity (see Greiner 2010: 151). Community is a networked whole of parts that are dispersed between various rural and urban places, the constituent sub-communities being connected by translocal ties. Translocal ties to the home community are a central element of identity construction of migrants and an important factor of cultural continuity despite displacement in the migration context. This identity construction is less place-bound than relational, based on networks of translocal links, and it is thus in continuity with the patterns of identity construction of mobile pastoralist Woɗaaɓe. It is based on a concept of translocal belonging in which kinship bonds prevail over the criteria of a specific, localized place. In this sense, migrants’ patterns of identity construction are characterized more by continuity than by disruption. In terms of integration into territorial and administrative structures, urban migration plays an ambivalent role. On the one hand, urban migrants selectively make use of the state’s offers (such as social services like schools and health facilities), and as the urban component of a wider translocal community, they make efforts to let their rural-based relatives profit from these benefits as well. On the other hand, even long-term urban dwellers are often not particularly sedentarized

The Translocal Dimension of Urban Migration  169

in the sense of ‘domestication’ by state structures. Neither the intra-urban, nor the rural–urban mobility is controlled by the state. Urban migrants move between these spheres largely without notice of administration and authorities. In their struggle to balance selective integration and the persisting identification with their society of origin, Woɗaaɓe simultaneously adapt to and resist the structuring impact of the state (Glick-Schiller 1990: 341f.).

Notes   1. For comparable examples, see also Greiner (2010:146, Namibia), de Bruijn and Brinkman (2011:49, Cameroon) and Acloque (2014, Mauretania).   2. See de Bruijn and Brinkman’s (2012: 47) critique of Olwig’s concept of ‘siting culture’.   3. There is some regional variation between the seasonal patterns of concentration and dispersal (compare, for example, Bonfiglioli 1998; Dupire 1962 and Boesen 2010), but the general characteristics are comparable.   4. As of 2010. Since flexibility and unsteady, precarious conditions are characteristic of the situation of most Woɗaaɓe migrants, the job-status or residential status of some people mentioned have changed since fieldwork was completed, in some cases even while fieldwork was conducted. In these cases, I indicate to which time the data given in the examples refer. If no period is mentioned, the status of the respective person has remained unchanged from the beginning of fieldwork until the time of publication.   5. See Boesen (2009: 76) for central Niger.   6. On the division of tasks between brothers as a joint economic strategy, see Hampshire and Randall (2005: 131).   7. As of 2010. In July 2011, Nano lost his job in Zinder and returned to Salaga.   8. See also Grémont (2011: 184) for the Tuareg in Niger and northern Mali.   9. The cheapest offers for a cellphone are of about 10,000 FCFA (ca.15€). One minute of communication costs about 100 FCFA (ca. 0,15€). 10. As of 2011. 11. See also Boesen (2009), Lassibille (2009) for central Niger.

Part V

Gassungol Woɗaaɓe The Translocal Network of the Ethnic Group

Chapter 10

The Translocal Community and Social Reproduction

As I have argued throughout this book, the translocal dispersal of social groups among the pastoral Woɗaaɓe has a double dimension: not only have historic migration processes led to descent groups being dispersed over spatially separated and sometimes far removed home ranges; the lineage faction that shares one home range also lives dispersed for a good part of the year due to the scarcity of pastoral resources and the desire to avoid overgrazing. This dispersal is periodically overcome by rainy-season gatherings (worso), which are the occasion for the ritual renewal of internal social bonds by jointly performing community constituting rituals (Schareika 2007, 2010b), notably name-giving ceremonies for new-borns (humturu) and clan endogamous betrothal marriages (kooɓgal) that are celebrated in an accumulated manner when the clan section is assembled. According to Islamic custom, name-giving ceremonies would be due seven days after the birth of a child. For the Woɗaaɓe, however, the participation of a representative part of the community in the ceremony is imperative for ritually acknowledging the integration of new-born clan members into the community (Schareika 2007, 2010b). Therefore, when pastoral groups are dispersed, only a relatively small and informal ceremony is generally held a week after the child’s birth (Bonfiglioli 1988: 142). A sheep might be slaughtered on this occasion as Islamic custom requires, yet the socially more relevant ceremony is carried out during the worso. As important as the worso, yet at the level of the ethnic group, are the ceremonial inter-clan meetings, ngaanka. It is no coincidence that these two institutions, worso and ngaanka, represent the major ceremonial framework for the social reproduction of the regional clan community and the ethnic community, respectively. Both are closely linked to marriage and thus to the physical

174  Space, Place and Identity

reproduction of society. This chapter explores the two principal forms of marriage in the institutional framework of meetings of the regional clan segment (worso) and inter-clan meetings (ngaanka), and the role of these two ceremonial institutions for social reproduction.

Two Conflicting Forms of Marriage Woɗaaɓe marriage patterns are predominantly endogamous when ethnic identity is taken as a parameter. With regard to clan identity, however, two opposed, almost inverse, principles are in play, corresponding to the two principal forms of marriage that are practiced. The first form, called kooɓgal, is a generally clan-­ endogamous union, arranged between the families of the couple by betrothal, often from early childhood. The preferred patterns are father’s brother’s daughter (FBD) marriage and cross-cousin marriage. A majority of kooɓgal marriages are thus between genealogically closely related kin. Through the repeated exchange of animals, kooɓgal marriage strengthens the bonds between kin from either the same or different lineages within a clan and thus reinforces the internal cohesion of the clan (Schareika 2007, 2010b). Te’egal marriage, which is generally clan-exogamous, is based on the elopement of married women with men from other clans and formalized as a proper form of marriage.1 While kooɓgal is the cultural ideal and thus of a higher social value (Loncke 2015: 312), te’egal is a marriage of choice, contracted by the couple itself instead of arranged by the families, and the relationships are therefore often said to be of a more affectionate nature. There are rare cases of te’egal within the same clan, yet they are widely dismissed as harmful for the unity of the group. The reason is that, by definition, te’egal is arranged with a woman who is already married according to the principles of kooɓgal. Since the consent of the woman is presumed, yet not that of her husband, violent reactions and acts of vengeance are common. However, they are generally acted out on an individual level, without mobilizing larger descent groups as might be assumed in a lineage society. The reason is that, despite the conflicts associated with the practice and although disapproved by the aggrieved husband in any particular case, te’egal in general is unanimously defended as a legitimate cultural principle governing the relations between clans. The interest in the practice is indeed general and multiple: for women, te’egal elopement constitutes a possibility of escaping from an unhappy marriage whereas divorce is not an option for them. For men, success in te’egal is a source of social renown. The option of individual choice, which te’egal offers other than kooɓgal, is an important argument for both women and men as it carries the promise of a more affectionate relationship. But the practice is also in the interest of the clan. A man’s success in te’egal strengthens his own clan to the disadvantage of another, since the wife and offspring of a te’egal union, if it remains stable,

The Translocal Community and Social Reproduction  175

reinforce a man’s patrilineage. This aspect potentially gives te’egal a demographic and thus clearly a political dimension. Perhaps the most important aspect, however, is that te’egal can level out a significant risk involved in kooɓgal: over time, the practice of closely endogamous marriage would lead to an increasingly closed social group with the risk of isolating itself and thus becoming socially and politically vulnerable. This is why an element of exogamy of some degree is desirable (Tylor 1889: 267; see also Lévi-Strauss 2002 [1947]). The offspring of te’egal relations have relatives in a different clan and thus a more extended social network (Schareika 2007: 187). Although a te’egal wife is integrated into the clan of her husband, she does not simply give up her former clan identity with all the social relationships it comprises. Rather, te’egal marriage links lead to multiple bonds and affiliations. Hence, the clan-exogamous marriages open up new options for interaction, create alliances and hence politically strengthen the ensemble of the regional clan segments (ibid.: 316). For the group, the advantages resulting from a larger network of relations and alliances translate into a stronger political position for negotiating access to pastoral resources, whilst for the individual the additional social bonds offer possibilities for reinforcing networks of social security, e.g. by animal loans (haɓɓanaaye). From a politico-economic angle, the opposing principles of clan-endogamous kooɓgal and clan-exogamous te’egal marriage can be interpreted in terms of a struggle between two fundamental principles for forging alliances among groups: on the one hand the strategy to keep the in-group closed and on the other, the strategy to define wider identities and to widen alliances – in short, the dialectics of inclusion and exclusion (Schlee 2004, 2009). Both principles can be advantageous in different situations, and actors strategically apply them according to their situational interests (ibid.). The Woɗaaɓe case is a good illustration as it shows both exclusive and inclusive strategies for strengthening social groups, working at two different societal levels – the patrilineage/regional clan segment and the ethnic group – exemplified by the two forms of marriage (see Table 10.1). From the lineage perspective, kooɓgal marriage strengthens the cohesion within the patrilineage or the regional clan segment; from an inter-clan perspective, te’egal strengthens the cohesion of the ethnic group by establishing multiple bonds, and ultimately networks of kinship and alliance that bind its segments together (Loncke 2015: 90). Table 10.1  The role of the two main forms of marriage for social cohesion. Marriage form

Institutional/ceremonial framework

Level of society concerned

Group-strategic principle

kooɓgal te’egal

worso ngaanka

patrilineage/clan segment ethnic group

exclusive inclusive

176  Space, Place and Identity

It has been pointed out that the uterine bonds across clan lines that result from te’egal are often of a very affectionate nature, because in contrast to agnatic relations they are free from rivalries about inheritance (Paris 1997; Schareika 2007: 184ff., 316). Te’egal marriage thus establishes networks of kinship relations between clans, relations ‘by mother’s milk’ (enɗam), which are strong binding forces. An expression like ‘Gam enɗam, ngoore!’ (‘in the name of the mother’s milk [that we have shared; i.e. in the name of the uterine bonds that exist between us], stop [quarreling]!’) expresses an almost congruent meaning of maternal links and harmonious kinship relations. And yet, paradoxically, these affectionate relations across clan lines originate in what is clearly conceived as acts of aggression. While women are never abducted without their consent, from a male perspective te’egal is perceived as ‘theft’ (nguyka), because women are ‘stolen’ from other clans (Boesen 2008a: 154). Paris (1997: 76) has thus aptly called te’egal an act of ‘predatory seduction’.

Ngaanka – Inter-clan Agreements on Mutual Predatory Seduction The seemingly anarchical practice of te’egal is in fact governed by a framework of rules and takes place in the form of a competition between clans. In order to better understand this aspect, it seems crucial at this point to take a closer look at the role of ngaanka.2 Ngaanka designates institutionalized ceremonial meetings that are based on reciprocal visits between two clans or their regional segments. Delegations from other clans can attend these meetings as guests, but what is primarily at stake are the bilateral relations between the visiting and the hosting clan. A clan can entertain ngaanka relations with several other clans, either from the same or from the opposite clan cluster. A core element of these relations is an agreement about the legitimacy of mutual te’egal between the respective clans. The ceremonies notably comprise performances by the young men in the emblematic geerewol dance. The declared aim of the participants is to seduce married women from the adverse clan and, if possible, to contract a te’egal marriage with one of them (Paris 1997). Indeed, the numbers of women ‘stolen’ by each party are a matter of much discussion after the ceremonies. Consequently, ngaanka is also referred to as ‘war’ (konu), i.e. ‘war’ about women (konu rewɓe) by means of seduction (ibid.; Schareika 2007: 296, 325). The active role of women in what is conceptualized by the men as ‘war’ and ‘theft’ of women must, however, be stressed. Despite their warrior-like rhetoric, the men largely depend on the decisions of the women. The rhetoric of war describes not so much the relations between the sexes, but rather those between the clans. Although the ceremonies are an important occasion for contracting te’egal marriages, the latter is not only legitimate in this framework. The ceremonies’ function is rather to establish and reconfirm bilateral agreements about the practice. The direct link between ngaanka and te’egal marriage is not only explicitly

The Translocal Community and Social Reproduction  177

Figure 10.1  Choosing ritual at the climax of a geerewol-dance – Abdenaser, Damergou region, October 2011 (photo: F. Köhler).

expressed in the verbal statements of the participants, but also symbolically in several elements of the ceremony, the most important being the reciprocal choosing of dancers by two or three young girls from the adverse clan at the end of each geerewol performance (see Figure 10.1) – an element that also stresses the active, and in fact initiative, part of the women. The ceremonies invariably end with the sacrifice of a bull, which is of fundamental importance (Paris 1997). The meat of the sacrificed bull is shared by all the community members.

Sharing the Meat of Sacrificial Animals Schareika (2007, 2010b) has argued that social relations among the Woɗaaɓe are established by a ritual exchange in which the sharing and joint consummation of the meat of sacrificial animals plays a central role. He has impressively demonstrated this for the rituals of the worso, during which kooɓgal marriage and filiation celebrations constitute and reproduce community at the level of the maximal lineage and clan. I follow Schareika’s argumentation, but I extend it to the level of inter-clan relations, where the same principle can be observed in the rituals of ngaanka. Kooɓgal marriage establishes new kinship relationships between the involved families and an important function of the rituals in relation to it is to publicly acknowledge and implement these newly established relationships as a social

178  Space, Place and Identity

reality at the level of the community (Schareika 2007, 2010b). Kooɓgal marriage involves numerous transactions in several steps, normally covering a number of years, often from shortly after the birth of the wife until the weaning of her first child and her definitive transfer to her husband (Dupire 1963: 59ff.; Maliki 1981: 124f.). These transactions can be interpreted in terms of a gift exchange that establishes social relations and thus constitutes a community. Schareika (2010b: 97) distinguishes between two principal forms of exchange involved in establishing these kinship relationships: the first is the transmission of rights in productive capital. These concern the animals that are transferred from the husband’s family to the wife (sendereeji), and for which she receives milking rights in order to feed herself and her children. The second form of gift exchange concerns the animals that are ritually slaughtered, ideally young bulls (ga’i kooɓgal). Such animal sacrifices, provided by the father of the husband, are carried out up to three times during the process of kooɓgal. The meat of the sacrificed animals is ideally shared among the ensemble of the worso community (i.e. at the level of the patrilineage or the regional clan segment).3 Schareika (2010b), referring to Rappaport’s (1999) theory of ritual, has given a convincing interpretation of the significance of the sharing of meat for establishing a collective which advocates and protects the marriage in accepting the claim that it entails. According to Rappaport, rituals create social relations by symbolically representing them (1999: 108). Thus, the ritual act of the sacrifice and the shared consummation of the meat establish the marriage as a social reality. Yet, participation in a ritual means a commitment to the symbolically expressed meaning that it contains. Furthermore, the public character of the ritual creates a social pressure to respect this commitment henceforth (ibid.: 123). The meaning of this, and in fact the importance of the act of the ritual sharing of the meat for the recognition of a marriage claim, is contained in expressions such as – ‘I have eaten the meat of her/his marriage’ (mi nyaami tewu kooɓgal makko), meaning ‘I participated in her/his marriage’. By referring to the sharing of the meat, the validity of the marriage is recognized. The implicit logic is not merely that of a testimony, although this aspect is also significant, but rather that of an obligation deriving from a gift for the person who accepts it. Mauss’ (1990[1925]) classic model of gift exchange is based on a dyadic relation between the one who gives and the one who receives, and, in addition, an ‘enigmatic’ third element, vaguely identified by Mauss as a force inherent in the gift, with the power to induce the obligation of returning it.4 Schareika, in his analysis of kooɓgal, adapts Mauss’ model by resolving the question of this mysterious force and proposing a triadic relation between the two parties engaged in the marriage transaction and a third party represented by the collective of the clan members. They all receive a share in the meat of the sacrificed animal and hence become obliged to assure – if necessary, by active intervention – that the reciprocal obligations between the two mainly concerned parties remain

The Translocal Community and Social Reproduction  179

respected (Schareika 2007: 133ff.). Schareika admits that the obligations of the third party remain diffuse (ibid.: 135), but they lead to an effective protection of the marriage link – by respect (the clan-members will not try to seduce the woman) and, if necessary, by defensive action (the clan-members will act in solidarity and help the husband to take his wife back in case she elopes with a man from an adverse clan).5 The ritual thus establishes a collective approval of the union. The collective of the clan members acts as a sanctioning force which protects the claim established by the sacrifice and publicly recognized by the sharing of the meat. Those who receive a share are symbolically indebted to the husband in the sense of an obligation to reciprocate. The offering of meat establishes a new social relation which elicits solidarity; the gift is thus returned in the form of solidary behaviour. At the same time, the sharing of the meat establishes a collective at the clan-level, comprising those who have participated in the ritual and thus consented to accept the union as binding. However, te’egal from members of other clans remains a threat because there is no community, nor any institution of legal assistance that could defend the kooɓgal union above the confines of the clan. The sacrifice on the occasion of a te’egal marriage principally serves the same purpose: the husband’s kin, by sharing the meat, testify to the union and accept the new husband’s claim on his wife (Dupire 1962: 250f.). The ritual is thus an important marker to establish the status of the woman as the legitimate wife of her new te’egal husband within his clan. Also, just as in kooɓgal, the wife in a te’egal marriage receives milking rights for animals from her husband’s herd (Maliki 1981: 125). The structurally similar way of implementation makes it quite clear that te’egal is a proper form of marriage and not any sort of concubinage, which is socially dismissed (Dupire 1970: 67). The animal sacrifice that serves to officialize the te’egal union does not invalidate the woman’s existing kooɓgal marriage but establishes an alternative claim in the clan of the new husband (Schareika 2007: 152f., 2010b: 111). The two marriages continue to coexist as claims, yet remain valid in different segments of the ethnic group. The concerned husbands, who belong to two different clans, share the premises regarding the principles of implementation of marriage claims, but they do not regard each other’s claims as binding for themselves. Rather, each group considers the union established by its own members as legitimate and there is no overarching institution in the form of a central power able to decide on the matter and to enforce a decision. Hence, te’egal challenges existing kooɓgal unions without being able to invalidate them and could thus be characterized as a counter-marriage (Schareika 2007: 152). This is perfectly coherent with the logic of the above interpretation: since the clan members of a woman from clan A do not partake in the sharing of the meat, her te’egal marriage with a man from clan B does not have consequences for the status of her existing kooɓgal marriage within her own clan A.

180  Space, Place and Identity

Structurally speaking, we thus have kooɓgal marriage, which is established on the level of one clan, and te’egal marriage, which is formalized in a similar way (by means of an animal sacrifice and the transmission of rights in productive capital), but at the level of an adverse clan – in full knowledge, yet deliberate disrespect of the claim that already exists in the other group. On this basis, we can now examine the element of the sharing of meat in the context of a third ritual occasion that involves an animal sacrifice: that of ngaanka. As mentioned above, ngaanka ceremonies principally end with the sacrifice of a bull (ngaari ngaanka), usually provided by the arɗo of the hosting clan-section. Its meat is grilled in parts, which are then reconstituted in their anatomical order on the hide before the whole animal is presented to the visiting clan (see Figure 10.2). The delegation of the visitors will then longitudinally cut the hide into halves and divide the different parts of the bull into two equal shares, one of which they present again to the hosts before the meat is shared within both clans and eaten by all participants (Paris 1997). The meat of the sacrificed bull (tewu ngaanka) is the functional equivalent of the meat in kooɓgal (tewu kooɓgal): its ritual sharing establishes a collective. In the case of ngaanka, however, this collective transcends the clan level; it comprises representative parts of at least two clans, visitors and hosts. The sharing of the meat establishes a commitment to the principles that are symbolically expressed in the different rituals of the ceremony. Central among these principles is the mutual approval of the reciprocal exchange of married women by te’egal as symbolized by the mutual ritual choosing of a dancer by a girl from the adverse clan. Ngaanka ritually implements a normative framework for the acceptance of mutual te’egal as a shared cultural practice and, hence, as a positive value and legitimate principle governing the marital relations between the two clans involved. By participating in the ritual, the members of the two clans commit themselves to this rule. This is congruent with the theory of Rappaport who maintains that: The primary function or metafunction of liturgical performances is not to control behaviour directly, but rather to establish conventional understandings, rules and norms in accordance with which everyday behaviour is supposed to proceed. Participation in a ritual in which a prohibition against adultery is enunciated by, among others, himself may not prevent a man from committing adultery, but it does establish for him the prohibition of adultery as a rule that he himself has both enlivened and accepted. (Rappaport 1999: 123) The ritual has thus established compliance regarding the mutual toleration of te’egal. But how is it assured that this principle is respected? Following Schareika’s interpretation of kooɓgal, where respect of the contractual claim is controlled by

The Translocal Community and Social Reproduction  181

a third party formed by the clan members, the issue in ngaanka is also to identify a collective that assures the respect of the ritual agreement between clans. I think the question about this collective (Mauss’ ‘enigmatic’ third element) can be answered in different ways. Ngaanka being a public ceremony that principally involves two clans, the collective that assures control of the agreements at stake is congruent with the ensemble of the participating members of these two clans. Participation in the rituals of ngaanka, in particular the sharing of the meat of the sacrificed bull, establishes a collective that comprises the members of the two clans involved, who will henceforth, by virtue of their repeated participation in the rituals and the indirect commitment it entails, accept and actually defend te’egal as a legitimate principle and ‘cultural rule’ because they have interiorized it as a part of tradition (ndonu). According to Schareika’s (2007) protocols of an interclan meeting, devoted to the question of whether te’egal should be abandoned between the two concerned clans or not, this becomes evident in the arguments put forth to defend the practice: it is the way of the tradition, literally ‘the way we have found’ (‘laawol ngol tawɗen’, ibid.: 366) and therefore a good or ‘clean’ one (‘laaɓngol’, ibid.). Hence, the only acceptable answer is to keep on following it (‘tokki laawol ngol’, ibid.) – which in essence is also the final outcome of the meeting. But another aspect also comes in here. Over time, the exchange of women by means of te’egal has concrete, physical results: it leads to uterine bonds, to cross-cutting ties, which in fact make relatives of the members of adversary clans. This becomes equally evident from Schareika’s (2007) data. One speaker refers to the members of the other clan as ‘those who gave birth [to us]’ (‘danyooɓe’, ibid.: 363) because among the mothers in one group are women from the other, who had been integrated by means of te’egal (ibid.: 309, 363f.). It is in this sense that the uterine bonds between clans, which are considered to be of outstanding value (‘enɗam feere woni’, ibid.: 343), are said to have created the Woɗaaɓe (‘enɗam aawi en’, ibid.: 371). Hence, te’egal is regarded as being at the basis of the shared identity and the abandonment of the practice is morally impossible because, ultimately, too many Woɗaaɓe are children of te’egal marriages or have been begotten by men who were socialized in a clan other than their fathers’, because their mothers had eloped in te’egal with another man when they were small children (ibid. 341). In this reading, the force that assures the application of the rule would be constituted by those individuals with social and kinship ties to two clans. By their concrete, physical existence they remind the community of the close familial bonds – ties of ‘blood’ and ‘milk’ (ƴiiƴam and enɗam) – between clans. It is these cross-cutting ties that make palpable the importance of te’egal for the unity of the ethnic group. Te’egal is at the same time a cultural heritage and an important factor for perpetuating the cohesion of the ethnic group by binding it together through links of kinship.

182  Space, Place and Identity

Figure 10.2  Preparing the presentation of the ngaari ngaanka – Kilaake, Damergou region, October 2011 (photo: F. Köhler).

As discussed above, as an ethnic group the Woɗaaɓe do not dispose of any level of interaction that integrates all of their constitutive segments. Above the level of the clan, or its regional segment, the ethnic groups exists at best as the sum of the bilateral relations maintained between segments of equal order. Te’egal and ngaanka are crucial institutions by which these relations are initiated, materialized, maintained and structured.

Rituals of Cultural Affirmation During the presentation of the sacrificed bull, a representative of the hosting clan has to enumerate, in a strictly prescribed order, the different parts of the animal (Paris 1997). The visitors will then return one half of all parts and present them, for their part, to the hosts. This mutual presentation and enumeration is a crucial element of the overall ceremony. It occurs towards the end of the meeting and the presentation by the visitors generally ends with a formula expressing a re-invitation, i.e. an approval of the maintenance of the mutual ngaanka relations. The enumeration of the parts of the sacrificed bull has a double function: firstly, it is public proof that no part is missing, that the party who is offering the animal has accomplished the sacrifice in the prescribed way, and that both parties will receive the share they are due. In this sense, the ritual stresses the principle of equity and balance which, on a symbolic level, expresses the necessity that the competition for women, which is at stake in ngaanka, remains equally balanced

The Translocal Community and Social Reproduction  183

between the participating clans. Secondly, the enumeration is a demonstration and a mutual test of ritual and, more broadly, cultural knowledge. Each clan aims to demonstrate that it is worthy of further engagement in ceremonial, and hence, marital exchange. The bull (also called ngaari fulfulde, the bull of the Fulɓe tradition) is a symbol of the pastoral way of life (ngaynaaka). Its presentation comprises, in addition to the parts of the animal itself, a number of emblematic items of pastoral life: an axe (jammbere), a calabash dipper (horde), a knife (laɓi), a milk beater (buruugal) and, most importantly, two strings of bark-fibre that are attached to the animal, one through the nasal septum and one to the tail (see Figure 10.3). They are called ba’ajol pulaaku or ba’ajol fulfulde and are symbols of the code of conduct and moral law which the Fulɓe call pulaaku and the Woɗaaɓe call mboɗangaaku. The cultural ‘way of the Fulɓe’ (laawol pulaaku) is sometimes metaphorically referred to as a rope or a net which maintains the unity of the ethnic group by binding its elements together (Maliki 1981: 130; Bonfiglioli 1988: 7; Schareika 2007: 319ff.). The metaphor gassungol Woɗaaɓe, as a designation for the ethnic group, is an expression of the same imagery (Paris 1997: 76; Lassibille 2006: 126; Loncke 2015: 94, 249; 2008: 219f.). Gassungol or boggol gassungol is the name of a special net of rope which is used to attach the load of the household goods to the pack animals when the camp is moved. The enumeration is thus a ritual evocation of essential symbols of cultural identity. The presentation of the bull can be interpreted as a ritual appreciation of the adverse clan, based on the mutual recognition of the other’s conformity with the cultural norms and values of mboɗangaaku (Paris 1997: 75). This mutual appreciation between clans, which is at the basis of the agreement surrounding the competition for women, is another central function of ngaanka.

Figure 10.3  A string of bark fibre is fixed to the nares of the ngaari ngaanka – Abdenaser, Damergou region, October 2011 (photo: F. Köhler).

184  Space, Place and Identity

The different ritual competitions in ngaanka – in dance, song, male beauty and ritual knowledge – are dramatic evocations of shared cultural values. Te’egal, as symbolized by the performances, is a central part of these values. Hence, participants in ngaanka make a commitment to this practice as a positive element and part of the cultural way of the Woɗaaɓe, the mboɗangaaku. The crucial point, then, is that the statement regarding the affirmation of te’egal is imbedded in a more general affirmation of a shared cultural identity. Competition (not only for women) as a basic structuring principle for the relations between clans is an important part of this shared cultural identity. However, their structural opposition does not so much separate the segments as tie them together. The aspects of competition and mutual recognition cannot be separated, just as the question of te’egal cannot be isolated from other elements of cultural identity. Ngaanka is a ‘total’ cultural phenomenon. This essence of Ngaanka can be grasped by Mauss’ concept of ‘prestation totale’ (Mauss 1990 [1925], 1947), which designates a contractual alliance between collectivities to engage in reciprocal exchange of things or services – among others, Mauss cites rituals, dance and women. These contracts are often characterized by an agonistic rivalry and competition, and, although seemingly based on voluntariness, they are in final consequence ‘strictly compulsory, on pain of private or public warfare’ (Mauss 1990 [1925]: 5). Dupire, without explicitly referring to Mauss, has similarly underlined the obligatory character of ngaanka, calling it a debt that does not expire (Dupire 1962: 312). The offering of a bull by the hosting clan induces a debt and the obligation for the visiting clan to return the gift, even after years, thus assuring the continuity of the relationship. Yet, the contract comprises an implicit agreement about another object of reciprocal exchange: the exchange of women by mutual te’egal marriage. A cycle of exchange between the groups is thus initiated – as in many societies – by the exchange of women (Lévi-Strauss 2002 [1947]) and a permanent alliance kept up by means of intermarriage (Tylor 1889: 267). The specificity of the case is that these women are already married in their clans and that the exchange takes the inverted form of negative gift-exchange, i.e. mutual ‘theft’. Nevertheless, it is still based on reciprocity, and is likely to establish social bonds. Each party has the right to take and the obligation to accept if the other takes in the same way. Mauss observed that peace ultimately relies upon the fulfilment and respect of the pact that binds the contracting parties (1947: 186). In the Woɗaaɓe case, peace is maintained at the price of conflicts, or, to put it more paradoxically, at the price of ‘war’: the fulfilment of the contract demands the acceptance of violations of basic social rules in the name of other rules, of a higher virtue. The adversaries become partners who, by reciprocal acts of giving and receiving, become mutually indebted to each other. Looking at ngaanka in terms of reciprocal exchange allows for a better understanding of the competing processes of opposition and social cohesion which are

The Translocal Community and Social Reproduction  185

at work. The mutual interest in the exchange of women and the social bonds that result from it bind the parties together in contractual relations of total prestation. Ultimately, this is the central motive for clans to engage in ngaanka contracts.

The Network of the Ethnic Group Godelier has further developed Mauss’ concept of prestation, and recognizes that one of the central functions of such ‘total’ social phenomena is to enable society to ‘represent itself (to others and to itself) and to reproduce itself as a whole’ (1999: 40). This clearly applies to ngaanka as well, which has been interpreted as central for perpetuating the cohesion of the ethnic group. As Dupire (1962: 312) has pointed out, it is the only ceremony which exists on the scale of the ethnic group. Boesen, without explicitly referring to Godelier, uses a very similar phrasing: ‘[Ngaanka] can be seen as (re-)creating the community as a whole – both through concrete matrimonial exchange (te’egal) and in the sense of a symbolic self-manifestation’ (Boesen 2008a: 158f., see also Lassibille 1999: 257). As noted above, the Woɗaaɓe symbolically allude to their ethnic group through the metaphor of a ‘net’ (gassungol) or network. Groups can be integrated into or drop out of this network, and the maintenance of ngaanka contracts – or the failure to maintain them – is a crucial element in these processes. In this sense they are not only of social but also of political relevance: mutual ngaanka relations define group-membership (Dupire 1962: 312; Lassibille 2008: 7ff.). The ngaanka contracts, with their implication of reciprocal exchange of women through te’egal inter-marriage, can be regarded as a condition sine qua non for a clan to become and to remain a part of Woɗaaɓe society. A direct contract with all segments at the same time, however, is not a necessary condition for such recognition and ethnic belonging. It is enough to connect with some segments by ngaanka opposition and for these again to connect with others in a way that forms the network of multiple, translocal inter-clan relations that defines the ethnic group (see Loncke 2015: 248f.). This principle is excellently captured in Paris’ (1997) seminal article on ngaanka: by analysing, over a period of more than twenty five years, all the pairings of clans who entertain mutual relations of ngaanka, Paris gives a comprehensive idea of how the fluid entity of the ethnic group is maintained in this sort of a loose network of bilateral relations, and of the role that ngaanka plays in this process. With the image of the gassungol Woɗaaɓe, the Woɗaaɓe have given their ethnic group a well-fitting, metaphoric designation. The group is conceived as a network, and ngaanka is what ties the knots together – through the kinship links that it creates and that, quite substantially, tie the clans together (Schareika 2007: 316). An interruption of all ngaanka relations and the abolition of te’egal by a particular clan would have the consequence that the clan would drop out of the network and, in the long run, lose its ethnic identity.

186  Space, Place and Identity

Therefore, ultimately the maintenance of the network of ngaanka contracts is what allows for the continuity of the group. It allows society in Godelier’s sense to represent itself and to reproduce itself as a whole. Hence, in ngaanka, ethnic identity is at stake. It is in this sense that ngaanka can be understood as an equivalent of the worso: both institutions play crucial roles for the constitution and reproduction of society on the levels of the ethnic group and the clan, respectively. And both institutions are expressions of the relational, as opposed to territorial, mode of space organization characteristic of pastoral Woɗaaɓe. The regional clan community (ritually reaffirmed in the joint worso celebration) and the ethnic group (constituted by the network of ngaanka contracts) can be understood as defining translocal and mobile social spaces (Retaillé and Walther 2012: 2f.). Worso and ngaanka are essential occasions for the periodic manifestation of these translocal social spaces, and of the reproduction of society. It is here, perhaps, that the notion of place as a meeting place where encounters and social interaction are performed becomes particularly palpable. The periodical meetings are place-making and community-producing in the sense that they temporarily materialize the translocal social spaces of the clan and the ethnic group in ritual frameworks and thus reproduce and reconfirm them (Loncke 2015: 249). When worso and ngaanka take place, the social spaces of the intra- and inter-clan networks manifest themselves for the time of the ceremonial encounters in a concrete place (Retaillé and Walther 2012: 12). When the meetings take place, they produce place,6 through the interaction of their participants and through their performances (Sheller and Urry 2006: 214). Regardless of the aspect of competition, ngaanka requires cooperation between clans and, hence, at the same time fosters mutuality and reciprocity. Fierce rivalry and a strong sense of belonging through the affirmation of a shared identity are here intrinsically linked. Schareika (2007: 315) has convincingly demonstrated that the consent to the admission of te’egal is the expression of a conscious political will for structuring the relations between clans. The aspect of marital exchange between clans as reinforcing the cohesion of the group is just as important in the ceremonies as is the ritual occasion for reaffirming common cultural values and enlivening them in cultural expressions that are crucial for the formulation and reproduction of ethnic identity. Te’egal is part of this identity and, hence, a positive value. In the symbolism of ngaanka, it is explicitly represented as part of the cultural values of the group, hence of identity. Ngaanka establishes common values and is thus an important factor for we-group stabilization (Elwert 2002: 47). On a different scale, the same holds true for the worso. This ultimately seems to be the reason why the periodical materialization, or spatial manifestation, of the fluid networks of clan and ethnic group in the ceremonial framework of worso and ngaanka, respectively, is essential for the maintenance and reconfirmation of a notion of community under conditions of

The Translocal Community and Social Reproduction  187

translocality. As Henri Lefebvre (1991 [1974]: 101) reminds us: ‘The form of social space is encounter, assembly, simultaneity’.

External and Internal Interfaces: Integration and Cohesion through Opposition As argued above, crucial identity boundaries not only define the Woɗaaɓe in relation to other ethnic groups (external interfaces), but also the different segments in relation to each other (internal interfaces). These boundaries are characterized by the dialectics of opposition and alliance typical of segmentary lineage societies, and they find their clearest expression in the institutionalized competition over women. The ethnic group is eventually defined and shaped along these boundaries – the interface between different clans – just as much as along the external boundaries – the interfaces with other ethnic groups. The structural opposition between the segments is defined by and expressed in the ngaanka contracts. The example thus illustrates the principle that the segments of lineage societies need the opposition of other segments to constitute and consolidate themselves (Evans-Pritchard 1940). In the case of ngaanka this is manifest in the clans’ ritual interdependence. By cooperating as rivals they mutually serve each other as the constituting other. While the institution of ngaanka thus, on the one hand, reiterates clan boundaries and clan identities by reinforcing the internal structural opposition between clans, on the other hand, it also assures the permeability and constant transgression of these boundaries by sanctioning the practice of te’egal inter-clan marriages that weave the bonds between clans that constitute the network of the ethnic group. But the mechanisms and processes of identification and we-group consolidation that are at work on the internal level within Woɗaaɓe society are only a part of wider processes of a similar logic that take place at the higher societal levels of the poly-ethnic society in the study region. Based on the analysis of internal interface situations, we can now come back to examples of external interface situations as analysed in Chapter 4 and compare their role for identity construction. Since the identity categories are in a taxonomic relation and form a continuum of inclusiveness and exclusiveness, an external identity boundary can at the same time also function as an internal boundary if appeal is made to a more inclusive category. Inversely, an internal boundary can become an external one at a more exclusive level of identification. The question of internal and external identity boundaries is thus ambiguous and always depends on the categories of identification to which appeal is made. In general, at the group boundary, two sorts of processes, the first inward-oriented and the other outward-oriented, take place. In an outward-oriented process, the we-group is consolidated, and the group-identity confirmed at the boundary through differentiation from an external other. Internally, consolidation is achieved through an inward-oriented process of common identification.

188  Space, Place and Identity

The simultaneous working of these principles at different societal levels has the effect that the processes of identification and differentiation are often ambiguous. Two special cases that illustrate this principle particularly well are the two social institutions of ngaanka inter-clan alliances, on the one hand, and the inter-ethnic joking relationship between Kanuri and Fulɓe, on the other. Mauss’ (2013 [1928]) interpretation of joking-relationships in terms of total social phenomena (see also Tamari 2006: 218f.), together with the above interpretation of ngaanka inter-clan contracts along the same analytical line, constitutes the basis for a comparison of the two social institutions. Both are ritualized inter-group relationships of agonistic and competitive character that at the same time express alliances, or contracts. The respective roles of these two institutions for the construction of identity and difference are ambiguous in the sense that both institutions have potential for both integration and the maintenance of difference at the inter-group level. Both are examples of inter-group alliances that are based on opposition as well as on reciprocity. In the case of the inter-ethnic joking relationship between Fulɓe and Kanuri, the competition in joking consolidates stereotypes and thus reinforces group boundaries and opposition, while at the same time facilitating exchange and communication. In the case of ngaanka, it is the competition over women that creates opposition, while the marital exchange and the resulting cross-cutting ties at the same time enhance cohesion between clans. The complex of ngaanka and te’egal thus ultimately reinforces the notion of a common identity. In both institutions – joking relationships and ngaanka – group consolidation is achieved through demarcation and differentiation from an opposed group, which thus serves as constituting other. At the same time, both institutions have a significant potential for integration of the groups involved within a more inclusive dimension of identification. The principles of differentiation and identification work at different levels of the society: in the case of ngaanka, inter-clan opposition consolidates the clans as we-groups that oppose each other in a shared ritual institution, thus at the same time consolidating the ensemble of clans by enhancing their commitment to the shared collective identity that defines the ethnic group. In the case of the inter-ethnic joking relationship between Kanuri and Fulɓe, we-group consolidation is achieved primarily at the ethnic level (in the case of the Fulɓe, the macro-ethnic level), while the shared ritual institution at the same time ties the opposed ethnic groups together by defining the framework of a meta-ethnic social space that includes both.

Notes 1. For discussions of Woɗaaɓe te’egal, see Bonfiglioli (1988: 44), Dupire (1962: 250ff., 1963: 68ff., 1970: 63ff.), Maliki (1981: 124f.), Paris (1997: 74ff.), Schareika (2007: 150ff., 2010b: 109ff.), Stenning (1959: 140ff.).

The Translocal Community and Social Reproduction  189 2. In the literature, the term is often spelled ngaanyka (Paris 1997; Loncke 2002; Boesen 2008a). 3. While I concentrate my analysis on the rituals of marriage, name-giving ceremonies (humturu), comprise the same elements, i.e. the transfer of animals to a new-born son by his father and the sacrifice of an animal, the meat of which is communally shared. A new-born child is thus established as a new member of the patrilineage. 4. For critiques of this problematic point in Mauss’ analysis, see Lévi-Strauss (1966), Sahlins (1974), Godelier (1999). 5. That people actually apply this logic in practice is expressed in a statement about kooɓgal marriage that Schareika (1994: 153f) recorded among Fulɓe in Benin, even though the sharing of meat occurs in an indirect form here, i.e. as the sharing of money that has been generated by selling an animal. All participants in a marriage receive a share of the money and it is explicitly referred to the importance of this act of sharing in order to assure the solidarity of the community in cases of conflicts that might later arise around the marriage. 6. Retaillé (2013: 66) points out that in French ‘to take place’ translates with ‘faire lieu’, literally to make place.

Chapter 11

Cultural Change and the Reproduction of Difference

Appadurai (1995: 213) has suggested that in a ‘deterritorialized’ world, the (re-) production of locality is increasingly a struggle, through which the capacity of communities to reproduce themselves is also put into question (ibid.: 215). Contrary to this vision, other authors have tried to show through concrete examples that translocal conditions need not be an obstacle for cultural reproduction (e.g. de Jong 1999), confirming ‘the capability of a translocal community to reproduce itself within the overwhelming contexts of the global’ (ibid.: 316). Among the Woɗaaɓe, mobility and migration are long established strategies in response to environmental insecurity. As argued above, contemporary translocal phenomena in connection with urban migration logically build on this history in a line of continuity, since migrant labour is itself in the first place a strategy that responds to environmental insecurity. In this sense, ‘deterritorialization’ and translocal modes of community production are not per se a new phenomenon and do not represent a particular challenge, or obstacle, for cultural reproduction. Nevertheless, the Woɗaaɓe elders would probably share Appadurai’s doubts. Although there is a remarkable continuity in the reproduction of difference in the face of urban migration and sedentarization, some significant changes are observable. The principal challenge for cultural continuity is not the dispersal of the community as such. More important is the fact that the exposure of urban migrants, but to a certain degree of all the society, to urban life – characterized by modernity, by a more rigid variant and stronger influence of Islam, and by a stronger presence of the state and its institutions – leads to disparate, multiple reference systems (e.g. religion, value systems, legal systems) which increasingly compete with indigenous values and cultural orientations. This chapter explores contemporary changes and continuities in the roles of the central cultural

Cultural Change and the Reproduction of Difference  191

institutions, the worso and ngaanka and the characteristic te’egal elopement marriage, which, as I have shown, is crucial in defining Woɗaaɓe identity and central in the cultural reproduction of the ethnic group.

Worso and Ngaanka in the Contemporary Condition Even though the institution of ngaanka and the practice of te’egal might seem archaic today and the latter is regarded with disdain by more Islamized neighbouring groups in Niger, they have not lost their relevance in the contemporary context, nor does te’egal seem to diminish in numerical importance.1 Ngaanka relations among Woɗaaɓe clans in the study region are of undiminished vitality and remain an important element in inter-clan relations. As noted above, urban migrants move to their pastoral home camps whenever they can, and in particular on the occasion of the big rainy season gatherings of the regional clan section, the worso, and the ngaanka inter-clan meetings. The efforts made by Woɗaaɓe in general, and by urban migrants in particular, in order to participate in these events again attests to the strong translocal orientation of migrants towards their home communities. Even after years in the city they do not cease to consider the ceremonial gatherings as prime occasions for sharing the life of the pastoral community to which they remain primarily oriented (see also Boesen 2004a). The mobility of Woɗaaɓe based in rural centres in order to participate in ngaanka meetings can be equally high. The local splinter group of Gojanko’en in Ganatcha, for instance, are rather far removed spatially from their clan mates and too small numerically to engage in ngaanka relations with other clan-groups on the local level. Thanks to the vital translocal network, this group, even though locally detached from the majority of their clan, is not isolated but stays connected with and integrated into the wider regional clan section. The availability of modern communication technologies facilitates the flow of information, and if news about a ngaanyka-meeting begins to circulate, they can reach segments in spatially removed areas such as the Koutous mountains even quicker today than they might have a decade ago between the dispersed pastoral camps within the Damergou region. Once the news spreads, important numbers of people will set out for the place designated for the meeting. Concerning the worso – the rainy season gathering during which the postponed ceremonies of marriage and filiation are celebrated – questions of continuity and change are rather ambivalent. In the previous chapter, I pointed out the importance of the worso for renewing social bonds and strengthening the cohesion within the regional clan section, and for affirming a shared identity (Maliki et al. 1984; Schareika 2007, 2010b). According to Bonfiglioli (1988: 49f., 135), a critical point in the process of clan segmentation is reached when the different factions no longer meet for this ceremonial occasion, but rather celebrate an independent worso each.

192  Space, Place and Identity

In the Damergou region, the Gojanko’en of the Kuskudu and the Mbuuldi maximal lineages continue to celebrate their annual worso. The faction that has migrated to Ganatcha, however, does not join in their annual gathering, principally due to the significant geographical distance. What is more, however, the Gojanko’en in the Koutous region also no longer celebrate a worso of their own either, but have abandoned the institution altogether: In the past, we celebrated a worso because only during the rainy season, the people were all together. For the rest of the year we were always dispersed. But now that we are always together in the surroundings of our well, whenever a child is born, we celebrate its name-giving ceremony right away. (Abdua Ɗawra, June 2011) The argument for this abandonment seems rational and pragmatic: where the core of the relevant community is constantly in a place and does not disperse as it did in the past, the worso is no longer a necessity. The ceremonies that in the past constituted the reason to convene the regional clan community for the worso, are now celebrated as they occur. Just as mobility is a necessary adaptation to ecological constraints rather than a cultural imperative, the worso as an accumulation of community-relevant ceremonies is in the first instance an adaptation to mobility and seasonal dispersal and can be given up in a context of sedentarization. In terms of social group formation and identification, however, the argumentation points to the advanced degree of dissociation of the Ganatcha faction from their clan mates in the Damergou region. The argument that the worso is no longer necessary because the community stays together on a permanent basis points to the fact that the relevant community is increasingly locally and no longer translocally defined. As far as ritual matters are concerned, the Gojanko’en from the Ganatcha faction have emancipated themselves from their wider lineage and clan community in the Damergou region. This does not mean that they are not welcome in the Damergou region if one of their visits coincides with the time of the worso there, or, vice versa. But the fact that no efforts are made for explicit invitations to the worso celebration, which would be unproblematic in a time of mobile communication coverage in most remote areas, can be regarded as a sign of increasingly local orientation and detachment from the geographically more distant parts of the clan community. This local orientation becomes evident also from the fact that the Gojanko’en of Ganatcha do invite friends from the wider local community to name-giving ceremonies or marriage celebrations, including not only other Woɗaaɓe but also some non-Woɗaaɓe neighbours. Inversely, the Gojanko’en of Ganatcha participate as visiting guests at the worso of their Yaamanko’en neighbours in the Koutous region, who, in contrast to the Gojanko’en, have not abandoned this institution

Cultural Change and the Reproduction of Difference  193

because relatively large numbers of lineage members live in the rangelands just north of Ganatcha. Concerning ngaanyka, things look slightly different: for lack of group size, the small Koutous faction relies upon the Damergou branch for participation at the ceremonies. Hence, for continued participation in ceremonial exchange across clans and thus ultimately for substantializing and maintaining ethnic identity and for being accepted as a part of the wider Woɗaaɓe society, the maintenance of translocal relations with other clan factions remains an essential concern. While the Gojanko’en of Ganatcha might hardly be informed about the worso of their clan mates in the Damergou region, they will make significant efforts in order to participate at a ngaanyka meeting in the other region. Does that mean that the worso has a lesser potential for group cohesion than ngaanka? Is it, as an element that only concerns clan-internal relations, of minor importance as compared to the more politically relevant ngaanka that concerns inter-clan relations and the reproduction of ethnic identity? The question as to whether the undiminished appeal of ngaanka can really serve as an indicator for the continuing relevance of this institution for inter-clan politics, is rather difficult to answer. In other regions of Niger, new manifestations of political interaction and representation have emerged and are still emerging today: ‘General assemblies’ (assemblées générales) of the Woɗaaɓe have been organized since the early 2000s in central Niger, explicitly addressing an outside audience and combining touristic and political aspects, aimed at networking with potential donors and at representing the Woɗaaɓe self-confidently towards political decision-makers. These meetings, which bring together the Woɗaaɓe at the regional level, are the result of increasing efforts of a cooperative and associative organization that is more developed in central Niger than further east. The assemblies have been described as being characterized by the same dialectic of cooperation and competition between clans that is characteristic of ngaanka as well (Lassibille 2009: 319), and that is ultimately a central principle of Woɗaaɓe inter-clan relations and a basis for the interaction among the constituent parts of the ethnic group. While for the mature men the political dimension of the meetings is central, for the young, they are an occasion for dancing (ibid.: 321) – and thus potentially for arranging te’egal relationships. In the study region, the model of the ‘general assembly’ has not yet been taken up, and for lack of tourists alone, ngaanka has nothing of a tourist event. The occasional western expatriates who might come to visit, generally only take glimpses at afternoon dances, take some pictures and have long disappeared before the central parts of the ceremonies take place. It can be argued, however, that for some urban-based Woɗaaɓe of the second generation of migrants, the interest in ngaanka meetings is also motivated by nostalgic motives and becomes to a certain extent folkloristic. Youth who have grown up in town have in many cases only a reduced knowledge and repertoire of dance and song, which are, however,

194  Space, Place and Identity

essential for an active participation in the performances. For these Woɗaaɓe migrants, ngaanka continues to play a role in the reaffirmation of ethnic identity, yet its community-constituting element remains only partially functional. An opposite trend, however, can also be observed: some urban grown-up youths begin to develop a new pride in their Woɗaaɓe identity and develop a keen interest in active participation in dance and song. This depends on the opportunity for regular visits in the pastoral home communities and thus points again to the importance of translocality as a factor of cultural continuity. Visits to the pastoral home community are particularly attractive for adolescent boys as they permit them to participate in the activities of the youth age-group (daɗɗo), and notably to engage in the nocturnal dances and the pre-marital sexual relations which are allowed by these activities. For urban youths, to acquire active knowledge of the traditional performative expressions can therefore be attractive because it is an essential prerequisite for being able to compete with rural age-mates for the attention of girls. Similarly, one reason for the undiminished interest of young men in ngaanka is thus quite bluntly: ‘gam rewbe’ (because of the women), i.e. because of the opportunities for contracting te’egal marriages, that it promises. There is, however, a strong gender bias with regard to this aspect. Boys are rather encouraged, when they visit their pastoral relatives, to join the daɗɗo activities, whereas urban-based girls who visit their pastoral relatives are likely to be kept at home at night and pre-marital sexual contacts are rather prohibited. In this regard, as with many aspects of the urban context, an adaptation to Islamic norms and practices can be observed. Similarly, urban migrants tend to prevent their wives from participating in ngaanka. However, the decrease in household-mobility plays a role here, too: because ngaanka ceremonies by principle take place in remote locations, they will no longer be ‘brought’ (hokki) to the increasingly significant parts of the population who are either urban-based or who permanently reside in rural centres. Hence, for the more sedentarized parts of the population, participation in the ceremonies would inevitably require travelling. This, however, tendentially excludes the less mobile parts of the population, notably older people and married women,2 from the collective experience of ethnic belonging that the ceremonies offer. As a consequence, ngaanka has at least in degree become less of a total social phenomenon, as it no longer reaches the entirety of the social group. The exclusion in particular of the latter group, i.e. the married women residing in towns or semi-sedentary centres, from participation in ngaanka, and by implication from the exchange circuit of te’egal marriage, however, signals an important qualitative change that subverts the very principle of ngaanka and te’egal. This does not mean, however, that te’egal marriage in general is decreasing in importance. As the following case study shows, it is rather that it continues to take place under changing conditions and in new contexts, leading to significant transformations and new controversies.

Cultural Change and the Reproduction of Difference  195

Te’egal Marriage in the Twenty-First Century The case study concerns Gado, a young man from the Gojanko’en clan of the Woɗaaɓe in the Damergou region. The events took place from mid-2008 to early 2012. A full transcription of this case’s account can be found in Köhler (2017a: text 16); references to this text are indicated in brackets with the number of the respective page.3 In his account, Gado related the story of his various quests for a wife, after the unilateral dissolution of his original kooɓgal marriage by his father-in-law (124). After an unsuccessful attempt at arranging a new kooɓgal marriage within his clan (125f.), he finally opted for a te’egal marriage with a young woman from the Ɓii Ute’en clan in the Diffa region, and, later, another one with a woman from the Jiijiiru clan from the Tchintabaraden area in central Niger whom he first met in Kano, where she was staying with her husband. His first successful te’egal marriage with Laadi, the Ɓii Ute’en woman, was remarkable insofar as Gado had never met her before the two finally arranged for eloping together. He first knew Laadi’s sister Mobappa and at an earlier moment the two even had planned for an elopement marriage. But since Mobappa had become pregnant in the meantime, there was no more question of eloping with her as she had entered a state of seclusion. Gado knew, however, that Mobappa had a sister, Laadi, and that, according to his Diffa-based lineage mates who knew her, the two were ‘all the same’ (ɓe fuu ɓe go’) (128). Gado managed to obtain Laadi’s mobile phone number from a lineage mate who, for his part, had obtained it from her sister during a visit. The relationship developed via mobile conversation alone to the point where the two arranged for Gado to travel to Diffa – a journey of about 475 kilometres – and took Laadi with him (128f.). After this first successful te’egal marriage, Gado and Laadi lived together for about a year and a half. Gado first continued to work in Zinder, but later the couple moved to Kano where Gado found new employment. This is where he met Jemmassu, a woman from the Jiijiiru clan, and the two agreed on elopement and te’egal (129). Gado sent Laadi back home, abandoned his job in Kano in order to avoid being tracked by the aggrieved Jiijiiru husband and brought Jemmassu, who joined him with a small child, to Zinder, where many of his lineage mates lived. However, unable to find work in Zinder, nor possessing any animals that could allow him to return to a pastoral livelihood, Gado was finally forced to move back to Kano although his new wife’s husband still lived there (130). The relative anonymity of the northern Nigerian metropolis indeed allowed him to stay undiscovered with her for some time, but finally his adversaries found them and Gado had to turn her in (130f.). Jemmassu’s husband decided to bring her to a different town to avoid further problems with Gado. Nevertheless, the woman managed to run away again and called Gado who picked her up on the road. However, the couple was tracked down by a group of Jiijiiru men quickly mobilized by Jemmassu’s

196  Space, Place and Identity

husband. They were caught and handed over to the police, and the couple had to spend a night in a cell (132). The Jiijiiru wanted the police to arrest Gado for a longer period, but he managed to convince the police to drop the case as a matter of cultural custom. Although Jemmassu was brought to relatives in the Tchintabaraden area this time, she and Gado did not abandon their plans for elopement. Jemmassu found an excuse to return to Nigeria to visit her father and she made use of the first opportunity to contact Gado on his phone. He instantly came to flee with her once more and the couple went back to Zinder, where they stayed at a house of one of Gado’s lineage mates. Here, the events took a dramatic turn: Jemmassu’s father and some of his lineage mates called Gado in Zinder to inform him that they have caught his younger brother in Kano and that they will hold him hostage until Gado returns Jemmassu to them (133). Gado refused to give in. However, a lineage mate and close friend, apparently distressed by the abduction, informed the Jiijiiru about Gado’s whereabouts and told them to come to Zinder and bring Gado’s brother in exchange for the woman. Gado and Jemmassu tried to flee to the Damergou region on a collective taxi, but upon their arrival in Tanout they were stopped at the checkpoint by the police who had, in the meantime, been informed by the pursuers via mobile phone. The Jiijiiru followed the couple on another taxi and, unable to keep Jemmassu, who was handed over to her lineage mates by the police (134), Gado returned to Zinder, stranded, yet not without hope, since Jemmassu remained determined to leave her husband for him. The case is interesting in several different regards. On the one hand, Gado’s reflections, commentaries and explanations give a good idea of emic conceptualizations of te’egal marriage and its rules (127, 130f.); on the other hand, the example shows how te’egal can be practically dealt with in the contemporary condition. The case again substantiates the significance of translocal kinship networks within and among Woɗaaɓe communities and how they have become spatially more extended in recent time and easier to mobilize. As a result of urban long-term migration, new translocal lineage communities have emerged that are scattered across different regions. Gado can rely on a network of lineage-mates residing as migrant workers in distant areas and functioning as providers of contacts with potential partners for a te’egal marriage. The homes of his lineage mates in urban centres like Zinder, Kano and Diffa serve as logistical bases for organizing elopement. The urban space with its relative anonymity has become a new arena for the contractualization of te’egal marriage. Modern transport facilities – notably collective taxis and overland buses – expand the options for elopement and often give it new dimensions. Mobile communication is used as a means to cope with these new dimensions of mobility and to fully exploit the opportunities that the expanded social space offers. To highlight the role played by the mobile phone alone, it should suffice to enumerate some of the occasions in Gado’s account at

Cultural Change and the Reproduction of Difference  197

which it is made use of. In the case of Gado’s te’egal marriage with Laadi, the first phase of their relationship could effectively be called a mobile phone relationship (Miller 2009: 25): the two never met in person before their elopement and everything was arranged on the phone (128). The logistic aspects of elopement were largely managed on the phone as well. During Jemmassu’s several flights from her husband, she was constantly in contact with Gado via phone and she repeatedly coordinated meetings with him on the road (131, 133). When tracked down by his persecutors, Gado talked to his lineage mates on the phone to get advice, and even when the Jiijiiru took Gado’s brother hostage, Gado communicated with him, as well as with the abductors, on the phone (133). And finally, after Gado’s ‘betrayal’ by his lineage mate – which also took place on the phone (133f.) – the pursuers went to the police and they communicated with the gendarmerie in Tanout in order to stop the taxi in which Gado and Jemmassu fled (134). The case study thus excellently demonstrates the mobile phone’s penetration into all spheres of everyday life, and the fact that it is perfectly well adapted to the contemporary contexts of new mobilities and translocality. But while the modern networks and the contemporary mobility and connectivity open up new options for arranging te’egal marriages, this also constitutes a considerable new challenge to the established normative and legal framework of te’egal. Gado’s case exemplifies the disparities between the urban and the pastoral realm and the conflicting roles of customary law and state law. In principle, there are different culturally appropriate reactions to cases of te’egal elopement. A fugitive couple will generally try to get out of reach of the clan mates of the woman, because it would be perfectly legitimate for the latter to take the woman back by force and beat up the perpetrator. However, many cases of te’egal become the object of negotiations in which conventional rules play a significant role. In one of the more explanatory parts of his narration, Gado described the customary procedure for dealing with such cases: either the woman’s father or one of her affinal relatives will pursue her and bring her back home. If a woman continually flees her husband to rejoin another man, however, the elders from the te’egal husband’s clan can eventually back the couple and refuse any further requests from the other party. Gado stressed the importance of customary rules in handling matters of te’egal. Violence as well as negotiations between elders are both accepted ways of dealing with conflicts evolving around te’egal. Involvement of the police, on the contrary, would a priori not be qualified as such (137). In fact, it is an important principle that te’egal cases should remain an internal affair between clans (Schareika 2007: 154, 186f.). The involvement of state authorities would be regarded by the elders as contrary to established custom (‘ɗum walaa nder ndonu meeɗen’, 135). As Schareika has shown, Woɗaaɓe today stress their political interest of keeping the settlement of te’egal disputes within the framework of their own customary institutions (2010a: 207). As long as they agree that cases of

198  Space, Place and Identity

te’egal are treated along customary lines and without involving state authorities, this will strengthen the position of the Woɗaaɓe leaders, since they keep a certain autonomy vis-à-vis the state. Today, however, and particularly in the urban context, the actual practice often looks very different. Actors are likely to turn to the institutions of state law if they feel that this might be in their interest, and police involvement has in fact become a regular means of reacting to te’egal elopement. In principle this is nothing surprising but rather an expression of the discrepancy between normative prescriptions and everyday practice which is characteristic of dealing with te’egal in general: the gap between the approval of te’egal in official normative discourse, and individual husbands who leave little doubt that they would try to take their wives back by any means if they were concerned themselves (Schareika 2007:154; 2010b: 113). The new element today is that actors and legal institutions exterior to Woɗaaɓe society increasingly come into play. The involvement of state authorities in te’egal cases is possible by the fact that the Nigerien criminal code provides a penalty of up to three months of prison and a fine of up to 100.000 FCFA (ca. 150 €) for adultery (République du Niger 2003 [1961], par. 286–89). The evidence of a woman who is caught in the company of a man other than her husband and denounced to the police by the latter is sufficient to put the accused couple into custody while the situation is being clarified. In practice, this means that people today can easily end up in jail for practices that in the past might have been settled in a sword fight. Settlement by the police has not replaced violence as a possible outcome of ill-conceived te’egal marriages, but rather, both exist simultaneously today. Gado himself offered an interpretation of the way his te’egal elopement with a Jiijiiru woman had been handled by the latter’s relatives. In so doing, he positioned himself not only as an advocate of customary practice but indeed as a moral guardian of Woɗaaɓe traditions (ndonu): Do you see what they did? They put me into the hands of the police . . . [T]hat is not done. Te’egal is not an affair for the police. (136f.) One might argue that when he reproached them of violating cultural norms, this was done with the intention of justifying his own actions and ultimately, to achieve his own ends. Gado was aware of the normative discourse among Woɗaaɓe elders that rejects the involvement of external authorities into te’egal affairs, and he used this as a moral argument against his adversaries. Obviously, due to his own role in the case, Gado did not have an interest in involving the police and his rhetoric might have sounded quite different, had he himself been the aggrieved husband. Actors may thus strategically make reference to cultural normative discourse when this seems opportune to justify their actions – regardless of how consistently ‘traditional’ they might act themselves in a similar

Cultural Change and the Reproduction of Difference  199

situation. In particular young urban grown-ups, for reason of their socialization in a modern urban context and their exposure to other moral values and legal systems, seem to be inclined to make use of instruments belonging to an opposing set of normative rules, particularly in situations where elders as the ‘guardians of the tradition’ are absent. One may interpret this as an opportunistic use of cultural concepts, either to justify actions taking place in a modern context, or to criticize one’s opponents precisely when their methods leave the normative framework. Woɗaaɓe with whom I have discussed the matter agreed that the urban space has its own rules and people do not tolerate te’egal the way they do in the pastoral realm: The bush and the town are not the same. Te’egal in the city provokes conflict. You cannot just take somebody’s wife in the city, because there are the institutions of the state, which will prosecute you. And besides, Islam forbids taking a woman without marrying her. In the city, religious observance is much more pronounced than in the pastoral camps, where many people have not completely embraced Islam. (Hamma Beleti, Zinder, June 2011) This statement addresses two aspects: the first is religion, the second is the question of urban civil rights in relation to the culture-specific practice of te’egal. The formulation ‘Islam forbids to take a woman without marrying her’ addresses the problem that te’egal, although a culturally perfectly established form of marriage among the Woɗaaɓe, is not in accordance with Islamic law and hence regarded as a horrifically pagan practice by Muslims from other ethnic groups, i.e. by society at large. A contemporary trend among Woɗaaɓe towards stronger Islamization has not led to abandonment or even to a significant weakening of the practice of elopement marriage. Many Woɗaaɓe do not seem to see a contradiction between considering themselves Muslims and practicing te’egal in the cultural framework of ngaanka. In town, however, Islamic morals have a stronger influence than in the pastoral realm because the social pressure of the multi-ethnic, predominantly Muslim society is more felt. The increasing impact of Islam on Woɗaaɓe marriage customs has led to the emergence of a new form of marriage, which combines elements of the customary kooɓgal betrothal marriage and that of local Islamic custom. Monetary bride-wealth payments become more important in relation to the transfer of animals. This is due also to socio-economic reasons. In a time of widespread pauperization, young men often cannot rely on their fathers any more to provide them with the animals necessary to contract a kooɓgal betrothal marriage and establish an independent household. The customary sacrifices and animal transfers are therefore in many cases widely reduced. While kooɓgal marriage used

200  Space, Place and Identity

to involve three bovine sacrifices, today the animals slaughtered are often only sheep and their number sometimes reduced to a single animal (Anderson and Monimart 2008/9: 78). But the monetization of marriage transactions is neither a uniquely urban phenomenon, nor merely due to economic factors; it is also an adaptation to the Islamic customs of neighbouring populations. Dupire already observed in the 1950s, that in more Islamized groups of Woɗaaɓe the transfer of cattle tended to be accompanied by actual bride-wealth payments (1962: 240f.). Although te’egal among urban Woɗaaɓe migrants is discouraged by the elders and a matter of much debate, this does not mean that the legitimacy of the practice as such is put into question by a majority of urban-dwelling Woɗaaɓe. Rather, as suggested above, it seems to be considered as problematic in the city because the external regulation instances of the state are more present and tend to be involved, which is seen as undesirable. The fact that te’egal marriage, despite these concerns and although in contradiction with the parallel development of growing Islamization, remains vital even in the city, can once more be explained with economic reasons: Since te’egal is based on elopement instead of negotiations between families, no payments to the woman’s kin are involved in this type of marriage. The customary sacrifice of generally only one ovine is economically relatively affordable and a te’egal marriage thus easier to accomplish than the clan-endogamous kooɓgal. However, the significance of te’egal marriage is thus to a certain extent changing in the urban migrant context from a prestigious secondary marriage to a less costly alternative for the less well-off. The question of how urban cases of te’egal elopement should be handled provides a source of big controversy among Woɗaaɓe. Even more so as ‘the Woɗaaɓe’ as a clearly bounded social and ethnic entity in Niger do not exist and negotiations or discussions between clans seldom take place in more than bilateral settings. In the urban context, the actual practice of individual actors is today more and more independent of the elders, not least because their influence, which in the pastoral realm functions as a controlling force, tends to be weaker in the city. However, this is a matter of degree and differs from one context to another, depending on the size and social composition of urban migrant communities. Whereas in Zinder relatively significant numbers of Gojanko’en elders would have been available who might have taken charge of the case, in Kano this was not the case and Gado had to act on his own, although with assistance by his lineage mates via mobile phone. This suggests that there is still a, however reduced, control by the elders, even in the urban context – and even via mobile phone. Their authority in negotiations of te’egal cases is still considerable and they can function as an instance of moral advice on young men. The most astonishing and perhaps the most telling aspect in the case study is the fact that in the end it was a lineage mate and close friend of Gado, who gave out the relevant information and thus enabled the Jiijiiru to track the couple down (133f.). Although this ‘betrayal’ might be surprising and was indeed

Cultural Change and the Reproduction of Difference  201

fervently criticized by the Gojanko’en elders, it can be explained if one takes into account that the Jiijiiru had taken a hostage, which was definitely an unsettling circumstance, as Gado confirmed: [T]hey caught my younger brother back in Kano and they locked him up. . . . They told me they would not let him go before I brought them their woman. I swear, what they did was contrary to the tradition! Take someone hostage? Unheard of! (133) Presumably, for Gado’s lineage mate these circumstances indicated that the case had left the framework of customary rules and taken on another dimension. Had the Jiijiiru attacked with swords and sticks, there is little doubt the friend would have defended Gado without hesitation. All this suggests that the urban setting creates a high potential of uncertainty, in which the limits between ‘traditional’ and ‘modern’ situations become blurred, and in which it is no longer clear which rules should be applied and which actors and institutions involved. This uncertainty is symptomatic of the situation of this society on the brink of change, characterized by the conflicting frameworks of customary practice, on the one hand, and the institutions of the state, on the other. The normative grey area that this situation of ambiguity causes can be used strategically by some, but the increasing tendency to involve external actors and institutions constitutes a potential threat to the authority of the customary instances of decision making. The Woɗaaɓe in Niger today live in a rapidly changing world. Their pastoral society which formerly existed in vivid interaction with the other local communities, but with clear-cut boundaries, following its own normative logics and applying its own legal principles, sanctioned by the decisions of its own institutions, is more and more confronted by the institutions of the modern nation state, without, however, being adequately integrated into it in a positive sense. Perhaps unsurprisingly, the example suggests that one important challenge for the cultural normative framework comes from the part of the Woɗaaɓe with an extensive experience of life in urban centres. The fact that numerous Woɗaaɓe, predominantly from the younger generation, are today exposed to urban life with its own rules and a stronger presence of state institutions has led to a situation of increasing uncertainty about which legal and normative framework to apply in a given situation. With the emergence and rapid spread of the modern devices of mobile communication and the improvement of public transport infrastructure, individuals can rapidly switch between the different spheres, and new concepts have begun to reach even the remotest areas. The new dimensions of mobility and connectivity have also registered an impact on cultural practices such as te’egal marriage, and actors make use of the new options in order to achieve their ends. In a context in which the appropriate normative practice in a given situation is an object of debate and can in fact no longer be clearly defined, the authorities

202  Space, Place and Identity

of the modern state become involved as a new external actor, and they can in fact be instrumentalized as a ‘weapon’ (Schareika 2007) in cases of inter-clan conflict about te’egal. Actors can opportunistically refer to the different discourses and their respective practices in what might be characterized as an attitude of normative forum shopping. The same individuals who might challenge the cultural normative framework by involving the state institutions in one situation may, in another, rhetorically refer to the same framework to argue against others whom they accuse of betraying it. Such interpretations, however, are no longer sanctioned by the elders but may rather be subjected to the principle that the ends justify the means. In view of such contradictions, the practice of te’egal, which has always been ambivalent in terms of integration and conflict, runs the risk of changing its character by degree from being also an element of cohesion and of cultural continuity towards being primarily a threat for cultural cohesion. The increasing involvement of external authorities threatens to render the institution of ngaanka with its inter-clan agreements about mutual te’egal marriage obsolete and thus ultimately to delegitimize this cultural practice.

Notes 1. Statistical data are not available, but the prevalence of te’egal has remained significant to this day. With regard to the study group, an informed estimate is that between a third and one half of the population at one point or other in their life makes at least one attempt at te’egal, although this does not mean that all these cases result in stable marriages. 2. As Boesen (2010) has convincingly demonstrated, the argument of lesser mobility of married women as compared to men is questionable for the case of the Woɗaaɓe and cannot be generalized. I contend, however, that in the case of ngaanka, the point can be maintained. Even in the mobile pastoral setting, married women would rather not be taken by their men to attend a ngaanka carried out by an adverse clan. Admittedly, however, the men’s efforts to keep their women at home during ngaanka festivities taking place in the area are only partially successful. 3. This section is derived, in part, from an article published in the Canadian Journal of African Studies on 22 November 2017 (Köhler 2017b), available online: https://www.tandfonline. com/doi/full/10.1080/00083968.2017.1387061.

Conclusion

In this book I have argued that, among the Woɗaaɓe, migration processes and placemaking strategies are closely linked to processes of group formation, and thus to questions of identification and social reproduction. Social groups form in a continuous process of fission of descent groups, on the one hand, and fusion of local groups after periods of co-residence or coordinated mobility within a particular area, on the other. The simultaneous processes of migration and group division have the effect that clans and lineages are fragmented and dispersed across space. Another dimension of this characteristic dispersal of social groups is a seasonally changing group morphology in adaptation to the scarce and irregular distribution of pastoral resources. The society is highly mobile and fluctuating between dispersal and concentration. In this situation, the capacity to maintain translocal social networks is both an important risk-management strategy, because social relations provide access to resources, and crucial for the maintenance of a notion of community. The concept of translocality has allowed me to grasp not only the phenomena related to modern migration processes, but more generally to social relations expanding across space. The significance of translocal social relations in the study group is also the reason why approaches that conceptualize space and place as relational and socially constructed have been given a particular consideration. In this framework, pastoral migration processes can be understood as producing complex networked social spaces. Nomadic pastoralist societies have been characterized as based on the principle that ‘both economic resources and political alliances result from the ability to establish social ties between places, rather than investing in productive activities in fixed locations’ (Retaillé and Walther 2012: 12). The Woɗaaɓe case

204  Space, Place and Identity

has confirmed this with the examples of translocal economic security networks based on haɓɓanaaye animal loans, for the economic aspect, and the translocal network of ngaanka inter-clan alliances, for the political part. Today, both the economic and the political aspect of translocal social alliance-making continue to be of major importance. The angle of translocality as a research perspective thus permits us to highlight aspects of continuity despite change in the mechanisms of social space construction. The conceptions of space and place as relational and based on mobility and connectivity offer a framework for interpreting the mobility of the pastoral Woɗaaɓe in terms of a network society avant la lettre. The study more generally of Sahelian societies with their cultures of mobility (Gallais 1975; Boesen 2004b, 2007b; Boesen and Marfaing 2006, 2007), can contribute to an understanding of the relation between place and the circulations and flows characteristic of contemporary phenomena of globalization (Retaillé and Walther 2012: 10). The current situation in the study group is characterized by significant transformations in livelihood patterns and spatial practices. I have argued that these transformations should be understood as both reactions to constraints and pro-active strategies. They are not merely a response to the diminishing of livestock, but also have a more strategic and even political dimension. This finding underlines the undiminished utility of Salzman’s adaptation and response model for the analysis of sedentarization processes in nomadic groups, which stresses ‘the options available to actors’ and ‘the voluntary shifts between available alternatives in pursuit or in defense of socially defined ends’ (1980: 14). To emphasize the strategic and progressive aspect of contemporary spatial practices in response to opportunities is not to deny that many decisions are also driven by constraints. The Woɗaaɓe engaged in these processes can be regarded as both victims and agents. The findings confirm two observations made in other contexts: first, they show that the concept of ‘Fulɓeness’ is fluid and negotiable (Dafinger and Pelican 2002: 3) and allows for the integration of new concepts and diverse livelihood strategies beyond pastoralism. With regard to the Woɗaaɓe this is remarkable insofar as their historical differentiation from other Fulɓe groups in the region has been the result of an ideological process of ethnicity, substantially based on nomadic pastoralist livelihood as a central criterion for identification (Bonfiglioli 1988). Second, the findings also confirm a tendency of wider significance among pastoralists in the region towards a more territorial appropriation of spatial resources. This does not imply that mobility is abandoned. Rather, it takes on new forms and new dimensions in the sense that it can span different spheres from the urban to the transregional and transnational, and, in some cases, even global. The ubiquitous phenomena of urban migration has added a new dimension to the dispersal of social groups, whilst also creating new forms of translocal connectedness.

Conclusion  205

The links between migrants and their pastoral home communities generally remain strong, and they build upon well-established patterns of translocality. Although long-term migrants develop multiple identifications between which they can shift, translocal ties to the pastoral home community remain central for the identity construction of migrants and represent an important aspect of cultural continuity despite displacement in the migration context. Migrants’ identity construction is less place-bound than relational, based on networks of translocal links, and it is thus in continuity with the patterns of identity construction of mobile pastoralist Woɗaaɓe. Migrants’ identities are thus often characterized by continuity rather than by disruption. In the case of the study group, this is also favoured by the relatively short distances from the pastoral home areas to the main target areas of migration, which allow for frequent visits and the maintenance of close community ties. Migrants’ choices about their target areas for migration are significantly determined by considerations of proximity and the availability of connective infrastructure. Mobile telecommunication and public transport in particular play crucial roles in supporting the new forms of mobility and translocality. To come back to the paradox formulated at the beginning of the introduction, i.e. the relation between sedentarization processes and the simultaneous increase in mobility at the level of the society as a whole, one could thus say that the improved options for mobility help the Woɗaaɓe to cope with the conditions of sedentarization and urbanity. With regard to the translocal orientation of migrants and their continuous linkages with their rural home communities, similarities to observations made in other contexts of regional migration (e.g. Olwig 1997; Greiner 2010) suggest an important insight to be gained from a systematic consideration of this literature, one that can enhance the understanding of short range migration in the study region more generally and beyond the particular focus of this book. An important issue that the book has addressed is about the relation between recent tendencies towards (both rural and urban) sedentarization and the question of integration into the structures of the modern nation state. Pastoral mobility has often been regarded by sedentary institutions as a strategy of state evasion and therefore as menacing, and numerous efforts have been made to settle pastoralists. The question of integration into structures of the modern state and into the wider society is therefore always of a particular concern for the analysis of pastoralist societies. My analysis has shown that, with regard to the study group, the thesis of nomadic state evasion through mobility can be maintained only with some qualifications. The interaction with the (pre-colonial, colonial and post-colonial) state was – and is still –largely characterized by an opportunistic double strategy of avoidance and cooperation. If, in the past, a strategy of state evasion through accentuated mobility might have been rewarding as it was likely to keep up a high degree of autonomy, today such a strategy would

206  Space, Place and Identity

lead to marginalization and undesirable exclusion. The state is no longer only perceived as a nuisance, embodied by tax-collectors and biased authorities, but it has also become an important distributor of the welfare of international aid and development funds. In order to participate in these distribution processes and gain access to the new resources, integration has therefore become an increasingly attractive consideration. However, since mobility generally remains strong and occurs widely without any control by state structures, the contemporary efforts for integration and the related tendencies towards sedentarization cannot be interpreted one-sidedly as a subjugation to the spatial patterns promoted by the sedentary state. Although tendencies of territorialization are gaining momentum, spatiality continues to be primarily dominated by the principles of relationality, translocality and mobility. As this book was in the process of publication, some of the realities it describes profoundly changed. While fieldwork was principally carried out in 2010–11, insecurity in the wider Sahel region has dramatically increased over the past years, especially since 2012, with severe consequences for rural populations in general and pastoralists in particular. The eastern Diffa region bordering Nigeria is particularly affected by these threats. Trans-border transhumance into Nigeria and into the Lake Chad area, which has long been an important pastoral option, has become highly risky due to the prevailing violence of the militant Islamist movement Boko Haram. What makes pastoralists particularly vulnerable in this regard is that non-state armed groups, albeit with different motives, favour the same peripheral areas with limited state control as they do. To avoid attacks or herd abductions, pastoralists are faced with the alternative of either abstaining from moving into these areas and missing crucial resources or engaging in equally dubious and dangerous agreements with outlaws, rebels and jihadists. As the Diffa region received huge numbers of refugees – among them also numerous mobile pastoralists – the situation is characterized by an extreme pressure on the pastures. At a less dramatic, yet still worrying scale, this is also the case in the study area. While the security situation as such is not particularly affected, increased pressure on the natural resources and resulting conflicts has become a perturbing issue. As there is little hope that the situation in Nigeria is going to change in the near future, these difficulties are likely to affect the living conditions in the study area for the years to come. In the urban context, the situation has also changed to the worse. The successful networking with expatriates had put numerous migrant workers from the study group into a relatively comfortable position for a number of years, and the time from the mid-1980s to the early 2000s can be considered as a ‘golden era’ for Gojanko’en watchmen in Zinder. Over time, however, the conditions of this kind of work have become considerably tougher and the risks involved higher, due to increasing urban insecurity. In Kano, in northern Nigeria, where many Woɗaaɓe migrants from the region work, insecurity has for a long time been more

Conclusion  207

of a concern than in Zinder (see Köhler 2017a: text 14,22ff.). More recently, however, the situation has seriously deteriorated. After the declaration of a state of emergency in three northern states of Nigeria in 2012, migrant workers from the study group who had been working in Kano for years had to leave the country when foreigners who were not officially registered, particularly from Niger and Chad, were expelled by the Nigerian government on the assumption that Boko Haram recruited some of its members from these countries. The transnational aspect of migration to cities in northern Nigeria, which had before been of rather secondary significance, suddenly mattered. Many Woɗaaɓe migrant workers returned to Niger, but the conditions for finding employment were at the same time degrading here as well. The long-time strategy of working for expatriates was made widely obsolete by the massive withdrawal of the latter in reaction to growing security threats. For the same reason, national or international organizations began to confide security tasks more and more to professional companies. The fact that the presence of expatriates strongly depends on political decisions makes the Woɗaaɓe networks with them rather vulnerable on the local level. Existing networks can continue to be of considerable economic importance, and they also begin to function better and better across space – once again very much facilitated by modern communication devices. However, the character of these network relations is increasingly changing from a resource-providing basis of a viable economic strategy to mere passive dependency. While numerous examples show that, in the past, people were able to successfully return to the pastoral economy if their job opportunities in the city were running out, today the situations in both the city and the pastoral realm are equally characterized by serious constraints. Given the current pace of change, it is difficult to assess whether the Woɗaaɓe’s flexibility and adaptive capacity will be sufficient to cope with the new challenges.

References

References

Abba, Souleymane. 1990. ‘La chefferie traditionelle en question’, Politique Africaine 38: 51–60. Acloque, Benjamin. 2014. ‘De la constitution d’un territoire à sa division: L’adaptation des Ahl Bârikalla aux évolutions sociopolitiques de l’Ouest saharien (XVIIe–XXIe siècles)’, Canadian Journal of African Studies 48(1): 119–43. Adamu, Mahdi. 1986. ‘The Role of the Fulani and Twareg Pastoralists in the Central Sudan, 1405–1903’, in Mahdi Adamu and A.H.M. Kirk-Greene (eds), Pastoralists of the West African Savanna. Manchester: Manchester University Press, pp. 55–61. Adepoju, Aderanti. 1995. ‘Migration in Africa: An Overview’, in Jonathan Baker and Tade Akin Aina (eds), The Migration Experience in Africa. Uppsala: Nordiska Afrikainstitutet, pp. 87–108. Aderinoye, R.A., K.O. Ojokheta and A.A. Olojede. 2007. ‘Integrating Mobile Learning into Nomadic Education Programmes in Nigeria: Issues and Perspectives’, International Review of Research in Open and Distance Learning 8(2): 1–17. Aina, Tade Akin. 1995. ‘Internal Non-Metropolitan Migration and the Development Process in Africa’, in Jonathan Baker and Tade Akin Aina (eds), The Migration Experience in Africa. Uppsala: Nordiska Afrikainstitutet, pp. 41–53. Amselle, Jean-Loup. 1987. ‘L’ethnicité comme volonté et comme représentation: A propos des Peul du Wasolon’, Annales. Économies, Sociétés, Civilisations 42(2): 465–89. ———. 1998[1990]. Mestizo Logics: Anthropology of Identity in Africa and Elsewhere. Stanford: Stanford University Press. Anderson, Benedict. 1991. Imagined Communities: Reflections on the Origin and Spread of Nationalism. New York, London: Verso. Anderson, Steve. 2007. La mobilité pastorale. Diffa, Niger: Ziviler Friedensdienst (ZFD)/ Deutscher Entwicklungsdienst (DED). Anderson, Steve, and Marie Monimart. 2008/9a. Recherche sur les stratégies d’adaptation des groups pasteurs de la région de Diffa, Niger oriental: Études de cas. London: IIED.

210  References ———. 2008/9b. ‘Recherche sur les stratégies d’adaptation des groups pasteurs de la région de Diffa, Niger oriental: Rapport de l’étude’, unpublished report. London: IIED. Appadurai, Arjun. 1986. ‘Theory in Anthropology: Center and Periphery’, Comparative Studies in Society and History 28(2): 356–61. ———. 1995. ‘The Production of Locality’, in Richard Fardon (ed.), Counterworks. Managing the Diversity of Knowledge. London, New York: Routledge, pp. 204–25. ———. 2003[1996]. ‘Sovereignty without Territoriality: Notes for a Postnational Geography’, in Setha M. Low and Denise Lawrence-Zúniga (eds), The Anthropology of Space and Place: Locating Culture. Malden, MA: Blackwell Publishing, pp. 337–49. Arzika, M. Sani, Bernard Bonnet and Zakara Gandou. 2007. Droits et pratiques des acteurs dans la gestion des situations conflictuelles liées à la mobilité pastorale. Republic of the Niger, Ministère des ressources animales /Agence Française de développement. Aubréville, André. 1949. Climats, forêts et désertification de l’Afrique tropicale. Paris: Société d’éditions géographiques, maritimes et coloniales. Azarya, Victor. 1988. ‘Reordering State-Society Relations: Incorporation and Disengagement’, in Donald Rothchild and Naomi Chazan (eds), The Precarious Balance: State and Society in Africa. Boulder, CO: Westview Press, pp. 3–21. ———. 1993. ‘Sedentarization and Ethnic Identity Among the Fulbe: A Comparative View’, Paul K. Eguchi and Victor Azarya (eds), Unity and Diversity of a People: The Search for Fulbe Identity. Osaka: National Museum of Ethnology, pp. 35–60. ———. 1994. ‘Civil Society and Disengagement in Africa’, in John W. Harbeson, Donald Rothchild and Naomi Chazan (eds), Civil Society and the State in Africa. Boulder, CO: L. Rienner Publishers, pp. 83–102. ———. 1996. Nomads and the State in Africa: The Political Roots of Marginality. Aldershot: Avebury. Azarya, Victor, Anneke Breedveld, Mirjam de Bruijn and Han van Dijk (eds). 1999. Pastoralists under Pressure? Fulbe Societies Confronting Change in West Africa. Leiden: Brill. Babiker, Mustafa. 2001. ‘Resource Competition and Conflict: Herder/Farmer or Pastoralism/ Agriculture?’, in Mohamed Salih, Ton Dietz and Abdel Ghaffar Mohamed Ahmed (eds), African Pastoralism: Conflicts, Institutions and Government. London: Pluto Press, pp. 134–44. Baier, Stephen. 1980. An Economic History of Central Niger. Oxford: Oxford University Press. Bako-Arifari, Nassirou. 1998. De la resurgence et de la re-justification de la chefferie en contexte de democratisation au Benin et au Niger. Berlin: Das arabische Buch. Barth, Fredrik. 1969. Ethnic Groups and Boundaries: The Social Organization of Cultural Difference. Oslo: Universitetsforlaget. Basch, Linda, Nina Glick Schiller and Christina Szanton Blanc. 1995. Nations Unbound: Transnational Projects, Postcolonial Predicaments, and Deterritorialized Nation-States. Amsterdam: Gordon and Breach. Bassett, Thomas J., Chantal Blanc-Pamard and Jean Boutrais. 2007. ‘Constructing Locality: The Terroir Approach in West Africa’, Africa: The Journal of the International African Institute 77(1): 104–29. Bayard, I., B. Bonnet, A. Dangouioua, A. Lambert, S. A. Ogunwale. 2001. Commercialisation des produits et sous-produits de l’élevage en zone pastorale. Montpellier: Institut de Recherches et d’Applications des Méthodes de développement (IRAM). Ben Arrous, Michel. 2004. ‘La translocalité, pour faire quoi?’, in Laurence Marfaing and Steffen Wippel (eds), Les relations transsahariennes à l’époque contemporaine: Un espace en constante mutation. Paris: Karthala, pp. 415–42.

References  211 Bender, Barbara. 2006. ‘Place and Landscape’, in Christopher Tilley, Webb Keane, Susanne Küchler, Mike Rowlands and Patricia Spyer (eds), Handbook of Material Culture. London: Sage Publications, pp. 303–14. Benfoughal, Tatiana, and Sébastien Boulay. 2006. ‘Sur les traces de quelques objets sahariens: Pistes de recherche croisée’, Journal des Africanistes 76(1): 9–24. Bernus, Edmond. 1974. ‘L’évolution récente des relations entre éleveurs et agriculteurs en Afrique tropicale: L’example du Sahel Nigérien’, Cahiers ORSTOM, série Sciences Humaines 11(2): 137–43. ———. 1975. ‘Human Geography in the Sahelian Zone’, in The Sahel: Ecological Approaches to Land Use. MAB Technical Notes. Paris: The Unesco Press. ———. 1995. ‘Pasteurs nomades: d’une fascination aveugle à une passion lucide’, in Jean-Marc Durou (ed.), Sahara: la passion du désert. Paris: Editions de la Martinière, pp. 93–6. Bertaux, Daniel, and Catherine Delcroix. 2000. ‘Case Histories of Families and Social Processes: Enriching Sociology’, in Prue Chamberlayne, Joanna Bornat and Tom Wengraf (eds), The Turn to Biographical Methods in Social Science: Comparative Issues and Examples. London: Routledge, pp. 71–89. Bierschenk, Thomas. 1992. ‘The Ethnicisation of Fulani Society in the Borgou Province of Benin by the Ethnologist’, Cahiers d’Etudes africaines 32(127): 509–20. ———. 1993. ‘The Creation of a Tradition. Fulani Chiefs in Dahomey/Benin from the Late 19th Century’, Paideuma 39: 217–44. ———. 1995. ‘Rituels politiques et construction de l’identité ethnique des peuls au Bénin’, Cahiers des Sciences Humaines 31(2): 457–84. ———.1997. Die Fulbe Nordbénins: Geschichte, soziale Organisation, Wirtschaftsweise. Hamburg: Lit. Bocquené, Henri. 1986. Moi, un Mbororo: Ndoudi Oumarou, Peul Nomade du Cameroun. Paris: Karthala. Boesen, Elisabeth. 1996. ‘Fulbe und Arbeit’, in Kurt Beck and Gerd Spittler (eds), Arbeit in Afrika. Hamburg: Lit, pp. 193–208. ———.1997. ‘Identité et démarcation: Les pasteurs peuls et leurs voisins paysans’, in T. Bierschenk and P.Y. Le Meur (eds), Trajectoires peules au Bénin. Paris: Karthala, pp. 21–48. ———. 1998. ‘Identité culturelle et espace culturé: Les Fulɓe entre brousse et village’, in E. Boesen, Christine Hardung and Richard Kuba (eds), Regards sur le Borgou: Pouvoir et altérité dans une région ouest-africaine. Paris: L’Harmattan, pp. 221–42. ———. 2004a. ‘Hirtenkultur und Weltkultur: Die moderne Mobilität der Wodaabe ZentralNigers’, in Kurt Beck, Till Förster and Hans Peter Hahn (eds), Blick nach vorn: Festgabe für Gerd Spittler zum 65. Geburtstag. Cologne: Köppe, pp. 210–23. ———. 2004b. ‘Identität und Pluralität: Die Fulɓe in Zentralniger’, Paideuma 50: 101–26. ———. 2007a. ‘Pastoral Nomadism and Urban Migration: Mobility among the Fulbe Wodaabe in Central Niger’, in Hans Peter Hahn and Georg Klute (eds), Cultures of Migration: African Perspectives. Münster: LIT, pp. 31–60. ———. 2007b. ‘Des localités nomades: Les “maisons” wodaabe en brousse et en ville’, in Elisabeth Boesen and Laurence Marfaing (eds), Mobilité et nouveaux urbains dans l’espace Sahara-Sahel: Un cosmopolitisme par le bas. Paris: Karthala, pp. 209–31. ———. 2008a. ‘Tribal Culture, World Culture, Youth Culture: Wodaabe Dancers on Multiple Stages’, Sociologus 58(2): 143–68.

212  References ———.2008b. ‘Gleaming like the Sun: Aesthetic Values in Wodaabe Material Culture’, Africa 78(4): 582–602. ———. 2009. ‘Organisation sociopolitique et territorialité dans le monde nomade du Niger: Le cas des Fulbe-Wodaabe’, in Sophie Caratini (ed.), La question du pouvoir en Afrique du Nord et de l’Ouest: Affirmations identitaires et enjeux de pouvoir, vol. 2. Paris: L’Harmattan, pp. 67–97. ———. 2010. ‘Wodaabe Women and the Outside World’, in Ulrike Freitag and Achim von Oppen (eds), Translocality: The Study of Globalizing Processes from a Southern Perspective. Leiden: Brill, pp. 25–54. Boesen, Elisabeth. and Laurence Marfaing. 2006. ‘Entre ville et désert: Mobilité, activités et urbanité dans l’espace Sahara-Sahel’, compte rendu d’un colloque international au Centre de l’Orient Moderne (ZMO) à Berlin les 10–12 décembre 2005. Revue européenne des migrations internationales 22(3). http://remi.revues.org/index4122.html. Boesen, Elisabeth, and Laurence Marfaing (eds). 2007. ‘Vers un cosmopolitisme par le bas?’, in Elisabeth Boesen and Laurence Marfaing (eds), Mobilité et nouveaux urbains dans l’espace Sahara-Sahel: Un cosmopolitisme par le bas. Paris: Karthala, pp. 7–22. Boesen, Elisabeth, Laurence Marfaing and Mirjam de Bruijn. 2014. ‘Nomadism and Mobility in the Sahara-Sahel: Introduction’, Canadian Journal of African Studies 48(1): 1–12. Bonfiglioli, Angelo (Maliki). 1985. ‘Evolution de la propriété animale chez les WoDaaBe du Niger’, Journal des Africanistes 55(1–2): 29–37. ———. 1988. Duɗal: Histoire de famille et histoire de troupeau chez un groupe de Woɗaaɓe du Niger. Cambridge: Cambridge University Press. ———. 1998. ‘In Search of Light. From Traditional Knowledge to Rural Development’, paper presented at the Seminar ‘Crisis and Culture in Africa – with Special Emphasis on Pastoral Nomads and Farmers in the West African Sahel’, Uppsala, Sweden 27–30 March 1998. The Nordic Africa Institute. https://sites.google.com/site/angelobonfi/traditknowl​ edge, accessed 21 August 2014. Bonte, Pierre. 1979. ‘Pastoral Production, Territorial Organisation and Kinship in Segmentary Lineage Societies’, in P. Burnham and R. Ellen (eds), Social and Ecological Systems. London: Academic Press, pp. 203–34. ———. 1999. ‘Ethnicity and Land Tenure in the Sahel’, in Steven P. Reyna and R.E. Downs (eds), Deadly Developments: Capitalism, States and War. Amsterdam: Overseas Publishers Association, pp. 213–34. ———. 2006. Anthropologie des sociétés nomades: Fondements matériels et symboliques (première partie). Paris: Université de Paris 8. Bonte, Pierre, and John G. Galaty. 1991. ‘Introduction’, in John G. Galaty and Pierre Bonte (eds), Herders, Warriors and Traders: Pastoralism in Africa. Boulder, CO: Westview Press, pp. 3–30. Botte, Roger, and Jean Schmitz. 1994. ‘Paradoxes identitaires’, Cahiers d’Etudes africaines 34(133–135): 7–22. Bourdieu, Pierre. 1986. ‘The Forms of Capital’, in J. Richardson (ed.). Handbook of Theory and Research for the Sociology of Education. New York: Greenwood, pp. 241–58. Boutrais, Jean. 1990. ‘Les savanes humides, dernier refuge pastoral: L’exemple des Wodaabe, Mbororo de Centrafrique’, Genève Afrique 28(1): 65–90. ———. 1994. ‘Pour une nouvelle cartographie des Peuls’, Cahiers d’Etudes africaines 34(133– 135): 137–46. ———. 1995. Hautes terres d’élevage au Cameroun. Paris: Orstom.

References  213 Bovin, Mette. 1974/5. ‘Ethnic Performances in Rural Niger: An Aspect of Ethnic Boundary Maintenance’, Folk 16–17: 459–­74. ———. 1985. ‘Nomades “sauvages” et paysans “civilisés”: Wodaabe et Kanouri au Borno’, Journal des africanistes 55: 53–73. ———. 1990. ‘Nomads of the Drought: Fulbe and Wodaabe Nomads between Power and Marginalization in the Sahel of Burkina Faso and Niger Republic’, in Mette Bovin and Leif Manger (eds), Adaptive Strategies in Arid Lands: Proceedings from a Seminar at the Scandinavian Institute of African Studies, Uppsala, Sweden, April 1989. Uppsala: Nordiska Afrikainstitutet, pp. 29–57. ———. 1998. ‘Nomadic Performance – Peculiar Culture? “Exotic” Ethnic Performances of the Wodaabe Nomads of Niger’, in F. Hughes-Freeland and M.M. Crain (eds), Recasting Ritual: Performance, Media, Identity. London: Routledge, pp. 93–112. Braukämper, Ulrich. 1971. ‘Zur kulturhistorischen Bedeutung der Hirten-Ful für die Staatswesen des Zentralsudan’, Paideuma 17: 55–120. ———. 1992. Migration und ethnischer Wandel: Untersuchungen aus der östlichen Sudanzone. Stuttgart: Steiner. Breedveld, Anneke, and Mirjam de Bruijn. 1996. ‘L’image des Fulbe: Analyse critique de la construction du concept de pulaaku’, Cahiers d’Etudes africaines 36(144): 791–821. Brickell, Katherine, and Ayona Datta. 2011. ‘Introduction: Translocal Geographies’, in Katherine Brickell and Ayona Datta (eds), Translocal Geographies: Spaces, Places, Connections. Burlington: Ashgate, pp. 3–20. Burnham, Philip. 1979. ‘Spatial Mobility and Political Centralization in Pastoral Societies’, in l’Equipe écologie et anthropologie des sociétés pastorales (ed.), Pastoral Production and Society. Cambridge: Cambridge University Press, pp. 349–60. ———. 1996. The Politics of Cultural Difference in Northern Cameroon. Edinburgh: Edinburgh University Press. Canut, Cécile, and Étienne Smith. 2006. ‘Pactes, alliances et plaisanteries’, Cahiers d’études africaines 184: 687–754. Catley, Andy, Jeremy Lind and Ian Scoones (eds). 2013. Pastoralism and Development in Africa: Dynamic Change at the Margins. London: Routledge. Ciavolella, Riccardo. 2010. ‘Frontiers of Mobility, Limits of Citizenship: Political Meanings of Mobility for some Fulani Groups in Mauritania’, in Tilo Grätz (ed.), Mobility, Transnationalism and Contemporary African Societies. Newcastle: Cambridge Scholars Publishing, pp. 72–89. Clauzel, Jean. 1992. ‘L‘administration coloniale française et les sociétés nomades dans l’ancienne Afrique occidentale française’, Politique Africaine 46: 99–116. Cohen, Anthony P. 2000. ‘Boundaries of Consciousness, Consciousness of Boundaries: Critical Questions for Anthropology’, in Hans Vermeulen and Cora Govers (eds), The Anthropology of Ethnicity: Beyond ‘Ethnic Groups and Boundaries’. Amsterdam: Het Spinhuis, pp. 59–80. Cohen, Ronald. 1978. ‘Ethnicity: Problem and Focus in Anthropology’, Annual Review of Anthropology 7: 379–403. Colin de Verdière, P. 1998. Conséquences de la sédentarisation de l’élevage pastoral au Sahel: Etude comparée de trois systèmes agropastoraux dans la région de Filingué (Niger). Montpellier, Hohenheim: Centre de coopération internationale en recherche économique pour le développement (CIRAD). Comte, Gilbert. 1988. L’Empire triomphant 1871–1936. Vol. 1: Afrique occidentale et ­équatoriale. Paris: Denoël.

214  References Cooper, Barbara M. 2005. ‘Oral Sources and the Challenge of African History’, in John Edward Philips (ed.), Writing African History. Rochester: University of Rochester Press, pp. 191–215. Dafinger, Andreas. 2013. The Economics of Ethnic Conflict: The Case of Burkina Faso. Woodbridge, Suffolk: James Currey. Dafinger, Andreas, and Michaela Pelican. 2002. ‘Land Rights and the Politics of Integration: Pastoralists’ Strategies in a Comparative View’, Max Planck Institute for Social Anthropology Working Papers No. 48. http://www.eth.mpg.de/pubs/wps/pdf/mpi-eth-wo​ rking-paper-0048. de Bruijn, Mirjam. 2000. ‘Rapports interéthniques et identité: L’example des pasteurs peuls et des cultivateurs hummbeeɓe au Mali central’, in Youssouf Diallo and Günther Schlee (eds), L’éthnicité peule dans des contextes nouveaux: La dynamique des frontières. Paris: Karthala, pp. 15–36. ———. 2007. ‘Mobility and Society in the Sahel: An Exploration of Mobile Margins and Global Governance’, in Hans Peter Hahn and Georg Klute (eds), Cultures of Migration: African Perspectives. Münster: LIT, pp. 109–28. de Bruijn, Mirjam, and Inge Brinkman. 2011. “‘Communicating Africa”: Researching Mobile Kin Communities, Communication Technologies, and Social Transformation in Angola and Cameroon’, Autrepart 57–58: 41–58. ———. 2012. ‘Research Practice in Connections: Travels and Methods’, in Mirjam de Bruijn and Han van Dijk (eds), The Social Life of Connectivity in Africa. New York: Palgrave Macmillan, pp. 45–63. de Bruijn, Mirjam, and Han van Dijk. 2003. ‘Changing Population Mobility in West-Africa: Fulbe Pastoralists in Central and South Mali’, African Affairs 102: 285–307. de Bruijn, Mirjam, Han van Dijk and Dick Foeken. 2001. Mobile Africa: Changing Patterns of Movement in Africa and Beyond. Leiden: Brill. de Bruijn, Mirjam, Francis Nyamnjoh and Inge Brinkman (eds). 2009. Mobile Phones: The New Talking Drums of Everyday Africa. Leiden, Bamenda: African Studies Centre, Langaa. de Certeau, Michel. 1984. The Practice of Everyday Life. Berkeley, Los Angeles: University of California Press. de Jong, Ferdinand. 1999. ‘The Production of Translocality: Initiation in the Sacred Grove in Southern Senegal’, in Richard Fardon, Wim van Binsberger and Rijk van Dijk (eds), Modernity on a Shoestring: Dimensions of Globalization, Consumption and Development in Africa and Beyond. Leiden, London: EIDOS, pp. 315–40. de St. Croix, F.W. 1972. The Fulani of Northern Nigeria: Some General Notes. Westmead: Gregg International Publishers. Delmet, Christian. 1994. ‘Sur la route du pèlerinage: Les Peuls au Soudan’, Cahiers d’études africaines 34(133–5): 473–81. ———. 2000. ‘Les Peuls nomades au Soudan’, in Youssouf Diallo and Günther Schlee (eds), L’éthnicité peule dans des contexts nouveaux: La dynamique des frontiers. Paris: Karthala, pp. 191–206. Diallo, Youssouf. 2001. ‘Processes and Types of Pastoral Migration in Northern Côte d ­ ’Ivoire’, in Mirjam de Bruijn, Han van Dijk and Dick Foeken (eds), Mobile Africa: Changing Patterns of Movement in Africa and Beyond. Leiden: Brill, pp. 153–68. ———. 2006a. ‘Scherzbeziehungen in Westafrika. Ursprung, Funktion und Entwicklung einer sozialen Institution’, Zeitschrift für Ethnologie 131: 171–81. ———. 2006b. ‘Joking Relationships in Western Burkina Faso’, Zeitschrift für Ethnologie 131: 183–96.

References  215 ———. 2008. Nomades des espaces interstitiels: Pastoralisme, identité, migrations (Burkina Faso – Côte d’Ivoire). Cologne: Köppe. Diallo, Youssouf, Martine Guichard and Günther Schlee. 2000. ‘Quelques aspects comparatifs’, in Youssouf Diallo and Günther Schlee (eds), L’éthnicité peule dans des contextes nouveaux: La dynamique des frontières. Paris: Karthala, pp. 225–55. Diallo, Youssouf, and Günther Schlee (eds). 2000. L’ethnicité peule dans des contextes nouveaux: La dynamique des frontières. Paris: Karthala. Dodd, Jerrold L. 1994. ‘Desertification and Degradation in Sub-Saharan Africa’, BioScience 44(1): 28–34. Donahoe, Brian, John Eidson, Dereje Feyissa, Veronika Fuest, Markus V. Hoehne, Boris Nieswand, Günther Schlee and Olaf Zenker. 2009. ‘The Formation and Mobilization of Collective Identities in Situations of Conflict and Integration’, Max Planck Institute for Social Anthropology Working Papers No. 116. http://www.eth.mpg.de/pubs/wps/pdf/ mpi-eth-working-paper-0116. Donner, Jonathan. 2006. ‘Die sozialen und wirtschaftlichen Implikationen der Mobiltelefonie in Ruanda: Eine Typologie unter dem Gesichtspunkt Telefonbesitz und Telefonzugang’, in Peter Glotz, Stefan Bertschi and Chris Locke (eds), Daumenkultur: Das Mobiltelefon in der Gesellschaft. Bielefeld: Transcript, pp. 41–59. Douglas, Mary. 1978[1966]. Purity and Danger: An Analysis of the Concepts of Pollution and Taboo. London: Routledge and Kegan Paul. Dupire, Marguerite. 1954. ‘Contribution à l’étude des marques de propriété du bétail chez les pasteurs Peuls’, Journal de la Société des Africanistes 24(2): 123–43. ———. 1962. Peuls Nomades. Etude descriptive des WoDaaBe du Sahel nigerien. Paris: Karthala. ———. 1963. ‘The Position of Women in a Pastoral Society’, in Denise Paulme (ed.), Women in Tropical Africa. London: Routledge, pp. 47–92. ———. 1970. Organisation sociale des Peul: Etude d’éthnographie comparée. Paris: Librairie Plon. ———. 1972. Les facteurs humains de l’économie pastorale. Etudes Nigériennes No. 6. Niamey: Centre Nigérien de Recherches en Sciences Humaines. ———. 1975. ‘Exploitation du sol, communautés résidentielles et organisation lignagère des pasteurs woDaaBe (Niger)’, in Théodore Monod (ed.), Pastoralism in Tropical Africa. London: Oxford University Press, pp. 322–37. ———. 1981. ‘Réflexions sur l’éthnicité peule’, in Itinérances . . . en pays peul et ailleurs: Mélanges réunis par les chercheurs de l’ERA 246 du CNRS à la mémoire de Pierre Francis Lacroix. Paris: Société des Africanistes, pp. 165–81. ———. 1994. ‘Identité ethnique et processus d’incorporation tribale et étatique’, Cahiers d’études africaines 34(133–5): 265–80. Dyer, Nat. 2008. Review of the Legislative and Institutional Environment Governing Livestock Mobility in East and West Africa. Desk review of the project Securing Pastoralism in East and West Africa: Protecting and Promoting Livestock Mobility. IIED/SOS Sahel UK. Elwert, Georg. 2002. ‘Switching Identity Discourses: Primordial Emotions and the Social Construction of We-Groups’, in Günther Schlee (ed.), Imagined Differences: Hatred and the Construction of Identity. Hamburg: Lit, pp. 33–54. Evans-Pritchard, Edward E. 1940. The Nuer. A Description of the Modes of Livelihood and Political Institutions of a Nilotic People. Oxford: Oxford University Press. Fabietti, Ugo, and Philip Carl Salzman (eds). 1996. The Anthropology of Tribal and Peasant Pastoral Societies: The Dialectics of Social Cohesion and Fragmentation. Pavia: Ibis.

216  References Faliu, Alain. 1980. ‘Eléments d’histoire du peuplement peul de la vallée du fleuve Sénégal’, Bulletin de l’I.F.A.N. 42(2): 257–76. Fay, Claude. 2006. ‘Sang, lait, distance et plaisanterie’, Cahiers d’études africaines 184: 755–78. Feyissa, Dereje, and Günther Schlee. 2009. ‘Mbororo (Fulɓe) Migrations from Sudan into Ethiopia’, in Günther Schlee and Elizabeth E. Watson (eds), Changing Identifications and Alliances in North-East Africa. Vol.II: Sudan, Uganda, and the Ethiopia-Sudan Borderlands. New York, Oxford: Berghahn, pp. 157–78. Fischer-Rosenthal, Wolfram. 2000. ‘Was bringt die Biografieforschung der Transformationsforschung?’, in Ingrid Miethe and Silke Roth (eds), Politische Biografien und sozialer Wandel. Giessen: Psychosozial-Verlag, pp. 27–39. Fortier, Anne-Marie. 2000. Migrant Belongings: Memory, Space, Identity. Oxford, New York: Berg. Fortes, Meyer. 1983[1969]. ‘Verwandtschaft und das Axiom der Amity’, in Fritz Kramer and Christian Sigrist (eds), Gesellschaften ohne Staat II: Genealogie und Solidarität. Frankfurt am Main: Syndikat, pp. 120–64. Fouéré, Marie-Aude. 2005. ‘Les métamorphoses des “relations à plaisanteries”’, Cahiers d’études africaines 178: 389–430. Freitag, Ulrike, and Achim von Oppen. 2005. ‘Translokalität als ein Zugang zur Geschichte globaler Verflechtungen’, ZMO Programmatic texts, No. 2. Berlin: Zentrum Moderner Orient. ———. 2010. ‘“Translocality”: An Approach to Connection and Transfer in Area Studies (Introduction)’, in Ulrike Freitag and Achim von Oppen (eds), Translocality: The Study of Globalising Processes from a Southern Perspective. Leiden: Brill, pp. 1–21. Friedmann, John. 2002. ‘Placemaking as Project? Habitus and Migration in Transnational Cities’, in Jean Hillier and Emma Rooksby (eds), Habitus: A Sense of Place. Aldershot and Burlington: Ashgate, pp. 299–316. Fuglestad, Finn. 1983. A History of Niger 1850–1960. Cambridge: Cambridge University Press. Galaty, John G., and Pierre Bonte (eds). 1991. Herders, Warriors and Traders: Pastoralism in Africa. Boulder, CO: Westview Press. Gallais, Jean. 1975. Pasteurs et paysans du Gourma: La condition sahélienne. Paris: CNRS Galy, Kadir Abdelkader. 2010. L’esclavage au niger: Aspects historiques et juridiques. Paris: Karthala. Geschiere, Peter, and Josef Gugler. 1998. ‘The Urban-Rural Connection: Changing Issues of Belonging and Identification’, Africa 68(3): 309–19. Glick Schiller, Nina. 1990. ‘“Everywhere We Go, We Are in Danger”: Ti Manno and the Emergence of a Transnational Haitian Identity’, American Ethnologist 17(2): 329–47. Glick Schiller, Nina, Linda Basch and Christina Szanton Blanc. 1995. ‘From Immigrant to Transmigrant: Theorizing Transnational Migration’, Anthropological Quarterly 68(1): 48–63. Godelier, Maurice. 1999. The Enigma of the Gift. Cambridge: Polity Press. Goffman, Erving. 1959. The Presentation of Self in Everyday Life. New York: Doubleday. Granovetter, Mark S. 1973. ‘The Strength of Weak Ties’, American Journal of Sociology 78(6): 1360–80. Grätz, Tilo, Barbara Meier and Michaela Pelican. 2003. ‘Zur sozialen Konstruktion von Freundschaft: Überlegungen zu einem vernachlässigten Thema der Sozialanthropologis (Schwerpunkt Afrika)’, Max Planck Institute for Social Anthropology Working Papers No. 53. http://www.eth.mpg.de/pubs/wps/pdf/mpi-eth-working-paper-0053.

References  217 Greenough, Karen. 2003. Development Agents and Nomadic Agency in the Damergou, Niger: Four Perspectives in the Development ‘Market’. Lexington: University of Kentucky. Greiner, Clemens. 2010. ‘Patterns of Translocality: Migration, Livelihoods and Identities in Northwest Namibia’, Sociologus 60(2): 131–61. Greiner, Clemens, and Patrick Sakdapolrak. 2013. ‘Translocality: Concepts, Applications and Emerging Research Perspectives’, Geography Compass 7(5): 373–84. Grémont, Charles. 2011. ‘Ancrage au sol et (nouvelles) mobilités dans l’espace saharo-sahélien: Des expériences similaires et compatibles’, L’Année du Maghreb 7: 177–89. ———. 2014. ‘Mobility in Pastoral Societies of Northern Mali: Perspectives on Social and Political Rationales’, Canadian Journal of African Studies 48(1): 29–40. Guichard, Martine. 1996. ‘“Les Fulbe du Borgou n’ont vaincu personne”: De la culture politique d’une minorité ethnique béninoise’, Ph.D. dissertation. Bielefeld: University of Bielefeld. _____. 2014. ‘Introduction’, in Martine Guichard, Tilo Grätz and Youssouf Diallo (eds), Friendship, Descent and Alliance in Africa: Anthropological Perspectives. New York, Oxford: Berghahn, pp. 1–15. Guichard, Martine, Tilo Grätz and Youssouf Diallo (eds). 2014. Friendship, Descent and Alliance in Africa: Anthropological Perspectives. New York, Oxford: Berghahn. Gulliver, P.H. 1975. ‘Nomadic Movements: Causes and Implications’, in Théodore Monod (ed.), Pastoralism in Tropical Africa. London: Oxford University Press, pp. 369–86. Gupta, Akhil, and James Ferguson. 1997a. ‘Discipline and Practice: “The Field” as Site, Method and Location in Anthropology’, in Akhil Gupta and James Ferguson (eds), Anthropological Locations. Berkeley: University of California Press, pp. 1–46. ———. 1997b. ‘Culture, Power, Place: Ethnography at the End of an Era’, in Akhil Gupta and James Ferguson (eds), Culture, Power, Place: Explorations in Critical Anthropology. Durham, London: Duke University Press, pp. 1–29. ———. 1997c. ‘Beyond Culture: Space, Identity, and the Politics of Difference’ in Akhil Gupta and James Ferguson (eds), Culture, Power, Place: Explorations in Critical Anthropology. Durham, London: Duke University Press, pp. 33–51. Hagberg, Sten. 1998. Between Peace and Justice: Dispute Settlement between Karaboro Agriculturalists and Fulɓe Agro-Pastoralists in Burkina Faso. Uppsala Studies in Cultural Anthropology. Uppsala: Acta Universitatis Uppsaliensis. ———. 2000. ‘Strangers, Citizens, Friends: Fulbe Agro-Pastoralists in Western Burkina Faso’, in Sten Hagberg and Alexis B. Tengan (eds), Bonds and Boundaries in Northern Ghana and Southern Burkina Faso. Uppsala Studies in Cultural Anthropology. Uppsala: Acta Universitatis Uppsaliensis, pp. 159–79. Hahn, Hans Peter, and Georg Klute (eds). 2007. Cultures of Migration: African Perspectives. Berlin: Lit. Hammel, R. 2001. Terroirs d’attache des pasteurs au Niger. London, Edinburgh: International Institute for Environment and Development (IIED). Hampshire, Kate, and Sara Randall. 1999. ‘Seasonal Labour Migration Strategies in the Sahel: Coping with Poverty or Optimising Security?’, International Journal of Population Geography 5: 367–85. ———. 2005. ‘People are a Resource: Demography and Livelihoods in Sahelian FulBe of Burkina Faso’, in Katherine Homewood (ed.), Rural Resources and Local Livelihoods in Africa. Oxford: James Currey, pp. 123–36. Hannerz, Ulf. 2003. ‘Being There. . . and There. . . and There! Reflections on Multi-Site Ethnography’, Ethnography 4(2): 201–16.

218  References Hardin, Garrett. 1968. ‘The Tragedy of the Commons’, Science 162 (3859): 1243–48. Hedberg, Charlotta, and Renato Miguel do Carmo. 2012. ‘Translocal Ruralism: Mobility and Connectivity in European Rural Spaces’, in Charlotta Hedberg and Renato Miguel do Carmo (eds), Translocal Ruralism: Mobility and Connectivity in European Rural Spaces. New York: Springer, pp. 1–9. Herskovits, Melville J. 1926. ‘The Cattle Complex in Eastern Africa’, American Anthropologist 28: 230–72, 361–380, 494–528, 630–664. Hickey, Joseph V. 1978. ‘Shifting Marital Alliances among the Bokkos Fulani’, Ethnology 17(1): 25–37. Hiernaux, P., and Bagoudou Maidaji. 2006. Evaluation des risques environnementaux liés à la gestion des ressources naturelles de la région de Zinder. Zinder (Niger): Projet de Sécurisation des Systèmes Pastoraux. Hirschman, Albert O. 1970. Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations and States. Cambridge, London: Harvard University Press. Holy, Ladislav (ed.). 1979. Segmentary Lineage Systems Reconsidered. The Queen’s University Papers in Social Anthropology, Vol. 4. Belfast: The Queen’s University of Belfast. ———.. 1996. Anthropological Perspectives on Kinship. London: Pluto Press. Homewood, Katherine. 2008. Ecology of African Pastoralist Societies. Oxford: James Currey Hopen, C. Edward. 1958. The Pastoral Fulbe Family in Gwandu. London: Oxford University Press. Horowitz, Michael M. 1972. ‘Ethnic Boundary Maintenance among Pastoralists and Farmers in the Western Sudan (Niger)’, in William Irons and Neville Dyson-Hudson (eds), Perspectives on Nomadism. Leiden: Brill, pp. 105–14. Horstmann, Alexander, and Günther Schlee (eds). 2001. Integration durch Verschiedenheit: Lokale und globale Formen interkultureller Kommunikation. Bielefeld: Transcript. IIED (International Institute for Environment and Development). 2006. ‘Innovation in Securing Land Rights in Africa: Lessons from Experience’, Briefing Paper. London: IIED. Ingold, Tim. 1987. ‘Changing Places: Movement and Locality in Hunter-Gatherer and Pastoral Societies’, in The Appropriation of Nature: Essays on Human Ecology and Social Relations. Iowa City: University of Iowa Press, pp. 165–97. ———. 2007. Lines: A Brief History. London, New York: Routledge. ———. 2009. ‘Against Space: Place, Movement, Knowledge’, in Peter Wynn Kirby (ed.), Boundless Worlds: An Anthropological Approach to Movement. New York, Oxford: Berghahn, pp. 29–43. Kane, Oumar. 2004. La première hégémonie peule: Le Fuuta Tooro de Koli Teŋella à Almaami Abdul. Paris, Dakar: Karthala. Khazanov, Anatoly M., and Günther Schlee (eds). 2012. Who Owns the Stock? Collective and Multiple Rights in Animals. New York, Oxford: Berghahn. Kintz, Danièle. 1982. ‘Pastoralisme, agro-pastoralisme et organisation foncière: Les cas des Peuls’, in E. Le Bris, E. Le Roy and F. Leimdorfer (eds), Enjeux fonciers en Afrique noire. Paris: Karthala, pp. 212–17. ———. 1985. ‘Archetypes politiques peuls’, Journal des Africanistes 55(1–2): 93–104. Klute, Georg, and Hans Peter Hahn. 2007. ‘Cultures of Migration: Introduction’, in Hans Peter Hahn and Georg Klute (eds), Cultures of Migration: African Perspectives. Berlin: Lit, pp. 9–27. Köhler, Florian. 2016. ‘The Sedentarization of Dwelling: Continuity and Change in the Habitat of Fulɓe Woɗaaɓe Pastoralists and Urban Migrants in Niger’, Text and Photo

References  219 Essays. Halle (Saale): Max Planck Institute for Social Anthropology. http://www.eth.mpg. de/pubs/series_fieldnotes/vol0015.html. ———. 2017a. ‘From Nomadic Pastoralism to Urban Migration: Transformation Processes in a Group of Fulɓe Woɗaaɓe in Niger’, Narrative Interviews and Complementary Material. Halle (Saale): Max Planck Institute for Social Anthropology. http://www.eth. mpg.de/pubs/series_fieldnotes/vol0020.html. ———. 2017b. ‘Managing Elopement on the Mobile Phone: Continuity and Change in Woɗaaɓe te’egal Marriage’, Canadian Journal of African Studies / Revue canadienne des études africaines 51(3): 341–58. DOI: 10.1080/00083968.2017.1387061. Kopytoff, Igor (ed.). 1987. The African Frontier: The Reproduction of Traditional African Societies. Bloomington, IN: Indiana University Press. Krader, Lawrence. 1959. ‘The Ecology of Nomadic Pastoralism’, International Social Science Journal 11(4): 499–510. Krätli, Saverio. 2007. ‘Cows Who Choose Domestication: Generation and Management of Domestic Animal Diversity by WoDaaBe Pastoralists (Niger)’, Ph.D. dissertation. University of Sussex: Institute of Development Studies. ———. 2008. ‘Cattle Breeding, Complexity and Mobility in a Structurally Unpredictable Environment: The WoDaaBe Herders of Niger’, Nomadic Peoples 12(1): 11–41. Krätli, Saverio, and Carolyne Dyer. 2006. ‘Education and Development for Nomads: The Issues and the Evidence’, in Carolyne Dyer (ed.), The Education of Nomadic Peoples: Current Issues, Tuture Prospects. New York, Oxford: Berghahn, pp. 8–34. ———. 2009. ‘Mobile Pastoralists and Education: Strategic Options’, Education for Nomads Working Paper 1. London: International Institute for Environment and Development (IIED). Kuper, Adam. 1982. ‘Lineage Theory: A Critical Retrospect’, Annual Review of Anthropology 11: 71–95. Lassibille, Mahalia. 1999. ‘L’homme et la vache dans l’esthétique des Peuls Woɗaaɓe’, in Catherine Baroin and Jean Boutrais (eds), L’homme et l’animal dans le bassin du lac Tchad. Paris: IRD, pp. 251–62. ———. 2006. ‘Les danses woDaaBe entre spectacles touristiques et scènes internationales: Les coulisses d’une migration chorégraphique’, Autrepart 40: 113–29. ———. 2008. ‘Stratégies chorégraphiques: L’exercice politique de la danse chez les Peuls WoDaaBe du Niger’, Journal des Anthropologues 112–113: 155–81. ———. 2009. ‘Les scènes de la danse: Entre espace touristique et politique chez les Peuls woDaaBe du Niger’, Cahiers d’études africaines 49(193–194): 309–35. Lefebvre, Henri. 1991[1974]. The Production of Space. Oxford, Cambridge: Blackwell. Lévi-Strauss, Claude. 1966. ‘Introduction à l’oeuvre de Marcel Mauss’, in M. Mauss, Sociologie et Anthropologie. Paris: Presses Universitaires de France. _____. 2002[1947]. Les structures elémentaires de la parenté. Berlin, New York: Mouton de Gruyter. Loftsdóttir, Kristín. 2000. The Bush Is Sweet: Identity and Desire among the WoDaaBe in Niger. Ann Arbor, MI: UMI Research Press. ———. 2001a. ‘A Difficult Time: Migrant Work and the WoDaaBe in Niger’, Cultural Survival Quarterly 25(3): 40–43. ———. 2001b. ‘Birds of the Bush: Wodaabe Distinctions of Society and Nature’, Nordic Journal of African Studies 10(3): 280–98. ———. 2001c. ‘Where My Cord is Buried: WoDaaBe Use and Conceptualization of Land’, Journal of Political Ecology 8: 1–24.

220  References ———. 2002a. ‘Knowing What to Do in the City: WoDaaBe Nomads and Migrant Workers in Niger’, Anthropology Today 18(1): 9–13. ———. 2002b. ‘The Place of Birth: Wodaabe Changing Histories of Origin’, History in Africa 29: 283–307. ———. 2002c. ‘Never Forgetting? Gender and Racial-Ethnic Identity during Fieldwork’, Social Anthropology 10(3): 303–17. ———. 2004. ‘When Nomads Lose Cattle: Wodaabe Negotiations of Ethnicity’, African Sociological Review 8(2): 52–76. ———. 2007. ‘Bounded and Multiple Identities: Ethnic Identifications of Wodaabe and Fulbe’, Cahiers d’études africaines 185: 65–92. Loimeier, Roman. 2011. ‘Die Rinderpest 1887–1898 in Afrika. Eine Tragödie in zwei Akten’, in Roman Loimeier (ed.), Seuchen im subsaharischen Afrika. Berlin: Lit, pp. 83–114. Lombard, Jacques. 1981. ‘Jeux de miroirs: Tendances individualistes et réseau communautaire chez les Peuls’, in Itinérances . . . en pays peul et ailleurs. Mélanges réunis par les chercheurs de l’ERA 246 du CNRS à la mémoire de Pierre Francis Lacroix. Paris: Société des africanistes, pp. 183–93. Loncke, Sandrine. 2002. ‘Lignages et lignes de chant chez les Peuls Woɗaaɓe du Niger’, Ph.D. dissertation. Paris: Institut National des Langues et Civilisations Orientales. ———. 2008. ‘Musique et transmission musicale dans une société nomade. L’exemple des Peuls Wodaabe du Niger’, Cahiers d’Ethnomusicologie 22: 203–22. ———. 2015. Geerewol: Musique, danse et lien social chez les Peuls nomades woɗaaɓe du Niger. Nanterre: Société d’ethnologie. Loncke, Sandrine, and Jean-Marc Durou. 2000. Les Bororo: Nomades Peuls du Niger. Paris: VILO. Lund, Christian. 1998. Land, Power and Politics in Niger: Land Struggles and the Rural Code. Hamburg: Lit. Maliki [also spelled Maaliki], Angelo B. 1981. ‘Ngaynaaka: L’élevage selon les WoDaaBe’, Discussion Paper, Niger Range and Livestock Project. Niamey: Republic of the Niger, Ministère du Développement Rural, USAID. ———. 1982. ‘Introduction à l’histoire des Wodaabe du Niger’, Rapport préliminaire/ Discussion Paper Number 3. Niamey: Republic of the Niger, Ministère du développement Rural, USAID. ———. 1988. Nomades peuls. Paris: L’Harmattan. Maliki, Angelo, Cynthia White, Louis Loutan and Jeremy J. Swift. 1984. ‘The WoDaaBe’, in Jeremy Swift (ed.), Pastoral Development in Central Niger: Report of the Niger Range and Livestock Project. Niamey: Republic of the Niger, Ministère du Développement Rural, USAID, pp. 255–530. Mansour, Moutari, and Su Fei Tan. 2008. Niger/Nigeria Desk Review. Project Securing Pastoralism in East and West Africa: Protecting and Promoting Livestock Mobility. IIED/SOS Sahel UK. Marcus, George E. 1986. ‘Contemporary Problems of Ethnography in the Modern World System’, in James Clifford and George E. Marcus (eds), Writing Culture. Berkeley: University of California Press, pp. 165–93. ———. 1995. ‘Ethnography in/of the World System: The Emergence of Multi-Sited Ethnography’, Annual Review of Anthropology 24: 95–117. Marfaing, Laurence. 2014a. ‘Quelles mobilités pour quelles ressources?’, Canadian Journal of African Studies 48(1): 41–57.

References  221 ———. 2014b. ‘Travailleurs migrants dans les villes du Sahara-Sahel comme Nouakchott et Bamako): La mobilité pour la resource’, in Elisabeth Boesen and Laurence Marfaing (eds), Mobilités dans l’espace ouest-africain: Ressources, développement local et integration régionale. Paris: Karthala, pp. 133–59. Massey, Doreen. 1994. Space, Place, and Gender. Minneapolis: University of Minnesota Press. ———. 2005. For Space. London: Sage Publications. Mauss, Marcel. 1947. Manuel d’éthnographie. Paris: Payot. ———. 1979[1904/5]. Seasonal Variations of the Eskimo: A Study in Social Morphology. London: Routledge & Kegan Paul. ———. 1990[1925]. The Gift. The Form and Reason for Exchange in Archaic Societies. New York, London: Norton. ———. 2013[1928]. ‘Joking Relations’, HAU: Journal of Ethnographic Theory 3(2): 317–34. Miller, Daniel. 2009. ‘What Is a Mobile Phone Relationship?’, in Erwin Alampay (ed.), Living the Information Society in Asia. Singapore: Institute of Southeast Asian Studies, pp. 24–35. Mohamadou, Eldridge. 1969. ‘Les Peuls du Niger oriental: Groupes ethniques et dialectes’, Camelang 2: 57–93. Moritz, Mark. 2006. ‘Changing Contexts and Dynamics of Farmer-Herder Conflicts across West Africa’, Canadian Journal of African Studies 40(1): 1–40. Moutari, Mansour Elhaji. 1999. ‘La problématique du foncier dans la zone pastorale de Zinder (Niger)’, in André Bourgeot (ed.), Horizons nomades en Afrique sahélienne: sociétés développement et démocratie. Paris: Karthala, pp. 425–33. Müller-Dempf, Harald. 2014. ‘Hybrid Pastoralists – Development Interventions and New Turkana Identities’, Max Planck Institute for Social Anthropology Working Papers No. 156. http://www.eth.mpg.de/pubs/wps/pdf/mpi-eth-working-paper-0156. Murphy, William P., and Caroline H. Bledsoe. 1987. ‘Kinship and Territory in the History of a Kpelle Chiefdom (Liberia)’, in Igor Kopytoff (ed.), The African Frontier: The Reproduction of Traditional African Societies. Bloomington, IN: Indiana University Press, pp. 123–47. Mutie, Pius Mutuku. 2013. The Art of Interethnic Coexistence: Some Evidence from Kenya. Saarbrücken: Lambert Academic Publishing. Niamir-Fuller, Maryam (ed.). 1999. Managing Mobility in African Rangelands: The Legitimization of Transhumance. London: Intermediate Technology Publications. Olwig, Karen Fog. 1997. ‘Cultural Sites: Sustaining a Home in a Deterritorialized World’, in Karen Fog Olwig and Kirsten Hastrup (eds), Siting Culture: The Shifting Anthropological Object. London: Routledge, pp. 17–38. Osman, Elhadi Ibrahim. 2009. ‘The Funj Region Pastoral Fulɓe: From “Exit” to “Voice”’, Nomadic Peoples 13(1): 92–112. ———. 2013. The Pastoral Fulbe in the Sudan Funj Region: A Study of the Interaction between State and Society. Bergen: Chr. Michelsen Institute. Ostrom, Elinor. 1999. ‘Coping with the Tragedy of the Commons’, Annual Review of Political Science 2: 493–535. Oumarou, Amadou. 2012. Dynamique du Pulaaku dans les sociétés Peules du Dallol Bosso (Niger). Paris: L’Harmattan. Oumarou, Boubacar. 2011. Pasteurs Nomades face à l’Etat du Niger. Paris: L’Harmattan. Packard, Randall M. 1987. ‘Debating in a Common Idiom: Variant Traditions of Genesis among the BaShu of Eastern Zaire’, in Igor Kopytoff (ed.), The African Frontier: The Reproduction of Traditional African Societies. Bloomington, IN: Indiana University Press, pp. 149–61.

222  References Paris, Patrick. 1981. ‘Le meeting de Jesus’, in Itinérances . . . en pays peul et ailleurs. Mélanges réunis par les chercheurs de l’ERA 246 du CNRS à la mémoire de Pierre Francis Lacroix. Paris: Société des africanistes, pp. 249–68. ———. 1990. ‘Les Bororos: Nomades Peuls du Niger’, in Marc Durou, Ténéré: Désert d’absolus. Paris: VILO, pp. 189–203. ———. 1997. ‘Ga’i ngaanyka ou les taureaux de l’Alliance: Description ethnographique d’un rituel interlignager chez les Peuls Vod’aab’e du Niger’, Journal des Africanistes 97(2): 71–100. Pelican, Michaela. 2003. ‘Interethnische Freundschaften in Nordwestkamerun: Ein Vergleich ethnischer und gender-spezifischer Konzepte und Praktiken’, Max Planck Institute for Social Anthropology Working Papers No. 56. http://www.eth.mpg.de/pubs/wps/pdf/ mpi-eth-working-paper-0056. ———. 2006. ‘Getting along in the Grassfields: Interethnic Relations and Identity Politics in Northwest Cameroon’, Ph.D. dissertation. Halle (Saale): Martin Luther University of Halle-Wittenberg. Peters, E.L. 1967. ‘Some Structural Aspects of the Feud among the Camel-Herding Bedouin of Cyrenaica’, Africa 37: 261–82. Peyre de Fabrègues, Bernard. 1987. ‘Aspects pastoraux du développement de l’élevage en zone Sahélienne dans le contexte de la période de sécheresse. Le cas du Niger’, in Bernard Peyre de Fabregues, Le développement rural: comprendre pour agir. Paris: ORSTOM, pp. 309–39. PSSP (Projet de Sécurisation des Systèmes Pastoraux). 2009. La mobilité pastorale dans la région de Zinder: Stratégies et dynamisme des sociétés pastorales. Niamey. Rain, David. 1999. Eaters of the Dry Season: Circular Labor Migration in the West African Sahel. Boulder, CO: Westview Press. Ranger, Terence. 1983. ‘The Invention of Tradition in Colonial Africa’, in Eric Hobsbawm and Terence Ranger (eds), The Invention of Tradition. Cambridge: Cambridge University Press, pp. 211–62. Rappaport, Roy A. 1999. Ritual and Religion in the Making of Humanity. Cambridge: Cambridge University Press. Reed, L.N. 1932. ‘Notes on Some Fulani Tribes and Customs’, Africa 5(4): 422–54. Republic of the Niger. 1954. Arrêté n°311-SA du 10 février 1954 fixant une limite Nord des cultures. ———. 1961. Loi n° 61-05 du 26 mai 1961 fixant une limite Nord des cultures. ———. 1987a. Arrêté No. 76/MI/MDI/DAPA DU 1er juillet 1987 fixant la composition des commissions chargées d’examiner les litiges relatifs à la circulation et le droit au pâturage du bétail dans les zones de cultures. ———. 1987b. Décret No. 87 – 077/PCMS/MI du 18 juin 1987 réglementant la circulation et le droit de pâturage du bétail dans les zones de cultures. ———. 1993. Ordonnance n° 93-015 du 2 mars 1993 fixant les Principes d’Orientation du Code Rural. ———. 1997a. Texte complémentaire du Code Rural Nigérien. Décret N° 97-007/PRN/MAG/E du 10 janvier 1997 fixant statut des terroirs d’attache des pasteurs. ———. 1997b. Décret N° 97-367/PRN/MAG/E du 2 Octobre 1997 déterminant les modalités d’inscription des droits fonciers au Dossier Rural. ———. 1997c. Décret N° 97-006/PRN/MAG/E du 10 janvier 1997 portant réglementation de la mise en valeur des ressources naturelles rurales. ———. 2003[1961]. Loi N° 2003-025 du 13 juin 2003 modifiant la loi N° 61-27 du 15 juillet 1961, portant institution du Code Pénal.

References  223 ———. 2004[1999]. Ordonnance n° 99-37 du 4 septembre 1999 portant Code Electoral modifiée et complétée par l’ordonnance N° 99-39 du 23 septembre 1999, la loi N°2003-32 du 17 juillet 2003, la loi 2003- 64 du 31 décembre 2003, la loi 2004-004 du 29 mars 2004, la loi n°2004-14 du 13 mai 2004 modifiant et complétant l’ordonnance n°99-37 du 04 septembre 1999 portant code électoral et la loi n°2004-22 du 02 juin 2004, modifiant et complétant l’ordonnance n°99-37 du 04 septembre 1999 portant code electoral. ———. 2005. Arrêté n° 098 /MDA/CNCR/SP du 25 novembre 2005 portant organisation, attributions et modalités de fonctionnement des Commissions foncières de communes, de villages ou tribus. ———. 2010. Ordonnance 2010-029 du 20 mai 2010 relative au pastoralisme. ———. 2013. Présentation des résultats préliminaires du quatrième (4ième) recensement général da la population et de l’habitat (RGP/H) 2012. Niamey: Ministère des Finances, Institut Nationale de la Statistique. Retaillé, Denis. 1984. ‘La mise en place d’une région en Afrique sahélienne autour du Koutous (Niger oriental)’, in C. Blanc-Pamard, J. Bonnemaison, J. Boutrais and V. Lassailly-Jacob (eds), Le développement rural en questions: Paysages, espaces ruraux, systèmes agraires. Paris: ORSTOM, pp. 181–203. ———. 1998. ‘L’espace nomade’, Revue de géographie de Lyon 73(1): 71–82. ———. 2006. ‘Espace mobile, territorialité sans territoire et urbanite sans ville’, in Denis Retaillé (ed.), La ville ou l’état? Développement politique et urbanite dans les espaces nomads ou mobiles. Rouen: Publications des Universités de Rouen et du Havre, pp. 217–42. ———. 2010. ‘De l’espace nomade à l’espace mobile en passant par l’espace du contrat: Une expérience théorique’. ADES-CNRS-Université de Bordeaux. http://halshs.archives-ouvertes.fr/halshs-00679163, accessed 21 August 2014. ———. 2013. ‘From Nomadic to Mobile Space: A Theoretical Experiment (1976–2012)’, in Judith Miggelbrink, Joachim Otto Habeck, Nuccio Mazzulo and Peter Koch (eds), Nomadic and Indigenous Spaces, Productions and Cognitions. Burlington: Ashgate, pp. 55–75. Retaillé, Denis, and Olivier Walther. 2011. ‘Spaces of Uncertainty: A Model of Mobile Space in the Sahel’, Singapore Journal of Tropical Geography 32(1): 85–101. ———. 2012. ‘New Ways of Conceptualizing Space and Mobility: Lessons from the Sahel to the Globalized World’, CEPS/INSTEAD Working Paper No. 2012-24. Esch-sur-Alzette: Centre d’Etudes de Populations, de Pauvreté et Politiques Socio-Economiques CEPS/ INSTEAD. Reyna, Stephen P. 2014. ‘Friendship in a World of Power’, in Martine Guichard, Tilo Grätz and Youssouf Diallo (eds), Friendship, Descent and Alliance in Africa: Anthropological Perspectives. New York, Oxford: Berghahn, pp. 161–79. Riesman, Paul. 1974. Société et liberté chez les Peul Djelgôbé de Haute-Volta. Essai d’anthropologie introspective. Paris, La Haye: Mouton. Sahlins, Marshall. 1961. ‘The Segmentary Lineage: An Organization of Predatory Expansion’, American Anthropologist 63(2): 322–45. ———. 1974. Stone Age Economics. London: Tavistock. ———. 1985. Islands of History. Chicago and London: The University of Chicago Press. Salifou, Maman. 2008. Historique de la décentralisation au Niger. Niamey: Ministère de ­l’intérieur, de la sécurité publique et de la decentralisation. Direction Générale de ­l’Administration Territoriale et des Collectivités Locales. Salih, M.A. Mohamed 1995. ‘Pastoralist Migration to Small Towns in Africa’, in Jonathan Baker and Tade Akin Aina (eds), The Migration Experience in Africa. Uppsala: Nordiska Afrikainstitutet, pp. 181–96.

224  References Sallah, Alhassane. 1999. ‘L’élevage extensif dans le Damergou (Niger)’, in André Bourgeot (ed.), Horizons nomades en Afrique sahélienne: sociétés développement et démocratie. Paris: Karthala, pp. 263–72. Salzmann, Philip Carl. 1978. ‘Does Complementary Opposition Exist?’, American Anthropologist 80(1): 53–70. ———. (ed.). 1980. When Nomads Settle. Processes of Sedentarization as Adaptation and Response. New York: Praeger, Bergin Publishers. ———. 2004. Pastoralists: Equality, Hierarchy, and the State. Boulder, CO: Westview Press. Sambo, Bodé. 2008. Historique et itinéraire de la mobilité du groupe d’éleveurs transhumants WodaaBe Suudu Suka’el de la commune de Tanout (Damergou) region de Zinder. Niamey: Université Abdou Moumouni, Laboratoire d’études et de recherche sur les dynamiques sociales et le développement local (LASDEL). Sandford, Stephen. 1983. Management of Pastoral Development in the Third World. Chichester: Wiley. Schareika, Nikolaus. 1994. Die soziale Bedeutung der Rinder bei den Fulbe (Benin). Hamburg: Lit. ———. 2003a. Westlich der Kälberleine. Nomadische Tierhaltung und naturkundliches Wissen bei den Woɗaaɓe Südostnigers. Münster: Lit-Verlag. ———. 2003b. Know to Move, Move to Know: Ecological Knowledge and Herd Movement Strategies among the Wodaabe of Southeastern Niger. Rom: FAO. ———. 2004. ‘Bush for Beasts: Lake Chad from the Nomadic Wodaabe Point of View’, in Matthias Krings and Editha Platte (eds), Living with the Lake: Perspectives on History, Culture and Economy of Lake Chad. Cologne: Köppe, pp. 171–90. ———. 2007. ‘Söhne des Feuers, Brüder der Milch: Politische Prozesse bei westafrikanischen Nomaden am Beispiel der Woɗaaɓe in Südostniger’, habilitation thesis. Mainz: Johannes Gutenberg-Universität. ———. 2010a. ‘Pulaaku in Action: Words at Work in Woɗaaɓe Clan Politics’, Ethnology 49(3): 207–27. ———. 2010b. ‘Rituell gezeugt: Verwandtschaft als symbolische Interaktion bei den Woɗaaɓe Südostnigers’, in Erdmute Alber (ed.), Verwandtschaft heute: Positionen, Ergebnisse und Perspektiven. Berlin: Reimer, pp. 93–117. ———. 2011. ‘In den Mühlen der Lineage: Friktionen und Fraktionen um das Projekt Getreidebank bei den Wodaabe Südostnigers’, in Nikolaus Schareika, Eva Spies and Pierre-Yves LeMeur (eds), Auf dem Boden der Tatsachen: Festschrift für Thomas Bierschenk. Cologne: Köppe, pp. 185–95. Schlee, Günther. 1984. ‘Nomaden und Staat: Das Beispiel Nordkenia’, Sociologus 34: 140–61. ———. 1990a. ‘Les nomades et l’etat au nord du Kenia’, Working Paper No. 149. Bielefeld: University of Bielefeld, Faculty of Sociology, Sociology of Development Research Centre. ———. 1990b. ‘Ritual Topography and Ecological Use’, Working Paper No. 134. Bielefeld: University of Bielefeld, Faculty of Sociology, Sociology of Development Research Centre. ———. 1997. ‘Ethnizitäten in neuen Kontexten: Urbane und hirtennomadische Migranten (Burkina Faso, Côte d’Ivoire, Benin, Kamerun, Sudan)’, Working Paper No. 282. Bielefeld: University of Bielefeld, Faculty of Sociology, Sociology of Development Research Centre. ———. 2000a. ‘Introduction: Une ethnicité dans des contextes nouveaux’, in Youssouf Diallo and Günther Schlee (eds), L’ethnicité peule dans des contextes nouveaux: La dynamique des frontières. Paris: Karthala, pp. 7–14.

References  225 ———. 2000b. ‘Les Peuls du Nil’, in Youssouf Diallo and Günther Schlee (eds), L’ethnicité peule dans des contextes nouveaux: La dynamique des frontières. Paris: Karthala, pp. 207–23. ———. 2001a. ‘Nomadismus’, in Jacob E. Mabe (ed.), Das Afrika-Lexikon: Ein Kontinent in 1000 Stichwörtern. Wuppertal: Peter Hammer/Stuttgart and Weimar: Metzler, pp. 454–457. ———. 2001b. ‘Einleitung’, in Alexander Horstmann and Günther Schlee (eds), Integration durch Verschiedenheit: Lokale und globale Formen interkultureller Kommunikation. Bielefeld: Transcript, pp. 17– 46. ———. 2003a. ‘Three Dyads Campared: Nuer/Anywaa (Ethiopia), Maasai/Kamba (Kenya), and Evenki/Buryat (Siberia)’, in Max Planck Institute for Social Anthropology. Report 20022003. Halle (Saale): Max Planck Institute for Social Anthropology, pp. 53–69. https:// www.eth.mpg.de/3026204/mpi-report-2002-2003.pdf, accessed 12 January 2017. ———. 2003b. ‘Integration und Konflikt’, Entwicklungsethnologie 12(1+2): 74–95. ———. 2004. ‘Taking Sides and Constructing Identities: Reflections on Conflict Theory’, Journal of the Royal Anthropological Institute 10: 135–56. ———. 2005. ‘Forms of Pastoralism’, in Stefan Leder and Bernhard Streck (eds), Shifts and Drifts in Nomad-Sedentary Relations. Wiesbaden: Reichert, pp. 17–53. ———. 2008. How Enemies Are Made: Towards a Theory of Ethnic and Religious Conflicts. New York, Oxford: Berghahn. ———. 2009. ‘Introduction’, in Günther Schlee and Elizabeth E. Watson (eds), Changing Identifications and Alliances in North-East Africa, vol. 1. New York, Oxford: Berghahn, pp. 1–13. ———. 2010. ‘Territorializing Ethnicity: The Political Ecology of Pastoralism in Northern Kenya and Southern Ethiopia’, Max Planck Institute for Social Anthropology Working Papers No. 121. http://www.eth.mpg.de/pubs/wps/pdf/mpi-eth-working-paper-0121. ———. 2011. ‘Fulbe in wechselnden Nachbarschaften in der Breite des afrikanischen Kontinents’, in Nikolaus Schareika, Eva Spies and Pierre-Yves LeMeur (eds), Auf dem Boden der Tatsachen: Festschrift für Thomas Bierschenk. Cologne: Köppe, pp. 163–84. ———. (ed.). 2012. Pastoralism in Interaction with Other Forms of Land Use in the Blue Nile Area of Sudan: Project Outline and Fieldnotes 2009–10. Halle (Saale): Max Planck Institute for Social Anthropology. https://www.eth.mpg.de/3513653/FN_Vol01_Past​ oralis_Interaction_web.pdf, accessed 12 January 2017. ———. 2013a. ‘Fulɓe Pastoralists in Eastern Sudan and Western Ethiopia: A Documentation’, http://www.eth.mpg.de/subsites/schlee_tagebuch_02/index.html, accessed 4 August 2015. ———. 2013b. ‘Collective Identification in Cities: Reflections on City Scale and Group Size’, Presentation at the Plenary Session of the Xth Congress of Russian Ethnographers and Anthropologists ‘Modern Cities and Social and Cultural Modernization of Russia’. Moscow, 2 July 2013. https://www.eth.mpg.de/3568333/2013_07_02_collective-identification_ en_web.pdf, accessed 5 August 2015. ———. 2013c. ‘Why States Still Destroy Pastoralism and How They Can Learn that in their Own Interest They Should Not’, Nomadic Peoples 17(2): 6–19. Schlee, Günther, Youssouf Diallo and Martine Guichard. 1999: ‘Abschlussbericht zu dem Projekt “Ethnizitäten im Vergleich: Abgrenzungs- und Assimilierungsprozesse bei den FulBe in der Elfenbeinküste, Burkina Faso, Benin und Kamerun” (Schl 186-8/1)’, unpublished report. Schlee, Günther, and Martine Guichard. 2007. ‘Fulɓe und Usbeken im Vergleich’, in Max Planck Institute for Social Anthropology. Department I: Integration and Conflict. Report

226  References 2007. Halle (Saale): Max Planck Institute for Social Anthropology, pp. 11–53. https:// www.eth.mpg.de/3035244/Abt1-Report-2007.pdf, accessed 12 January 2017. ———. 2013. ‘Fulɓe and Uzbeks Compared’, in Peter Finke and Günther Schlee (eds), CASCA – Centre for Anthropological Studies on Central Asia: Framing the Research, Initial Projects. Halle (Saale): Max Planck Institute for Social Anthropology, pp. 25–62. https://www.eth.mpg.de/3374775/FN_Vol06_CASCA_web.pdf, accessed 12 January 2017. Schlee, Günther, and Karin Werner (eds). 1996. Inklusion und Exklusion. Die Dynamik von Grenzziehungen im Spannungsfeld von Markt, Staat und Ethnizität. Cologne: Köppe. Schultz, Emily A. 1984. ‘From Pagan to Pullo: Ethnic Identity Change in Northern Cameroon’, Africa 54(1): 46–64. Schwartz, H. Jürgen. 2005. ‘Ecological and Economic Consequences of Reduced Mobility in Pastoral Livestock Production Systems’, in Elliot Fratkin and Eric Abella Roth (eds), As Pastoralists Settle: Social, Health, and Economic Consequences of Pastoral Sedentarization in Marsabit District, Kenya. New York, London: Kluwer Academic Publishers, pp. 69–86. Scott, Michael F., and Brendan Gormley. 1980. ‘The Animal of Friendship (Habbanaae): An Indigenous Model of Sahelian Pastoral Development in Niger’, in David Brokensha, Dennis M. Warren and Oswald Werner (eds), Indigenous Knowledge Systems and Development. Lanham: University Press of America, pp. 93–112. Sheller, Mimi, and J. Urry. 2006. ‘The New Mobilities Paradigm’, Environment and Planning 38: 207–26. Siele, David, Jeremy Swift and Saverio Krätli. 2013. ‘Reaching Pastoralists with Formal Education: A Distance-Learning Strategy for Kenya’, in Andy Catley, Jeremy Lind and Ian Scoones (eds), Pastoralism and Development in Africa: Dynamic Change at the Margins. London: Routledge, pp. 206–14. Smaldone, Joseph P. 1977. Warfare in the Sokoto Caliphate: Historical and Sociological Perspectives. Cambridge: Cambridge University Press. Smith, Michael P., and Luis E. Guarnizo (eds). 1998. Transnationalism from Below. New Brunswick, London: Transaction Publishers. Sökefeld, Martin. 2000. ‘Translokalität und Identität: Das Problem räumlicher Grenzen in der Ethnologie am Beispiel der Stadt Gilgit, Nordpakistan’, Zeitschrift für Ethnologie 124: 51–72. Spittler, Gerd. 1981. Verwaltung in einem afrikanischen Bauernstaat: Das koloniale FranzösischWestafrika 1919–1939. Freiburg: Atlantis Verlag. ———. 1993. Les Touaregs face aux sécheresses et aux famines: Les Kel Ewey de l’Aïr (Niger). Paris: Karthala. Stenning, Derrick J. 1957. ‘Transhumance, Migratory Drift, Migration: Patterns of Pastoral Fulani Nomadism’, The Journal of the Royal Anthropological Institute 87(1): 57–73. ———. 1959. Savannah Nomads: A Study of the WoDaaBe Pastoral Fulani of Western Bornu Province, Northern Region, Nigeria. Oxford: Oxford University Press. ———. 1966. ‘Cattle Values and Islamic Values in a Pastoral Population’, in Ioan M. Lewis (ed.), Islam in Tropical Africa. Oxford: Oxford University Press, pp. 387–400. ———. 1988[1965]. ‘The Pastoral Fulani of Northern Nigeria’, in James L. Gibbs (ed.), Peoples of Africa. Prospect Heights. Illinois: Waveland Press, pp. 361–401. Sutter, John W. 1990. ‘Commercial Strategies, Drought and Monetary Pressure: Wo’daa’be Nomads of Tanout Arrondissement, Niger’, in Carl Salzmann and John G. Galaty (eds), Nomads in a Changing World. Naples: Istituto Universitario Orientale, Dipartimento di Studi Asiatici, pp. 339–82.

References  227 Swift, Jeremy. (ed.). 1984. Pastoral Development in Central Niger: Report of the Niger Range and Livestock Project. Niamey: Republic of the Niger, Ministère du Développement Rural, USAID. Tamari, Tal. 2006. ‘Joking Pacts in Sudanic West Africa: A Political and Historical Perspective’, Zeitschrift für Ethnologie 131: 215–43. Tarrius, Alain. 1994. ‘Zirkulationsterritorien von Migranten und städtische Räume’, in Mirjana Morokvasic and Hedwig Rudolph (eds), Wanderungsraum Europa: Menschen und Grenzen in Bewegung. Berlin: Sigma, pp. 113–32. Thébaud, Brigitte. 1990. ‘Politique d’hydraulique pastorale et gestion de l’espace au Sahel’, Cahiers des Sciences Humaines 26(1–2): 13–31. Tilley, Christopher. 1994. A Phenomenology of Landscape: Places, Paths and Monuments. Oxford, Providence: Berg Publishers. Tonkin, Elizabeth. 1998. Narrating our Pasts: The Social Construction of Oral History. Cambridge: Cambridge University Press. Turton, David. 2011. ‘Wilderness, Wasteland or Home? Three Ways of Imagining the Lower Omo Valley’, Journal of Eastern African Studies 5(1): 158–76. Tylor, Edward B. 1889. ‘On a Method of Investigating the Development of Institutions; Applied to Laws of Marriage and Descent’, The Journal of the Anthropological Institute of Great Britain and Ireland 18: 245–72. United Nations Development Programme (UNDP). 2014. Sustaining Human Progress: Reducing Vulnerabilities and Building Resilience. Human Development Report 2014, Summary. New York: United Nations. van Santen, José. 2000. ‘“Garder du bétail, c’est aussi un travail”. Les relations entre pasteurs peuls et agriculteurs du centre du Bénin et du Nord-Cameroun’, in Youssouf Diallo and Günther Schlee (eds), L’ethnicité peule dans des contextes nouveaux: La dynamique des frontières. Paris: Karthala, pp. 129–59. Vansina, Jan. 1980. ‘Memory and Oral Tradition’, in Joseph C. Miller (ed.), The African Past Speaks: Essays on Oral Tradition and History. Folkestone: Dawson, pp. 262–79. VerEecke, Catherine. 1993. ‘Sub-National Fulbe Identity in Nigeria? Responses to Political Change in Post-Independence Times’, in Paul K. Eguchi and Victor Azarya (eds), Unity and Diversity of a People: The Search for Fulbe Identity. Osaka: National Museum of Ethnology, pp. 139–61. Vincent, Joan. 1974. ‘The Structuring of Ethnicity’, Human Organization 33(4): 375–79. Waibel, Gabi. 1998. Seßhaftwerdung und sozialer Wandel bei den Tuareg Zinders (Niger). Hamburg: Institut für Afrika-Kunde. Watson, Elizabeth E. 2010. ‘A “Hardening of Lines”: Landscape, Religion and Identity in Northern Kenya’, Journal of Eastern African Studies 4(2): 201–20. Werbner, Pnina. 1997. ‘Introduction: The Dialectics of Cultural Hybridity’, in Pnina Werbner and Tariq Modood (eds), Debating Cultural Hybridity. Multi-Cultural Identities and the Politics of Anti-Racism. London, New Jersey: Zed Books, pp. 1–26. _____. 1999. ‘Global Pathways. Working Class Cosmopolitans and the Creation of Transnational Ethnic Worlds’, Social Anthropology 7(1): 17–35. Werthmann, Katja. 1997. Nachbarinnen: Die Alltagswelt Muslimischer Frauen in einer Nigerianischen Großstadt. Frankfurt (Main): Brandes & Apsel. White, Cynthia. 1987. ‘Changing Animal Ownership and Access to Land among the WoDaaBe (Fulani) of Central Niger’, in P.T.W. Baxter and R. Hogg (eds), Property, Poverty and People: Changing Rights in Property and Problems of Pastoral Development. Manchester:

228  References Department of Social Anthropology and International Development Centre, University of Manchester, pp. 240–74. Wilson-Fall, Wendy. 2000. ‘Conflict Prevention and Resolution Among the Fulbe’, in I. William Zartmann (ed.), Traditional Cures for Modern Conflicts: African Conflict ‘Medicine’. Boulder, CO: Lynne Rienner Publishers, pp. 49–65. Wood, John C. 2009. ‘Roads to Nowhere: Nomadic Understandings of Space and Ethnicity’, in Günther Schlee and Elizabeth E. Watson (eds), Changing Identifications and Alliances in North-East Africa. Volume I: Ethiopia and Kenya. New York, Oxford: Berghahn, pp. 225–40. Woodburn, J.C. 1972. ‘Ecology, Nomadic Movement and the Composition of the Local Group among Hunters and Gatherers: An East African Example and Its Implications’, in Peter J. Ucko, Ruth Tringham and G.W. Dimbleby (eds), Man, Settlement and Urbanism. Cambridge: Schenkman, pp. 193–206. Younfa, Alhassane, and Les équipes des programmes ZFD (‘Ziviler Friedensdienst’) et LUCOP (‘Lutte contre la Pauvreté’). 2010. Le processus d’élaboration du SAF au Niger: Capitalisation des expériences du DED Niger. Niamey: Gesellschaft für Internationale Zusammenarbeit. ZFD (Ziviler Friedensdienst). 2008. L’élevage mobile dans les régions de Zinder et de Diffa – Niger: Petit Atlas analytique et synthétique. Niamey: Deutscher Entwicklungsdienst. ———. 2011. Petit guide pratique de conseils aux éleveurs mobiles et à ceux qui défendent leurs intérêts: Quelques conseils concernant les pratiques de l’élevage mobile, le Code Rural et la décentralisation (version 5). Niamey: Gesellschaft für Internationale Zusammenarbeit.

Index

Index

A Abuja, 123, 126 Ader, 6, 21, 25, 32, 35ff., 43ff., 50, 79, 131, 143 administrative chiefs. See customary authorities Agadez (region), 21, 26, 35ff., 43ff., 56, 79, 135 agriculture, 3, 31, 49, 55ff., 59f., 77, 81, 85ff., 99, 101, 108, 124, 126, 129, 153 expansion of, 32, 49, 85 agro-pastoralism, 1, 6, 31, 60, 63, 66, 74, 77ff., 81, 103, 155 aid distribution, 7, 39, 90f., 109ff., 111, 116, 123, 155, 206 al buhari, 39 Alijam, 21, 25, 27, 72, 79, 107 animal theft, 88 Arabs, 56f., 70ff., 74, 108, 134 arɗo, 5, 32, 34, 47, 50, 51n7, 111, 113 assimilation, 60f., 64, 73, 74, 147, 166 associations. See civil society associations B banga-banga, 39, 91 Barth, Fredrik, 7, 64f., 67, 138 biggal, 82ff., 118 Ɓii Ute’en, 26, 27n1, 27n2, 38, 47, 51, 70, 72, 107f., 141f., 195

bilinguality, 147 Bilma, 56 biographical approach, 15f. Birni N’Konni, 31, 32 borders, 9, 39, 50, 102f., 123, 139, 206 Borno, 25, 27n7, 60 Boultoum, 56 boundaries ethnic/social, 7f., 13, 17, 26, 60ff., 64ff., 134, 138, 144ff., 187f., 201 spatial, 57f., 102 Bouza, 32f., 55 C cattle corridors, 102 centre. See villageization Chadawanka, 32, 35 chiefs. See customary authorities civil society associations, 70, 112ff., 115 clan clusters, 21, 23ff., 27n1, 69, 72, 107, 176 hierarchies, 48, 101 structure, 17, 22, 25, 31 clustering, urban, 139ff. Code Pastoral. See pastoral legislation Code Rural. See land tenure law coexistence (of ethnic groups), 55ff., 60, 62 collective identification. See identification colonial conquest, 31f., 50n4

230  Index colonial politics, 32, 34f., 87, 92 colonizers, interaction with, 28, 31ff., 41f., 205 commissions foncières. See land tenure commissions communication, 61, 143, 146, 151, 160ff., 167 technologies, 160ff., 167, 169n9, 191, 196, 201, 205, 207 community home, 147, 150ff., 164ff., 191, 194, 205 lineage/clan, 104ff., 111f., 134, 140f., 145, 148, 163, 173ff., 185ff., 192 local, 77ff., 92, 94, 96, 102ff., 116, 118, 163, 192 translocal, 92, 104, 155, 160ff., 167f., 173ff., 190, 196 urban, 125, 138ff., 145ff., 151f., 162, 166ff., 200 competition between and within kinship groups, 104, 104, 112ff., 116, 136, 164, 176, 182ff., 187f., 193 political, 110, 116 about resources, 38, 49, 96, 100f., 130 complementarity, 6, 18, 23, 56f. 62, 100ff., 125, 155, 157, 168 conflict, 12, 55ff., 60ff., 70f., 85ff., 89, 100ff., 110, 144, 174, 184, 189n5, 197, 199, 202, 206 conflict resolution, 60f., 70f., 86, 88, 197 co-residence, 7, 24, 27, 73, 78, 104f., 203 crop damages, 55, 60, 62n1, 82, 85f., 100ff. cultural change, 8, 15, 99, 118ff., 147, 164ff., 190ff., 204 cultural continuity, 8, 18, 120, 163, 168, 190ff., 202, 204f. cultural reproduction, 1, 8, 138, 190f. customary authorities and aid/development, 109, 154 allocation of land by, 87f., 95, 108 appointment of, 34f., 50n3, 50n6, 51n10, 110f. functions of, 34, 50n7, 86, 95, 116 multiplication of offices, 110, 113 relations with, 47, 88, 108f. sedentary vs. mobile, 34, 95, 108 customary law, 70, 96, 197f., 201

D daangol pulaaku, 69 daɗɗo, 105, 194 Damagaram Takkaya, 39, 79, 83, 86, 153, 158 Damergou, 2, 6, 14, 21, 25ff., 36f., 44ff., 51n10, 56, 69, 77, 79, 84, 86, 88, 97, 106f., 110, 112, 115, 130, 142, 152, 156ff., 161f., 191ff. date de libération des champs, 57 Degereewol, 21, 25, 27n1, 72, 107 desertification, 85, 93, 116 deterritorialization, 10, 190 development projects, 90ff., 104, 109f., 114f., 117, 134, 136, 140, 153ff., 158, 164 Diffa, 1f., 6, 27n2, 70, 114, 128, 131, 141ff., 156, 160f., 195f., 206 dispersal of social groups contemporary factors for, 13, 92, 156, 162f., 167f., 190ff., 204 due to migrations, 10, 25ff., 31, 103, 173, 203 seasonal, 11, 32f., 93, 151, 169n3, 173, 203 divorce, 174 drought, 5, 10, 18, 21, 29, 37, 39, 43ff., 59, 79, 90f., 117, 123f., 135, 167 duɗal, 79 E elopement marriage. See marriage endogamy, 5, 144, 173ff., 200 environmental degradation, 116ff. ethnic identity, 5, 64ff., 107, 143f., 185f., 193f. ethnic markers, 67, 147 ethnicity, 13, 60f., 63ff., 70ff., 138f., 148, 185f., 204 Evans-Pritchard, Edward, 23 exchange economic, 10, 56ff., 62, 155, 168 marital, 5, 24, 174ff., 184f., 188 reciprocal, 58f., 132f., 146, 178ff., 184ff. ritual, 163, 177ff., 193 social, 17, 56ff., 62, 72, 100, 138, 146, 160, 163ff., 188 exclusion, 41, 49, 64, 68, 87, 90, 97, 109, 118, 130, 175, 194, 206

Index  231 ‘exit’ and ‘voice’, 40ff., 115. See also Hirschman, Albert exogamy, 5, 174f. expatriates, 1, 5, 10, 128ff., 135ff., 139f., 152, 155, 164, 193, 206f. F farmer–herder conflicts, 32, 55, 57, 62n1, 79, 85f., 101 farming. See agriculture firstcomer principle, 47ff., 51n16, 87, 135f. flexibility, 7, 57, 66, 73n6, 95, 119f., 123, 130, 166f., 169n4, 207 food sharing, 145f. 177ff. (ritual) fragmentation of lineage groups, 13, 23, 31f., 104, 111, 203 political, 35, 110f., 112ff. friendship, 42, 58, 61, 132ff., 146, 160 frontier, 40 Fulɓe, 1, 3, 5, 8, 17, 24ff., 29ff., 35, 40, 42, 50n7, 51n11, 56ff., 60f., 63, 64ff., 70ff., 80, 86f. 112, 114, 119, 126, 130, 134, 139, 144, 148, 160, 183, 188, 189n5, 204 Fulfulde, xi, 65, 67f., 72, 93, 120n11 G Ganatcha, 14, 28, 39, 47, 77, 78ff., 84ff., 94, 97, 102ff., 115, 118f., 134ff., 153ff., 158ff., 191ff. gassungol Woɗaaɓe, 18, 183, 185 geerewol, 176f. globalization, 8, 10, 13, 57, 190, 204 Godelier, Maurice, 185f. Gojanko’en, 5, 6, 14, 21, 26, 27n1 27n2, 30ff., 44, 50n6, 69, 72, 77, 78, 91ff., 103ff., 107, 112f., 125, 129ff., 139ff., 148n2, 191ff., 195, 200f., 206 group relations inter-ethnic, 55ff., 17, 55ff., 58ff., 62f., 64ff., 70ff., 79, 86, 100f., 138f., 144ff., 189f. intra-ethnic, 4, 5, 18, 23ff., 27n1, 48, 104, 139ff., 173ff., 176ff., 182ff., 185ff., 189f., 191ff. group size local, 116, 193

pastoral, 24 urban migrants’, 139ff., 146, 152, 200 H Haaɓe, 17, 64f., 67, 69, 71, 73n2, 74n9, 133f., 139, 148, 150 haɓɓanaaye. See loan animal exchange Hamani, Diori, 42 Hardin, Garrett, 116 Hausa ethnic group, 29, 56f., 64, 71, 72, 108, 127, 145, 148 language, xi, 93, 146f. Herskovits, Melville, 120n13 Hirschman, Albert, 40, 115 history, 3, 8, 15, 16f., 24, 26, 28, 35, 48, 56, 65f., 96, 135, 143 of migrations, 6, 27, 28ff., 44f. oral, 28, 48, 50n5, 50n6, 136 home range (geographical attachment), 26, 38, 43ff., 98, 106, 173 I identification, 8, 13, 17f., 60, 105, 143f., 146f., 150f., 155, 157, 165f., 169 categories, 7, 69, 72, 134, 143, 147, 187 collective, 1, 7, 13, 143, 188, 60 65, 67ff., 143, 187f., 192, 202ff. dimensions of, 17f., 69, 71ff., 105, 147, 188 situational changing of, 17, 69, 71, 205 identity change, 74n7 Ingal, 35f., 46 inheritance, 24, 39, 110, 119, 124, 176, 199 insecurity, 31, 88, 133, 141, 190, 206 integration among social groups, 17, 24, 26, 55ff., 58ff., 69, 72f., 100, 102, 134, 144ff., 166, 187f., 202 into state structures and society, 1, 7, 39ff., 72, 90ff., 99, 109, 114f., 168f., 205f. inter-ethnic relations. See group relations interstitiality, 40f., 130 Islam. See religion J Jiijiiru, 27n1, 48, 72, 107f., 129, 140, 143, 195ff. joking relationships, 60f., 139, 188

232  Index K Kanem Empire, 56 Kano, 2, 6, 123, 126, 128, 139, 142, 152, 157, 195f., 200f., 206f. Kanuri, 38f., 56f., 60ff., 64, 71f., 81, 86, 101, 108f., 139, 148, 153, 158f., 188 kinship network(s) (see networks) as a principal of social organization, 16, 26, 61, 72, 111, 147ff., 176ff. relationships, 5, 24, 26, 88, 144, 168, 176ff. kooɓgal. See marriage Kopytoff, Igor, 40, 47 Koutous (region), 2, 6, 12, 14, 21, 26, 28, 39, 44, 62, 77ff., 84ff., 97, 106, 162, 191ff. settlement history, 56f. Kuskudu, 21f., 26, 32ff., 36ff., 69, 72, 77, 78, 84, 92, , 94, 97, 103, 107f., 110, 112, 129f. 134f., 139ff., 161ff., 192 L laamiiɗo (politico-administrative office), 34, 50n7, 107 land appropriation, 12, 49f., 77, 87, 90, 97, 99ff., 120, 130, 204 land rights, 79, 95, 97, 98n12, 119f. land tenure commissions, 97, 98n12 land tenure law, 96, 98n12, 102 land use planning, 102 legal pluralism, 201f. lineage(s), 4, 21f., 23ff., 27n1, 27n5, 32ff., 41, 44, 47, 69, 72, 78f., 88f., 104f., 131, 136, 139, 140ff., 155, 162f., 174f., 177f., 187, 189n3 communities, 47, 104, 106, 134, 145, 148, 196 segmentation, 24, 103ff., 106f., 110f., 192, 203 system, 23, 69f. livelihood diversification, 1, 6f., 18, 31, 59, 62, 66, 77, 81, 99, 119, 123f., 168, 204 loan animal exchange (haɓɓanaaye), 25, 49, 51n15, 109, 119, 175, 204 locality, 8f., 12, 14, 62, 91f., 94, 103, 105, 111, 138f., 147, 153, 166, 190

localized communities, 77, 79, 94, 96, 103ff., 116, 118, 155, 163, 192, 201 M Maiduguri, 141 manure contracts, 58f., 101 marginality, 41, 72, 90, 93, 113f., 127, 130, 206 markets, 56, 58, 62, 69, 81, 88, 153, 155, 157ff. marriage, 46, 88, 98n6, 191 by betrothal (kooɓgal), 5, 46, 105, 173ff., 189n5, 195, 199f. changing practices of, 195ff., 198ff. by elopement (te’egal), 5, 71, 72, 104f., 174ff., 191, 194, 195ff., 202n1 patterns of, 61, 65, 105, 148, 174, 197 rituals of, 72, 146, 177ff. Mauss, Marcel, 151, 178, 181, 184ff. mboɗangaaku, 65, 68, 72, 73n4, 183f. Mbororo’en (Weeweɓɓe), 66 Mbororo, 63, 66, 73, 73n5, 74n7 Mbuuldi, 21f., 27n3, 32ff., 37, 41, 69, 72, 94, 112, 192 medicine trade, 126, 133, 144, 151 migration cultures of, 10, 123, 166 ecological aspects of, 11, 21, 37ff., 43, 84ff. historic, 3, 6ff., 14, 21, 28, 29ff., 35ff., 44, 143, 173 pastoral, 7, 11, 26, 42f., 96, 105, 203 rural–urban, 1, 5ff., 13, 77, 106, 123ff., 138ff., 149ff., 152ff., 157ff., 162ff., 167ff., 190, 204 socio-political aspects of, 17, 26, 44, 58, 84ff., 107ff., 136, 138ff., 149ff., 152ff., 157ff., 162ff., 167ff., 192, 196, 205 mobile phones. See communication technologies mobile space, 12, 43, 93, 151, 186 mobility, 7ff., 10ff., 13ff., 28ff., 39ff., 42ff., 55, 80ff., 84, 88f., 90ff., 95, 99, 106f., 107, 144, 154, 156, 160, 167, 190ff., 194, 202n2, 203ff. circulatory, 167f. intra-urban, 130 modern, 1, 6, 8f., 138, 149f., 156, 157ff., 162, 164f., 168, 196f., 201, 205ff.

Index  233 pastoral, 1, 3, 11f., 15, 24ff., 30ff., 59, 67, 77ff., 80f., 86, 88, 90, 97, 106, 116ff., 130, 151, 159, 164, 168, 205 rural–urban, 7, 152ff., 157ff., 164f., 169 monetization, 31, 58, 101, 200 multi-sited fieldwork, 13 N nation state, 1, 9, 25, 201, 205 Ndoovi’en, 65ff., 73n3 neighbourhood, 8f., 11, 26, 60, 67, 85f., 103, 138ff., 144ff., 152f., 192, 200 as a factor favouring integration, 62, 72, 86, 145f. network(s), social, 4f., 10, 12, 18, 27, 38, 45ff., 51n15, 106, 109, 124, 128, 131f., 138, 154f., 157ff., 164, 168, 175, 191, 193, 203f., 205 between clans, 11, 18, 72, 171ff., 185ff., 204 based on friendship, 2, 134f., 158f. based on kinship, 2, 10f., 125, 131f., 139ff., 175, 196 translocal (see translocal networks) ngaanka, 5, 18, 25, 67, 69, 72, 105, 173ff., 177, 180ff., 185ff., 187ff., 191ff. Ngel Tireeji, 38, 47, 51n11, 77, 84, 108, 134, 161, 163 Niamey, 91, 128, 131, 136, 142f. 148n2, 153 Nigeria, 2, 3, 6, 29, 37, 39, 56, 79, 89n6, 114, 123, 126f., 139, 141, 145, 195f. 206 Njapto’en, 25, 26, 27n1, 27n2 nomadism, 1, 3, 6, 10, 11, 15, 31, 35, 39, 63, 66, 77ff., 80f. 82, 88, 90f., 92f. , 95, 117, 118ff., 149, 161, 164, 203ff. northern limit of cultivation, 78, 85 O oral history. See history overgrazing, 116, 118, 173 P participation political, 15, 113 social, 17, 144f. pastoral legislation, 57, 85, 96, 117 pastoral reserves, 79, 102, 118

pastoral wells. See wells pastoralism, 3, 6, 17, 24, 35f. 49, 57, 77ff., 82ff., 91, 93, 96, 116, 119, 124, 155, 160, 165, 203 techniques of, 82ff., 117 pastoralists’ associations, 70, 112, 117 patron-client relations, 132 placemaking, 7f., 11f., 17, 43f. 49f., 51n11, 56, 99ff., 103f. 130, 150, 164, 203 political affiliation, 38, 107ff. changes in, 107, 110 and ethnicity, 107ff. and local proximity, 108f. polytaxis, 68, 147, 148n5 public transport, 1, 45, 88, 157ff., 167, 196, 201, 205 pulaaku, 64ff., 69f., 72, 73n1, 73n4, 74n10, 114, 183 R reciprocity, 5, 16, 58f., 132f., 146, 176ff., 184ff., 188 religion Christian, 120n12 Islamic, 17, 29, 41, 70ff., 74n7, 173, 190, 191, 194, 199f. 206 remittances, 124, 132, 136, 154ff. return migration, 21, 37, 151, 165f., 169n7 Rinderpest epizootic (1890s), 31, 50n4 ritual, 5, 26, 57, 67, 72, 105, 145, 163, 173f., 177ff., 182ff., 186, 187f., 189n3, 189n5, 191ff. rural–urban migration. See migration S Salaga, 14, 38f 77, 84ff., 93f., 97, 101, 108, 112, 131, 134, 156ff., 161, 163, 165, 169n7 school(s), 41, 42, 80, 90ff., 99f., 104, 109, 112, 116, 128f., 132, 146, 147, 151ff. 163ff., 168 sedentarization, 1, 7, 12, 21, 31, 58, 66f., 77ff., 90, 92, 99, 101, 103, 116ff., 118ff., 164, 190, 192, 204ff. segmentary opposition, 23, 27n5, 69, 112, 143, 184f., 187f. seniority principle, 34, 47f., 66, 101 sharing systems of land use, 49, 57f. 62, 101 situational identification. See identification

234  Index social cohesion, 103f., 175, 184 social group formation, 7, 13, 17f., 25, 26, 31, 44, 50, 69, 71, 103f., 114, 192, 203 social mobility, 7, 137 social reproduction, 1, 8, 18, 203, 160, 173f., 186 social space, 11f., 17f., 27, 43, 50, 55, 58, 62, 63ff., 72f. 132, 186ff., 196, 203f. Sokoto Empire, 24, 29ff., 40, 44, 50n6, 50n7 Jihad, 3, 24, 29f. solidarity, 23, 60, 69, 106f., 112, 136, 142, 148, 179, 189n5 soro, 67 state evasion, 32, 39ff., 90, 92, 113, 127, 144, 205 stereotypes, 60, 71, 126, 133f., 188 stranger–host relations, 58f., 158f. Suudu Suka’el, 27n1, 48, 114, 129, 142, 143, 158 switching of identifications. See identification T Tahoua, 6, 26, 142 Tanout, 37, 79, 88, 109, 142, 153, 157, 159, 161, 196 taxes, 29, 31, 33ff., 40, 41, 62, 108f., 116, 128, 206 Tchintabaraden, 26, 27n2, 195f. te’egal. See marriage Termit, 56 territorialization, 7, 49f., 57f., 77, 99ff., 103ff., 204, 206 terroir d’attache, 96f., 102, 118 Tesker, 36ff., 39, 49, 56, 79, 88 tigu. See medicine trade tourism, 133f., 193 traditional chiefs. See customary authorities tragedy of the commons, 116, 118 transhumance, 42, 45, 47, 51n13, 58, 82, 83, 86, 206 translocal networks, 2, 11f., 38, 45ff., 27, 46, 49, 106, 132, 149f., 152ff., 167f., 185ff., 191, 193, 196, 203, 204ff.

translocality, 6, 8ff., 11f., 13f., 18, 27, 28, 46, 92, 104, 131, 149f., 152ff., 157ff., 160ff., 167f., 173, 190ff., 197, 203ff. transnationalism, 9f., 132, 139, 141, 204, 207 Tuareg, 32, 35f., 56f., 65, 70, 72, 73n2, 86ff., 94, 98n8, 101, 108, 113, 120n2, 127, 133f., 137n2, 137n3, 148, 169n8 Tubu, 49, 56f., 73n2 U Usman Dan Fodio, 24, 29, 50n6 V villageization, 38, 77, 90ff., 100, 103f., 111, 120, 194 and materialization of infrastructure, 103, 164f. and political offices, 110ff. and the role of aid and development projects, 90ff., 164f. virilocality, 164 W warfare, 29f. watchmen, 1f. 5, 125ff., 130ff., 135, 139ff., 145, 152, 155, 163, 165, 206 wayfaring, 11, 43 wells, 7, 21, 35ff., 39, 43, 48, 70f., 74n12, 77, 79, 83, 86ff., 91, 96f., 99, 101, 103, 106, 109f., 119, 126, 158 acquisition of, 38f., 43, 47f., 79, 86, 96f., 103f., 111, 119, 135, 155 Woɗande, 65, 68 worso, 151, 173f., 175, 177f., 186, 191ff. wuumre, 79, 105 Y Yaamanko’en, 27n1, 27n2, 79, 87, 104f., 192 Z Zinder (city), 2, 6, 14, 39, 79, 91, 115, 123ff., 128ff., 130ff., 135, 139ff., 145f., 151ff., 157ff., 162f., 165, 167, 195f., 200, 206f. Zinder (province), 2, 5f., 21, 91, 94, 103, 108, 126