Informal Institutions and Citizenship in Rural Africa: Risk and Reciprocity in Ghana and Cote d'Ivoire (Cambridge Studies in Comparative Politics)

  • 41 412 4
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

Informal Institutions and Citizenship in Rural Africa: Risk and Reciprocity in Ghana and Cote d'Ivoire (Cambridge Studies in Comparative Politics)

Informal Institutions and Citizenship in Rural Africa Risk and Reciprocity in Ghana and Côte d’Ivoire This book challen

1,411 452 2MB

Pages 312 Page size 329.999 x 499.999 pts Year 2010

Report DMCA / Copyright

DOWNLOAD FILE

Recommend Papers

File loading please wait...
Citation preview

Informal Institutions and Citizenship in Rural Africa Risk and Reciprocity in Ghana and Côte d’Ivoire

This book challenges previous assumptions about institutions, social capital, and the nature of the African state by investigating the history of political and economic change in villages on either side of the Ghana– Côte d’Ivoire border. Prior to European colonial rule, these Akan villages had very similar political and cultural institutions. By the late 1990s, however, Lauren M. MacLean found puzzling differences in the informal institutions of reciprocity and indigenous notions of citizenship. Drawing on extensive village-based fieldwork and archival research, MacLean argues that divergent histories of state formation not only shape how villagers help each other but also influence how local groups and communities define citizenship and then choose to engage with the state on an everyday basis. She examines the historical construction of the state role in mediating risk at the local level across three policy areas: political administration, social service delivery, and agriculture, highlighting the importance of the colonial and postcolonial state in transforming informal institutions. Lauren M. MacLean is an assistant professor of political science at Indiana University. She earned her Ph.D. in 2002 from the Department of Political Science at the University of California at Berkeley and then completed a Robert Wood Johnson Foundation Scholars in Health Policy postdoctoral fellowship at the University of Michigan (2002–2004). Her work has been published in Comparative Studies in Society and History, the International Journal of Public Administration, the Journal of Modern African Studies, and Studies in Comparative International Development.

To Jason, Jasper, Skylar, and Benjamin

CAMBRIDGE STUDIES IN COMPARATIVE POLITICS

General Editor Margaret Levi

University of Washington, Seattle

Assistant General Editors Kathleen Thelen Massachusetts Institute of Technology Erik Wibbels Duke University Associate Editors Robert H. Bates Harvard University Stephen Hanson University of Washington, Seattle Torben Iversen Harvard University Stathis Kalyvas Yale University Peter Lange Duke University Helen Milner Princeton University Frances Rosenbluth Yale University Susan Stokes Yale University Other Books in the Series David Austen-Smith, Jeffrey A. Frieden, Miriam A. Golden, Karl Ove Moene, and Adam Przeworski, eds., Selected Works of Michael Wallerstein: The Political Economy of Inequality, Unions, and Social Democracy Andy Baker, The Market and the Masses in Latin America: Policy Reform and Consumption in Liberalizing Economics Lisa Baldez, Why Women Protest: Women’s Movements in Chile Stefano Bartolini, The Political Mobilization of the European Left, 1860--1980: The Class Cleavage Robert Bates, When Things Fell Apart: State Failure in Late-Century Africa Mark Beissinger, Nationalist Mobilization and the Collapse of the Soviet State Nancy Bermeo, ed., Unemployment in the New Europe Carles Boix, Democracy and Redistribution Carles Boix, Political Parties, Growth, and Equality: Conservative and Social Democratic Economic Strategies in the World Economy Catherine Boone, Merchant Capital and the Roots of State Power in Senegal, 1930--1985 Catherine Boone, Political Topographies of the African State: Territorial Authority and Institutional Change Continued after the Index

Informal Institutions and Citizenship in Rural Africa Risk and Reciprocity in Ghana and Côte d’Ivoire

LAUREN M. MACLEAN Indiana University

cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Dubai, Tokyo Cambridge University Press 32 Avenue of the Americas, New York, NY 10013–2473, USA www.cambridge.org Information on this title: www.cambridge.org/9780521192965 © Lauren M. MacLean 2010 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2010 Printed in the United States of America A catalog record for this publication is available from the British Library. Library of Congress Cataloging in Publication Data MacLean, Lauren M. Informal institutions and citizenship in rural Africa : risk and reciprocity in Ghana and Cote d’Ivoire / Lauren M. MacLean. p. cm. – (Cambridge studies in comparative politics) Includes bibliographical references and index. 1. Ghana – Rural conditions. 2. Villages – Ghana. 3. Borderlands – Ghana. 4. Cote d’Ivoire – Rural conditions. 5. Villages – Cote d’Ivoire. 6. Borderlands – Cote d’Ivoire. I. Title. II. Series HN832.A8M33 2010 307.76′209667–dc22 2009047385 ISBN

978-0-521-19296-5 Hardback

Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party Internet Web sites referred to in this publication and does not guarantee that any content on such Web sites is, or will remain, accurate or appropriate.

Contents

List of Tables and Figures List of Abbreviations Acknowledgments

page ix xi xiii

p a r t i : t h e t r a ns f o r m a t i o n o f i n f o rm a l i n s t i t u t i o n s o f s o c i a l r e ci p r o c i t y i n g h a n a a n d c ôt e d ’ i v o i r e 1. Introduction 2. The Informal Institutions of Reciprocity: A Quantitative Puzzle and Analysis 3. Local Conflicts Over the Meaning of Reciprocity: A Qualitative Analysis of Change

3 40 65

p a r t i i : l e g a c i e s o f t h e st a t e r o l e i n m e d i a t i n g r i s k i n g h a n a a n d cô t e d ’ i v o i r e 4. The Legacies of the Colonial Administrative State in Constructing the Citizen, Family, and Community Roles

99

5. The Construction and Retrenchment of State Social Service Provision and the Unintended Consequences for Reciprocity

120

6. The Empire of the Young: Contrasting Legacies of State Agricultural Policy for Local Capitalism and Reciprocity

164

vii

viii

Contents

p a r t i i i : i n f o r m a l i n s t i t u t i o n s o f r e c i p r o c i ty a n d t h e pr o s p e ct s f o r de m o c r a t i c c i t i z e n s h i p 7. Transformations of the Informal Institutions of Reciprocity and the Implications for Citizenship

199

8. Conclusion

227

Appendix

247

Bibliography Index

263 287

Tables and Figures

tables 1.1. How the History of State Formation in the Region Stimulates Divergence in Informal Reciprocity page 8 1.2. Similar Country Cases: A Comparison of Basic Indicators in the Late 1990s 33 3.1. First-Mentioned Source of Help If Respondent Was Sick and in Need 84 4.1. Divergent Normative Frameworks of British and French Colonial Administrative Rule 100 5.1. Comparison of National-Level Health Expenditures and Infrastructure in Ghana and Côte d’Ivoire 132 5.2. Comparison of National-Level Education Expenditures and Infrastructure in Ghana and Côte d’Ivoire 133 5.3. Comparison of Social Indicators Over Time in Ghana and Côte d’Ivoire 146 6.1. Annual Cocoa Production in Tons in Ghana and Côte d’Ivoire 173 6.2. Comparison of Help Given from Cocoa versus Tomato Farmers 180 7.1. Comparison of Local Conceptualization of State 204 7.2. Comparison of Local Conceptualization of Citizens’ Rights 207 7.3. Comparison of Local Conceptualization of Citizens’ Duties 209 7.4. Perceptions of Most Important Individual or Group Leader for Village Development and Policymaking 214 A.1. Comparative Analysis of Country, Village, and Interviewer 257 Effects on the Total Value of Help Given to All Social Ties A.2. Regression Model Results for Value of Help Given to All Social Ties 258 ix

x

Tables and Figures

A.3. Regression Model Results for Value of Help Given to the Nuclear Family Only A.4. Regression Model Results for Value of Help Given to the Extended Family Only A.5. Regression Model Results for Value of Help Given to All Social Ties Split by Country

259 260 261

figures 1.1. Map of Study Region in Africa 1.2. Map of Fieldsites in Similar Regions of Côte d’Ivoire and Ghana 1.3. The Argument 6.1. Linking Economic Production to Informal Institutions of Reciprocity

4 5 10 179

List of Abbreviations

CAISSTAB CPP GOCI GOG IFAD IMF FPI NDC NGO NPP PAMSCAD PDCI PNDC RDR SAP USAID

Caisse de Stabilisation et de Soutien des Prix des Productions Agricoles Congress People’s Party Government of Côte d’Ivoire Government of Ghana International Fund for Agricultural Development International Monetary Fund Front Populaire Ivoirien National Democratic Congress Nongovernmental Organization New Patriotic Party Programme of Actions to Mitigate the Social Costs of Adjustment Parti Démocratique de la Côte d’Ivoire Provisional National Defense Council Rassemblement pour la Démocratie Républicaine Structural Adjustment Program United States Agency for International Development

xi

Acknowledgments

This book would not have been possible without the intellectual and moral support of many different people in many different parts of the world. Tom Callaghy inspired a passion for both teaching and research as a professor and senior thesis advisor during my undergrad days at Penn and has continued to guide and mentor me at each stage of my career. In my first job after college at World Resources Institute in Washington, DC, Tom Fox, Kirk Talbott, and Peter Viet encouraged me to do and learn as much as I could, including facilitating my first trip to West Africa. Another WRI colleague, Dorm Adzobu, his wife, Annie, and their children welcomed me so warmly and generously to Ghana that it has been difficult to travel anywhere else. During graduate school and beyond, my doctoral committee advisors at Berkeley, David Leonard, Michael Watts, Robert Price, and Kiren Chaudhry, have provided invaluable intellectual support and advice. I also benefited from critical feedback from many others at various stages of the project, including Christopher Ansell, Ruth Collier, Akhil Gupta, Frank Hirtz, Jonah Levy, the late Donald Rothchild, and Richard Roberts. The “Berkeley mafia” of graduate student colleagues continues to thrive through virtual communication, so countless thanks go to Regina Abrami, Dennis Galvan, Markus Goldstein, Ken Greene, Evan Lieberman, Khalid Medani, Aaron Schneider, and Rob Weiner. In particular, I owe thanks to Melani Cammett and Julie Lynch, who provided incisive readings of the manuscript at critical points. A special thank you to Teri Caraway, who provided extensive comments and was my link to an academic community at Northwestern University in Chicago. I also am grateful to Bruce Clayton (Northwestern University) and Antje Schwennicke (Indiana University) for their statistical expertise and support. xiii

xiv

Acknowledgments

This project would not have been possible without the support and efforts of many friends and colleagues in Ghana and Côte d’Ivoire. I am grateful for the support of Joseph Ayee and everyone in the Department of Political Science at the University of Ghana at Legon. I wish to thank E. GyimahBoadi and the Centre for Democracy and Development for their support during my research stay and afterward. In Côte d’Ivoire, I would like to thank Laurent Vidal, Pierre Janin, Virginie Briand, and other colleagues then at IRD’s (formerly ORSTOM) Centre de Petit Bassam for the research affiliation, intellectual guidance, and logistical support. Needless to say, I could not have conducted this research without the invaluable contributions of my research assistants: Kweku Dickson and Faustina Sottie in Ghana, and Fulgence Kanga and Célestin Mian in Côte d’Ivoire. Not only was the quality of the data they collected outstanding, but they also taught me volumes about the local culture and politics during the many meals and late night discussions we shared. Mr. Ernest Appiah deserves special thanks for providing his expertise and managerial support in all aspects of the project in Ghana. The most tremendous debt I owe is to the four village communities who welcomed a total stranger into their midst. Thank you to Nana Amaankwaah and Nana Fosu Ababio II in Ghana as well as Nana Bouadou Dogi in Côte d’Ivoire for approving and making arrangements for our research team. Thank you to our “tuteurs” M. Affouafou N’Doli, M. Bonzou, and M. Assande Kablan as well for all they did to facilitate a welcoming home. I feel so honored to have been included in the lives of so many families during my stay; in particular, I wish to acknowledge the friendship of Augustine Addai-Appiah, Janet Antiwiwaa, Ibrahim Amadu Sissao, Noel Koffi Kouassi Assande, Jacques Ehui Assa, and Thomas Tanoh Kablan. Thank you also to Mr. Joseph Fosu-Oppong; Georgina Darko; the late Agbatou Adiko; the then–deputy mayor of Abengourou, Kouame Amoakon; and the then–sous-préfet, Victor Gnangbi for their genuine concern for my safety and happiness in finding the right villages for the study and for their extensive efforts to make the appropriate introductions. I also want to thank my American families in Africa: the Schefflers in Ghana; and the Belle Isles, Bissets, Rayners, and Sikes in Côte d’Ivoire. Next, I would like to acknowledge the financial support I have received for this project: the National Science Foundation, the Social Science Research Council International Pre-Dissertation Fellowship Program, the Fulbright Hays Doctoral Dissertation Research Abroad Program, the Institute for the Study of World Politics, the University of California at Berkeley African Studies Center, the University of California at Berkeley’s

Acknowledgments

xv

Department of Political Science, the Robert Wood Johnson Foundation, Indiana University’s Department of Political Science, Indiana University’s Workshop on Political Theory and Policy Analysis, as well as Indiana University’s Center on Philanthropy. Many colleagues have provided support as I worked to revise the dissertation and turn it into a book. At the University of Michigan, Ann Lin, my mentor while doing the RWJ post-doctoral fellowship, helped me think through the book’s arguments. At Indiana University, Jeff Isaac read and gave valuable comments on several drafts. Lin Ostrom and Mike McGinnis supported a workshop “book party,” at which I received constructive criticism from colleagues across disciplines including Beth Buggenhagen, Mike Ensley, Sheldon Gellar, John Hanson, Jeff Hart, Les Lenkowsky, Bill Schuerman, Regina Smyth, Bev Stoeltje, Armando Razo, David Reingold, Jean Robinson, Beate Sissenich, and Ann Marie Thomson. Thank you also for the comments and cheerleading from Eileen Braman, Gardner Bovingdon, Kon Dierks, Ilana Gershon, Gina Lambright, Marissa Moorman, and Hillel Soiffer. I am also grateful for the diligent research assistance of Megan Hershey, Mike Radcliffe, Katie Scofield, and Martha Wilfahrt. At Cambridge University Press, I am so thankful for the support of my editors, Lew Bateman and Margaret Levi, throughout this process. In particular, I owe a tremendous debt to associate editor Bob Bates, who provided insight and encouragement at every stage. I learned so much from the review process as an author, scholar, and teacher. I valued all the constructive criticisms and suggestions offered by the external reviewers. I also would like to thank Emily Spangler and the production staff at Cambridge. I would also like to thank several journals for permission to reprint material that has already been published. Some material from Chapters 4 and 5 are published with kind permission from Springer Science+Business Media, from the article, “Constructing a Social Safety Net in Africa: An Institutionalist Analysis of Colonial Rule and State Social Policies in Ghana and Cote d’Ivoire,” Studies in Comparative International Development, Vol. 37, Issue 3 (2002), pp. 64–90. Several chapters in the book draw from material originally published in “State Social Policies and Social Support Networks: The Unintended Consequences of State Policymaking on Informal Networks in Ghana and Cote d’Ivoire,” International Journal of Public Administration, Vol. 26, Issue 6, pp. 665–691, which is reprinted by permission of Taylor & Francis Ltd. (http://www.tandf.co.uk/journals). Some material from Chapter 6 has already appeared in “Empire of the Young: The Legacies of State Agricultural Policy on Local Capitalism and

xvi

Acknowledgments

Social Support Networks in Ghana and Cote d’Ivoire,” Comparative Studies in Society and History 46, 3 (July 2004), pp. 469–496. And parts of Chapter 7 draw on material previously published in my article, “Mediating Ethnic Conflict at the Grassroots: The Role of Local Associational Life in Shaping Political Values in Cote d’Ivoire and Ghana,” Journal of Modern African Studies, Vol. 42, No. 4 (2004), pp. 589–617. Finally, I would never have finished this book without the endless support and grounding perspective of my husband, Jason MacLean, and our children, Jasper, Skylar, and Benjamin. They always cheered me on, even when the project involved difficult separations for overseas fieldwork, or time tied to my computer chair finishing another draft. The support and interest of my parents, in-laws, grandparents, aunts, uncles, and cousins proved to me that the extended family is not necessarily a vestige of the past here in the United States. I also owe a huge debt to my “village community” of friends and dream teams of caregivers for our children. This book is dedicated to them all.

part i THE TRANSFORMATION OF INFORMAL INSTITUTIONS OF SOCIAL RECIPROCITY IN GHANA AND CÔTE D’IVOIRE

chapter 1 Introduction

African political economies are not always and everywhere in crisis. Indeed, over the past 100 years, the countries of Ghana and Côte d’Ivoire in West Africa have been considered successful models of democratic and economic development at different points. We cannot understand the varied paths of the Ghanaian and Ivoirian political economies by solely focusing at the macro level on state weakness, or by exclusively concentrating at the micro-level on the deficits of social capital or missing institutions. In this book, we travel to similar villages in the countryside on either side of the Ghana–Côte d’Ivoire border to see how the history of people’s experience of state-building has fundamentally transformed economics and politics from the ground up. A little more than 100 years ago, prior to European colonial rule, an outsider traveling to the four Akan villages selected for this study in the forest zone of West Africa would have found communities with very similar political histories, economies, social organization, and cultures. As the Asante Empire began to centralize and expand in the mid-seventeenth century in what is now Ghana, these Akan groups resisted incorporation.1 Instead, they chose to migrate farther westward of the Asante capital of Kumasi to what is now the southwestern part of Ghana and the southeastern part of Côte d’Ivoire. (See Figures 1.1 and 1.2.) The Akan peoples in these 1

A variety of Akan peoples had migrated to what is now Ghana around the thirteenth century AD. The Akan linguistic group includes the Akuapem, the Akyem, the Asante, the Baoule, the Brong, the Fante, and the Nzema peoples of Ghana and Côte d’Ivoire. They were organized as small states until a few groups began to centralize and expand their kingdoms, most notably, the Asante Empire. By 1874, the Asante Empire included over 100,000 km2 and approximately three million subjects. See, for example, Wilks (1993, 1975); McCaskie (2001, 1995b); Arhin (1976); and Fortes (1969).

3

Informal Institutions and Citizenship in Rural Africa

4

Europe

M

O

RO

CC

TUNISIA

O

ALGERIA

L I B YA

WESTERN SAHARA

CAPE VERDE ISLANDS

Middle East

EGYPT

MAURITANIA MALI

N I G E R CHAD

RE

A

DJIBOUTI

LI A M

A A

UG

MALA W

ZAMBIA

BOTSWANA

BI

Q

MOZ

NAMIBIA

AM

I

ZIMBABWE

R

ANGOLA

GA SC A

GO

SO

MA DA

Cabinda

A ND

KENYA DEMOCRATIC RWANDA REPUBLIC OF THE BURUNDI CONGO TANZANIA

E

RO ON

ME GABON

C

EQUATORIAL GUINEA SÃO TOMÉ & PRINCIPE

ETHIOPIA CENTRAL AFRICAN REPUBLIC

ON

O

A NA

T OG

COTE D’IVOIRE

LIBERIA

GH

EA

SIERRA LEONE

NI G ER IA

CA

GUI N

E RI T

S U D A N

N

BURKINA FASO

U

L GA

THE GAMBIA GUINEABISSAU

E

BENI

SEN

SWAZILAND 0 0

500

1000 500

1500 1000

2000

2500 km

SOUTH A F R I C A LESOTHO

1500 miles

fi g u r e 1 . 1 . Map of Study Region in Africa

fieldsite villages thus shared similar village chieftancies, matrilineal family systems, and customary systems of land tenure, inheritance, and justice.2 Even today, village residents of these Akan regions continue to identify as “one family,” avowing that they fundamentally have shared the same precolonial history, politics, and culture. Yet, when I arrived in these villages in 1998–1999, I found striking differences in the local politics and cultures. In particular, I was surprised by the extent and types of variation in the informal institutions of social reciprocity and indigenous notions of citizenship. First, informal reciprocity – the ways that village residents exchanged help and social support with their nuclear and extended family, clan, friends, neighbors, ethnic group, or others – was quite different in the Akan areas now on either side of the Ghana–Côte d’Ivoire national border. In the Ghanaian region, fewer people were 2

For example, see Firmin-Sellers’ (1996) comparative analysis of the development of land tenure systems in the same two regions as this book.

0

20

40 20

80

60 40

100

120

60

Dawadawa

140 160 km 100 miles

80

Bondoukou

Sampa

U

0

Wenchi

U R O

M’Bahiakro

Dimbokro

a

Bandam

Divo

O

R

B

Debiso

Nkawkaw

Koforidua Enchi

Aboisso

Grand Bassam

Lake Volta

Kumasi

Prestea

Suhum Oda

DunkwaOn-Ofin

Tarkwa Brédibéri

A

Obuasi

Agboville

Abidjan

N

Bibiani

no Ta

Adzope

A

é

Gagnoa

H

Agogo

mo Co

N zi

Akoupé

Donkorkrom

Bi a

Toumodi

Ejura Mampong

Bechem

G

A B E

Abengourou

Kwadjokrom

Atebubu

Sunyani

N G O

D ’ I VOI R E

F

A

H

N

Yakasse

C Ô T E

Yamoussoukro

A

G

Berekum

Dakro Bouaflé

-

Accra

Pra

Bouaké Béoumi

Yeji

O

Dunkwa Cape Coast

Half Assini Axim Cape Three Points

fi g u r e 1 . 2 . Map of Fieldsites in Similar Regions of Côte d’Ivoire and Ghana

Sekondi-Takoradi

Winneba

6

Informal Institutions and Citizenship in Rural Africa

exchanging any kind of help at all, and when they did, it was a much lower level of support. The village residents in the Ghanaian region gave a little bit of help to a much wider array of social ties, particularly friends. In contrast, in the Ivoirian region, greater numbers of village residents gave more significant amounts of help, but this was given to a much narrower group of people, particularly members of the immediate nuclear family. Second, local conceptualizations of citizenship also differed in remarkable ways in these two similar Akan regions. In the Ghanaian region, village residents articulated a community-oriented notion of citizenship, whereas in the Ivoirian region, villagers described an individualized, entitlementbased sense of citizenship. Furthermore, the Ghanaian patterns of citizenship and politics seemed to facilitate the resolution of ethnic conflict at the grassroots, whereas the Ivoirian patterns magnified the salience of ethnic cleavages upward to the national level. None of the donors, policymakers, or scholars I had interviewed beforehand in the capital cities of Accra and Abidjan would have predicted such divergence between these analogous regions in what are today the separate countries of Ghana and Côte d’Ivoire. These experts on the ground as well as the existing scholarship did draw attention to the profound differences between other subnational regions within the two countries, for example, expected differences between the more Muslim, patrilineal northern regions of each country and the more Christian, matrilineal southern regions.3 But even in making those distinctions, the assumption was reinforced that these crossborder Akan regions should be more alike than different. And yet, the differences I found on the ground between these Akan villages were striking. This book seeks to answer two connected questions. First, why did the informal institutions of reciprocity differ in such surprising ways in similar Akan villages on either side of the Ghana–Côte d’Ivoire border? If Ghanaian and Ivoirian villagers from these crossborder regions indeed considered themselves to be “one family,” it is puzzling why they helped their families, friends, and neighbors in such distinctive ways. Second, what were the consequences of these different informal institutions of reciprocity for the practice of citizenship? If their precolonial cultures and political histories were so similar, why did Ghanaians and Ivoirians from these regions conceptualize citizenship and participate in local and national politics so differently? 3

In a study of another region in southern Ghana, Hill (1963) argued that different ethnic culture and lineage systems resulted in varied structures of collective organization for agricultural production.

Introduction

7

I argue that diverse histories of colonial and postcolonial state formation have stimulated these puzzling local-level differences in the informal institutions of social reciprocity and citizenship in rural Ghana and Côte d’Ivoire. In 1884, competing to expand their colonial empires, the European powers met in Berlin and agreed to carve “this magnificent African cake” into pieces.4 With little knowledge of precolonial African political systems or cultural differences, colonial rulers created highly heterogeneous, multiethnic states. For example, Côte d’Ivoire now encompasses as many as sixty different language groups, while Ghana includes at least thirty-five language groups.5 In many instances, these arbitrary colonial boundaries unwittingly split ethnocultural groups into two separate nation-states. This is indeed the case for the Akan groups described previously: culturally and politically similar to begin with, they now live on either side of the Ghana–Côte d’Ivoire national border. Through this comparative study, we can see how the historical and political process of building a nation-state from the colonial era to the present was experienced differently on the ground and profoundly shaped the informal institutions of reciprocity. In order to understand the puzzling variation in informal reciprocity at the local level, we must consider the historical construction of the state role in mediating risk in three different policy areas: (1) political administration; (2) social infrastructure and service delivery; and (3) economic policy in agriculture. (See Table 1.1.) These divergent histories of state formation not only shape the nature of reciprocity operating on the ground but also influence how local groups and communities define citizenship and then choose to engage and interact with the state on an everyday basis. (See Figure 1.3.) My thesis is that a more centralized and expansive role for the colonial and postcolonial state in Côte d’Ivoire ironically stimulated a greater volume of informal reciprocal exchange during the late 1990s. But this reciprocity was more heavily concentrated on the nuclear family. This pattern of informal reciprocity in this region of Côte d’Ivoire strengthened the development of a uniethnic, entitlement-based notion of citizenship that extended out from the

4

5

King Leopold of Belgium is credited with describing Africa in this way prior to the Berlin Conference. Hochschild (1998: 58). The number of language groups is actually a subject of debate and supports Posner’s (2005) argument that formal state institutions can shape the range of social identities available. In Ghana, the 1960 census recorded over 100 different language and cultural groups. Today, only nine are state-sponsored languages, and twenty-six others remain as non-official languages. This evidence suggests a process of state-supported consolidation similar to what Posner found in Zambia.

t a b l e 1 . 1 . How the History of State Formation in the Region Stimulates Divergence in Informal Reciprocity State Role in Mediating Risk

Description of State Role

Level of Risk Mediation by State

Variation in History of State Formation in Region

Effect on Informal Reciprocity

1. Political Administration (Chapter 4)

Political construction of the state role vis-a ` -vis the family and citizen.

B/A GHANA Low

B/A GHANA Decentralized colonial and postcolonial administrative state that supplements only preexisting extended family and community systems. ABENGOUROU REGION CÔTE D’IVOIRE Centralized colonial and postcolonial administrative state generates high expectations that state supports individual citizens in nuclear families. B/A GHANA An historically longer investment in social services is eroded earlier by unstable and low-quality service provision since the late 1960s so lower expectations of service utilization when state retrenched in the mid-1980s.

B/A GHANA More diversified reciprocity.

ABENG CÔTE D’IVOIRE High

2. Social Service Delivery (Chapter 5)

General provision of state infrastructure and delivery of social services to subsidize health, subsidize education, and alleviate poverty.

B/A GHANA Low

ABENG CÔTE D’IVOIRE High

3. Economic Policy (Chapter 6)

Economic policy interventions to support production, mediate market volatility, and tax productivity.

B/A GHANA Low

ABENG CÔTE D’IVOIRE Medium

ABENGOUROU REGION CÔTE D’IVOIRE A shallower investment in social services historically, but more stable and higher quality since late 1960s until mid-1990s, generating higher expectations of service utilization when state retrenched. B/A GHANA State intervenes actively in agricultural policy to monopolize purchasing and extract very heavily from export crop production stimulating shift to shortterm, labor-intensive vegetable production for domestic markets. ABENGOUROU REGION CÔTE D’IVOIRE State intervenes in agricultural policy, but allows some private purchasing, more moderate taxation of export crop production and facilitates labor in-migration supporting expansion of long-term investments in land-intensive cocoa and coffee production for export markets.

ABENGOUROU CÔTE D’IVOIRE More concentrated reciprocity.

B/A GHANA Lower quantity of reciprocity exchanged. More diversified reciprocity.

ABENGOUROU CÔTE D’IVOIRE Higher quantity of reciprocity exchanged. More concentrated reciprocity.

B/A GHANA More diversified reciprocity in particular among young.

ABENGOUROU CÔTE D’IVOIRE More concentrated reciprocity that links older and younger generations.

Informal Institutions and Citizenship in Rural Africa

10

CÔTE D’IVOIRE

Étatiste Colonial State

Concentrated Local Reciprocity Between Generations

National EntitlementBased Citizenship

Étatiste Postcolonial State Adopts SAP

Mobilized but Narrowly Concentrated Local Reciprocity Between Generations

Uni-Ethnic National EntitlementBased Citizenship

Decentralized Colonial State

Diversified Local Reciprocity Between Generations

Community Duty-Based Citizenship

Decentralized Postcolonial State Adopts SAP

Diversified Local Reciprocity Among the Young

Multi-Ethnic Community Duty-Based Citizenship

Shared Precolonial State, Informal Institutions of Reciprocity, and Citizenship

GHANA

Precolonial

Colonial Divergence

Postcolonial Divergence

fi g u r e 1 . 3 . The Argument

village up to the national capital. In contrast, in the Ghanaian region, a more decentralized and restricted role for the colonial and postcolonial state was associated with a much lower volume of informal reciprocal exchange. In this region in the late 1990s, neither the central state nor informal social relations were providing a robust social safety net, particularly for the very young and very old. Yet, the reduced central state support had spurred a diversification of informal reciprocity across different social ties in the Ghanaian region. This diversification of reciprocity among the nuclear extended family and, importantly, friends of various ethnic, gender, and class backgrounds reinforced a more multiethnic, duty-based conception of citizenship focused on the local, village political community in this region of Ghana. My argument thus highlights both costs and benefits – social, economic, and political – of the different histories of state formation in each case. The book does not conclude simply that a particular type of state intervention had ideal consequences on all fronts. Rather, the book’s contribution is to emphasize the nuances of how colonial and postcolonial states have transformed informal reciprocity and citizenship in different ways over time. Much of the recent literature on institutions or social capital downplays the role of state legacies in transforming informal social institutions and political culture, however. For theorists of the new institutions or social capital, a focus at the micro-level occludes a structural analysis of the historical role of the state. I argue that these perspectives are insufficiently political and as a result mistakenly characterize Africa as more homogeneous than the empirical reality shown here. On the other hand, historical institutionalists have often overemphasized the historical development of formal

Introduction

11

state institutions, such as different constitutional and electoral rules, at the national level and have hence missed important interactions at the microlevel – in particular, the role of informal institutions. The existing scholarship that does examine both the state and the informal social and political institutions at the local level has portrayed the interaction in relatively dichotomous terms as a zero-sum relationship. The dominant theme is that the African state is weak.6 Some scholars have suggested that as states weaken, vibrant informal networks grow and expand.7 Others have theorized the causal story in reverse: that the existence of strong informal networks allows the option of refusing to be subordinated and thus weakens the state from below.8 In contrast, I draw theoretical inspiration from disciplinarily diverse literatures on institutions and agrarian change to show how the colonial and postcolonial state has had profound effects on village social institutions and political cultures.9 Based on intensive fieldwork in these regions, I use an historical and ethnographic analysis of micro-level interactions to empirically investigate how the history of state power has varied and actually mattered at the village level.10 The construction of the colonial and postcolonial state does not necessarily destroy informal reciprocity wholesale. Rather, the different legacies of state-building are the key determinants in the transformation of informal institutions of social reciprocity and citizenship over time and in particular places in rural Africa. By uncovering local patterns of social and political exclusion and the ways that rural people were linked (or not) to the broader national political system, we begin to shed light on one of the most troubling and important puzzles in the region.11 Why, after decades of political instability and economic collapse, has Ghana emerged as a model of democratic consolidation while the 6

7

8 9

10

11

See Herbst (2000); Chabal and Daloz (1999); Bayart, Ellis, and Hibou (1999); and Olowu and Wunsch (1990). See Chazan (1983); Azarya and Chazan (1987); Dei (1992); Cheru (1997); Pellow and Chazan (1986); MacGaffey (1991); and Tripp (1997). See Hansen (2004) for a critique of the idealization of the informal economy by donors and policymakers. See Hyden (1980); Bayart (1993); Bayart et.al. (1999); and Reno (1998). Scott (1976, 1985); Bates (1989, 1990, 1997); Hyden (2006); Watts (1983); Boone (2003); and Mamdani (1996). Fieldwork was conducted in January 1994, April–August 1997, and October 1998–October 1999 in two similar villages of Tano District in the Brong-Ahafo region of Ghana and in two similar villages in the Abengourou region of Côte d’Ivoire. Fictional names are used for these villages in the footnotes to protect the anonymity of the sources. Social exclusion is conceptualized here more broadly as a process of being marginalized or left out of social relationships, not simply formal citizenship status within the nation-state as frequently theorized in the literature on the advanced industrialized countries (Gore et.al. 1995).

12

Informal Institutions and Citizenship in Rural Africa

postindependence stability and prosperity of the “Ivoirian miracle” has been shattered by the outbreak of ethnoregional civil war? Both Ghana and Côte d’Ivoire shared very similar ethnic, religious, and regional divisions, where a mostly Christian, Akan South has dominated a more impoverished, largely Muslim North in the economic and political realms. Ironically, however, faced with similar, overlapping cleavages, local ethnic conflict has had very different consequences for the nation. Fortunately, the majority of the book’s data was collected in the field from October 1998–October 1999, immediately prior to the outbreak of political violence in Côte d’Ivoire. In fact, the fieldwork in Côte d’Ivoire was completed just 1 month prior to the first coup d’état and collapse of the long-dominant Parti Démocratique de Côte d’Ivoire (PDCI) regime.12 This book thus provides a snapshot of a culture on the brink of civil war, illuminating key differences in the nature of local-level social exclusion and the interpretation of democratic citizenship.

puzzling contrasts in informal social reciprocity and indigenous notions of citizenship in similar regions of ghana and coˆ te d’ivoire A brief comparison of the early morning breakfast routine in two similar villages in each of these Akan regions of Ghana and Côte d’Ivoire poignantly reveals the unanticipated differences in the informal institutions of social reciprocity and indigenous notions of citizenship. The existing scholarship did not predict these differences. Weak but Diversified Informal Social Reciprocity and Community-Oriented Citizenship in the Ghanaian Region In the Ghanaian villages, in what was historically a cocoa region, many young farmers recently switched to a more backbreaking and volatile new cash crop: tomatoes. Even before the sun rose, many of the younger parents, aunts, uncles, and cousins left home to water their new tomato farms. As a result of the early 12

A series of coup d’états derailed the Ivoirian political system beginning in the winter of 1999. By 2002, the country had exploded into a violent civil war. Finally in March 2007, the rebels and central government reached a peace agreement that aimed to hold new elections by November 2008. By October 2008, the requisite timetable for demobilization, voter registration, and state redeployment in rebel areas was proving difficult to achieve. In early November 2008, the elections were cancelled. The new timetable for elections was November 29, 2009 but they were again postponed. The new election dates had not yet been announced at the time of this writing but were expected possibly in early 2010. UNIRIN (2008) and Coulibaly and Lewis (2009). See also Geschiere (2009).

Introduction

13

morning labor required to farm tomatoes, there was no one left at home to cook for the extended family. Most of the multifamily courtyards and outdoor cooking areas were eerily quiet. The youngest and the eldest were on their own, left with coins to buy their early morning meals at a number of informal eating spots around the villages. Typically, the women cooking at these “chop shops” bustled while children squatted alone – a few feet apart from each other – on small wooden benches eating their breakfast without talking. Elder men from the dominant Akan ethnic group lamented their loneliness and economic vulnerability, contending that the extended family system had weakened markedly in recent decades. The quantitative amount of social reciprocity exchanged was not very high in the Ghanaian villages – rarely enough to cover even one child’s school fees, or to handle one medical emergency. But what really concerned these Akan elders was the rise of the new “empire of the young,” where youth – both from Akan and nonindigenous minority ethnic groups, and both male and female – wielded newly attained wealth and power in the village. Many complained that the very young and old were increasingly left out. The variation in reciprocity at the local level shaped the contrasting ways that Ghanaians in these regions thought about citizenship and participated in politics from the village all the way up to the nation-state. Village residents in this region of Ghana mentioned much more frequently their duty to provide communal labor as something they owed to the state. Ethnic minorities often participated as elected leaders on village committees to help resolve village problems. As a result, ethnic grievances were mediated for the most part at the grassroots, and ethnic conflict has not been projected onto the nation.13 Ghana has had the chance to consolidate its democracy without being torn apart by ethnically based political mobilization and violence. Indeed, since permitting multiparty politics in 1992, Ghana has held five relatively peaceful national elections, including a nonviolent transfer of political power from the formerly authoritarian incumbent to the main opposition party in 2000, as well as another alternation of power after an extremely tight runoff in December 2008. Stronger but More Narrowly Concentrated Informal Social Reciprocity and Individualized Notions of Entitlement in the Ivoirian Region Notwithstanding their shared precolonial Akan culture, the informal institutions of reciprocity differed in surprising ways in the Ivoirian villages when compared with the Ghanaian ones. The morning routine had not changed 13

MacLean (2004).

14

Informal Institutions and Citizenship in Rural Africa

there, as it had in Ghana. The continued predominance of cocoa production in this region of Côte d’Ivoire meant that at least a few female family members remained at home to cook breakfast and care for the extended family’s young and old in the Ivoirian villages. Very few prepared food “chop shops” were open for business in this region of Côte d’Ivoire because most everyone shared their meals in larger, extended family courtyards. The Akan elders in Côte d’Ivoire did not share the same worries and complaints as they did so often in the Ghanaian region. The preceding different patterns of informal reciprocity shaped the ways that Ivoirians thought about their role as citizens and engaged with the state in everyday politics. Since Ivoirian reciprocity was more narrowly concentrated on the nuclear and immediate extended family, Ivoirians were less connected to the larger village community through social interaction. Ivoirians frequently struggled to think of one thing that they might owe the state, but very readily rattled off a long list of rights, almost always private goods that would be consumed entirely by individuals as opposed to shared by the community as a whole. Ethnic minorities had little voice in resolving local issues, and interethnic conflicts were quickly passed to the next level of a highly centralized government to decide. Not only were young, nonindigenous Muslim groups increasingly excluded from economic gains but they were also locked out from the local political game. These ethnic grievances have been channeled upward, where they then have threatened to undermine the very unity of the nation.

current theoretical explanations Already, this brief look at everyday life in villages in similar regions of Ghana and Côte d’Ivoire reveals unanticipated differences. To investigate these empirical puzzles in Ghana and Côte d’Ivoire, I will ground my argument in a multidisciplinary literature broadly-oriented toward understanding institutions and agrarian change.14 While recent scholarly attention to social capital and institutions has reaffirmed the value of analyzing the micro-level, politics and history remain undertheorized. Robert Bates similarly critiques the new institutionalists for failing to adequately theorize politics.15 Bates summarizes: The image conveyed in the new institutionalism is that of economic actors, frustrated in their efforts to transact in markets, structuring non-market institutions that will

14

15

See Hall and Taylor (1996) for a review distinguishing between the three schools of institutional theory (i.e., the new institutional economics, historical institutionalism, and sociological institutionalism). See also DiMaggio (1998) and Bardhan (1989). Bates (1997: 40–1).

Introduction

15

enable them to transcend their dilemma and thereby attain welfare-enhancing outcomes. The reality is that non-market institutions are often created in the legislature or the court room or by economic actors who anticipate the appeal of others within such political arenas . . . In attempting to construct an economic theory of non-market institutions then, the new institutionalism commits major errors of omission: it underplays or ignores the importance of politics.16

Bates thus highlights the need to address both the relative power of different political groups who contest the choice between different nonmarket institutions with different distributional consequences as well as the role of the state in shaping the political setting where that struggle takes place.17 I argue that it is vital to add a deeper appreciation for politics and history to our institutionalist analysis in order to predict and understand the variation that exists on the ground. Conceptualizing the Informal Institutions of Reciprocity Before proceeding, it is important first to define the informal institutions of reciprocity at the center of this study. Reciprocity was relatively ignored in the political science literature until social capital attracted a flurry of scholarly attention.18 In fact, this new interest in social capital has expanded beyond academia to be embraced by powerful development organizations. The World Bank in particular has organized several conferences and published numerous edited volumes on the conceptualization and relationship of social capital to development.19 Most recently, World Bank attention has been dedicated to trying to measure social capital quantitatively.20 While none of these scholars and practitioners agrees on a single definition of social capital or theory of social capital’s origins and causal effects, Robert Putnam’s work continues to be the most frequently cited.21 Putnam defines

16 17

18

19

20 21

Bates (1997: 42). For a similar point about how the state shapes political preferences, see also Weyland (2002: 75). On social capital, see Lin (2001); Fukuyama (1995); Putnam (1993, 2000); Coleman (1988, 1990); and Bourdieu (1986). Krishna (2002: 4) uses the concept of social capital but emphasizes the role of informal associations and mutual support networks in his work on India. Some of the theoretical work on civil society overlaps with that on social capital but has focused more on formal organizations than the informal social relations under analysis here. See, for example, Varshney (2002); Berman (2001); and Bratton (1989). See Naryan et.al. (2003); Grootaert and van Bastelaer, eds. (2002a, 2002b); and Dasgupta and Serageldin, eds. (2000). See Grootaert et.al. (2004). The editors of a World Bank volume concluded that: “ . . . social capital means different things to different people.” Dasgupta and Serageldin (2000: x).

16

Informal Institutions and Citizenship in Rural Africa

social capital as “the connections between individuals and the norms of reciprocity and trustworthiness.”22 Like Putnam’s, many definitions of social capital emphasize social relationships and norms of reciprocity as a key component, but the lack of clarity in this loaded concept of social capital undermines the measurement and analysis of reciprocity. The editors of a World Bank volume on social capital launched a similar critique of this lack of definitional clarity arguing that “a concept that encompasses too much is at risk of explaining nothing.”23 A major weakness of the current use of social capital is that the attention to local social relations and culture is almost always divorced from a broader structural analysis of the political economy.24 Fine has made one of the most incisive critiques of how this “plump and benevolent” definition of social capital is too “fungible” and can be applied to anything anywhere, thus obscuring the real power and conflict involved in the political economy of development.25 My conceptualization of reciprocity is quite different and intentionally emphasizes politics and history. I deliberately choose not to use the term social capital in this book. Instead, I draw on recent scholarly debates in anthropology to develop a more dynamic and contextualized approach.26 I employ the most minimalist definition of reciprocity to allow an investigation of if, how, and when exchanges take place and change in particular places and historical moments. I do not assume whether these norms and relations of reciprocity have positive consequences for societal cohesion, economic growth or democracy.27 In particular, I show how broader 22 23 24

25 26

27

Putnam (2000: 19). Grootaert and van Bastelaer (2002b: 5). One important exception is Steger’s (2002) immanent critique of Putnam’s Bowling Alone which argues that globalization is actually a more powerful explanation for the decline in social capital in the United States. See also Evans (1996) work on the synergy between the state and social capital. Fine (2003, 2001). In an excellent historical overview of the literature, Graeber (2001) concludes that anthropologists still have not resolved their dilemmas about theories of value. Key classic works are Malinowski (1950); Mauss (1954); and Levi-Strauss (1969). The scholarly discussion continues today with work by Strathern (1988), and important critiques by Weiner (1980, 1992) and Gregory (1982). See also works by economic sociologists such as Granovetter (1985) and Blau (1964) on social networks and social exchange. There is of course a voluminous anthropological literature on kinship. On the nature of Akan kinship in Ghana, see work by Rattray (1923, 1929), Danquah (1928), and Fortes (1950, 1969) and critique by McCaskie (1995b). See also Nukunya (1992) on the family in Ghana, and Vimard (1987, 1993) on the family in Côte d’Ivoire. Fine (2003) and Levi (1996) critique Putnam and other social capital theorists for assuming that these networks are inherently positive and neglecting the potentially “unsocial” dark side.

Introduction

17

structures of state power may fundamentally shape local reciprocity in a dialectical, back-and-forth interaction over time.28 This politically and historically contextualized approach allows a more effective measurement of reciprocity and then facilitates the subsequent empirical investigation (rather than assumption) of the consequences for citizenship and democracy.29 Hence, reciprocity is defined here simply as long-term ties of exchange, or give-and-take, between individuals and groups over time. The number and type of people involved in exchanges is open and potentially dynamic. The reciprocal arrangements may include nuclear and extended family, friends, neighbors, other villagers, members of the same or different ethnic groups, and so on.30 Furthermore, the exchanges may involve people who live in different localities across geographic space. For example, some village residents reported financial help they received from uncles who lived in larger urban areas. Surprisingly, international remittances from family members living overseas were not significant in either region.31 Exchanges are also not simple dyadic relationships between two individuals at one time period; they may involve multiple triangulations across generations. Platteau makes a compelling argument that these informal institutions are guided by a norm of “balanced reciprocity” over the long term.32 As such, an element of uncertainty exists for participants as they invest in and negotiate their social relationships.33 Building on Goran Hyden’s recent typology of the “economy of affection,” I characterize an individual’s overall pattern of reciprocal relations as (1) horizontal versus vertical and (2) diversified versus concentrated.34 First, more horizontal reciprocity involves people of similar age, wealth, and/or power, whereas more vertical reciprocal relations are between people of vastly different ages and political/economic resources. Although the horizontal and symmetrical nature of reciprocal exchange between kin and 28

29 30

31

32 33 34

Tarrow (1996) critiques Putnam for making social capital in Italy seem quite insulated from state power. Adcock and Collier (2001). My conceptualization of reciprocity would include rotating savings groups or “tontines” among friends or neighbors. On tontines in Cameroun, see Henry, Tchente, and GuillermeDieumegard (1991) and Janin (1995). In Manuh’s (2006) excellent edited volume on international migration, Addison (2006: 119) reports that remittances from Ghanaians were bigger and more stable than foreign aid and foreign direct investment since 1990. In the same volume, Mazzucato, Bart, and Nicholas (2006: 150–1) find that international remittances are geographically concentrated in the regions of Ashanti and Greater Accra and are more likely to be received by richer individuals and households. Platteau (1997: 767–8). Bates (1990). Hyden (2006: 78–83) contrasts vertical with lateral and open with closed in his typology.

18

Informal Institutions and Citizenship in Rural Africa

friends is often romanticized, the vertical and asymmetrical patterns of exchange between the rich and the poor predominate.35 Second, I consider to what extent individuals spread their reciprocity over a large number of different types of social relationships.36 Thus, reciprocity relations are more diversified when individuals spread their gift giving relatively equally among a large number of diverse social categories and more concentrated when reciprocity is focused on a fewer number of specific social relations. Another important point of distinction in the conceptualization of reciprocity is not simply who is tied together, but what the nature of these ties is. These relationships have political value.37 For example, gifts may be exchanged in return for political loyalty, votes, and support. And, conversely, when social reciprocity is eroded, political loyalties may be undermined. Clearly, these reciprocal arrangements have economic benefits in terms of providing social insurance in high-risk ecological environments or across the life cycle.38 But scholars should not lose sight of the potential power differentials among givers and receivers, and the political value placed on fulfilling obligations. Again, by using a minimalist definition of reciprocity, I can explore the meaning of these relations of reciprocity to the participants themselves, rather than assuming a purely instrumental logic. Measuring the Informal Institutions of Reciprocity So if these relationships are so complex, contested, contingent, and long term, how can we hope to study reciprocity empirically? It would be useless to attempt to record a balance sheet of debts and credits between two people, or to try and map a holistic network at one point in time or space.39 For this 35

36

37 38

39

Polanyi (1944) romantized horizontal reciprocity while Scott (1976: 88); Hyden (1980, 2006); and Berry (1985) highlight vertical asymmetry. Several studies by agricultural economists have challenged the assumptions of unitary households and complete risk-sharing within families and communities in different parts of Africa. Udry and Conley (2004) characterize the networks in contemporary Ghana as “sparse” and surprisingly low-density. Kazianga and Udry’s (2006) study of Burkina Faso finds that even during a severe drought, there were almost no village-level risk pooling mechanisms, and that individuals relied almost entirely on self-insurance. Markus Goldstein (2000) emphasizes the lack of pooling of information and resources between husbands and wives in the same household in his study of another region of Ghana. Dercon and Krishnan (1996) find incomplete informal risk-sharing in their study of Ethiopia and Tanzania. See Scott (1976); Hyden (1980); and Watts (1983). Bates (1990) has shown how investments made in children or cattle serve as old-age insurance. See also Bates and Curry (1992). I thank Beth Buggenhagen for her point that attempting to represent reciprocity as a social network would give the complex and dynamic social relationships a “misplaced concreteness.”

Introduction

19

reason, the text refers to the relations of reciprocity, not networks. To shed light on these changing sets of norms and relationships, multiple methods of data collection were combined. First, I documented the verbal articulation of normative beliefs about reciprocity. I define norms here as codes of conduct agreed upon as appropriate within a certain group.40 For example, many villagers spoke of the reciprocity between the young and old as something that was just understood as a natural part of the cycle of life. As one woman explained, “I raised my children from when their teeth first came in. They also can help me until my teeth fall out.”41 She was expressing a norm of reciprocity here, emphasizing the long-term exchange between parents and their children. Second, I listened carefully in interviews, focus groups, and everyday conversations for evidence of norms being contested or enforced regularly with the sanctioning of inappropriate behavior as an informal institution. The breach of informal rules was not necessarily enforced by the central state but was often self-enforcing by the group through joking, gossip, social stigmatization, or even violence.42 One sanction that was cited by villagers as occurring more frequently in the recent past was the use of witchcraft to punish a young nephew or niece who had succeeded but not helped his or her extended family member with needed financial support.43 The rise in the frequency of sanctioning via witchcraft accusations in both Ghana and Côte d’Ivoire might suggest that currently a struggle is being fought between individuals and groups over the underlying normative framework, a topic to be explored further in Chapter 3. This book reveals how African normative orders are contested and transformed at various rates and differentially for diverse groups within a society. Finally, to analyze these unwritten and invisible norms and informal institutions, much of this book examines how individuals actually give help to each other and receive help from each other at the village level. For example, to whom did an elderly widow in Ghana turn for help over the past year when she was sick and in need of medical attention? These relationships reflect normative frameworks and informal institutions. I collected

40 41 42

43

See Ostrom (2000) and Crawford and Ostrom (1995). In-depth interview with elderly woman by author, Kyere, Côte d’Ivoire, August 20, 1999. Scott (1985) explores non-violent sanctions in his analysis of “the weapons of the weak.” Bates (1989) argues that the less successful Kikuyu initiate the Mau Mau rebellion to reclaim their rights and punish the more successful Kikuyu who chose to withdraw from kinship networks. On the rise of witchcraft accusations in different parts of Africa, see Comaroff and Comaroff (1999); Geschiere (1997); and Meyer (1995).

20

Informal Institutions and Citizenship in Rural Africa

information about the extent and structure of social exchanges among a randomly sampled group of villagers. Again, my approach differs from traditional social network analysis in that I do not conceive of these social relations as a network that can be mapped as a whole entity. Instead I attempted to take a snapshot of the number and types of reciprocal exchanges in a particular context and to investigate the meaning of these individual ties. Explaining Informal Institutional Change So how does the current literature explain such different changes over time in the informal institutions of reciprocity in these regions of Ghana and Côte d’Ivoire? Like the scholarship on social capital, much of the new institutionalist literature has paid insufficient attention to politics. Douglass North has portrayed the evolution of the formal “rules of the game” such as property rights or contracts as an endogenous process emerging incrementally from individual interaction over time.44 New institutionalists have viewed the impetus for the rise of nonmarket social institutions as a nearly automatic response to market failure.45 Kenneth Arrow summarizes the core argument well when he writes: “When the market fails to achieve an optimal state, society will, to some extent at least, recognize the gap, and non-market social institutions will arise attempting to bridge it.”46 This process of institutional origins is starkly apolitical. Politics are only minimally acknowledged in the work of North and Jack Knight at the micro-level in terms of differences in individual bargaining power and distributional preferences.47 Importantly, the state is largely absent in shaping the parameters for distributional conflict.48 The new institutionalists hence see the informal institutions of reciprocity of the lineage, extended family system, or village community in Africa as a response to the market failure to provide social insurance.49 It is true that in most African economies, market-based insurance does not exist, so 44 45

46 47 48

49

North (1990). See, for example, Greenwald and Sitglitz (1986). See also Dercon (2003) and Stiglitz (1986). Engerman and Sokoloff (2002) emphasized the role of initial factor endowments and geographic preconditions on the subsequent evolution of institutions. See also Greif (1992, 1997). Arrow (1971: 137). See North (1990: 16) and Knight (1992). Bardhan (2004: 14) also highlights how many new institutionalists narrowly consider the role of the state. Widespread market failures in credit in developing countries present similar problems and lead to underinvestment in education and health and efficiency-equity losses for the poor. See Dercon (2003); Deaton (1992); and Stiglitz (2005). See also Alderman and Paxson (1992).

Introduction

21

individuals invest resources in these nonmarket institutions of reciprocal exchange in order to secure access to future support when in need.50 It is not simply the failure of the market that stimulates the rise of these nonmarket social institutions of reciprocity, however, but also the failure of the public sector. Since most African states cannot provide security or social insurance, individuals turn to their families, lineage, ethnic group, or religious associations for social support.51 But if the market failure for social insurance has been persistent and similar in both Ghana and Côte d’Ivoire, then why have the reciprocity institutions (or responses to market failure) changed over time in such different ways in the two regions? Theories about risk highlight potential sources of variation in the informal institutions used to cope with similar types of market failure.52 The basic intuition here is that if individuals or households faced higher levels of risk, they would develop strategies to spread that risk, for example, increasing investments in a broader social network.53 The assumption here, of course, is that a diversification of social investments spreads risk by increasing an individual’s access to resources across a broader range of social ties that are diversely positioned in the ecological, economic, and political environment; and are thus less likely to experience misfortune all at the same time. I explore this possibility along with the counterfactual: that diversification represents a reduction in the ability to cope with risk as reciprocity is exchanged with more and more distant and perhaps less reliable ties. Several scholars have pointed out that one of the most important sources of risk is the natural environment. For example, where rainfall is more adequate and reliable, households face less risk and thus do less extensive pooling of their access to cattle.54 Robert Bates has extended this argument to show how ecological risk combines with economic production to stimulate variations in kinship systems.55 Thus, in forest ecological zones with 50

51

52

53

54 55

Sara Berry has shown how African farmers invest in and constantly renegotiate social institutions to insure access to land and labor and cope with economic failures. Berry (1989: 41–55). See also Berry (1993). Montiel, Agenor, and Haque (1993) examine the pooling of savings through credit societies where capital markets are absent or very weak. Popkin (1979) shows how individuals turn to their churches or the opposition Communist party when the Vietnamese state fails to provide public works and secure property rights. For a review of the literature on risk management and coping strategies in rural developing countries, see Dercon (2002); Anderson (2001); and Morduch (1999). See also Platteau (1997) for a critical analysis of the normative basis and outcomes for informal risk-sharing arrangements. Shipton (2008) shows how the Luo of Kenya have developed several strategies, including investing in their lineage, as a way of spreading risk. Hukansson (1989). Bates (1990).

22

Informal Institutions and Citizenship in Rural Africa

high integration into world markets for high-value exports, households face lower risks and thus favor more individualist accumulation of wealth and access to resources. In contrast, in savannah ecological zones with cattle herds, households face higher risks and as a consequence prefer greater pooling of wealth and collective access to resources. This study’s design compares two regions from similar ecological zones but finds surprising innovations in local economic production that do stimulate variations in the patterns of reciprocity. I build on Bates’ insight to show how the nature of economic production is itself a response to the historical perception of risk vis-a` -vis the central state. Risk comes not simply from nature or from economic location but also from the political relationships between the center and periphery. I will thus add to Mary Douglas’ insistence that we understand how the social context shapes risk perception by analyzing the historical and political context of informal institutional change.56 The informal institutions of reciprocity do not emerge ahistorically and apolitically; instead, the past history of state intervention profoundly shapes their origin and subsequent development over time. Of course, historical institutionalists have long argued that the history of formal state institutions is critical to understanding crossnational differences in democracy, state capacity, and economic development.57 In much of this work, the politics and history of the state role is indeed taken very seriously, as I argue it should be. The problem, however, is that this focus on formal, macro-level institutions has been at the expense of the consideration of informal, micro-level institutions. Even recent scholarly efforts to highlight the role of informal institutions in comparative politics have presented the interactions between formal and informal institutions as different sorts of zero-sum relationships.58 Thus, Helmke and Levitsky’s typology outlines four different types of formal-informal interaction: (1) complementary; (2) substitutive; (3) accommodating; and (4) competing.59 This typology of interactions is conceived as a potential rivalry between informal and formal institutions rather than a mutual transformation.60 This zero-sum conceptualization is reinforced further when scholars focus rather narrowly on the interaction of formal and informal institutions attempting to govern 56 57

58

59 60

Douglas (1985). Some examples of works that emphasize cross-national differences in formal institutions are Immergut (1992); Collier and Collier (1991); Berger and Dore (1996); and Thelen (2004). For a theoretical overview, see Helmke and Levitsky (2004). In African politics, see Hyden (2006); Galvan (2004); and Bratton (2008). Helmke and Levitsky (2006: 14). Migdal (2001) highlights a process of mutual transformation but between state and society.

Introduction

23

the same set of issues or operating in the same political/policy domain. In contrast, I advocate using a more holistic approach where formal institutions from very different policy domains actually shape the political struggle over the rewriting of the institutional rules.61 The contemporary zero-sum conceptualization actually echoes the arguments of an earlier generation of comparative politics scholars who examined the interactions between the colonial and postcolonial state and the informal institutions of reciprocity. Goran Hyden, in his early work on the failure of “ujaama” socialism in Tanzania, conceptualized peasant economies and the state in a zero-sum oppositional relationship.62 According to Hyden, peasants in Africa still had the ability to “exit” through their participation in the “economy of affection.”63 As a result, peasants wielded the power and rendered the state weak and impotent as a “balloon suspended in mid-air . . . punctured by excessive demands and unable to function.”64 Much of the recent scholarship on the African state has followed in this vein, characterizing local informal social networks as powerful and capable of permeating or subverting a weak African state.65 Later, scholars such as Naomi Chazan, Deborah Pellow, and Aili Mari Tripp also depicted a zero-sum relationship between the African state and informal networks but with the causal arrow going in the opposite direction from Hyden.66 These scholars argued that as the authoritarian state collapsed in the late 1970s and 1980s, informal voluntary associations and social networks surged to fill the gaps. James Scott is one of the few theorists who portrayed the state as capable of profoundly influencing these informal social relations in rural areas.67 His theoretical and methodological approach has inspired this book on many levels, but our arguments differ in an important way. Scott argued that as the centralized colonial state and capitalist market expanded, the informal system of reciprocities among landlords and peasants in rural Southeast Asia

61

62 63 64 65 66

67

Lauth (2000) and Helmke and Levitsky (2006). Galvan (2004) emphasizes an intentional bottom-up process of syncretism in rural Senegal where new institutions, cultures and identities are actually recrafted. See also a provocative paper by Hoon (2009). Hyden (1980). Hyden (1980: 9). Hyden (1983: 19). Bayart (1993); Reno (1998); and Oluwu and Wunsch (1990). Chazan (1983); Pellow and Chazan (1986); and Tripp (1997). See also Dei (1992) and MacGaffey (1991). Scott (1976, 1985). Indeed, in Scott’s more recent book Seeing Like a State (1998), he examines cases where local people were incapable of resisting an authoritarian state’s high modernist designs.

Informal Institutions and Citizenship in Rural Africa

24

eroded. When these long-term networks of reciprocity were undermined, peasants rebelled, or, more frequently, in the face of a strongly repressive state, they engaged in petty or symbolic protest within the refuge of peasant popular culture.68 In place of Scott’s emphasis on the breakdown of the “moral economy,” I show how moral economies are not necessarily disappearing everywhere in the same way or at the same rate.69 Instead, these relations of reciprocity were frequently reconstituted, often including new types of members with varying implications for the structure of local and national power.70 In the next section, I develop this argument about how the state and informal institutions of reciprocity actually transform each other over time.

the argument The book’s argument is that variations in the history of state formation produced important differences in the reorganization of the informal institutions of reciprocity.71 I move away from drawing a rigid dichotomy of rivalry between informal and formal institutions to a more dynamic conceptualization of transformation. In doing so, I call attention explicitly to the complex and overlapping relationships between formal and informal institutions and how the distinctions may be blurred as they are experienced on the ground.72 In many developing societies with lower levels of literacy (often a minority of the population) and strong oral traditions, it is not very useful to define all unwritten rules as informal.73 The key is not whether these rules are actually written, but the degree to which knowledge of the rules is open, visible, and transparent. Public codification is essential for formal institutions, but that codification process may be oral and not 68 69

70

71

72

73

Scott (1985). Watts (1983) acknowledges the possible persistence of “moral economies” but highlights the production of rising inequality in his analysis of the political economy of famine in Northern Nigeria in the 1970s. In a classic debate with Scott, Popkin (1979) argued that the central state was not necessarily a threat to peasant community and welfare but actually an ally in the peasants’ efforts to revolutionize the agricultural economy and gain freer access to markets. Outside of political science, a small sub-group of studies have looked at the intersection of formal state policies and laws with the informal arrangements for what they term social security. See Benda-Beckmann et.al. (1988); Midgley (1994); and Hirtz (1995). I found Dennis Galvan’s diagram that highlights the overlapping of the various layers of the institutional superstructure and infrastructure very useful in thinking through my own conceptualization here. See Galvan (2004: 17). See also Ostrom (2005). Most scholars distinguish informal institutions negatively by their lack of written codification and state-enforcement. See Helmke and Levitsky (2006: 5); Hyden (2006); Platteau (1994, 1997); and Lauth (2000: 24).

Introduction

25

written.74 The greater uncertainty and lack of transparency for informal rules also creates a larger opening for negotiation and contestation. Thus, where other scholars ground informal institutions very tightly to a glacial conceptualization of cultural change, I show how the process of informal institutional change, while not usually revolutionary, can be intensely political and relatively rapid, occurring within a span of a decade in the fieldsite regions, not over centuries.75 An example from Ghana may illustrate this more fluid conceptualization of formal and informal institutional rules. If a village resident had a dispute with another villager, there may be several sets of rules for how to resolve the conflict. Following the written constitution and legislation of the central state, the village resident could file a complaint to be heard by the elected village-level unit committee. This would be a formal institution enforced by the central state. The village resident could also decide to approach the village chief for help. There are specific rules on how to do this, too, and everyone in the village knows them. While unwritten, these are highly formal rules that have been passed down orally for generations that specify who to contact, what type and amount of payment would be required, and how the resolution process would unfold. These are more than norms or “customs” because if the village resident did not abide by them, he or she would be sanctioned. As such, they are also formal institutions that are enforced by the local chieftancy. Finally, a village resident might be able to contact the village chief through his favored brother. There is also a commonly shared understanding in the village that a gift of gin would be brought to the brother to open the dialogue. The rules for how to approach the chief through this brother are also unwritten, but are considerably less transparent and more fluid. For example, these informal rules may change if the brother lost favor with this chief or with the installation of a new chief. In contrast, the formal rules for how to approach the chief are more akin to “oral constitutions” and would be less likely to change.76 This book’s contribution is not to create a 74

75

76

Mershon (1994: 50) similarly argues that informal rules are typically unwritten but highlights that they are “less visible and less explicit.” According to North (1990), since most cultural changes are incremental, informal constraints only change on the margins over a long period of time, and thus lag behind changes in formal rules. Knight (1992) also emphasizes that informal rules tend to persist even after the formal rules are changed. Platteau (1994: 804) also concludes that history and cultural change move slowly; as a result, it will take multiple decades for “adequate” formal institutional change to improve economic outcomes. While not their primary purpose, Dunn and Robertson (1973) make a similar distinction between the “formal procedural rules” and the informal management of transactions within the stool councils in their study of Brong-Ahafo region.

26

Informal Institutions and Citizenship in Rural Africa

more precise taxonomy of informal rules, but instead to highlight the blurred boundaries and interactions between rules that are more and less formal across different types of governance systems – that of the central state, chieftancy, lineage, village community, and so on. Building on insights from Catherine Boone’s work on the origins of state formation in West Africa, I argue against a distinctly “African” or even “Ghanaian” or “Ivoirian” pattern of state formation that would produce one common pattern of reciprocity.77 I agree emphatically with Boone’s point that the history of state formation varied in important ways across the continent, but even within countries across different subnational regions. The study compares the Brong-Ahafo region of Ghana with the Abengourou region of Côte d’Ivoire.78 My work engages with historical sources to unpack the causal mechanisms underlying institutional change.79 From my comparative historical analysis, I build a theory showing how particular aspects of the history of state formation might be expected to stimulate certain broad types of transformation in the informal institutions of reciprocity. (See Table 1.1.) The historical role that the state has played in mediating (sometimes magnifying) risk vis-a` -vis individuals, families, and communities in these regions shapes the transformation of the informal institutions of reciprocity in the contemporary period. The state mediates risk in three important and often unrecognized ways. To begin, the nature of the state institutions for political administration shapes informal reciprocity. Here, the initial formation of central state institutions to govern the colonial territories entails a normative construction of what the state role is vis-a` -vis the family and the citizen. The new formal political institutions of the central state did not necessarily crush or replace the preexisting rules wholesale.80 Rather, the new formal state institutions and informal institutions transformed each other over time in a dialectical process of everyday interactions

77 78 79

80

Boone (2003). Firmin-Sellers (1996) compares different communities in the same regions. This serious attention to history and subnational variation contrasts dramatically with work by Acemoglu, Johnson, and Robinson (2001) that neither investigates how colonial institutional legacies are transmitted over time, nor finds any significant variation across or within the entire continent of Africa. See also Acemoglu and Robinson (2006). Political scientist Young (1994: 77–140) theorized the colonial state as the “Bula Matari” or crusher of rocks, but I draw here on the perspective of legal anthropology. For useful reviews of legal anthropology’s contributions, see Moore (2001) and Merry (1988). See also BendaBeckmann et.al. (1988); Rudoph and Rudoph (1967); and Moore (1986).

Introduction

27

that differed in the two regions.81 Thus, in this area of Brong-Ahafo, Ghana, the more decentralized colonial and postcolonial administrative institutions have been built on the idea that the state only supplements preexisting extended family and community systems of social welfare.82 The result of these more-decentralized administrative institutions was a more diversified system of informal reciprocity. In contrast, in the Abengourou region of Côte d’Ivoire, the more centralized colonial and postcolonial administrative institutions stimulated higher expectations for the state to provide more extensive social support to individual citizens in smaller nuclear families. The result of this higher mediation of risk by the Ivoirian state was a more concentrated system of informal reciprocity. In this way, state formation is not only a political process in terms of negotiating the actual formal institutions for governance but also an intensely political process in constructing the normative conceptualization of what the state role should be vis-a` -vis a certain type of citizen, family, and community. The second aspect of the state’s role in mediating risk is perhaps more obvious and the least contentious of the three. Here, the actual day-to-day provision of infrastructure and delivery of social services to subsidize health and education and alleviate poverty is highlighted.83 Again, the conceptualization of the state role in mediating risk is broader here than a small number of development economists who have focused narrowly on how public “safety nets” might “crowd out” informal risk-sharing systems.84 I then argue somewhat counterintuitively that an historically more active state role in mediating social risk actually stimulates a higher level of lessdiversified relations of reciprocity. This is clearly not a simple zero-sum relationship in which state retrenchment or collapse leads automatically to a resurgence of informal networks and voluntary associations.85 The economic crisis that began in the 1970s has basically continued for more than 81

82

83 84 85

Colson (1974), for example, highlighted how communities strategically navigated among the constraints and opportunities of overlapping formal state and informal institutions. Dunn and Robertson (1973) also emphasized the conflict, contestation, and negotiation within and between “tradition” and “modern” over time in the Ahafo region of Ghana. Posner (2005) makes a similar argument showing how the British colonial institutions of indirect rule in Northern Rhodesia (now Zambia) constructed tribal identities and created incentives for rural and urban Africans to invest in these tribal identities. For example, the British government had specific policies to encourage the maintenance of ties between urban migrants and their rural “tribal” communities in order to ensure access to a safety net when wage workers were unemployed or retired. On Latin America, see Kaufman and Nelson eds. (2004). Dercon (2002: 155); Albarran and Attanasio (2002); and Attanasio and Rios-Rull (2000). See, for example, Chazan (1988); Azarya and Chazan (1987); Bratton (1994); and Dei (1992).

28

Informal Institutions and Citizenship in Rural Africa

two decades. And what were initially theorized as short-term coping strategies have evolved in different ways over time in different localities.86 In the Abengourou region of Côte d’Ivoire, state investments in social service infrastructure and delivery began later than in colonial Ghana, but were more stable and of a higher quality between the late 1960s until the mid-1990s. Because of the state’s more active role in mediating social risk, the informal institutions of reciprocity were less diversified in Côte d’Ivoire than in Ghana. Furthermore, Ivoirians in this region had higher expectations of service utilization and drew on their reciprocity networks more intensely when the state retrenched in the mid-1990s. In contrast, Ghanaians had much lower expectations of what the state should or historically had done since the late 1960s. Moreover, after a longer period of economic instability during the 1970s, many Ghanaians had already diversified and then depleted their informal reciprocity assets. The third component of the state’s role in mediating risk in the economy is also rarely considered in combination with the first two. I argue that the history of the state role to support production, mediate market volatility, and tax productivity shaped economic production decisions directly and the informal institutions of reciprocity indirectly. Thus, where states historically pursued highly extractive policies, individuals chose less land-intensive production strategies with shorter-term yields and subsequently diversified their investments in the informal institutions of reciprocity. This was the case in this region of Ghana. In contrast, in Côte d’Ivoire, the state historically pursued a more moderate level of taxation and facilitated labor in-migration that stimulated the expansion of long-term investments in cocoa and coffee production for export markets. As a result, the informal institutions of reciprocity in Côte d’Ivoire were more concentrated, but continued to link the younger and older generations within extended family systems. Overall, my argument highlights the need to examine historically all three of the preceding areas of state formation – political administration, social service delivery, and economic policy – in order to understand how the state’s role in mediating risk has shaped the construction and reconstruction of the informal institutions of reciprocity. Ironically, in the Abengourou region of Côte d’Ivoire, a local experience of a strongly centralized state that actively mediated risk in all three areas actually stimulated a higher quantity of reciprocal exchange, but more narrowly concentrated on the nuclear family. 86

Guyer, Denzer, and Agbaje (2002) critique this conceptualization of short-term coping strategies as inadequate for understanding the contemporary situation in urban Southern Nigeria.

Introduction

29

Trade-offs clearly existed in terms of the relative benefits of economic security versus political exclusion, trade-offs that will be explored in more detail in the next stage of the argument. Stage Two: How Variations in Informal Institutions of Reciprocity Shape Citizenship and Politics In stage two of my argument, I show how this reorganization of the informal institutions of reciprocity in turn shapes the ways that Ivoirians and Ghanaians conceptualized citizenship and practiced politics at the local and national levels.87 I hypothesize that differences in the history of public policies in three key areas – political administration, social service delivery, and agriculture – shaped and reconstituted local patterns of reciprocity, which then reshaped the boundaries of political community and exercise of citizenship. This indirect pathway through social relations reinforced the policies’ direct interpretive effects on ideas of political membership, local perceptions of government, and the meaning of indigenous civic obligations.88 The direct and indirect effects of state policies on citizenship in the two regions are depicted in Figure 1.3. The claim is significant in its linkage of social and political inclusion and exclusion, but it must be duly qualified. The hypothesis is not that the changing informal institutions of reciprocity are the only causal factors influencing local individuals’ views on their rights, duties, and avenues for participation in politics. Instead, the idea is that it is an extremely important and heretofore understudied set of factors in a longer causal chain. Thus, key differences in the historical patterns of state formation are influencing local practices of citizenship and politics through an indirect path of the transformation of informal reciprocity.89 In contrast with most of the existing literature on citizenship, I seek to investigate the very boundaries of the political community or communities (plural). I want to know not simply how the nation is imagined but also what is imagined by whom. Benedict Anderson’s work on the construction of 87

88 89

In her study of rural China, Tsai (2007) examines how informal institutions of moral obligation within solidary groups then increase the likelihood of public service provision where there is no avenue for democratic mechanisms of accountability. My work remains more focused on how these informal institutions of reciprocity shape the political participation of the village residents themselves rather than the political elites. See Pierson (1993) and Landy (1993) on interpretive effects of policies on mass publics. The direct path between the history of state formation and citizenship are explored by Mamdani (1996). He examines the effects of the bifurcated state on citizenship in Uganda and South Africa.

30

Informal Institutions and Citizenship in Rural Africa

nationalism has been hugely influential in highlighting the social construction involved in identifying as a larger national community.90 I certainly highlight the constructed nature of political identity, but I pay explicit attention to the development of political communities at other nonnational levels, for example, the construction of subnational and/or transnational communities.91 By expanding the focus beyond the parameters of the nationstate, we can also consider the possibility of simultaneous identification in more than one community. This builds explicitly on Peter Ekeh’s earlier theoretical work that highlights the existence of and tension between the primordial and civic publics.92 My hypothesis of linking state formation and changing patterns of informal reciprocity to divergences in indigenous conceptions of citizenship is again unconventional. Most of the existing literature focuses on the effects of more traditional political science variables such as the role of the state in guaranteeing rights (T. H. Marshall); the development of capitalism (Marx); the character of local government and voluntary associations (Tocqueville); and the rise of transnationalism or multiculturalism (Held and Archibugi; Kymlicka, et.al.)93 But in contrast with Marshall’s theory, social rights appear to have preceded civil and political rights in Africa, and then each set of rights has been, at various points, fully revoked or partially compromised. While civil and political rights were formally accorded to all Africans at independence in the early 1960s, a lot of political attention was focused on social rights; in particular, the right to obtain an education and to a lesser extent, health care.94 Less than one decade later, in many of the newly independent African countries, civil and political rights were quickly lost as military officers

90 91

92 93

94

Anderson (1983). On subnational identities, Posner (2005) examines how institutions, in particular, party system rules, shape which type of ethnic cleavage is mobilized, that is, language, tribal, religious, racial, and so on. On transnational political networks and identities, see Keck and Sikkink (1998) and Archibugi and Held (1995). Ekeh (1975). Marshall and Bottomore (1992). See Marx’s early essay “On the Jewish Question” (originally published in 1843) for a discussion of the distinction between the rights of man and the rights of citizen, in Marx (1978: 40–6). See also “Capital” (Marx 1978: 438); “Manifesto of the Communist Party” (Marx 1978: 486–91); and “The German Ideology” (Marx 1978: 170–4). On Tocqueville, see (Tocqueville 1945: 71; Tocqueville 1990: 115–19). On supranational citizenship, see Archibugi and Held (1995). On multiculturalism, see Young (1989); Kymlicka (1995); and Kymlicka and Norman (2000). See Ndegwa (1998: 358) on how different development paradigms of the 1960s and 1970s supported the expansion of social rights in Africa.

Introduction

31

launched coup d’états and established authoritarian regimes.95 In the current neoliberal era, many African countries have reinstated multiparty democratic systems where civil and political rights are officially guaranteed, but the retrenchment of the welfare state undermines the fiscal basis for social rights as they have been conceived in the past. Again, citizenship rights did not simply progress in one linear universal direction; instead, there were multiple contingent patterns. While this study pays careful attention to how reciprocity may bridge (in Ghana) or overlap with (in Côte d’Ivoire) class differences, the development of class consciousness does not appear to be the primary explanatory factor for differences in local citizenship identities and political practice, as Marx argued. Furthermore, the two cases share the same level of capitalist development and degree of incorporation into the global capitalist economy, yet we see striking differences in the ways local people perceive their political community and relationship to the state. Of the previous four potential rival explanations for Ghana and Côte d’Ivoire’s divergence in local citizenship and political practice, Tocqueville’s theory is the most difficult with which to contend. In fact, I have argued elsewhere that the nature of associational life does differ in important ways in these regions of Ghana and Côte d’Ivoire, and contributes to differences in the resolution of ethnic conflict.96 Again, my argument here is not that differences in the relations of reciprocity are the only determinant of citizenship. Rather, the variation in the patterns of reciprocity is a key factor that combines with the experience gained in local political associations and in local political institutions over time to shape indigenous concepts of citizenship. Finally, more recent theories about the growth of transnational institutions and/or the rise of multiculturalism since the 1980s do not explain the puzzling variations seen in Ghana and Côte d’Ivoire. Divergence exists during similar time periods of transnationalism and multiculturalism. Ghana and Côte d’Ivoire participate in similar fashions in regional and international political institutions. Furthermore, the makeup and political mobilization of cultural groups within both nation-states are roughly 95

96

The first postindependence coup d’état occurred in tiny Togo in January of 1963. Coups or military regimes were subsequently experienced in Benin and Congo-Brazzavile in 1963; Uganda and Gabon in 1964; Zaire and Benin again in 1965. But it was the military’s ouster of President Kwame Nkrumah in Ghana and President Nnamdi Azikiwe in Nigeria in 1966 that shocked the world. The rest of the 1960s and 1970s continued to be a period of great political instability in much of sub-Saharan Africa. MacLean (2004).

32

Informal Institutions and Citizenship in Rural Africa

similar. When ethnic groups have mobilized differently in Côte d’Ivoire, it is to demand political control of the nation, not special minority rights within the nation. These differences in ethnic politics are more tightly linked to the variation in the social patterns of reciprocity and exclusion than the politics of multiculturalism. My two-stage argument reveals the important causal linkages between the variations in the history of state formation, informal institutions of reciprocity, and citizenship. I reject the notion that the African state is uniformly weak or failed, and that informal institutions are the only rules that matter on the ground. I also reject the idea that the colonial state barely touched African societies or is no longer relevant to African cultures today. But I go beyond a simple story of path-dependent colonial legacies to focus squarely on the repeated interactions between African states and societies over time in the mediation of risk. I demonstrate in rich empirical detail the role of history and politics in this iterative process of informal institutional change.

research design and methodology Matched Case Research Design My arguments are developed by employing an innovative matched case design and using a combination of qualitative and quantitative methods. This research design compares the political history of two similar regions on either side of the Ghana–Côte d’Ivoire border. These comparative cases were very carefully selected in order to control for such key causal factors as culture, geography, economic production, poverty levels, and level of infrastructure. What is particularly unique about this design is the possibility of controlling for precolonial culture. These two regions both belong to the same overarching Akan ethnocultural group that was arbitrarily divided by the colonial border at the turn of the twentieth century.97 These precolonial cultural systems still remain highly salient in everyday life in the contemporary period and might be expected to affect local norms and informal institutions of social reciprocity. In addition to this shared culture, the cases selected share many other similarities on the international, national, regional, and village levels. First, both countries share similar positions in the international political economy. Ghana and Côte d’Ivoire both served as coastal trading posts for various European powers as early as the sixteenth century; experienced over a 97

See Dunning (2008) on the strengths and limitations of such natural experiments.

Introduction

33

t a b l e 1 . 2 . Similar Country Cases: A Comparison of Basic Indicators in the Late 1990s Basic Indicators

Ghana

Côte d’Ivoire

Population

18.9 million $1,661 Urban 34 Rural 36 1,900

14.7 million $1,881 Urban 39 Rural 33 1,955

62.6% 53 60 79% 71% 43.6 3.1

54.8% 53 48 75% 61% 43.9 2.7

GDP per capita based on PPP (1998) Gini coefficients, 1991–1996 Area under permanent crop production in thousands of hectares (1980) Rural population as percentage of total (1998) Life expectancy at birth (years in 1980) Life expectancy at birth (years in 2006) Primary school enrollment (1980) Literacy rate (2000) Age 0–14 as percentage of total population Age 65+ as percentage of population Source: World Bank (2001a, 2007); and UNESCO (2007).

half-century of European colonization (by the British in Ghana and the French in Côte d’Ivoire); gained independence in 1957 and 1960, respectively; continued the export of cash crops even after colonial rule; suffered from desperate debt burdens by the late 1970s; and were pressed to adopt World Bank/IMF economic reform by the early 1980s and multiparty political systems in the early 1990s. Second, Ghana and Côte d’Ivoire share many similarities at the national level.98 They are both relatively small countries with young, predominantly rural populations and similar indicators of poverty, social infrastructure, and social development.99 (See Table 1.2.) They also both have a similar ethnic and religious composition that overlaps with regional infrastructural inequalities. Thus, Ghana and Côte d’Ivoire both have a predominantly Akan (the most populous ethnic group), Christian South that has received greater investments in economic and social infrastructure in comparison with a predominantly non-Akan, Muslim North. 98

99

This cross-national comparison has a long history in the study of African politics. See, for example, Crook (1990) and Foster and Zolberg (1971). The one striking difference in social indicators in Table 1.2. is the declining life expectancy in contemporary Côte d’Ivoire. This will be explored further in Chapter 5 but most likely is caused by an increase in HIV/AIDS rates rather than a significant difference in social service infrastructure.

34

Informal Institutions and Citizenship in Rural Africa

Of course, one key difference at the national level is the level of political stability after independence. After Ghana became the first African nation to obtain independence in 1957, President Kwame Nkrumah was ousted by a coup d’état in 1966. From the mid-1960s to the mid-1980s, Ghana was devastated by a series of coup d’états and authoritarian regimes that practically collapsed the economy. In comparison, since its independence in 1960, Côte d’Ivoire’s economy and infrastructure had boomed during three decades of stable leadership under President Félix Houphouët-Boigny and his hegemonic PDCI. This difference will be explored in later chapters through the analysis of the experience of state policies over time. Third, the regions selected, the region of Brong-Ahafo in the western part of Ghana, and the department of Abengourou in the eastern section of Côte d’Ivoire, share many important commonalities. To begin, they are approximately the same distance (from four to six hours drive) from the national capital and thus have experienced similar levels of exposure to “Western” education, culture, and infrastructure that emanated from the coasts upward.100 They have analogous geographies, started exporting cocoa at similar periods, share similar precolonial political systems, and continue to be predominantly populated by the same Akan ethnic groups who maintain a loose political alliance to the current Ashanti kingdom. On a broader scale, these regions represent areas that are ethnically diverse with relatively long histories of involvement in cash crop agriculture for international export. Finally, the fieldsite villages were carefully chosen based on extensive discussions with regional and local government officials. Two villages from each region were selected based on how well they represented the larger region; as well as their similar population sizes; agricultural economies; and level of economic, social, and political organization. These villages were all relatively remote and somewhat difficult to access. They were between twenty minutes to an hour’s ride from the regional or even district capitals and at least three miles off any major paved road on a bumpy dirt road. Despite the distance and rough road, all the villages received “taxi” service several times daily, even during the rainy season. A matched case research design demands this sort of rigorous selection of fieldsites in order to control for many alternative explanations of the variations in informal reciprocity and citizenship. The study essentially controls for many aspects of ethnic culture, precolonial political history, levels of 100

This research design essentially controls for the main explanatory factors of modernization theory. See, for example, Lerner (1958) and Apter (1965).

Introduction

35

wealth and integration into the global economy, and levels of education and exposure to “Western” culture. In this way, the role of different patterns of state formation and histories of policymaking during the colonial and postcolonial periods are revealed and highlighted. While the data for this book remain limited to specific regions of each country precisely because these fieldsite villages and regions were chosen explicitly for their common precolonial history, culture, ecology, and economies, the differences that exist today can reasonably be supposed to represent important qualitative differences between the two countries. The fact that the villages chosen were such a distance from the coast and national capitals makes it even more interesting to witness distinct national differences between these two particular regions. In many places throughout the text, I refer in shorthand to the differences between Ghanaians and Ivoirians to facilitate the flow of the text for the reader. I am not making the claim that the patterns of informal reciprocity, citizenship, or everyday politics would look similar in all other regions of Ghana or Côte d’Ivoire. Other regions would vary in their precolonial culture, geography, and political economy as well as in their experience of state formation over time. Hence, those regions would undoubtedly have different starting points as well as different processes of institutional transformation. Multimethod Approach Qualitative and quantitative methods were combined to collect data on the experience of state formation and the nature of reciprocity and citizenship at the local level. In order to comprehend complex processes of social, economic, and political transformation over the past 100 years, the project required nearly two years of intensive fieldwork at the village level. I lived full-time in the four fieldsite villages in order to participate in and supervise all the data collection. While there was often no running water or electricity, the village residents did everything possible to make my stay welcoming and comfortable. In each village, it was often remarked that I was the first white person to have spent the night. My full-time presence allowed me to participate and observe the everyday activities of the village, including buying food at the village market or kiosks, walking to help collect water, attending producer cooperative meetings, paying my respects at village funerals, visiting the schools, eating at various villagers’ homes and local “chop shops,” attending all the different church services, going to weed a neighbor’s farm, and so on. This rich ethnographic observation was invaluable in three ways. First, my ethnographic observations yielded significant data in its own right;

36

Informal Institutions and Citizenship in Rural Africa

I discovered things by participating and observing that I did not learn by asking questions. Second, this ethnographic data helped provide the context for interpreting other data sources I was collecting. And, third, my participation in village activities facilitated a rapport with most village residents that undoubtedly increased the validity of my other data sources. Although I speak a basic level of Asante Twi (one of the Akan languages spoken in Ghana and closely related to Agni, the Akan language spoken in Côte d’Ivoire), I worked very closely with two research assistants in each country.101 Almost all the interviews were conducted in the respondent’s first language, usually Twi or Agni.102 Each day, I would take turns accompanying one of the research assistants while the other did their rounds independently.103 We lived and ate all our meals together, so I was able to closely monitor the work as it was completed. The research assistants not only helped with translating and conducting interviews but also participated in discussions about how to interpret what we were discovering. They also played critical roles throughout the field research in making sure that our entry and stay in the village did not create any problems or pose a burden to anyone.104 Much of the key data is drawn from a survey of approximately 100 individuals in each of the four fieldsite villages. I used a multistage stratified random sampling technique to select the samples in each village, which is explained in more detail in the Appendix. One explicit objective of this stratified sampling methodology was to ensure adequate numbers of women and non-Akan survey respondents because they could have systematically different patterns of reciprocity. In addition to the structured survey questionnaire, in-depth interviews and oral histories were conducted with a wide range of respondents, including but not limited to the village chief, village elders, queenmother and royal family members, local political leaders, local party representatives, teachers, health care providers (including traditional birth attendants and herbalists), 101

102

103

104

The assistance of Kweku Dickson, Faustina Sottie, and Ernest Appiah in Ghana and Célestin Mian and Fulgence Kanga in Côte d’Ivoire is very gratefully acknowledged. Each of my research assistants was able to speak more than one language so we were always able to communicate with any respondent. Many of the interviews with nonindigenous migrant farmers were conducted in Dioula or Hausa. The survey questionnaire recorded the particular research assistant who conducted the interview as well as whether I was present or not. No systematic interviewer bias was revealed in the statistical analysis. This is discussed in detail in Chapter 2. For example, the research assistants helped me make arrangements to ensure a safe water supply for our team. They also helped communicate our wish to provide plenty of extra money to the people designated to cook our meals.

Introduction

37

local religious leaders, local cooperative or other association leaders, youth leaders, and informal savings organizers or moneylenders. Extensive in-depth interviews were also conducted with officials from various government, nongovernmental, and donor organizations in the local, district, regional, and national capitals. I used a question guide for the in-depth interviews, but essentially followed the respondent’s cues and conversation flow. In each village, I also organized a women’s and a men’s focus group. The focus groups were separated by gender so that each group would talk more freely. Men and women often ate, worked, and socialized separately in the villages; and older men would tend to dominate public gatherings. The focus groups were generally held late in our three-month village stay so that we had a sense of who to invite, and the participants were more familiar with us. We invited a group of eight to ten people who had a diversity of roles in the village, for example, teacher, nonindigenous migrant farmer, local politician, elder or subchief, church leader, youth leader, wealthy farmer, and so on. We aimed to assemble a small group of people who we thought to be relatively lively, articulate, and well-respected in the village in order to create an amiable atmosphere for discussion. There were only a couple of participants who attended, but then surprised us by never seeming comfortable enough to talk. We held the focus groups either late in the afternoon or in the early evening after people had returned from the farm and immediately before or after supper had been prepared. The participation rate was approximately 75 percent, with participation slightly higher for men than women.105 One of the research assistants facilitated the focus group using a question guide, but again followed the flow of the discussion. The other research assistant taped the discussion and took written notes to aid in the subsequent transcription. Finally, to look at how reciprocity and citizenship have changed over time, I also analyzed archival sources and collected oral histories. The archival research was conducted in Accra, Sunyani, Kumasi, Abidjan, Dakar, Paris, and London. Reflecting the different colonial administrative approaches, most of the relevant British colonial records were more localized and available in Accra, Ghana, whereas almost all the French colonial records were centralized in the national repository in Paris, France. Since 105

The group meeting had a lower participation rate than individual interviews because they were less accommodating of last-minute changes in individual schedules and plans. I also think that some individuals were slightly more intimidated to participate in front of others in a group meeting. There was probably a gender bias in that more women were intimidated than men which partially explains the higher male participation. The other part of the explanation is that men generally had more free time at home.

38

Informal Institutions and Citizenship in Rural Africa

most of these archival sources were written from the perspective of the colonial rulers, I also made substantial use of oral histories from village residents. In this way, I could collect information about individual, family, and village responses to earlier periods of economic crisis, for example, the Great Depression of the 1930s or the brushfires of the 1980s.

plan of the book’s arguments My objective is to explain why reciprocity and citizenship differ so profoundly in two such similar regions of Ghana and Côte d’Ivoire. The next two chapters of the book illuminate the variation in the informal institutions of reciprocity. In Chapter 2, we gain entrée inside four different African villages in order to document quantitatively with original survey data important differences in the informal institutions of reciprocity in these regions of Ghana and Côte d’Ivoire. Despite a nearly identical cultural landscape, the chapter systematically documents puzzling contrasts in the levels of giving, extent of participation, and structure of social ties in informal social support networks among village residents in Ghana and Côte d’Ivoire. In Chapter 3, the differences in the local relations of reciprocity are explored using qualitative data and analysis. The chapter challenges accepted stereotypes about Africa – in particular, the vibrancy of the extended family system and cohesion of village reciprocity – by examining local discourses on the meaning of these concepts. The second section of the book focuses on how different histories of the state formation in the mediation of risk have transformed and stimulated the previous variations in informal reciprocity. Chapter 4 highlights how the variation in the degree of administrative centralization since the colonial era has stimulated differences in the structure of informal reciprocity in Ghana and Côte d’Ivoire. Chapter 5 traces differences in the institutional histories of state social service delivery to explain the surprising differences in the quantitative values of social support given in Ghana and Côte d’Ivoire. In Chapter 6, a more recent divergence in the structure of the relations of reciprocity in Ghana and Côte d’Ivoire is explained in terms of the state’s historical role in mediating economic risk through agricultural policies and changing local-level production strategies. The third section of the book analyzes the far-reaching consequences of contrasts in informal reciprocity for citizenship. Chapter 7 examines how these striking changes in the role of the state and informal institutions of reciprocity shape indigenous notions of citizenship and political participation. As the state redefines its role in political administration, the provision of

Introduction

39

social services, and the regulation of agriculture, the boundaries of family and community are contested and redrawn at the local level in different ways in the two cases. Finally, the book concludes by examining how this empirical study has helped build new theories with potential relevance beyond the boundaries of these small African villages in West Africa.

chapter 2 The Informal Institutions of Reciprocity: A Quantitative Puzzle and Analysis

entry into the four african villages For the first five to six hours of the journey from the capital cities of Ghana (Accra) and Côte d’Ivoire (Abidjan) to the fieldsite villages in the BrongAhafo and Abengourou regions, I traveled in relative comfort in a large, Greyhound-style bus. Once we arrived at the district or regional capital, my two research assistants and I were told to “wait small” to fill up a compactsize taxi (usually squeezed with four in the back and three up front). As we clattered down the pitted dirt road for the last half hour of our trip, I noticed the extra bolts and duct tape that kept various parts of the car secure and glimpsed the ground below through several small holes in the floorboard. I supposed it was in the interest of fuel economy that the engine was turned off, and we coasted down a couple of the hills on our way. None of these things seemed interesting, or even noteworthy, to my research assistants as they quietly scanned the horizon for signs that we may have entered our first fieldsite. Of course, this “official arrival” beginning in October 1998 was not the first time I had visited each of these fieldsite villages. On a scoping trip a year earlier (during the spring and summer of 1997), I had followed the chain of command from the capital cities of the nation, region, and district to gather information about the differences among the various regions, districts, and villages; and to obtain official permission to visit and perhaps later conduct my study. Once the districts had been chosen, I visited several villages before settling on two similar communities that seemed fairly representative of that district or region as a whole. I chose to conduct fieldwork in a smaller number of villages in order to maximize the reliability of the data I collected 40

The Informal Institutions of Reciprocity

41

as well as my ability to interpret it by spending a longer time in each village.1 This smaller number of village cases made it impossible to use a random sampling strategy, so I instead made a careful purposive selection. After I returned to the United States, I exchanged letters with the appropriate authorities in each village to request their approval to return the following fall (1998) and spend three months in their village conducting my research. As we rolled into each of the villages, we were greeted first by throngs of children who crowded around the three of us. Some were grinning and eager to meet all these “strangers,” especially me, the young, white American woman.2 Others were more guarded but still curious. And then a few were absolutely petrified but had clearly been dragged to see their first “ghost” by older brothers or sisters who squealed with delight as they struggled to wriggle away. Within an hour of our arrival, we brought our gifts of “Southern Comfort” whiskey (from “my people”) and the more familiar bottles of schnapps to our official meeting with the village chief and elders.3 After the libations were poured to the ancestors, we discussed the purpose and logistics of my visit. In particular, there was great concern expressed how I might cope with a lack of electricity and/or running water, how I would cook, what foods I would eat, what water I could drink, and where I would bathe and go to the toilet. In each village, an agreement was reached that the news of my small team’s arrival and the official approval of our research should be spread throughout the village as soon as possible. The next morning, in Ghana, we awoke before 5 A.M. to the clang of the “gong-gong” beater making his rounds with the news before the village residents had left for farm. Despite the fact that we had already been officially permitted to enter the village and carry out the study, we did not begin our survey and other interviews immediately. The first two days, while taking a census of housing

1

2

3

The fieldwork included a three-month stay in each of four villages total. While I share the perspective of many anthropologists that this was a relatively short period of time to try and understand issues that were complex, personal, and in the process of changing, I felt that the benefit I gained from a controlled comparison was worth any loss in additional contextual detail. The concept of a “stranger is someone who arrives in the village but originates from another place. Even Ghanaians from another geographic area or sub-ethnic group are considered “strangers.” The person could be passing through town momentarily or could settle for a more extended period of time and still remain a “stranger.” For example, my two Ghanaian research assistants were considered “strangers” even at the end of our three-month stay. In both of the Ghanaian villages, these meetings were public and relatively well-attended by a diverse group of village residents. In contrast, in Côte d’Ivoire, the meetings were more private with a small, limited attendance.

42

Informal Institutions and Citizenship in Rural Africa

units and drawing a map of the village, we went door to door, greeting and introducing ourselves to each of the village residents in their homes in the traditional idiom. Those villagers who might have been away in the fields when we came by were swiftly told by their neighbors and others that we had made the effort to stop by. It was also greatly appreciated that I personally greeted and introduced myself in the local language of Ashanti Twi, even if later I relied on my research assistants to translate more challenging and nuanced conversations.4 Very quickly, the warm welcome I had received since my arrival was deepened, and I came to be known affectionately by everyone throughout my stay as “Sister Akua.”5 At this point, it is important to be clear that I was always an outsider in these four village communities, and a highly visible one at that. But with time and effort, respondents seemed able to trust me enough to share more candidly their thoughts and experiences. While perhaps not as dangerous as discussing human rights violations, our conversations still involved talking with me about vulnerability, hard times, exclusion and conflict within their families, between friends and in the larger political and economic system. Villagers often spoke about sorcery and witchcraft, too. Albeit exceedingly rare for most political scientists, it is essential to provide the preceding information about the context of the data collection.6 The considerable time spent negotiating my entry into the fieldsite communities and developing a rapport with my respondents strengthened the validity of the data I was able to obtain.

a picture of the four villages Beyond facilitating the sampling and building rapport with interview respondents, the time dedicated to conducting a census of housing units was valuable for developing a picture of each of the villages as a whole. This was important to verify that the villages were approximately similar and representative of the region, as anticipated during the case selection process.

4

5

6

In the Ivoirian villages, the Akan language spoken was Agni but almost everyone spoke or understood Twi. The Akan have one first name after the day of the week of their birth, thus Akua is for women born on Wednesday. My last name became the family name of my host family in each village. Often in interviews, however, respondents would substitute their own family name to “Sister Akua” to further strengthen our imagined familial ties. Anthropologists have long discussed their “positionality” and their personal role in the research process much more candidly and systematically than political scientists. A useful exception is the chapter by Shehata (2006).

The Informal Institutions of Reciprocity

43

Each village was at least three miles from any major, paved road with a bumpy, dirt road that ran through the center. Compact sedans and, less often, a larger van would ply the route between each village and the regional or district capital several times a day. This meant that transportation was available, but usually required waiting at least an hour or more for the car to arrive and fill up, depending on your luck. The waiting area for the taxi was in most cases right next to the village market in which in the early morning a handful of women sold vegetables, dried fish, bread, and small sachets of oil, sugar, salt, and soap from little tables. Each village also had a number of small covered kiosks or boutiques that sold a slightly more expanded number of household items such as matches. At the time of the fieldwork, the two Ghanaian villages had no electricity but were on the verge of being connected to the power grid. In one of the Ghanaian villages, one village resident, who had worked abroad in Italy, owned a generator that was used weekly to show videos on his television to crowds of adults and children who brought their own wooden benches to sit and watch in the middle of an open field. The two Ivoirian villages both had electricity, so wealthier households enjoyed refrigerators and sometimes fans in their homes. Although one of the Ivoirian villages was on the cusp of receiving a telephone line at my departure, during the time of the fieldwork there was no phone service, cellular or land-based, in any of the fieldsite villages. Each village had at least two elementary schools and one middle school. The schools were generally located on the edge of the village. None of the villages had a full-scale clinic within its boundary, but they all had one within five miles, which was considered walking distance if necessary, and a hospital within one hour taxi travel time. The fieldsite villages ranged in population from approximately 1,500– 1,800.7 There were between 120–240 housing units in each village. There were slightly more women than men living in the Ghanaian villages, and slightly fewer in the Ivoirian villages. In this area and with these ethnic groups, polygamy was an ordinary practice with between 13–20 percent of respondents reporting that they were in a polygamous marriage. On average, polygamous husbands reported having two wives, but in rare instances, a very prosperous man might have more than two. In the Akan system, wives usually left their families and moved to the village of their husbands. The sampled respondents (who had any children) counted an average of 4.6–5.4 children. The average age of the sampled respondents was thirty seven in 7

An exact census of the population was impossible to obtain in any of the villages.

44

Informal Institutions and Citizenship in Rural Africa

Ghana and Côte d’Ivoire, but this is a rough estimate given the difficulty for some people in remembering their age. The average age of the village population is of course much younger due to the large number of children under eighteen years of age in each household that were not counted during the sampling procedures. The architectural styles of the houses varied from large concrete villas to tiny mud-and-thatch huts. Some houses were large, multihousehold compounds, and others were single-household cottages. The cooking area was usually in the middle, open-air courtyard of the compound or by the side of the house. Often husbands would give each of their wives separate “chop money” from which to supplement their homegrown produce and cook their family’s meals. Thus, multiple households might cook in the same place but not in the same pot. Cowives sometimes worked together, and at other times, they worked independently. Even a wife and her husband may not necessarily pool their resources into a blended nuclear family account but instead each parent would have separate income streams and be responsible for particular expenses for the family and children.8 The idea of pooling is further complicated in these regions by the fact that the predominant Akan ethnic group is matrilineal, where the line of descent is traced through the mother. The abusua is the extended family, whose members descend from a common female ancestor. The abusua ancestors actually own the land, and the descendants occupy and take care of that which is passed down. According to Akan tradition, abusua property is thus inherited from a deceased father to his eldest living brother, or then to his sister’s son (n.b., not to the father’s children). In the Akan system, therefore, uncles and nephews have particularly strong ties. For all these reasons, I do not assume pooling within a certain kind of household, but instead investigate the patterns of giving and receiving among individuals.9 Houses in the village were arranged according to the curves of the landscape, but in roughly a linear pattern, with dirt roads and paths shooting off from the main road. In each of the fieldsite villages, a significant migrant population existed. The sampling procedure distinguished only the nonAkan migrants, primarily Muslim individuals originating from the northern regions of Ghana and Côte d’Ivoire; or from the Sahelian countries farther north, such as Burkina Faso, Mali, and Niger.10 Hence, the data suggest that

8 9 10

See Goldstein (1999) on intra-household and individual social insurance mechanisms. See Moore (1992) on gender relations within the household in Africa. Some residents belonged to Akan groups from other regions of Ghana and Côte d’Ivoire, for example, Fanti from the coast of Ghana or Baoule from the center of Côte d’Ivoire.

The Informal Institutions of Reciprocity

45

non-Akan migrants accounted for between 11–21 percent of the village populations in Ghana, and between 25–37 percent in Côte d’Ivoire. The higher percentage of non-Akan migrants in Côte d’Ivoire is to be expected given the country’s historically more open immigration policy, the implications of which are discussed in more detail in Chapter 6. While some of the migrants may have arrived within the past year, many had lived in the village for decades and did not return to their hometown or country of origin very frequently. These long-term migrants often owned their own houses or compounds in a neighborhood of mostly foreigners known as the “zongo” on the edge of the village. More-recent migrants often rented single rooms in larger Akan compounds or lived in the farm cottages close to the Akan-owned plantations where they worked. The four villages were selected originally because the two regions were both known as long-time cocoa producers. According to national policymakers and local farmers, cocoa had started first as a cash crop farther south and then been adopted in these regions sometime around the 1920s or 1930s. By the late 1990s, these regions’ cocoa plantations were rather old, and the new cocoa frontier had moved farther southwestward in both countries. Even though policymakers and farmers still described both regions as big cocoa producers, to my surprise, I soon discovered through my survey research that many of the Ghanaian farmers had shifted out of cocoa to produce other crops, most notably tomatoes. The dynamics and implications of this unexpected and dramatic change in cash crop production, occurring in Ghana but not in Côte d’Ivoire, is also discussed later in Chapter 6. Among farmers, there was a hierarchy that extended down from big plantation owners, through smaller farm owners, to family members working on a relative’s farm, to migrant sharecroppers, and to day laborers. Finally, it is important to describe the political and social hierarchy found in the four villages. In each village, an Akan chief, subchiefs, and a council of elders met at least occasionally to discuss village matters and resolve problems. The queenmother also participated in village governance to varying degrees. Akan village residents could be considered a royal, commoner, or slave. These designations were not usually discussed openly, particularly the status of slave, but were known to everyone and still considered salient. The non-Akan migrant populations also had at least one chief, who was clearly subordinate but communicated to different degrees with the village Akan chief. In addition to the institutions of village chieftaincies, the Ghanaian villages had unit committees, in which village residents of diverse ethnicities, ages, genders, and socioeconomic status were elected to govern village affairs. Serving as a unit committee member or officer unquestionably

46

Informal Institutions and Citizenship in Rural Africa

conferred a higher status on those villagers. There was no such village-level representative body in the Ivoirian villages. The headmaster and teachers were also usually held in relatively high esteem, except that they were frequently “strangers” from other ethnic groups and hometowns so never truly incorporated into the local structures of power. The only other relevant local organizations were a handful of voluntary associations, in particular, numerous active churches in Ghana and several producer associations in Côte d’Ivoire.11 The proliferation of churches within the past two decades in Ghana was due to the rise of evangelical and charismatic Protestant churches that drew congregants away from the historically dominant mainline Protestant churches. Following the French model, the Catholic church was dominant but not very active in the Ivoirian villages. In addition to their participation in Christian and Islamic religions, many village residents simultaneously maintained indigenous religious beliefs in the power of ancestors and a pantheon of gods and spirits. Overall, in spite of a few unanticipated characteristics that will be explored later, I was able to confirm that the four fieldsite villages were relatively typical for the region and approximately similar to each other.

design and administration of survey to investigate informal institutions of reciprocity Sampling of Individual Village Residents Having explored the villages overall, the next step was to choose the individual village residents who would participate in the survey component of the study. This survey allowed me to take a snapshot of the informal institutions of reciprocity of a representative sample of village residents. With a survey sample size of approximately 100 adult individuals from each village (400 total), I could use statistical analysis to explore intravillage and crossregional variation, but still ensure ample time to conduct in-depth interviews, focus groups, and observations both in the village and the district or regional capitals. To choose 100 individuals in each village, I used a multistage, stratified, and random sampling technique. (See the Appendix for additional notes on sampling and statistical data analysis.) The total population was the adult (eighteen years old and over) village population, which included some remote settlements on the village outskirts of largely

11

MacLean (2004) compares these local voluntary associations and political culture.

The Informal Institutions of Reciprocity

47

nonindigenous migrant workers who were considered by village residents to be a constituent part of their village.12 Measurement of the Outcome to Be Explained In order to measure the variation in the informal institutions of reciprocity, I documented individuals’ actual behavior in giving and receiving informal social support. In essence, the strategy was to reveal the underlying rules about exchanging help indirectly by analyzing the actual transfers made by respondents to every conceivable type of social tie.13 After extensive discussions with other researchers and reviews of relevant questionnaires,14 a survey instrument was designed that asked respondents what help they had received and given over the past twelve months with all types of social relations in order to document the actual patterns of local-level exchange.15 I use the self-reported value of help given for hospital fees, medicine, school fees, school supplies, and clothing by respondents to their family, friends, and others as the outcome to be explained in the quantitative analysis. Of the more than seventeen different types of help investigated, these five types were generally the most memorable and reliable because the crisis was life-threatening, the gifts were given or received at the same particular times of year (school start, harvest, Christmas, and/or Easter), and (except for clothing) the gifts were almost always given in cash.16

12

13

14

15

16

One element that was routinely mentioned as justification for a settlement being a part of the village was that the settlement’s children walked every day to attend the village school. The survey codes included: spouse; children; mother/father; grandparents; sister/brother; cousin; aunt/uncle; other extended family member; friend; association member; church/ religious organization member; government; village chief; agricultural worker; stranger; and other. Reviews of survey questionnaires used by Goldstein (1999) and Adams (1991a, 1991b) were particularly helpful, as was the work of Osseo-Asare (1991). Personal discussions with Chris Udry and in particular Markus Goldstein as he left the field in Ghana in 1998 and with Frank Hirtz and the von Benda-Beckmanns were also extremely valuable. While social relations of reciprocity are not reducible to monetary exchanges, transfers of cash or in kind value are believed to be leading indicators of the depth and extent of relationships. It was not feasible in the scope of this project to collect data on other types of social support, such as time spent visiting, counseling, or helping others to find work for example. Osseo-Asare (1991) does collect and analyze a wider range of support in her work, but the project relies primarily on frequency counts and thus, it is not possible to segment the data with regard to the reasons for transfers of time or money, the exact social category of the giver or receiver, or the amounts of time or money as is done in this study. Even though clothing was almost always given in kind, particular kinds of cloth had set prices that were commonly known and easily remembered.

48

Informal Institutions and Citizenship in Rural Africa

It is worth pausing here to emphasize that the survey data analysis focused only on the giving side of the exchange. While the dependent variable for the book as a whole is “reciprocity,” I cannot analyze this concept quantitatively. As detailed in the introduction, reciprocity is a give and take, but not necessarily between two discrete people. The reciprocal exchange is often diffuse over a large number of geographically distant people (i.e., not contained within the parameters of just one village) and generalized over a long period of time. In the survey, I did collect both help given and help received, but I do not conceptualize the relations of reciprocity as a discrete network that can be mapped in its entirety at any particular point in time. In subsequent chapters, I will use qualitative data to capture the dynamic and generalized aspects of reciprocity that are not accurately reflected in this quantitative snapshot. But for our purposes now, the closest and most reliable quantitative measure available is “help given.”17 The second issue worth highlighting here is that the concept of different types of family systems (i.e., nuclear versus extended family systems) is not indigenous to the areas studied. In initial interviews and pretests, Ghanaians and Ivoirians struggled to conceptualize divisions between family members or even comprehend such categories as “nuclear” or “extended.” In the local Akan language in both countries, the word “abusuaa” technically refers to the person’s lineage (from which spouses are excluded), but is commonly used to mean a larger extended family that includes family members from both spouses.18 No Akan word exists for a smaller “nuclear” family unit. In addition to being culturally bounded, the composition of the “nuclear family” also changes over the life cycle. In particular, because the survey sample included respondents ranging in age from eighteen to eighty years of age, a respondent’s parents or siblings could be of different relative importance depending on his or her age. In order to deal with concepts that do not originate in the local context and change over the life cycle, I investigate empirically how village residents define the strength and lines of inclusion for various family members by asking for the specific social ties (i.e., child or aunt) rather than imposing my preconceived “Western” aggregate categories on the respondents. Based on the responses in the survey and semi-structured interviews, I employ throughout the book the narrowest “Western” concept of the “nuclear family,” which includes only the respondent’s spouse and children. Thus, a reference to the “extended family” could potentially include the respondent’s parents, siblings, grandparents, aunts, uncles, cousins, and other family members. 17 18

See Appendix for discussion of why recall of help given was more reliable than help received. Nukunya (1992: 11).

The Informal Institutions of Reciprocity

49

conventional wisdom disconfirmed with survey data: informal institutions of reciprocity differed across the ghana–coˆ te d’ivoire border The following analysis of survey data reveals a quantitative puzzle that disconfirms the conventional scholarly and popular portrayals of the African village. I found surprising differences in the patterns of reciprocity in culturally similar villages across the Ghana–Côte d’Ivoire border. In many of my discussions and interviews prior to fieldwork, academics and policymakers anticipated that given such tremendously similar ethnocultural, social, and ecological landscapes, the informal institutions of reciprocity would be very similar. Even several village residents echoed this “conventional wisdom,” joking that their neighbors across the border were part of “the same family” or “one people,” and essentially questioned the logic of the study’s research design. The very fact that there are significant differences between the countries supports my earlier point that these norms and informal institutions of reciprocity should be problematized and not simply assumed to exist in some idealized form. The survey results also refute the conventional wisdom concerning the vibrancy of local reciprocity, particularly in rural areas.19 The village residents’ local safety net had weak threads and patches of exclusion. But, even if these informal institutions of reciprocity were weak, they nonetheless did still continue to exist. Contrary to other predictions in the literature,20 they had not been totally destroyed, but instead were being reshaped by the central state. So how did these informal institutions of reciprocity actually differ? They differed remarkably in terms of (1) the level of social support exchanged; (2) the percentage of village residents participating at all; and (3) the extent to which transfers were diversified among a wide variety of social ties. Overall, the relations of reciprocity in this region of Ghana were relatively shallow but diversified among friends, whereas those in this region of Côte d’Ivoire were deeper but more concentrated on giving to the nuclear family. Higher Values of Giving in Côte d’Ivoire than Ghana The first and most striking difference in the relations of reciprocity was that the quantity of help given to the nuclear family, extended family, friends, or “others” was much higher in this region of Côte d’Ivoire than in this region

19 20

See Chazan (1983) for example. Scott (1976), for example, suggests that these networks might be eroded dramatically.

50

Informal Institutions and Citizenship in Rural Africa

of Ghana.21 A comparison of the mean value of help given (hereafter understood to be help given for hospital fees, medicine, school fees, school supplies, and clothing unless otherwise specified) to all social ties reveals that annual giving in this region of Côte d’Ivoire ($117) was more than four times higher than giving in this region of Ghana ($29). To a reader from an advanced industrialized country, these amounts may not seem outstanding in absolute terms. But, when median agricultural incomes ranged from $333 in the Ivoirian fieldsite region to $346 in the Ghanaian one, this allocation to help others was substantial.22 Thus, the average respondent in the Ivoirian villages gave more than 35 percent of their annual income while the average respondent from the Ghanaian villages gave more than 8 percent. This giving was also occurring during a period of economic hard times when “there was no longer any money”. In focus group discussions and in-depth interviews, men and women in both regions unanimously agreed that “life is hard.” They explained their lack of means as due to the fact that the cost of living had increased while prices for their farm produce had decreased. In this way, the quantitative level of giving was both absolutely low and relatively impressive at the same time. These striking differences in levels of giving were not merely due to differences in purchasing power parity between the two countries. In fact, all the figures in the dataset were already adjusted for purchasing power parity, meaning that the Ghanaian values reported were actually increased by 2.1 from their original levels.23 The differences were also not due to differences in the rural cost of living in this region of Ghana versus this region of Côte d’Ivoire that might not be included in the World Bank’s national-level dataset. Using price surveys conducted in village and regional markets in Ghana and Côte d’Ivoire, the study analyzed the costs of hospitals, medicines, school fees, school supplies, and clothing; and confirmed that any existing price differentials for these types of help did not, on balance, equal the large difference in levels of giving between the two cases. When respondents were asked in another survey question to estimate what portion of their overall monthly income they usually gave in help to 21

22

23

“Others” includes: an association; church; government; village chief; agricultural worker; landlord; “stranger”; and other. The median agricultural income is used here due to the presence of a few very large producers that skew the mean in both countries. The mean agricultural income was $614 in Ghana and $963 in Côte d’Ivoire. The study uses the World Bank’s (2001b: 44) national-level data on consumer prices to calculate the purchasing power parity ratio.

The Informal Institutions of Reciprocity

51

others, the median percentage value for Ghanaian respondents was 10 percent, while the median value for Ivoirians was 30 percent. This general estimate thus confirms the specific behavior reported: residents of this region of Côte d’Ivoire tended to give more than three times more in value every month or year than from the culturally similar, neighboring region of Ghana. Higher Extent of Participation in Reciprocal Exchange in Côte d’Ivoire than in Ghana In addition to higher values of giving in Côte d’Ivoire, the informal institutions of reciprocity were more extensive, involving a much wider section of the population in Côte d’Ivoire than in Ghana. Thus, not only did the average Ivoirian from this region give a higher value of help to all social ties; many more Ivoirians gave at least something. While more than 94 percent of all respondents in the Ivoirian region reported helping someone (with any type of help), only 61 percent of respondents in the Ghanaian region helped anyone at all. The figures for help with hospital fees, medicine, school fees, school supplies, and clothing indicate even more dismal rates of participation in reciprocal relationships by Ghanaians (35 percent) in comparison with Ivoirians (81 percent). Several respondents, when asked why they did not give any help to anyone, explained that no one came to them because “they know that I have nothing on me.”24 Others described how their lack of ability to give help prevented them from receiving help that they might have needed. One nineteen-year-old single Akan male described the implicit, long-term reciprocity involved in receiving help by acknowledging that he did not receive help because “they know that I have nothing to give.”25 Thus, it is the high number of respondents in this region of Ghana that were basically not participating in any social exchanges, the “non-givers,” that reduced the earlier mean values of giving by all respondents. An added comparison of the mean value of help given by those respondents who were able to give anything closes the gap, but still shows that Ivoirians from this region gave more than Ghanaians. Of those who gave something, Ivoirians gave a mean value of $144, still 1.7 times higher than the mean value in Ghana of $84.

24

25

Survey interview with a 60-year-old widowed Akan female (anonymous) by author. Makwan, Ghana. Similar quotes in survey interviews in Barima, Ghana. Survey interview with a 19-year-old single Akan male (anonymous) by author. Makwan, Ghana.

52

Informal Institutions and Citizenship in Rural Africa

More Concentrated Structure of Giving in Côte d’Ivoire than in Ghana Finally, in addition to being higher in value and more extensive, reciprocity had a different structure in this region of Côte d’Ivoire than in this region of Ghana. The total amount of giving was split differently among the various social categories of the nuclear family, extended family, friends, and others in the two regions – with giving more concentrated in the Ivoirian region than in the Ghanaian one. To begin, while most people gave the majority of their help to their nuclear family, this happened to an even greater degree in Côte d’Ivoire than in Ghana. Thus, Ivoirians as a group gave almost 70 percent of their total help to the nuclear family and only 24 percent to the extended family, whereas Ghanaians gave 46 percent to the nuclear family and 35 percent to the extended family. The pattern of a more concentrated network in Côte d’Ivoire and a more diversified one in Ghana is already beginning to emerge. The percentage figures are slightly different when one examines the average individual allocations instead of the proportions of the total group. Thus, average Ivoirian individuals split their help almost evenly between the nuclear and extended family, giving 45 percent of their total help to the nuclear family and 40 percent to their extended family; whereas average Ghanaians gave a little more than 14 percent to the nuclear family and 51 percent to their extended family. Again, because the extended family encompasses a more numerous and wider range of social ties, these allocations indicate that the social networks were more diversified in Ghana than they were in Côte d’Ivoire. Although the previous percentage values are insightful, they should be kept in perspective. Just because it appears that the nuclear family was more dominant than the extended family in Côte d’Ivoire than in Ghana does not necessarily mean that the extended family was weak in Côte d’Ivoire. It should be remembered that the absolute values in help given to both the nuclear family and to the extended family in Côte d’Ivoire were higher than the nuclear or extended family in Ghana. The analysis disaggregating this broad category of the extended family is also revealing. In both regions, giving to siblings was the biggest proportion of help given to the extended family, much larger than giving to parents or grandparents. There are some important crossregional differences to investigate here, however. First, giving to siblings was even more pronounced in the Ghanaian region (17 percent) than in the Ivoirian one (11 percent). And second, giving to parents (in absolute and percentage terms) and grandparents was higher in Côte d’Ivoire than Ghana. These

The Informal Institutions of Reciprocity

53

figures seem to indicate a pattern of stronger intergenerational solidarity in this region of Côte d’Ivoire than in Ghana, which will be explored in more detail in Chapter 6. One notable difference between these two regions was the frequency of giving and receiving to aunts and uncles. Both regions in Ghana and Côte d’Ivoire were predominantly Akan and thus matrilineal societies, in which the relationship between uncles and nephews was traditionally very strong. Ironically, however, in both regions of Ghana and Côte d’Ivoire, the average frequency of giving help to aunts and uncles was very low: only 2.6 percent. One might anticipate that uncles would give to their nephews more than they would receive. For instance, uncles often serve as patrons, paying for their nephews’ school fees and school supplies, giving cloth as a Christmas gift, or even paying hospital or prescription costs in the case of a health crisis. This definitely seems to be the case in Côte d’Ivoire (16 percent), but, interestingly, much less so in Ghana (2 percent). This data reinforce the notion that intergenerational reciprocity was stronger in this region of Côte d’Ivoire than in Ghana. Another important variation in the structure of giving in Ghana and Côte d’Ivoire concerns the help given to friends. Ghanaian respondents allocated a larger percentage of their total help to friends than did Ivoirians. Again, this does not mean that friends were insignificant in Côte d’Ivoire because the absolute values of help given were actually slightly higher for the total sample of all respondents in Côte d’Ivoire. Only that, as a proportion of the total amount of help given, friends seemed to be relatively more important in Ghana than in Côte d’Ivoire. This finding will be reinforced in Chapter 3’s qualitative analysis where Ghanaian respondents cited friends as sources of and outlets for help more frequently in open-ended questions, informal interviews, and focus groups than Ivoirians. As one Ghanaian man summed up the typical response: “It’s all to do with friendship. I go to my friend when I’m in need.”26 Meanwhile, in Côte d’Ivoire, friends generally do not take precedent. As one woman explained, “If this were a friend [who needed help], I would refuse. Because my family is more important than my friend.”27 The same appears to be true for the category labeled “others.” For the total sample of all respondents, the average Ghanaian individual gave nearly four times more help to “others” than the average Ivoirian.

26

27

Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Women frequently cited friends as their primary source of help after their husband. Survey interview with a 30-year-old married Akan female (anonymous) by author. Kyere, Côte d’Ivoire.

54

Informal Institutions and Citizenship in Rural Africa

How Reciprocity Differed for Women, the Non-Akan, and the Poor, in Ghana and Côte d’Ivoire Having portrayed the surprising overall distinctions in the level, frequency, and structure of giving in such similar regions of Ghana and Côte d’Ivoire, I now explore how reciprocity differed for women, the non-Akan, and the poor in these areas. Reciprocity Among Women in Ghana and Côte d’Ivoire Overall, in each region, women had different patterns of reciprocity than men.28 But importantly, these gender differences varied in significant ways between the regions. This again suggests that there were more fundamental differences between the countries that went beyond the gendered patterns explored in the following sections. First, women gave less help on average than men in both regions of Ghana and Côte d’Ivoire. The mean amount of help given by women in the Ghanaian villages was only 34 percent of the help given by the average man. The average help given by women in the Ivoirian villages was more than in Ghana but still only 44 percent, less than half, of what was given by the average man in that region. These differences were statistically significant in Côte d’Ivoire and borderline in Ghana.29 Women were also less likely than men to give large quantities of help. For example, women were only between 25–27 percent of those respondents who gave more than $200 of help in both the Ghanaian and Ivoirian villages.30 Second, in both regions, slightly more women than men gave no help at all. For example, of those respondents who gave no help, 57 percent were women in the Ghanaian villages, and 54 percent were women in the Ivoirian villages. While the number of women not giving anything at all was higher than expected given the sample, these differences were not statistically significant. While help given is the primary outcome explored in this chapter, it is worth noting that women were less likely to be excluded completely from receiving help than men. Fewer women than expected

28 29

30

Manuh (1994). The results of an independent sample t-test comparison of means was p= 0.105 in the Ghanaian region, and p= 0.007 in the Ivoirian region. There was only one woman who gave more than $200 in Ghana so this did not facilitate a statistical test of significance. In Côte d’Ivoire, a cross-tabulation shows that fewer women than expected gave more than $200 of help and that this result was statistically significant (Pearson chi-square= 4.60*).

The Informal Institutions of Reciprocity

55

received $0 of help, and this difference was statistically significant in both Ghana and Côte d’Ivoire.31 Not only was the quantity and extent of participation different for women in these regions, so was the structure of the help given. In Ghana, women were even more diversified than men, giving a larger percentage of their total help to their extended families, friends, and others. In Côte d’Ivoire, women were also more diversified in their giving than men, but they concentrated their help given on their extended families in particular and did not emphasize friends as much as in Ghana. These findings support Markus Goldstein’s previous work in another region of Ghana that suggests that women have different patterns of reciprocity than men and are not necessarily pooling resources exclusively within the household.32 Reciprocity and the Non-Akan in Ghana and Côte d’Ivoire Similar to the previous analysis of gender, the patterns of reciprocity between Akan and non-Akan respondents varied in important ways. But, again, these differences between ethnic groups varied in significant ways between the regions. Generally, the ethnic differences were statistically significant more frequently in Côte d’Ivoire than in Ghana. These results indicate that there were more fundamental differences between the countries that went beyond the ethnic variations explored in the following sections. First, perhaps not surprisingly, non-Akan respondents gave less help on average than the indigenous Akan in both countries. In Ghana, non-Akan respondents gave 24 percent of the help given by Akan. In Côte d’Ivoire, nonAkan respondents gave more but still only 37 percent of what the average Akan respondent gave during that year. These differences between Akan and non-Akan were statistically significant in both the Ghanaian and Ivoirian regions.33 Like women, non-Akan were also less likely to receive large quantities of help of more than $200 for the year in both countries but the result was statistically significant only in Côte d’Ivoire.34 Second, slightly more than expected non-Akan gave no help at all in both countries. Only in Ghana was this difference statistically significant however.35 When it comes to receiving help, non-Akan were more likely to be 31

32 33 34 35

In a cross-tabulation of $0 help received and gender, Pearson chi-square= 3.98* in Ghana and 15.42*** in Côte d’Ivoire. Goldstein (1999). The independent sample t-test results show p= 0.02 in Ghana and p= 0.001 in Côte d’Ivoire. In Côte d’Ivoire, Pearson chi-square= 11.85***. The result of cross-tabulation of zero help given and non-Akan is Pearson chi-square= 3.94*.

56

Informal Institutions and Citizenship in Rural Africa

excluded absolutely from receiving help as well. These differences were statistically significant in both Ghana and Côte d’Ivoire. This result stands in contrast to women who tended to receive more help than men. Third, the structure of help given by non-Akan respondents paralleled in important ways the overall variations found in these regions of Ghana and Côte d’Ivoire. In general, non-Akan respondents gave help to a broader variety of social ties than the non-Akan in Côte d’Ivoire who concentrated more than two-thirds of their help exclusively on their nuclear families (much like the Akan in Côte d’Ivoire who gave 70 percent). The non-Akan did give slightly higher proportions of their total help to extended family members (56 percent) in the Ghanaian region than did the Akan (34 percent). This might possibly reflect a very broad conceptualization of extended family by the non-Akan to mean more than the lineage but rather the ethnic group as a whole. The help given to friends was also a significant outlet by the nonAkan in the Ghanaian (13 percent of the total amount of help given) and Ivoirian regions (10 percent of the total). Reciprocity and the Poor in Ghana and Côte d’Ivoire Before proceeding to analyze the patterns of reciprocity among the poor, we must compare the distribution of wealth in these two regions of Ghana and Côte d’Ivoire. In villages with very low levels of formal sector employment, wealth was measured by combining information about housing and asset characteristics into a wealth index. (See the Appendix for a detailed discussion of this index.) An in-depth analysis of the distribution of wealth in the two regions shows some critical differences that a simple comparison of the means obscures. The mean scores on the wealth index were practically equal in these two regions.36 An analysis of the wealth index by quartiles, however, suggests that the distribution of wealth was actually bimodal in this region of Côte d’Ivoire while normal in this region of Ghana. In Côte d’Ivoire, the income distribution was much more polarized between a larger number of very rich (36 percent) and very poor (29 percent). In contrast, in Ghana, the majority of respondents were concentrated in the middle/lower part of the wealth index range with 36 percent of respondents characterized as not-so-well-off and 26 percent in the middle/high quartile. Because of these crossnational differences in the distribution of wealth, it was important to interact “country” and “wealth” in the statistical analysis. 36

Ghana has a mean score of 2.56 with a standard deviation of 0.99, and Côte d’Ivoire has a mean score of 2.72 with a standard deviation of 1.23.

The Informal Institutions of Reciprocity

57

So who were the wealthy and the poor in these two regions of Ghana and Côte d’Ivoire? The wealthy were disproportionately Akan men in both regions. In Ghana, the richest people were 68 percent men and 76 percent Akan. In Côte d’Ivoire, the richest were 65 percent men and 89 percent Akan. Interestingly, the story of who belonged to the poorest wealth category differed in more marked ways between the two regions. For example, the poor were disproportionately female in Ghana with 68 percent of the poorest being women despite equal sampling by gender. In contrast, the poorest in Côte d’Ivoire were split exactly evenly between the genders. However, the poor were disproportionately non-Akan (78 percent) in Côte d’Ivoire, but not in Ghana (21 percent). In contrast with Côte d’Ivoire, the non-Akan in the Ghanaian region were relatively evenly spread across wealth categories. Again, important crossnational distinctions persist as we dig deeper into the data. The next question is how did wealthy and poor people participate differently in the informal institutions of reciprocity? Overall, the differences between the participation of various wealth categories and the level of help given were only statistically significant for the Ivoirian region but not for the Ghanaian one.37 For every single category of wealth in Ghana, the most frequent level of help given was nothing at all. Even 65 percent of the richest group of people in Ghana gave $0. When the rich did give, it was more frequently a higher amount, with 18 percent giving between $50 and $200 of help. Ironically, while 51 percent of the poorest wealth category in Ghana gave nothing, this was the lowest percentage among the four wealth categories.38 Thus, even though the poorest in the Ghanaian region had the least to give, they were still giving something more often than other wealth groups. The amount given by the poorest was not the lowest value, either (less than $5); instead, 30 percent of the poorest gave between $5 and $50. In contrast with the Ghanaian region, in Côte d’Ivoire, a much smaller percentage of every wealth category was not giving any help at all to anyone. The poorest two wealth categories were basically tied, with about 26 percent not giving anything while the top two wealth categories each had about 15 percent not giving anything. The most frequent level of help given for three of four wealth categories, notably including the poorest, was a relatively high

37

38

The Fisher’s exact test is used because of the low number of cases (>5) in several cells of the cross-tabulation. In Ghana, p= 0.168, and in Côte d’Ivoire, p= 000. 68 percent of the not-so-well-off and 76 percent of the middle/high wealth category gave nothing.

58

Informal Institutions and Citizenship in Rural Africa

amount at $50–200 of help given. The highest percentage of not-so-well-off gave a slightly lower amount of $5–50 of help. Not surprisingly, in the Ivoirian region, the richest most frequently gave the highest quantities of help over $200. To conclude this section, despite intraregional variations between women and men, Akan and non-Akan, and the wealthy and poor, the most persistent and significant patterns of variation were at the country level between Ghana and Côte d’Ivoire.

why were the informal institutions of reciprocity so different?: the statistical analysis The puzzle to be solved now is why was reciprocity so different in these two similar regions of Côte d’Ivoire and Ghana? Why did Ivoirians from this region give so much more help than Ghanaians through these informal institutions of reciprocity? And, what explains the more nuanced, but still important, differences in the structure of giving help to others? Why did Ghanaians allocate a bigger portion (of a lower level of total help given) to the extended family, friends, and others than in Côte d’Ivoire where help was relatively more concentrated on the nuclear family? In this section, the results of a multiple regression model are discussed to evaluate my hypothesis about the key role of the state as well as a range of rival explanations for the puzzling local variations.

Primary Hypothesis My primary hypothesis to be tested was that differences in formal state institutions restructured the informal institutions of reciprocity in these similar regions in nationally distinct ways. Hence, the divergent state legacies of political administration (discussed in Chapter 4), social service provision (discussed in Chapter 5), and agricultural policy (discussed in Chapter 6) have stimulated important differences in the quantity and quality of the informal institutions of reciprocity in these regions of Ghana and Côte d’Ivoire. Specifically, I estimated that the historically more centralized intervention in all three areas of public policy in Côte d’Ivoire would produce a higher level of help given primarily to the nuclear family during state retrenchment. Because of my inability to quantify the preceding hypothesized causal mechanism, I simply included “country” as an explanatory variable in the regression model.

The Informal Institutions of Reciprocity

59

Rival Explanatory Variables Based on my reading of the relevant theory, I also included two types of rival explanatory variables in the regression model. The first type of explanatory variables was a set of variables that influenced an individual’s ability to give help. For example, the most obvious and important explanation of the higher level of giving in Côte d’Ivoire is that Ivoirian villagers were simply wealthier. The other ability variables included were education level, number of children who completed secondary or higher levels of education, number of children who earned a salary, and age. The second type of explanatory variables reflected an individual’s opportunity to give help. The most prominent example here would be that villagers with a greater number of children would give a higher level of help. The other opportunity variables included were gender, ethnicity, lineage activity level, and participation in local voluntary organizations. Of course, several of these variables could arguably influence both an individual’s ability and opportunity to help, which was considered in the analysis. The Results The model was relatively powerful at explaining the variation in the informal institutions of reciprocity overall. (See Tables A.2., A.3., A.4., and A.5. in the Appendix.) Whether all of the explanatory variables worked well together as a model is less important, however, than whether country was consistently significant and how influential it was relative to other variables. The results of the statistical model showed that the most consistently significant and substantively powerful variable was country, which will be discussed last. I turn first to a discussion of the other variables that were consistently significant but less influential, including variables related to a respondent’s children, gender, education level, and wealth (interacted with country). The Role of Children in Social Support The number, salary, and educational level of the respondent’s children were all important factors but affected reciprocity in different ways. Second only to country differences, the number of children with a regular salary was the most substantively powerful of the three variables. Interestingly, the effect was negative. Thus, if a respondent had a child with a regular salary, the respondent gave significantly less help ($95 less for each child) than someone without such a regularly employed child. This shows that only when a

60

Informal Institutions and Citizenship in Rural Africa

respondent’s child obtains regular, formal sector employment does the flow of support reverse to benefit the parent. The analysis suggests that the support from salaried children was even more powerful and significant in the Ivoirian region than in the Ghanaian one. (See Tables A.2. and A.5.) This point is supported by the qualitative evidence as exemplified by one young Ivoirian female’s comment: “If the elderly person has well-off children, then these children will take good care of him and send him money at the end of every month. This happens only if you worked hard in your youth.”39 Meanwhile, when respondents had a greater number of children overall, or had more children who had completed secondary school or higher, they gave more help to others. Ironically, the magnitude of effect was larger for respondents who had relatively highly educated children, most likely reflecting the elevated costs of supporting a student in another city and/or an underemployed school graduate. This trend counters the normative concept of long-term reciprocity that most villagers articulated when they justified the sacrifices made to educate their children. “You have to borrow [if you don’t have any money for school fees] because the children are the hope of tomorrow, thus, you can’t let them trail behind . . . ”40 Thus, even though Ghanaians and Ivoirians from these regions had similar numbers and types of children, more Ivoirians gave more help to their children than did Ghanaians. Since the two regions shared very similar profiles in terms of the composition of children, the variables concerning the respondents’ children explained variation within regions but not between them. Something else must have been spurring the differences in the informal institutions of reciprocity in these two regions of Ghana and Côte d’Ivoire. The Role of Gender Gender also played an important role in shaping the amount of help given to all ties and the help given to the nuclear family. Predictably, the effect was negative, meaning that if a respondent was female, she gave $41 less help than if the respondent were male. What is more insightful, however, is that while being female reduced the amount of help given in both regions, the influence was substantively more powerful and statistically significant only in Côte d’Ivoire. Thus, females in this region of Côte d’Ivoire were less able to give help than those in this region of Ghana. Evidently, gender played little role in the variation of help given to the extended family. This will be 39

40

Survey interview with a 23-year-old female (anonymous) by author. Opanin, Côte d’Ivoire, September 1999. Survey interview with a 70-year-old male Akan cocoa farmer (anonymous) by author. Kyere, Côte d’Ivoire, August 1999.

The Informal Institutions of Reciprocity

61

reexamined during the later discussion of the gender component of changing production systems. The Role of Education Level Education level was statistically significant in shaping the amount of help given to all ties as well as the help given to the nuclear family. Thus, with increasing levels of education, respondents gave $21 more help to all social ties and $18 to their nuclear families. Education level also did not appear as a significant variable in the help given to the extended family. It is important to note that the magnitude of the substantive effect of education level appeared relatively modest in comparison with the other variables. Furthermore, the magnitude of effect was greater in Côte d’Ivoire than it was in Ghana for giving to all ties. The Role of Wealth The role of wealth in shaping reciprocity was not at all as simple as one might expect. Wealth was never statistically significant or powerful in any of the analyses on its own. Only the interaction between wealth and country had an effect on the level of help given to all ties and to the nuclear family. Thus, it was not simply being wealthier that increased reciprocity on its own. Rather, if respondents were wealthier and lived in the Ivoirian region, they gave about $13 more help than inhabitants of similar wealth status from the Ghanaian region. This suggests that the distinct way wealth was distributed in the two regions, which was discussed earlier, shaped the informal institutions of reciprocity. I will argue in subsequent chapters that the primary causal factor was the different legacies of state formation. The contrasting histories of state public policy intervention shaped the distribution of wealth in the two regions that then affected the informal institutions of reciprocity. Hence, wealth was not a purely economic outcome but fundamentally a political one, very much entangled with the past experiences of state formation in each of the regions. Even so, the interaction of country and wealth had a relatively modest magnitude of effect on the amount of help given compared to the other variables. Again, the statistical analysis confirmed that an important country effect needed to be further explored. Other Variables of Secondary Importance Before examining the effect of country itself, several variables, which scarcely appeared or did not appear at all, will be reviewed briefly. Even though these variables were not the primary causal determinants of the overall variation in reciprocity, their inconsistent patterns do shed some light on smaller differences. For example, although age did not appear consistently in the analysis, when and where it appeared was insightful. Age was statistically significant

62

Informal Institutions and Citizenship in Rural Africa

only for help given to the nuclear family, and the effect was very small there. Thus, with each additional year of age, respondents gave $1.45 less help to their nuclear family. This finding makes intuitive sense because parents would give less and less help to their children as they became more financially independent. What is more puzzling is how increasing age had opposite effects when the analysis was conducted separately for each country. Thus, in the Ivoirian region, the older the respondents were, the more help they gave; whereas in the Ghanaian region, the older the respondents were, the less help they gave overall. Furthermore, the role of age was statistically significant only in Ghana. This point of evidence will be taken up again later in Chapter 6 in the discussion of changing production systems and intergenerational reciprocity. The index of participation in local voluntary organizations appeared marginally significant for help given to all social ties, but the relative magnitude of effect was, like age, not very impressive. Thus, when respondents had an increase in participation in local voluntary organizations, they gave more help but a relatively small amount ($2). For help given to the nuclear and extended family, this variable was neither statistically nor substantively important. Furthermore, when the analysis is conducted separately by country, this weak variable completely disappeared. While respondents from the Ghanaian region had a slightly higher level of participation in local voluntary organizations than in the Ivoirian region, the difference was so small that it was statistically insignificant.41 Overall, this does support the hypothesis extrapolated from the civil society literature that increased activity in local voluntary organizations would lead to increasing social capital and denser social networks. However, this factor was clearly not the primary causal determinant of variation in informal reciprocity. Finally, two variables were notable in their total absence. First, the Akan ethnicity variable was never statistically significant for any outcome. Second, whether a respondent’s lineage was active did not influence an individual’s level of giving.42 The Role of Country Differences What offers the most promise of explaining the puzzling differences in informal reciprocity was “country,” the most consistently significant and powerful variable in the statistical model. Unlike the respondents’ children, gender, education level, or wealth, country is highly significant for help given 41

42

A t-test comparing the mean scores for the index of participation in local voluntary organization indicates a lack of statistical significance at 0.783. A t-test comparing the means of the index across the two country cases revealed a significance level of 0.424.

The Informal Institutions of Reciprocity

63

to all social ties, help given to the nuclear family, and help given to the extended family. This point was confirmed when the model was rerun with the data split by country, and the patterns were consistent. Even after splitting the data by village, the same general patterns from the original model reappeared. Country was far and away the most substantively powerful variable, too. Remarkably, even with the addition of the country/wealth interaction term, country persisted as the most influential factor. Since the “country” variable was highly significant and powerful in its effect for all three outcomes, the statistical analysis validates that the proposed hypothesis about the legacies of state intervention is possible. Even though the project’s unique research design of using two very similar crossborder regions controls for many variables that might otherwise be within “country,” a lot remains to be explained. The statistical analysis is used in this study not as the end result, but instead, merely to show that there is a causal dynamic related to “country” that needs to be further unpacked and explained in the next section of the book.

conclusion The informal institutions of reciprocity thus differ in surprising ways in otherwise very similar regions of Côte d’Ivoire and Ghana. Many more people participated in informal social exchanges and at a much higher level of giving in Côte d’Ivoire than in Ghana. The social ties were therefore deeper and more extensive in Côte d’Ivoire and shallower and more restricted overall in Ghana. Survey research probing the quantity and structure of actual helping behaviors proved to be a powerful method of systematically comparing relationships that were often taken for granted. Self-reports of how an individual gave help to others over the past year revealed indirectly fundamental differences in the informal institutions of reciprocity between these two regions of Ghana and Côte d’Ivoire. In addition to facilitating a more systematic documentation of the puzzling outcomes, the quantitative analysis also provided some initial clues as to what explains the differences. The multivariate regression model helped to eliminate powerful alternative explanations. It was not simply that Ivoirians were more well-to-do and thus able to help where Ghanaians were too cashstrapped to do anything. Instead, “country” portends to have the most influence in shaping this puzzling variation in reciprocity between the cases. While compelling and certainly useful, this quantitative analysis still leaves a lot of questions unanswered. The systematic documentation on the

64

Informal Institutions and Citizenship in Rural Africa

quantitative amount of help given over the past year to certain types of people does not provide clues as to the qualitative meaning of these informal institutions of reciprocity. Or, how these relationships may have changed over people’s lifetimes. The next chapter takes advantage of the strengths of qualitative data analysis to investigate what reciprocity means to different village residents and how these things may be different today than in the past.

chapter 3 Local Conflicts Over the Meaning of Reciprocity: A Qualitative Analysis of Change

In much of Africa, the meaning of reciprocity within the family, friends, and village community is quite openly and regularly discussed in the course of daily life. In Ghana and Côte d’Ivoire, reciprocity appears often as the subject of a Sunday sermon or the theme of a provocative bumper sticker in bushtaxis, “tro-tros,” and buses. Perhaps because the informal institutions underpinning support to families and communities are under such pressure and are in the process of being contested, they are frequently at the center of stories, gossip, jokes, and even tales of witchcraft.1 This chapter uses interpretive qualitative methods to analyze these public elements of popular discourse in combination with transcriptions of informal interviews, focus group discussions, and survey interviews.2 I explore how the meaning of reciprocity has narrowed and become more market-oriented in recent years but in surprisingly different ways for the village residents of culturally similar regions of Ghana and Côte d’Ivoire.

informal institutions of reciprocity at the village level: what it means to help At the village level, the informal institutions of reciprocity might be defined more simply and accurately as whether and how people help each other over time. But even the very notion of “help” was different for people in various 1

2

See Meyer’s (1995) article on obsessions with the Devil as a critique of the changing systems of mutual family assistance during economic crisis in Ghana. Yanow and Schwartz-Shea (2006).

65

66

Informal Institutions and Citizenship in Rural Africa

economic, social, and cultural contexts. While respondents immediately and frequently referred to monetary assistance, many local people in both the Ghanaian and the Ivoirian regions mentioned a variety of nonmonetary helps that were considered equally (and sometimes more) important. For example, many respondents in both regions referred to the help given by a relative who paid school fees for their child or, even more significantly, provided room and board to a child attending secondary school in another city. In general, Ghanaians took these nonmonetary types of support for granted more frequently than Ivoirians, particularly when there was a reciprocal exchange between two parties. For instance, one Ghanaian woman reported initially that she had not helped anyone. Yet later she described helping her sister on the farm, basically providing free agricultural labor.3 The respondent did not think that this really counted as help because she was being given food in return. Many Ivoirian respondents, on the other hand, made a point of declaring that help was not limited to money but also included these types of “in kind” support; for instance, moral support, advice, and assistance with school admissions or employment. It’s not only with money that one helps. One can also help with manners, to give you advice. This is the thing that is most important. Moral support is the most important.4

Overall, Ivoirians were more cognizant of the value of what was given and received for all categories of help, whether monetary or in kind. Nevertheless, it is important not to envision either of these rural areas as nonmonetized subsistence economies in which extended families grew their own food and provided for all the members’ needs without much need for cash. In both fieldsite regions, individuals complained of high prices and reported significant needs for cash to pay for a range of necessities such as matches, sugar, soap, meat and fish,5 clothing, school fees, school supplies, medicine, hospital fees, and transportation. As one Ghanaian woman explained: Help is not only money . . . but with those of us in this village who are farmers, everyone has food and you can get some money to buy a cloth, but you can’t wear only that one cloth and the food alone too can’t help so it’s the money that we need.6

3 4 5

6

Survey interview (anonymous) by author. Barima, Ghana. Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999. An expression used in Ghana to describe a poor person is someone who doesn’t have any fish with his kenkey (a staple made of steamed corn dough). Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999.

Local Conflicts Over the Meaning of Reciprocity

67

Some farmers might also need cash to pay for agricultural workers. For instance, one Ghanaian woman, who was at least seventy years old, complained that the small bit of income earned from her cocoa farms had to be spent on day laborers.7 Thus, even though help “in kind” was an important component of local transfers, village residents in rural areas continually needed cash, even to maintain a locally acceptable level of food security.

three types of help: gifts, “soft” loans, and loans with interest In both Ghana and Côte d’Ivoire, village residents distinguished among three types of help: (1) help that was given freely as a gift; (2) help that was given as a “soft” loan, without interest; and (3) help that was given as a loan with interest added. In both regions, help that was given freely with no expectation of any repayment or interest was relatively rare, except in the case of the help given to their spouses and children. This finding directly challenges the conventional wisdom on the operation of kinship in Africa. The overall increase in the marketization of reciprocity was so unexpected that it was almost incredible. Particularly shocking was the notion (discussed as follows) that family members might conceivably charge interest. Many village residents emphasized that these changes were not welcome, but had come about reluctantly over the past two decades as a result of the prolonged economic crisis. As one Ghanaian male explained, “It is the hardship. I would like to give a help free [as a gift] but if I do, I’d be in hardship so I give help as a loan. But not to my brothers and sisters.”8 An Ivoirian man described how because of recent hard times, loans have also become more prevalent in Côte d’Ivoire: “If you have a problem that you can’t handle, truly, it’s difficult to find someone else [to help]. Often, one bases it on the work that you do but it’s given to you as a loan.”9 Despite sharing this overall stiffening of the terms of reciprocity, the comparative analysis revealed important differences. More respondents in the Ghanaian region reported giving and receiving loans without interest than did respondents in Côte d’Ivoire, where the third type, loans with interest, predominates. This second type of help was not even considered a loan in Ghana because there was no interest involved.10 7 8 9 10

Survey interview (anonymous) by author. Tape recording. Barima, Ghana. Men’s focus group (anonymous). Tape recording, Barima, Ghana. Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. This is true despite the fact that the help may be so formalized as to involve a witness who documents the exchange.

68

Informal Institutions and Citizenship in Rural Africa

But, say you’re hard-pressed for 1 week or 3 days . . . I don’t think the person (lender) would add anything (interest). We give like that all of the time.11

An Ivoirian man described the same type of loan conditions even more explicitly: “Others look at your work. If they know that you can’t pay, they don’t give it to you, even if you have a witness.”12 Thus, there were distinctions made in the terms of the help given in both Ghana and Côte d’Ivoire, but interestingly, the criteria used appear to differ. In Ghana, the decision was based primarily on the relationship with the recipient; in Côte d’Ivoire, the terms depended on the recipients’ work status and thus their ability to repay. Many more Ghanaian respondents than Ivoirians described how, when they were personally unable to give or lend money, they helped those in need by connecting them to a person who could give a loan. In this way, the respondents acted as a type of unofficial cosigner.13 This role of guarantor was apparently strengthened if someone personally accompanied the needy one to the perceived source of help rather than simply suggesting someone to him or her.14 Other respondents mentioned having the head of the family guarantee payment of the expenses later or having the family “go in for a loan.” One eighty-five-year-old Ghanaian cocoa farmer described how in the past, he would have been extremely responsive in trying to help, “even to the extent of borrowing [on behalf of the recipient],” but now, with his own impoverished situation, he had trouble even imagining someone coming to ask for help.15 Many of these interest-free loans in Ghana seem to be under the understanding of long-term reciprocity, i.e., that the person will pay the original loan amount back eventually when they have more money, often after a crop has been harvested.16 In Ghana, the time period could be up to one year, but generally was shorter.17 Other loans could be paid back in other terms; for example, repaying money by weeding the lender’s farm.18 Sometimes the 11 12 13

14

15

16 17 18

Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Survey interview with 30-year-old single Akan male (anonymous) by author. Makwan, Ghana. Several respondents expressed their fear of leading someone to ask for a loan since they would be indirectly implicated in the repayment. Survey interviews (anonymous) by author. Barima and Makwan, Ghana. Survey interview with an 85-year-old Akan male cocoa farmer (anonymous) by author. Makwan, Ghana. Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Ibid. Ibid.

Local Conflicts Over the Meaning of Reciprocity

69

loan terms were so “soft” in Ghana that the borrower was not entirely sure whether or not he had to repay the money at all. In contrast, in Côte d’Ivoire, respondents described how loan terms had become more stringent over time. First, respondents were no longer able to repay a loan in kind as they could in the past when “he would go work as a laborer in order to reimburse a credit” or “reimburse . . . perhaps with a bottle of gin.”19 Second, most loans were not interest-free as they tended to be in Ghana, but included interest. One man explained: “Before our relatives gave loans without interest; now, there are more loans with interest.”20 Others agreed: Now the interest is 100% while in the past, there wasn’t any interest and money wasn’t more considered than man. Now, one considers money more than the man, therefore, there are a lot more loans with interest than presents.21 Before our relatives got along very well and they helped each other. But, [now] if you want a loan, it’s with interest. There are no longer any presents. People don’t love each other anymore.22

These comments provide evidence not only of crossnational differences between these two regions of Ghana and Côte d’Ivoire but also suggest changes in the informal institutions of reciprocity over time. An Ivoirian woman who exclaimed that to charge interest on a loan in the past “would have been truly obnoxious towards the person in need” sums up the normative difference over time.23 By the late 1990s, charging interest was expected as the unwritten “rules of the game” and was thus not considered offensive. In contrast, in Ghana, the subject of interest on loans was still somewhat repugnant and distasteful to discuss. Even the variation in interest rates charged on loans in Ghana and Côte d’Ivoire revealed key differences in the evolution of these informal reciprocity institutions in the two regions. In the Ghanaian region, on the one hand, some local people served as informal moneylenders in the village with lower interest rates, usually around 50 percent, while the “big” moneylenders in the nearest towns charged interest from 75 percent to as high as 100 percent per month and required collateral.24 Thus, in the Ghanaian villages, where people knew 19 20 21 22

23

24

Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999. Ibid. Ibid. Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999. Similar responses from the Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Women’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Men’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999.

70

Informal Institutions and Citizenship in Rural Africa

each other, interest rates were significantly lower than in town. In the Ivoirian region, on the other hand, interest on loans appeared to be relatively uniform at the higher level of 100 percent in the village and in town.25 The more “commercial” and usurious practices of the town dominated the informal village systems of credit in Côte d’Ivoire. While few Ghanaians or Ivoirians from these regions reported having to seek loans from the big, well-known moneylenders based in nearby towns and cities, a significantly higher number of Ivoirians mentioned going to the bank in times of need than did Ghanaians.26 The bank was mentioned not only as a source of loans but also as a place where some Ivoirians deposited savings of money that had been set aside. Many more respondents in Côte d’Ivoire than in Ghana described how they relied on their own personal resources for help in times of need, and no one else. As for money, one should not count on someone . . . The little money that we’re going to have [from the upcoming cocoa harvest], the students are going to take the large part. But, it’s necessary to set aside something for the unexpected; you never know. If you have a sum of 100,000 CFA and someone tells me to buy meat for 1,000 CFA, why will I refuse? Because I have to reserve something to avoid going and asking someone when I have a problem. In summary, we must get ourselves organized.27

Several respondents mentioned that they would work harder at their economic activities in order to be able to save: “to have enough money to guarantee me in the future; to not have to ask for help.”28 One forty-fiveyear-old cocoa farmer in Côte d’Ivoire described where he would get help if he were in need: “Myself. The bank. That’s why I work harder at my job [farming].”29 In response to a survey question that asked what types of strategies respondents used to cope with crisis, 56 percent of Ivoirians replied that they worked harder at their primary job, whereas fewer than half the number of Ghanaians (24 percent) replied similarly. In addition, very few respondents in Ghana reported maintaining a savings account in a commercial bank in town. Most respondents in Ghana also explained that they did not go to commercial banks for loans because banks did not generally give credit to nonsalaried workers.30 Furthermore, when they did, there were 25

26 27 28 29 30

Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Women’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Survey interviews (anonymous) by author. Kyere and Opanin, Côte d’Ivoire. Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Ibid. Survey interviews (anonymous) by author. Kyere, Côte d’Ivoire. Only a few teachers based in the villages had ever requested and received a loan from a commercial bank.

Local Conflicts Over the Meaning of Reciprocity

71

significant transaction costs in terms of processing delays of up to five months and kickbacks required in addition to the processing fees.31 Since highly formalized lines of credit may not have been utilized as often in this region of Ghana as in this region of Côte d’Ivoire, it appeared that informal channels of credit were used relatively more frequently in Ghana than in Côte d’Ivoire. A significantly higher number of Ghanaians (40 percent) reported that they bought food and supplies on credit in times of crisis compared with Ivoirians (1.6 percent). In addition, a slightly higher number of Ghanaians reported borrowing from family and friends (56 percent) than did Ivoirians (45 percent).32 These informal sources of help continued to be important but also appeared to be narrowing and increasingly market-oriented. The next four subsections investigate how village residents defined and prioritized their obligations to help different categories of family members, friends, and village neighbors.

the meaning of family Before examining how and when people believe they should help their family, it is essential to understand how respondents might define the family differently in these regions of Ghana and Côte d’Ivoire. By using a more open-ended measurement strategy, I could illuminate the nuances in the indigenous conceptualizations and categorizations of family. I argue against the notion of any conceptualization of family that is shared typically in “the African village.” These concepts differed in important ways between these two regions on either side of the Ghana–Côte d’Ivoire border. In general, in the Ghanaian region, the family seemed to be defined relatively more broadly by respondents than in the Ivoirian region. Ghanaians usually referred to the extended family when queried about their notion of family: Here in Africa, our family is not like the American or European type of family. Your mother, aunties, uncles and all their families are your family. About 300 people can constitute a family . . . 33

When Ghanaians defined the extended family, it was usually in matrilineal terms to include the mother’s siblings; in particular, the mother’s uncle as “the family.” In a survey interview, one respondent explained how he had 31

32 33

Men’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. This difference is of borderline statistical significance (0.082). Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999.

72

Informal Institutions and Citizenship in Rural Africa

sought a loan from one of his uncles to pay the medical bills of another uncle who had suffered a motor accident.34 On the one hand, this shows the continuing relevance of the matrilineal system in Ghana; on the other, these ties may not be as strong as they were in the past because the help that was given was in the form of a loan. In contrast, Ivoirians seemed more cognizant of the concept of the nuclear family. An expression in French (“la petite famille,” or the little family) was used frequently in conversations to refer to the mother, father, and their children. “Because times are hard, it’s necessary to first look after your ‘little family.’”35 Along with this recognition, more respondents readily acknowledged without hesitation that they would help their nuclear family first.36 “ . . . If it’s not an urgent problem, you must first help those who brought you into the world.”37 This was never admitted as quickly or openly in Ghana. Furthermore, Ivoirians seemed more at ease making distinctions between the nuclear and extended family system, including what type of support was required for each. For the nuclear family, it’s an obligation to help. It’s required (forcé) while for the extended family, it’s simply help. You are not forced because there are the others.38

There was an idea implicit in the previous statement and others like it that because the nuclear family was smaller, the responsibility could not be shirked. But with the larger extended family, there were always other members who could provide help to the needy in your stead. Individuals were more tempted to “free ride.” Just as the overall concept of the nuclear family differed in Ghana and Côte d’Ivoire, so did the components. While a respondent’s children came first above all other ties in both Ghana and Côte d’Ivoire, the way that this obligation was articulated varied in the two countries. In Ghana, one fortyyear-old father of six children ranging in age from one to eighteen justified his refusal to help a relative: “I would rather use the money that I would give to the relative (extended family member) to send my child to the hospital so I’ll not give any money to the relative.”39 Another Ghanaian respondent explained the normative framework even more clearly in terms of what is considered “foolish” or “sane”: 34 35

36 37 38 39

Survey interview with a 37-year-old Akan male (anonymous) by author. Makwan, Ghana. Survey interview with a 22-year-old Akan woman (anonymous) by author. Kyere, Côte d’Ivoire. Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire. Ibid. Ibid. Survey interview (anonymous). Tape recording. Makwan, Ghana.

Local Conflicts Over the Meaning of Reciprocity

73

People will think I am foolish to give money to a relative in need while my own child is sick in the house. In such a case, I will use my little money to take my child to the hospital. My relative will wish that I give him the little money that I have . . . But no sane average person would help the relative in such an instance.40

Ivoirians also emphasized taking care of their children’s needs first. In Côte d’Ivoire, one father explained simply, “It’s the children we bring unto the world that we help the most.”41 Nevertheless, generally, the norm was not articulated boldly like Ghanaians as an absolute obligation that any “sane” person would acknowledge. Rather, even the support of one’s children was framed in relatively more instrumental terms; again, in the language of exchange. “If one doesn’t help our children, they can’t succeed . . . then I am left alone.”42 Many Ivoirians spoke of their children in terms of whether they had succeeded in finding work or regular salaried employment yet. “If you have a child who works and who is well-placed, he can bring you money every month. But, if he hasn’t yet found work, you will always be suffering.”43 Another Ivoirian woman expressed her frustration at her children’s lack of employment despite the fact that she supported them through school: They haven’t yet found work. Above all, for the expenses that you made in order for them to repay you for your efforts. Despite all of these sacrifices, we continue to suffer.44

Thus, Ivoirian respondents tended to articulate the social support that was given and received to and from their children in terms of a long-run exchange, whereas Ghanaians described the relationship simply as an unquestionable, moral duty. While children were frequently discussed as both key recipients and donors of social support, spouses were rarely mentioned in either Ghana or Côte d’Ivoire. Often, as much can be learned from what people do not say as what they do. There are two reasons for the absence of much discussion of help given or received to spouses. First, this type of support was taken for granted more than any other. More than the relationship between parent and child, which changes and inverts over time, the relationship between spouses is one based on mutual support, sharing, and reciprocity: “If you have a husband, it is his duty to help you, because our tradition demands that your 40 41 42 43

44

Survey interview with a 45-year-old married Northern male (anonymous). Makwan, Ghana. Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire. Women’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, September 1999. Women’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Ibid.

74

Informal Institutions and Citizenship in Rural Africa

husband helps you in all situations.”45 Second, the evidence suggests that spouses did not pool risk together exclusively as a household. Instead, men and women responded to crises individually through participation in different types of social networks. Men and women had access to different types of resources with different gendered obligations to their children, parents, and other family members. One Ghanaian woman summarized how social networks worked for women in the village: If you have a husband, your crisis time help comes from him. But at times, your husband’s help is too little. In such a situation, family members may help, mainly your mother or your father. Apart from these, that is why we attach ourselves to other people like a friend, so that when the main helpers fail, they [friends] may help.46

Spurred on by this grassroots analysis, I now turn to an examination of the norms of friendship.

the varied meaning of friendship In both Ghana and Côte d’Ivoire, some common expressions appeared on bumper stickers on taxis and buses that revealed the consequences of hard times on friendship. In Ghana, one would frequently see a phrase “Chop time, no friends.” In Côte d’Ivoire, a bumper sticker declared, “In times of suffering, there are no friends.” While both of these expressions indicated a similar ambivalence toward sharing with friends in hard times, the notion of friendship varied in Ghana and Côte d’Ivoire.47 First, many more Ghanaian respondents agreed that the help of friends was the most important, above all others. Ghanaians also articulated the reciprocity of friendship in clear terms. For example, one twenty-five-yearold Akan male explained that he would help a friend before a relative because “this relative would not have helped me if I were in need, whereas a friend will always help.”48 Another Ghanaian male confirmed this difference between a friend and a brother: It is true, that friends are more reliable than family because at times, when you tell your brother of your need, he would not mind you, but a friend is someone you are free with, and one does not feel shy when telling a friend of anything. When a brother is not able to help you, it brings shame. There is actually a difference between a friend

45 46 47

48

Women’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Ibid. Adams and Plaut (2003) discuss how culture shapes conceptions of friendship in their social psychology study of Ghanaians and Americans. Survey interview (anonymous) by author. Makwan, Ghana.

Local Conflicts Over the Meaning of Reciprocity

75

and a brother. A brother is better than a friend, but if a brother is not able to help you, it brings hard feelings.49

Thus, friends were not only more sensitive to needs but were also in an easier position to refuse if they were unable to help. Interestingly, even when Ghanaians were trying to help a relative, they would frequently mention seeking the assistance of a friend to help that relative.50 In Ghana, friends were not only more reliable as sources of help but also as recipients. One respondent explained that this was because the friend “will know I am having difficult times” so they will be more likely to repay a loan (rather than leave him with the debt). In contrast, the family member might not be able to pay and might default easier because you are family.51 Meanwhile, in the Ivoirian region, friendship was less central. Whereas Ghanaians tended to jump quickly to explain why people give more help to their friends, Ivoirians raised issue with the assumption, doubting whether or not it was true. Some respondents did cite the help that they gave and received to and from friends, but the majority placed a lower emphasis on their friends than on their close family, explaining that you were not “obliged to help” friends.52 “There are friends that are good, but they cannot achieve the status of a brother.”53 Another Ivoirian woman declared: “If you see a family member who suffers, you can’t drop him to help a friend.”54 One Ivoirian used a vivid metaphor to describe the difference between family and friends: You’re born with your family. It’s like your hair. It’s something natural; you can’t get rid of it. Friends . . . They are like a beard. You can shave it . . . 55

There was a common assumption that help to friends was less of a priority because the respondent’s friend “also has a family who can help him.”56 When Ivoirians did speak of friends, it was clear that very few were “close friends,” intimate enough to merit helping. Furthermore, it appears that friendship did not change the terms of help in Côte d’Ivoire. Help was given as a loan, and only if it was deemed that the friend could repay:

49 50 51 52 53 54

55 56

Men’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Survey interviews (anonymous) by author. Makwan, Ghana. Survey interviews (anonymous) by author. Barima, Ghana. Survey interviews (anonymous) by author. Opanin, Côte d’Ivoire. Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999. Women’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Interview (anonymous) by author. Kyere, Côte d’Ivoire, July 14, 1999. Survey interviews (anonymous) by author. Opanin, Côte d’Ivoire.

76

Informal Institutions and Citizenship in Rural Africa

If I go ask for a loan from a friend, and he knows that I can pay, he will give it to me. By contrast, if he knows that I don’t have anything, not one foot of cocoa plantation, he won’t give it to me. Now, there are a lot of criteria, even if it is a friend.57

Interestingly, when a friend in Côte d’Ivoire became so close that a respondent would help him, he “becomes a brother.”58 This type of situation was described as if it certainly were rare indeed.59 In contrast, Ghanaians continued to make important distinctions between their brothers and friends, even when they placed the friendship in higher regard. This demonstrated the respondents’ comfort with placing value on a friendship for its own sake in Ghana, versus transforming the relationship to a family one in Côte d’Ivoire. Whereas Ghanaians’ notion of friendship seemed more abstract, generalized, and emotional, Ivoirians tended to give two purely instrumental justifications for anyone to help a friend. First, Ivoirians frequently explained that if someone helped a friend, it was because the friend had already helped that person a lot. Second, they described how helping friends was less risky with regard to the dangers posed by witchcraft: It’s because they [the relatives] are witches and if you want to help them, they will destroy you [kill you with witchcraft]. That’s the reason certain people help their friends because they are scared of their relatives.60

More than one respondent described how someone might prefer to help a friend over a needy family member because the relative might try to kill him with witchcraft. According to local beliefs, in order for witchcraft to be effective, it must be someone who is sufficiently close, often an extended family member.61 Witchcraft is an important part of the contemporary culture in both Ghana and Côte d’Ivoire. Witchcraft is not an irrational set of primitive beliefs that has been steadily eroded and undermined by colonial contact or 57 58 59

60 61

Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Survey interviews (anonymous) by author. Opanin, Côte d’Ivoire. Women’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Ibid. Repeated observation summarized in notes in field journal, vol. 3, September 21, 1999. Geschiere (1997) explores the relationship between witchcraft and kinship in his book on Cameroon. He summarizes (1997: 212): “Nearly everywhere in Africa, it is inconceivable, still today, to formally refuse maintaining family ties: the family remains the cornerstone of social life, and one cannot live without its intimacy. Yet it is precisely this intimacy that harbors deadly dangers since it is the very breeding ground of witchcraft.”

Local Conflicts Over the Meaning of Reciprocity

77

some process of modernization.62 In fact, attention to witchcraft appears to be on the rise as a way of explaining and coping with new tensions, conflicts, and rising inequalities.63 During my fieldwork, village residents described how increasingly “greed,” “jealousy,” and “envy” were the source of problems, and not just within the extended family. Ghanaians and Ivoirians explained how individuals who had achieved material success but not shared appropriately were often either accused of being a witch or were the victims of witchcraft. In Ghana, a wave of incidents in which strangers were accused of casting spells to shrink someone’s genitalia resulted in multiple deaths. In Côte d’Ivoire as recently as 2004, an indigenous village chief declared that long-term migrant cocoa workers from the North were using witchcraft against the indigenous population to get them to sell their valuable cocoa land.64 This sparked a wave of violent attacks and counterattacks between the migrant and indigenous population, resulting in more than a dozen deaths. Witchcraft was thus a sanctioning mechanism for violating norms; for example, not fulfilling social obligations to the family or what was considered to be unequal or unfair economic accumulation. Witchcraft was also routinely used by both Ghanaians and Ivoirians to explain the root causes of a traffic accident, illness and disease, an economic misfortune, and so on.65 So, for example, after a Ghanaian villager described how heavy rain had contributed to a terrible car accident that killed a neighbor, he went on to explain how it was known that the accident victim had been cursed by his older aunt. Village residents known to be dying of HIV/AIDs were also more often described as being the victims of witchcraft. Similarly, if someone received poor grades on an exam, lost a job, or suffered losses in their business, a jealous family member was usually to blame. These beliefs in witchcraft were articulated with conviction by people of all ages, education levels, socioeconomic status, and rural or urban location 62

63

64 65

See Geschiere (1997) on what he calls “the modernity of witchcraft.” During the colonial period, the French and British both attempted to criminalize the most visible indigenous practices of witchcraft. Both French and British colonial rulers made witchcraft punishable with imprisonment, fines and/or beating. See Gray (2000) for a study of the Akan response to British colonial law in the Gold Coast. See Van Gijsegem (2006) for an analysis of the application of the French legal code to punish witchcraft in Côte d’Ivoire. While I did not set out to collect data on the incidence of witchcraft, village residents frequently declared that witchcraft accusations had become more prevalent. See Comaroff and Comaroff’s (1999) provocative piece theorizing the increase in occult economies in South Africa. See the introduction in Moore and Sanders (2001) for a skeptical critique of this line of argument. This was reported in The New York Times (Sengupta 2004). Evans-Pritchard (1937) explored how witchcraft was often the root cause for an unfortunate event in his classic study of the Azande.

78

Informal Institutions and Citizenship in Rural Africa

in Ghana and Côte d’Ivoire. Regardless of their current religion or devotion to their faith, Ghanaians and Ivoirians continued to openly incorporate indigenous beliefs about witchcraft into a syncretic worldview. It was believed that a witch would “eat” or disembowel their victims. Revealing a curious blended belief in the power of witches and Western medical technology, one village resident (who was Christian and literate) proclaimed that the X-ray of a victim of witchcraft had been shockingly empty. This was visual proof using “modern” technology that a witch had devoured the victim. Witchcraft was not always “bad,” however. Ghanaians and Ivoirians would also talk of seeking assistance from someone who would use “good” witchcraft to protect the individual from the malevolence of others. In spite of the previous commonalities, however, witchcraft was still conceptualized, invoked, and handled in dynamic and distinct ways in these two regions of Ghana and Côte d’Ivoire. These differences are discussed in more detail in this chapter’s later sections on the motivations for giving help and consequences of not doing so.

informal institutions of village-wide reciprocity While there were significant differences in the earlier concepts of the family and friendship in Ghana and Côte d’Ivoire, the notion of village reciprocity in the abstract appeared to be relatively similar in all four of the villages. In both Ghana and Côte d’Ivoire, respondents consistently described the informal institutions of village reciprocity as “helping each other mutually” or “loving each other.” But, beyond these common abstract ideals, respondents highlighted important differences in their experience of village reciprocity in their particular villages. While Ghanaians described the high and low points of village reciprocity at different time periods, Ivoirians made more blanket statements, downplaying or even dismissing the presence of village reciprocity altogether.66 One villager in Kyere, Côte d’Ivoire, claimed: The village of Kyere does not have solidarity because we don’t like each other. We wish that someone else would have a problem or unhappiness so that we can mock him or talk about him all over town, even around a jug of palm wine.

A woman in Opanin, Côte d’Ivoire described how the reality of village reciprocity differed from the pretense:

66

Men’s and women’s focus groups (anonymous). Tape recordings. Kyere, Côte d’Ivoire, August 1999.

Local Conflicts Over the Meaning of Reciprocity

79

One pretends to love each other, but when we have to do certain things together, it’s in this moment that we see that we don’t love each other. So, it’s like we love each other superficially, but not deeply. That’s how this presents itself.67

Even when Ivoirian respondents would disagree and insist that village reciprocity existed, it was articulated in a clearly limited way. For instance, one Ivoirian man insisted that villagers do love each other. But, then he qualified the statement, adding, “Each person has his group of people that love him.”68 Another woman echoed, “Yes we love each other, but it’s between neighbors that we visit each other, because it depends on the behavior of each of us . . . Certain ones don’t love me.”69 This is hardly the mythologized image of a harmonious villagewide collective. In contrast, Ghanaians described working together as a village to overcome difficulties or accomplish certain objectives moderately more than Ivoirians. Nevertheless, while there were a few more examples of respondents from the Ghanaian villages helping each other, reciprocity was not consistent in either country, and certainly fell short of the shared normative ideal. In order to reveal the parameters of village reciprocity, I briefly compare how respondents reported handling four types of crisis events that might be expected to stimulate village reciprocity: (1) funerals; (2) a natural disaster; (3) an elderly villager in need; and (4) communal labor for village development.70 While none of the villages demonstrated coordinated action to provide mutual help in all instances, the Ghanaian villages appeared slightly more cohesive than the Ivoirian ones. Ironically, in both villages, the most successful villagewide organizational efforts were those to collect money to hold funeral celebrations. The norm in both countries seemed to be that help with funerals was more obligatory than any other type of help. In fact, district-level data show that respondents in Tano District allocated more money for funeral expenses than they did for clothing, and just as much as they did for education.71 There were two very different reasons usually given for this. First, the deceased person needed a

67

68 69

70

71

Women’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999. Women’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. The development projects that required communal labor were not “crisis events” as were the others, however, they did necessitate village-wide cooperation and coordinated action to be achieved successfully. GOG (1996: 40).

80

Informal Institutions and Citizenship in Rural Africa

funeral in order to rest peacefully. Second, help given for a funeral would be “reimbursed” as people who attended the funeral would give donations such that often the costs could be recouped.72 One Ghanaian respondent explained an Akan proverb well-known in both countries: The reason why we say ‘families love corpses’ is that when one is ill, the family would deny the person health care, but when the person dies, because money can be made from the corpse, the family would go to greater lengths to ensure a good funeral.73

Funeral collections were particularly well-organized in Côte d’Ivoire, an exception to the general finding of moderately greater village reciprocity in Ghana. The Ivoirian villages designated someone for each of the geographic sections of the village to maintain a book that listed each prorated individual donation clustered by family head. These men would thus go systematically throughout their section of the village and take up a collection to give to the grieving family. Each name, as well as the total amount collected by “courtyard,” would be announced publicly at the end of the funeral ceremonies. If a person did not pay the required amount, a fine was assessed and collected at the next funeral. While village reciprocity was displayed positively for funerals in both Ghana and Côte d’Ivoire, on the negative side, there was little evidence of any coordinated villagewide action in either country in response to a natural disaster.74 In 1983, in both Ghana and Côte d’Ivoire, bushfires destroyed many cocoa plantations and most of both regions’ food crops. Several people in Ghana and Côte d’Ivoire characterized the time as a “devastating” period of “extreme hunger” or “famine.”75 In Ghana, the threat to food security was compounded by the influx of Ghanaians deported from Nigeria that same year. Some emergency foodstuffs were sent by each national government, but instead of being distributed freely, Ghanaian and Ivoirian villagers reported having to stand in long lines to buy small quantities of rice.76 Even when families had money saved, it was difficult to find enough food to buy.77 Since this crisis affected the basic food security of almost every household in 72 73 74

75

76

77

Survey interviews (anonymous) by author. Barima, Ghana. Men’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Hirtz (1995) also found in the Philippines that the study village did not organize collective action for relief after a natural disaster. Survey interviews (anonymous) by author. Barima and Makwan, Ghana and Kyere and Opanin, Côte d’Ivoire. Survey interviews (anonymous) by author. Barima and Makwan, Ghana, and Kyere and Opanin, Côte d’Ivoire. Many respondents described how they would try to fill up on palm soup because the other staples were so scarce.

Local Conflicts Over the Meaning of Reciprocity

81

every village, it serves as a window to compare the mechanisms of village reciprocity in practice. Contrary to the shared norm of villagers helping each other, most Ghanaian and Ivoirian respondents denied that there was any significant coordination among families in the village to help one another during this crisis.78 One Ghanaian man mentioned standing in a queue to buy corn dough to cook themselves in their homes instead of allowing the kenkey seller to cook it herself.79 Ivoirian women also described waiting in line to buy one kilogram of rice. Several Ivoirians explicitly recalled having to present their identity cards when they were buying rice, a mechanism that reinforced the idea of responding to individuals, not collectives, in need.80 Several Ghanaian women mentioned that it was by “God’s grace” that they were able to manage.81 Several people confirmed the following woman’s account of how the endurance of terrible hardship eroded reciprocity: During that period, if you didn’t have money, you just faced hardships. No one helped anybody. With so many children in a household, everyone was crying his own cry. As for that period, there was no help anywhere. When you get something, you give it to the children whilst you sleep on an empty stomach.82

Another Ghanaian man reiterated how people were most concerned about their own household’s survival and did not mobilize to help the community at large: I don’t think [the community mobilized to help one another] because during that time, everyone was fighting for his own. We used to cook “atsetree” so the little that I have, I can’t share with the community.83

The few people who said that they helped others helped their own children, parents, and siblings to eat.84 No one in Ghana or Côte d’Ivoire mentioned organizing as a village “moral economy” to ensure that everyone had a minimum amount to eat; instead, help was more individually based.85 In fact, several Ghanaian villagers noted that during this period, there were

78 79 80 81

82 83 84 85

Men’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Ibid. Survey interviews (anonymous) by author. Kyere, Côte d’Ivoire. Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Women’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Women’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Survey interviews (anonymous) by author. Kyere and Opanin, Côte d’Ivoire.

82

Informal Institutions and Citizenship in Rural Africa

widespread “thefts of farm produce,” the very antithesis of pulling together in reciprocity.86 While the comparison of the first two examples of crisis events revealed similarities between the two cases, the comparison of the third and fourth points indicates moderate but significant differences. In Côte d’Ivoire, the boundaries of village reciprocity were more restricted than in Ghana when respondents agreed that the village as a collectivity does not take care of needy individuals, such as an elderly person; that is, the responsibility of the family. “If you have an elderly man living with you in your house, you take care of him. But, it’s not the villagers who are going to take care of him.”87 Another elderly man in Côte d’Ivoire confirmed that the villagers were ultimately not responsible; an individual’s family was responsible for an elderly person’s care: Maybe, one can give the elderly person a little food or the first care but as much as his own family member with means doesn’t come, he is going to die, because, we also, we have our own charges and our families too . . . 88

While Ghanaians also placed primary responsibility for the care of elderly people on the family, most respondents did not reject outright any responsibility of villagers, as was done so explicitly by Ivoirians. One woman explained that anyone in the village could care for an elderly person in need, even if there was a family member with modest means who refused to help: When such a thing happens, anyone at all, not necessarily a family member, can decide to provide for the needs of that older person since we are each other’s keeper. There therefore is no need for someone to talk to the person who has refused to take responsibility.89

Other Ghanaians reinforced this idea of village involvement, saying, “People in the community give such older people [who are not being well taken care of] money or food.”90 Several Ghanaian respondents explained that everyone in the village held a collective responsibility: “We can’t just walk past another person’s suffering in the village.”91 While the norm regarding village reciprocity vis-a` -vis an elderly villager in need seemed slightly stronger in 86 87 88

89 90 91

Survey interviews (anonymous) by author. Makwan, Ghana. Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999. Survey interview with a 70-year-old Akan man with thirteen children (anonymous) by author. Kyere, Côte d’Ivoire. Survey interview with a 40-year-old Akan woman (anonymous) by author. Barima, Ghana. Survey interviews (anonymous) by author. Makwan, Ghana. Ibid.

Local Conflicts Over the Meaning of Reciprocity

83

Ghana, several respondents nevertheless maintained that a few elderly individuals had died from lack of help.92 Finally, Ghanaians also reported a higher level of reciprocity than Ivoirians with regard to communal labor for village development. Throughout the period of the fieldwork, able-bodied male villagers were called on a rotating basis at 5 A.M. by “gong-gong” to help dig holes for the newly arrived electric poles. Ghanaian women often provided support to this communal effort by making meals for the workers. Respondents in Ghana also reported some cooperative farming, but it was not villagewide and instead was usually organized through an ethnic or religious association.93 In contrast, communal labor was never witnessed during the fieldwork period in Côte d’Ivoire. Apparently, the village chief in Kyere experienced difficulties in simply calling the elders for a meeting, with many preferring to work on their own individual farms instead. The only communal work group cited in Côte d’Ivoire was organized by the village women in Kyere to sweep and clean the village, but even this was no longer functioning. The catechist for the Catholic church in Opanin concluded that the lack of village reciprocity was proven by the fact that the final elements of the church construction remained unfinished since 1993.94 Overall, while the normative ideal of village reciprocity seemed remarkably similar in both Ghana and Côte d’Ivoire, a comparison of crisis events uncovers some subtle but important differences in the way that villagers coordinated to help each other in the two regions, with Ghanaians displaying a slightly higher level of reciprocity than Ivoirians. Since those in need did not have a cohesive, villagewide safety net to rely upon as a last resort, who did they turn to, and who said yes?

where to turn for help: the reliability of family versus friends and others Overall, Ivoirians relied on a much narrower network of social ties, primarily the nuclear family, whereas Ghanaians had a wider social network, receiving social support from a wider variety of sources, including the nuclear family, the extended family, and in particular, friends. In a survey question that asked what source would give the respondent help if they were sick and in

92 93 94

Survey interviews (anonymous) by author. Barima and Makwan, Ghana. Survey interviews (anonymous) by author. Barima, Ghana. Interview with catechist for the Catholic church (anonymous) by author. Opanin, Côte d’Ivoire, September 12, 1999.

Informal Institutions and Citizenship in Rural Africa

84

t a b l e 3 . 1 . First-Mentioned Source of Help If Respondent Was Sick and in Need

Source of Help Nuclear family Extended family Friend Moneylender Rich person Church Association Chief Government Other

B/A Region of Ghana (%) 23 20 28 9 7 8 0 0 0 4

Abengourou Region of Côte d’Ivoire (%) 59 11 7 0 1 0 6 2 0 14

need, the overwhelmingly dominant first response of the majority of respondents in Côte d’Ivoire was the nuclear family (59 percent). (See Table 3.1.) The nuclear family was thus cited as the first source of help by more than two-and-one-half times more Ivoirians than Ghanaians (23 percent). Basically, the nuclear family was perceived to be the pillar of support in Côte d’Ivoire, whereas it served as one of several approximately equal pillars of assistance available in Ghana. Interestingly, the previous difference was not due solely to an enormous difference in the reliance on the extended family by Ghanaians. While Ghanaians (20 percent) mentioned the extended family almost twice as much as Ivoirians (11 percent), more significant for Ghanaians, in striking contrast to Ivoirians, was the role of friends in providing assistance. Nearly 28 percent of the Ghanaian respondents replied first that a friend would provide help if they were sick, whereas only 7 percent mentioned likewise in Côte d’Ivoire. Friends were thus the single most important source of help for Ghanaians, mentioned more frequently than both the nuclear and extended families.95 In lieu of support from friends, Ivoirians described help from their siblings, spouse, and parents as most reliable.

95

In a similar survey question placed later in the questionnaire as a crosscheck, friends were again cited as the single most reliable source of help by over 17 percent of Ghanaians whereas only 5 percent of Ivoirians responded in this way. The second question was open-ended and followed the extensive probing of actual transfers (detailed in Chapter 2) asking which person or group was most reliable, that is, most likely to provide help when needed.

Local Conflicts Over the Meaning of Reciprocity

85

In addition to friends, Ghanaians had several other significant sources of help. The church seemed to be important in this region of Ghana, particularly for Akan women. In contrast, the role of the church was almost never mentioned in Côte d’Ivoire. Also, local “rich people” provided important sources of help in the Ghanaian region, indicating another informal network of assistance that was not cited in the Ivoirian region. Finally, moneylenders served as a source of help to some, an option that was never mentioned in Côte d’Ivoire. In contrast, associations were a relatively more important source of help in this region of Côte d’Ivoire, but it was almost exclusively for Akan men. This distinction most likely reflected the role of cocoa producer organizations in providing short-term loans to those who had paid their membership dues. In comparison with the support given by the church or loans given by local wealthy individuals in Ghana, this credit system was relatively more formally institutionalized. It is interesting to note here that two potential sources of help were almost never mentioned by anyone in either region: village chiefs and the government.

to whom to give help: prioritizing among family, friends, and neighbors When it came to deciding to whom to give help, Ghanaians had greater difficulty in prioritizing among competing claims. In response to a survey question describing a hypothetical scenario in which the respondent had limited means and was faced with appeals for money from several people all at once, Ghanaians had a more difficult time placing any one individual or type over anyone else. After struggling with the question, the majority of Ghanaians (55 percent) responded that they would give a little bit in equal amounts to everyone. As one Ghanaian respondent explained: “Giving a little bit to everyone is better because it will afford everyone involved the opportunity to at least do something about the difficulty.”96 The second most frequent response in Ghana was that they would give nothing to anyone (15 percent). Many of these Ghanaians reported not being able to help anyone because they would not be “able to satisfy everyone.”97 One thirty one-year-old Akan woman explained: “I can’t give to one person and leave the others to their fate so I’ll tell them all it’s unfortunate, but I can’t help.”98 96 97 98

Survey interview (anonymous) by author. Barima, Ghana. Survey interview with young Akan student (anonymous) by author. Barima, Ghana. Survey interview (anonymous) by author. Barima, Ghana.

86

Informal Institutions and Citizenship in Rural Africa

In contrast, Ivoirians were split between giving everything they had to the person in the greatest need, for example, to a sick child (44 percent), or giving everything to their closest family member (32 percent). One Ivoirian man explained his choice of prioritizing his closest family: “One says that wellordained charity begins at home. One must begin first to do good at home before going outside.” Another Ivoirian man explained: “It’s the children that we bring unto the world that we help the most.”99 Fewer than 2 percent of Ivoirians replied that they would not give anything to anyone. Ivoirians thus seemed less comfortable refusing to help than Ghanaians, a finding that will be discussed in the following sections.

why help others? Particularly given the crushing hardships that individuals faced on a daily basis in these villages in Ghana and Côte d’Ivoire, why did people believe that they should help others in need? The motivation for helping others, whether it was pure altruism or perhaps an aspiration for social prestige, was frequently the subject of local proverbs and popular sayings in both countries. However, a close comparative analysis revealed significant differences in these normative frameworks in similar cultural regions of Ghana and Côte d’Ivoire. In general, Ghanaians articulated the reason for helping others in moral or spiritual terms, whereas Ivoirians framed their support in terms of economic exchange. For Ghanaians, helping others seemed to be a question of personal values and morals; that it was “right” to help other people in need. In an openended survey question that asked respondents to explain why they give help to other people, the most frequent response for Ghanaians was: to show kindness or generosity (46 percent). In contrast, only 15 percent of Ivoirians replied that they gave help to show kindness. The majority of Ivoirians replied that they gave help in order to increase their chances of receiving help in return (58 percent). Thus, help was not conceived in Côte d’Ivoire simply as being “right” or “good,” but in more strategic terms as an explicit economic exchange. While a significant number of Ghanaians (30 percent) also viewed help as an exchange, they totaled only about one-half of those responding similarly in Côte d’Ivoire. In informal interviews and focus groups as well, more people in Ghana mentioned altruism as their primary motivation for helping others than in

99

Men’s focus group (anonymous). Kyere, Côte d’Ivoire, August 1999.

Local Conflicts Over the Meaning of Reciprocity

87

Côte d’Ivoire. One woman described the force of the moral obligation she feels to help others: There are some people who might not be related to you in any way but when they tell you their problems and you can see that you can help, then you have to give them whatever they asked for.100

Another person also explained how you must help someone in need: “If only you have it, there’s no need to allow the person to go without giving it out. You have to give some.”101 Many Ghanaians also couched their desire to help someone else in spiritual terms, as the Christian and “merciful” thing to do.102 As one Ghanaian Christian explained: “I now go to church, and it is my spiritual duty to help those in need.”103 It was a rare exception when an Ivoirian mentioned God as the reason for helping someone else. It is important to note here that since the onset of British colonial rule, Protestant churches dominated an active religious scene in Ghana, which has become even more fervent with the rise of new evangelical and charismatic churches since the late 1970s. Meanwhile, the Roman Catholic Church dominated a more subdued religious environment in Côte d’Ivoire. This difference in the general religious climate could contribute to the greater tendency for Ghanaians to describe their motivations to help in religious terms than did Ivoirians.104 Ghanaians often described how the moral obligation to help someone depends on that person’s “good character.”105 Ghanaians reported giving help “to the one who is honest and has good morals.”106 It seemed that to Ghanaians, an integral component of good character was being diligent and hardworking.107 Ironically, many Ghanaians mentioned that “sick” or “mad” people would not receive any help, even though they may need it to survive.108 Thus, it seems that Ghanaians felt morally obliged as long as the person occupied the same moral universe. 100 101 102 103 104

105

106 107 108

Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Ibid. Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Survey interviews (anonymous) by author. Barima, Ghana. The book earlier rejected a related hypothesis that religion influences the level of help given, but it could still shape why that help is given. Survey interviews (anonymous) by author and Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Survey interview (anonymous) by author. Barima, Ghana. Ibid. In Ghana, mental illness was usually blamed on someone, often the mentally ill person himself or herself for smoking marijuana or drinking too much distilled liquor.

88

Informal Institutions and Citizenship in Rural Africa

In contrast, when character was discussed in this region of Côte d’Ivoire, “good” qualities were more restricted and straightforwardly tied to an individual’s economic conduct than in Ghana. One respondent contrasted “those who are nice” and receive more help with “the poor” who receive less.109 Good “behavior” in Côte d’Ivoire was directly linked to the person’s demonstrated record of repaying credit.110 Respondents described how “the one who doesn’t behave well, who doesn’t show respect, and who doesn’t reimburse loans” will get less help while “the one who behaves well and who reimburses loans” will get more.111 “Good behavior” in Côte d’Ivoire also signified someone who generally managed his money well.112 Even the words that were used in Ghana and Côte d’Ivoire belie differences in the underlying normative frameworks. The term used most frequently in Ghana was “character” which implies a person’s inner moral core or being, whereas “behavior,” used in Côte d’Ivoire refers to an outwardly visible history of concrete actions and deeds. A second main point of difference between the two cases is that helping in the Ghanaian region was not articulated in terms of an explicit economic exchange as frequently as it was in Ivoirian region. There is a shared expression in the local Akan language that describes the concept of reciprocity as “hand go and hand come,” but it was rarely referred to in Ghana and frequently referred to in Côte d’Ivoire. Both Akan and non-Akan Ivoirians repeatedly used another equivalent expression as well: “I give to receive in the future because the right hand washes the left hand and vice versa.”113 When reciprocity was mentioned in Ghana, it was discussed in a very long-term, more generalized way. For instance, one man said that he had helped provide lodging and food for a stranger to the village because he has children and family living in other towns that might one day need help.114 In Côte d’Ivoire, on the other hand, the reciprocity described was much more immediate and specific.115 “The one who can help me when I have difficulties 109 110 111

112 113

114 115

Survey interview (anonymous) by author. Opanin, Côte d’Ivoire. Survey interviews (anonymous) by author. Kyere, Côte d’Ivoire. Survey interview with 21-year-old Akan male (anonymous) by author. Kyere, Côte d’Ivoire. Survey interview (anonymous) by author. Opanin, Côte d’Ivoire. Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Survey interview with 25-year-old married non-Akan woman (anonymous) by author. Opanin, Côte d’Ivoire. Expression repeated in informal interviews and in Women’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Survey interviews (anonymous) by author. Barima, Ghana. There were a couple of Ivoirian respondents that replied that they did not expect to be reimbursed immediately. One said that he gave help for the future, for his children and

Local Conflicts Over the Meaning of Reciprocity

89

is the last one that I helped.”116 Another respondent maintained, “I will help first the extended family member because if I don’t help them, they won’t be able to help save me when I have problems.”117 The appeals were judged accordingly, too, so that if someone could help a respondent in the near future, the respondent would be more likely to help him. One thirty-fiveyear-old Akan woman in Côte d’Ivoire explained that with limited means, she would help only her “close family because it is they who can secure me . . . ”118 Similarly, Ivoirians often mentioned that if someone could repay easily, they would be given help more readily.119 I present evidence later in Chapter 8 that this type of specific reciprocity reinforced class-based social exclusion in Côte d’Ivoire. Third, fear of witchcraft was also cited as the reason for giving help to others much more frequently in Côte d’Ivoire than in Ghana.120 Interestingly, this motivating factor did not appear at all in responses to the survey question asking specifically about why people helped each other. Respondents mentioned witchcraft only in response to longer, more probing questions that asked what respondents would do in an imaginary situation in which their resources were limited (or even insufficient), so they could not satisfy one or more appeals from needy family or friends. As one forty-year-old married Akan man in an Ivoirian village explained: I will give to him [a relative in need]. It’s obligatory that I give to him. If I don’t give to him, that will go badly and witchcraft will intervene.121

Thus, not only do people face having their reputation ruined; worse, they believe that they risk injury or death: I will do everything to come to him with help because, here, if a family member traveled to come see you, you must help him by giving him one-third of what you have. If not, he will hurt you.122

116

117

118 119 120

121 122

grandchildren, “not for me immediately.” But this was rare. Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Survey interview with married Akan male (anonymous) by author. Kyere, Côte d’Ivoire. Similar observations recorded in Women’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999. Survey interview (anonymous) by author. Opanin, Côte d’Ivoire. Survey interviews (anonymous) by author. Kyere, Côte d’Ivoire. Geschiere (1997) also notes important regional variations in the incidence and nature of witchcraft in Cameroon and calls for more comparative studies. Survey interviews (anonymous) by author. Kyere, Côte d’Ivoire. Survey interview (anonymous) by author. Opanin, Côte d’Ivoire.

90

Informal Institutions and Citizenship in Rural Africa

Both in interviews and in everyday conversation, many Ivoirians told of how a failure to fulfill an obligation to a family member often leads to cursing or bewitchment.123 Ironically, Ivoirians also spoke of how becoming involved in helping a family member can lead to bewitchment as well, so they preferred to help those outside of their family. Respondents sometimes spoke of having to adopt a “hard” attitude toward family so that the witches would be afraid of cursing them. In contrast, witchcraft was never mentioned in Ghana in the survey interviews or focus groups in response to the same questions about giving help with limited or insufficient means available. Again, this is not to say that witchcraft was not a part of the belief system in Ghana; it was. However, the origin of a curse was more often attributed to a “stranger” instead of someone close such as a relative or friend as was done in Côte d’Ivoire.124 Finally, the study found that relatively few people in Ghana described giving help as a way to attain social prestige or “name.” The few people in Ghana who helped for this reason could not continue to do so in grave economic times. The motivation for social prestige seemed moderately more prevalent in Côte d’Ivoire, in which respondents described helping a brother in need in order to “be bigger in the eyes of others.”125 Nevertheless, an aspiration for social prestige appeared to be the least significant factor motivating people to help others in both Ghana and Côte d’Ivoire. The survey data confirm this conclusion finding that only 0.6 percent of respondents in Ghana and 2.3 percent of respondents in Côte d’Ivoire cited gaining prestige as the reason why they give help to others.

what happens when help is not forthcoming . . . What happens if a Ghanaian or an Ivoirian villager simply could not help a person who came to him in need? Or, what if he refused? If there is a commonly shared norm that help should be given in certain instances or to 123 124

125

Survey interviews and focus groups (anonymous), Kyere and Opanin, Côte d’Ivoire. For example, during the fieldwork in Ghana, one of my research assistants, a “stranger” to the village, was accused of shrinking a woman’s breasts. After a meeting of the chief, elders, and all parties involved, the accusation was declared unwarranted, and a public apology was made. Even a false accusation can have serious consequences, however, as many “strangers” lost their lives to mob justice before the proper authorities could intervene during the previous year in Ghana. The observation of a more frequent ascription of witchcraft to strangers in Ghana is particularly interesting given earlier work by Bannerman-Richter (1982) and Assimeng (1989) that describes the operation of witchcraft only within the “abusua.” Survey interview with 65-year-old Akan man who described himself as “wealthy” (anonymous) by author. Opanin, Côte d’Ivoire.

Local Conflicts Over the Meaning of Reciprocity

91

certain types of people, then being unable or unwilling to help in those situations should produce conflict and tension. The study finds that these points of discord were not the same in Ghana and Côte d’Ivoire. In general, the lack of ability or refusal to help stimulated negative feelings of sorrow or shame on the part of the individual in Ghana, but less interpersonal conflict than in Côte d’Ivoire. First, many more Ghanaians described how the lack of fulfillment of their moral duty to help brings sorrow. One Ghanaian woman described how she at first refused to help her sister, even when she knew that she had some money saved: Giving part of my money to her wasn’t something I wanted to do so I told her I had none. Looking at her going empty-handed made me feel so sorry so I called her and gave her part of the money.126

Another Ghanaian woman agreed that you feel sorry when you do not help someone, but added that when the other person knows that you have the money and you still refuse, they are hurt. She implicitly acknowledged the potential fear of witchcraft by adding that they “won’t do anything against you though. He’ll just feel hurt.”127 Second, consistent with Ghanaians’ tendency to couch their moral duty in spiritual terms, when that moral duty went unfulfilled, it was considered a sin. A Ghanaian woman relayed: “If that person has some but says he won’t give you, he has sinned. It’s a sin against God.”128 Another Ghanaian woman explained: If you don’t have it, even God knows you don’t have it. But if you have even a little, you must help with that bit . . . Otherwise, you’ll also have difficulty the next day, it’s not good . . . it’s a sin . . . 129

God was mentioned as the ultimate judge in Ghana much more frequently than in Côte d’Ivoire. A Ghanaian woman declared: If that person has no fear of God in him, no matter what anyone says, he will just not take care of the elderly (in need of help). But, on the other hand, if the person loves God, he will live up to what’s expected of him.130

A twenty-eight-year-old, non-Akan man living in Ghana agreed that “looking after an older person is a service to God. If one refuses to do it, nobody would say or do anything about it.”131 126 127 128 129 130

131

Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Ibid. Ibid. Ibid. Survey interview with a 56-year-old Akan female who takes care of an elderly member of her family (anonymous) by author. Makwan, Ghana. Survey interview (anonymous) by author. Makwan, Ghana.

92

Informal Institutions and Citizenship in Rural Africa

While the first two points demonstrate how Ghanaians were most influenced by the fulfillment of their own personal moral or religious code, the next point shows that Ivoirians appeared more threatened by the potential of social stigmatization. The varied responses to one of the survey questions revealed this difference well. The question asked respondents to imagine a situation in which a relative pleads for assistance at a time when the respondent’s own means are inadequate to pay for the health and education of their own nuclear family. The majority of Ghanaians seemed to want to find a way out of the difficult situation without being able to offer much help. More than 43 percent would take the family member to ask for loans elsewhere, whereas another 40 percent would simply explain their own dire circumstances but offer no help. The overwhelming majority of Ghanaians were comfortable refusing to help their family member when they themselves were in hardship. In contrast, the majority of Ivoirians (82 percent) replied that they would give the family member some of the little money they had. “If I have it, I will give it to him. Even if I don’t have all of it, I will share the little I have with him . . . ”132 Ivoirians seemed less willing to refuse, even when their own situation was difficult, because of the greater conflicts that would ensue. A follow-up to the question revealed that Ivoirians were more fearful of not helping their needy family members, even when it was difficult to do so. More than 91 percent of Ivoirians answer that their relatives would call them “wicked,” “selfish,” or “evil” if they refused to help. One twenty-five-yearold Ivoirian male explained: I will give (the family member) one half (of what I have) because this is a member of the family . . . If I don’t help him, he will talk badly of me . . . The situation would be different (with a friend) because I can tell him to wait until I have some money and I can explain to him my situation.133

Another respondent described how a person would cut visible ties and essentially go into hiding when they were no longer able (or willing) to help. “When they come [to the village], they pass unseen or even hide themselves discreetly and then return without being seen.”134 In both Ivoirian villages, village residents bemoaned the fact that several wealthy and influential “sons” had cut ties with the larger extended families, basing their residence in the regional or national capital and only very rarely visiting 132

133 134

Survey interview with a 30-year-old married Akan female (anonymous) by author. Opanin, Côte d’Ivoire. Survey interview (anonymous) by author. Kyere, Côte d’Ivoire. Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999.

Local Conflicts Over the Meaning of Reciprocity

93

the village, usually late at night, or surreptitiously. No one in Ghana mentioned this type of social stigma in response to that particular question. In general, Ghanaians seemed relatively more understanding if someone was unable or unwilling to help. Several people mentioned that if a relative’s money was “his own,” he could do whatever he wanted to with it.135 If one is doing his own work to get his own money and is a family elder (abusua opanyin) and you come to him for help and he doesn’t help you, you can’t say anything to him because it’s his own money and he worked to get his own money so he is using it to care for his wife and children.136 The family is there as a family, but it is like a forest, every tree has its position. Every person is struggling for his children, so you can’t rely on them; that my family member is working hard so I should relax.137

Several respondents described how there were no consequences if a family member decided not to help another relative. In contrast, Ivoirians described the anger and hostility that would necessarily ensue in the relationship. “He will be angry and he might curse you . . . because to him, you were bad.”138 Many Ghanaians mentioned that the elders of the family would call the person with modest means and advise him or her to “live up to expectations.”139 Ghanaians also described how a friend, the village chief, or a pastor might also call and talk to the person about taking care of a relative in need. Nevertheless, Ghanaians emphasized that the family member could not be “forced,” but they must help on their own “initiative” or from their own “free will.”140 Some Ghanaians framed this as meaning that helping when you had the means was “not an obligation,” whereas others said that you could “not be forced to do the right thing,” implying that the norm would be to help.141 Several people in Ghana said that if someone (who could help but did not) were “bad” or “wicked,” then they would not heed the advice or pressure from others anyway. Most frequently, Ivoirian respondents described how this difficult situation might be resolved through particular individual efforts, not necessarily the coordination of the family or community.142 While Ghanaians frequently

135

136 137 138 139 140

141 142

Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Women’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Survey interviews (anonymous) by author. Opanin, Côte d’Ivoire. Survey interviews (anonymous) by author. Barima, Ghana. Survey interviews (anonymous) by author. Barima, Ghana, Women’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Survey interviews (anonymous) by author. Barima, Ghana. Survey interviews (anonymous) by author. Kyere and Opanin, Côte d’Ivoire.

Informal Institutions and Citizenship in Rural Africa

94

mentioned the role of a friend, chief, or pastor in mediating the potential conflict, this wider network of social ties was rarely if ever mentioned in Côte d’Ivoire. Perhaps because of the lack of conflict resolution mechanisms, Ivoirians expressed discomfort at breaking the expected norms of helping. Finally, many Ghanaians expressed a depressed realism that anyone asking for help would understand their refusal simply by seeing their “plight” or “condition.”143 Many respondents in Ghana described how no one would even come to them for help because it is known that they do not have anything to give. “They know that I have nothing on me.”144 In this way, because their resources were so obviously and perhaps, publicly, limited, Ghanaians could not even conceive of potential social stigmatization. Almost no one described this kind of situation in Côte d’Ivoire. Social stigma and the potential for conflict seemed much more real and forbidding.

conclusion The meaning of reciprocity – in terms of who, how and why people helped each other – differed at the village level in surprising ways in these two regions. First, the informal institutions of reciprocity appear to have changed over time in both regions of Ghana and Côte d’Ivoire. Village residents in both regions described how the reciprocity of today was different than it had been in the recent past. In particular, in both regions, the norms of reciprocity were narrowing over time. It seemed more difficult now to get help when in need. For example, in both cases, there was an increasing prevalence of loans (sometimes with quite high rates of interest), instead of outright grants of aid, even to members of the extended family. The practice of reciprocity as a village collectivity also fell far short of the ideal in both regions. This is not simply an older generation’s nostalgia for the “good ole days,” but changes that were noted by youth and people of all ages. This chapter thus begins to demonstrate empirically the importance of history in our examination of the informal institutions of reciprocity. Finally, the informal institutions of reciprocity were changing in different ways in the two regions. Even the above overall trend of restricted reciprocity occurred in particular patterns in the Ghanaian fieldsite villages versus the Ivoirian ones. These patterns suggest the importance of the Ghana–Côte 143 144

Survey interviews (anonymous) by author. Makwan, Ghana. Survey interview with 60-year-old widowed Akan female with four kids (anonymous). Makwan, Ghana. Nearly the exact same words were used by many others of a wide range of ages, marital status, and dependent load, for example, a 19-year-old single Akan man said: “They know I have nothing to give.”

Local Conflicts Over the Meaning of Reciprocity

95

d’Ivoire border in an area considered to have a blurred boundary. Overall, Ghanaians articulated broader and more inclusionary institutions of reciprocity than did Ivoirians. For example, the concept of the family was defined relatively more broadly in Ghana than in Côte d’Ivoire, where “la petite famille” was a familiar and frequently invoked term. Furthermore, in Ghana, reciprocity was considered a moral and often a religious duty that included a significant place for friends and others; whereas in Côte d’Ivoire, even help to one’s children was framed as a more calculated economic exchange. These differences were truly surprising because the regions and villages compared in the book shared so many similarities – in particular, the same broad Akan cultural history. Now both the quantitative analysis in Chapter 2 and the qualitative interpretive analysis in Chapter 3 have highlighted important crossnational differences. But, exactly how and why do national-level political institutions matter? In the next three chapters, I use a deeply historical approach to dissect the state, revealing the mechanisms of articulation between the formal state and informal institutions of reciprocity over time in these two regions of Ghana and Côte d’Ivoire. I explore how colonial state legacies have influenced the postcolonial state’s mediation of risk through policies of political administration (Chapter 4), the provision of social services (Chapter 5), and agriculture (Chapter 6). Then, in the book’s final section, I investigate the implications of these interactions for local notions of citizenship and democracy.

part ii LEGACIES OF THE STATE ROLE IN MEDIATING RISK IN GHANA AND CÔTE D’IVOIRE

chapter 4 The Legacies of the Colonial Administrative State in Constructing the Citizen, Family, and Community Roles

The informal institutions of reciprocity differed in puzzling ways in these similar Akan villages on either side of the Ghana–Côte d’Ivoire border. More people were participating and exchanging significantly higher quantities of help, but to a narrower group of social ties in the Ivoirian region than in the Ghanaian one. But the variation in informal reciprocity was not the only surprise I found during my stay in these four villages. Elders in these regions told very different stories about the village chiefs during the colonial era and how they responded to the hard times of the Great Depression. In the Ghanaian villages, elder respondents recalled how their chiefs were powerful and sometimes organized resistance to disliked British colonial policies. In contrast, in the Ivoirian villages, older residents remembered their chiefs as representing French colonial interests, frequently describing the chiefs’ participation in conscribing forced labor. As I conducted my fieldwork, I then recognized legacies of the past in the different response to everyday problems in the two regions. In the Ghanaian villages, residents approached the village-level unit committee and/or the village chief to resolve an urgent problem. Meanwhile, in the Ivoirian villages, people reported a problem to the sous-préfet, located more than an hour away, or sought even more distant advice from a “big man” in Abidjan. In this next part of the book, I analyze the connections between these puzzling differences in state-building and reciprocity. I argue that fundamentally different historical experiences of state-building in political administration, social policy, and agricultural policy have produced surprising local variation in informal institutions of reciprocity in these villages. Despite sharing many common pressures over time, the Ghanaian and Ivoirian states 99

100

Informal Institutions and Citizenship in Rural Africa

t a b l e 4 . 1 . Divergent Normative Frameworks of British and French Colonial Administrative Rule

Normative Viewpoints Role of state in economy Organization of state Nature of society to be governed Nature of family Impact of political administration on informal institutions of reciprocity

British in B/A Region of Ghana

French in Abengourou Region of Côte d’Ivoire

Liberal Decentralized Ethnic communities

Statist Centralized Individual subjects/potential citizens Extended family system Nuclear family system Diversification of reciprocal Concentration of reciprocal ties across wider lineage ties within nuclear family village and ethnic group and proximate extended within rural areas family linking to urban capitals

have pursued historically different roles in the mediation of risk vis-a` -vis these rural regions, with profound effects on the local cultures and politics. To begin Part II, Chapter 4 explores the long-term legacies of the states’ differing approaches to political administration in these regions of Ghana and Côte d’Ivoire. Political administration includes how to administer, govern, and maintain basic order and security in the community. It involves determining who has the power to make decisions in the society. As such, the comparative analysis of the different approaches and experiences of political administration in these two regions is the essential starting point for this book’s explanation. In this chapter, I compare the basic differences in the concepts of the state role, nature of the society to be governed, and the organization of the family, and how those differences were experienced on an everyday basis in these two regions over time. I explore how these contrasting histories of political administration were linked to remarkable differences in the informal institutions of reciprocity in these regions of Ghana and Côte d’Ivoire. Beginning in the late 1880s, during the colonial era, the British and French had contrasting normative views about who they were governing and how best to go about doing it. (See Table 4.1.) The differing perspectives held by the British and French on the nature of citizenship, the family, and community led to the construction of different types of formal state administrative institutions, and ultimately different social and agricultural policies (discussed in more detail in Chapters 5 and 6), in a complex iterative process over time. Thus, despite similar overarching goals and objectives, the British

The Legacies of the Colonial Administrative State

101

administrative system in colonial Ghana ended up looking quite different from that organized by the French in colonial Côte d’Ivoire. The British developed a more decentralized system, in which the colonial state interacted indirectly via-a` -vis different ethnic communities, while the French established a strongly centralized administration, in which the colonial bureaucracy intervened more directly to govern what the French perceived as future individual citizens living in nuclear families. Nearly 100 years later, in the early 1980s, despite similar pressures from international financial institutions to decentralize, the Ghanaian and Ivoirian state’s approach to political administration has continued to diverge. In Ghana, the government has attempted to devolve decision-making to the district and even village level whereas in Côte d’Ivoire, rhetorical efforts to decentralize have thinly concealed an outward extension of central state authority over the outlying areas, never allowing any direct village-level decision-making. In this chapter, I show how the local experience of these differences in formal state administrative policies shaped the informal institutions of reciprocity. In the later chapters of Part II, I proceed by examining how these different experiences in politically governing local communities early in the twentieth century critically shaped the subsequent construction of the state apparatus for social service delivery and agricultural policy in the colonial Gold Coast (renamed Ghana at independence) and Côte d’Ivoire.

precolonial state-building: similar norms and formal institutions from the asante kingdom in ghana and coˆ te d’ivoire Early European explorers to Africa drew barren maps that detailed only the nascent European trading presence on the coastal geography, but left the continent’s interior essentially void. Certainly, a few lions and elephants were drawn wandering around the vast empty territory, but none of the indigenous political boundaries was depicted. Later missionaries and colonial officials would similarly describe Africa as lacking centralized states, emphasizing in the narratives sent back to Europe that Africans lived so primitively in small villages.1 State-building in Africa did not begin with European colonialism, however. Nor were the underlying norms of the state, family, and community established then. In fact, a long history of state-building predates the colonial 1

Crowder (1968: 12) makes the argument that this view of Africa as anarchic was an important justification for the imposition of colonial rule and expansion of missionary activity.

102

Informal Institutions and Citizenship in Rural Africa

presence.2 The Akan peoples migrated to the forest regions of what is now central Ghana in the thirteenth century A.D. and were organized as small states. Later, around the seventeenth century, the Asante kingdom was involved in countless wars with the Denkyira and various other Akan subgroups, attempting to enlarge its borders and consolidate power over its subjects.3 Indeed, these conflicts provoked several groups to flee from the central part of Ghana in the seventeenth century to what is now the more western part of Ghana and eastern Côte d’Ivoire. Since the colonial rulers drew the current borders of the African state somewhat arbitrarily, precolonial experiences with state-building were highly varied within each country case. The frontiers of precolonial states did not coincide with colonial territories. In fact, many kingdoms, political alliances, and ethnic groups stretched across the new boundaries. This is precisely the case in the regions of Ghana and Côte d’Ivoire, where the research for this book was conducted. In both regions, most residents belong to the overarching Akan ethnic group and trace their origins back to the Asante Empire, having fled conflict in search of a peaceful territory to settle. The migrant groups that settled in each country then established a centralized kingdom based on a hierarchy of chieftaincy. In both cases, some form of loose political alliance was maintained with the Asante kingdom, which continues even today.4 Even though there is less known about the precolonial norms of the state, family, and community in these areas, McCaskie has argued convincingly that the Asante state played a role in shaping the operation of kinship in Akan society in precolonial times.5 Paralleling this book’s argument for the more contemporary period, McCaskie emphasizes the interaction between kinship and the state over time. This argument stands in contradistinction to earlier anthropological work that predicted the disappearance of kinship 2

3 4

5

African historians first examined the written archival records of the various colonial powers so initial scholarly attention focused on the African encounter with colonialism. A second generation of African historians has combined sources including oral histories, missionary records, newspapers, and travel writing to analyze state formation in earlier precolonial periods. See McCaskie (1995a, 1995b) and Wilks (1993). McCaskie (2007). Notably, when the Asantehene, or Asante king, died during my fieldwork stay, the village chiefs and local paramount chiefs all traveled to Kumasi to attend the funeral. There are a few anthropological studies that attempt to analyze the precolonial structures of state power, but rarely do they discuss state formation in terms of norms. On the Agni in Côte d’Ivoire, see Perrot (1982). On the Asante in Ghana, see Wilks (1993); Busia (1951); Ackah (1988); and Rattray (1923, 1929). McCaskie does attempt to tackle what he calls anthropologists’ questions about culture and beliefs of the Asante in precolonial time. See in particular McCaskie (1983, 1995b).

The Legacies of the Colonial Administrative State

103

institutions with the centralization of the state.6 McCaskie does indeed acknowledge the highly coercive interventions of the centralizing Asante state, highlighting the resultant tensions between the normative principles of communal egalitarianism associated with the lineage in contrast with the individualistic emphasis on accumulation espoused by the state. But McCaskie also shows how the Asante state rarely legislated in areas that challenged the aman mmu, or the “immemorial custom that ordered a community.”7 Furthermore, where the Asante state had challenged customary actions by the lineage, these measures actually strengthened “jural corporateness by clarifying its boundaries and so underscoring its central, non-negotiable integrity.”8 Similarly, McCaskie demonstrates that access to citizenship in the Asante state was bound up with lineage membership and further, how the state managed to structure whether and how individuals were incorporated into the lineage: In the case of Asante citizens the state acted, on the one hand, as the guarantor of the principle of jural corporateness. But on the other hand, the state constantly intervened in the definitions accruing to individual cases and kin networks within the facts of jural corporateness. It continually adjusted the parameters of legal status in relation to its own project, recasting people in a succession of more or less privileged roles inside jural corporateness, while at the same time retaining final rights of arbitration over the boundaries of incorporation and expulsion.9

McCaskie’s work thus presents a convincing argument that even the precolonial patterns of state formation interacted with and reshaped the informal institutions of reciprocity at the village level. What is unique about this study’s case selection, however, is that we can essentially control for precolonial culture and politics by comparing Akan groups that have similar cultural backgrounds and political histories vis-a` -vis the Asante state. Like McCaskie, I argue that precolonial politics did matter for the development of local norms, but in these two crossborder regions, the political histories were so similar that the effects on the informal institutions of reciprocity were most likely analogous. I thus proceed by concentrating on how more recent periods of state-building since the colonial era have shaped notions of kinship and community.10

6 7 8 9 10

McCaskie (1995b: 74–143). McCaskie (1995b: 87). McCaskie (1995b: 88). McCaskie (1995b: 90). The study is obviously facilitated during the colonial period by the existence of written colonial records maintained in the archives. While the majority of records found are written by colonial officials either based in the colonies or at times in the metropole, many of the documents referenced in the book cite the dialogue or correspondence of Africans or are

104

Informal Institutions and Citizenship in Rural Africa

colonial state-building: contrasting british and french approaches to political administration With the onset of the colonial era, key differences in state formation began to emerge between these two regions. The political institutions designed by the British and the French to administer their new colonies differed in remarkable ways. While the British promoted a more indirect, socially mediated relationship between the citizen and the state, the French advocated a direct relationship.11 While these differences are inadequately described by the classic (and contested) typology of “indirect” and “direct” rule in Anglophone and Francophone Africa, it is an equally unsatisfactory oversimplification to conclude that because both the British and French at times relied on chiefs, the systems were virtually the same.12 I am not reifying oldfashioned stereotypes of colonial rule, but instead examining the kinds of administrative policies that were actually on the ground and how they were perceived by indigenous Africans at various points in time.13 The goals of the British and French colonial systems were conceptualized differently by the colonial rulers and experienced differently by African societies.14 These fundamental contrasts in the British and French colonial conceptions of the state and citizenship were experienced even more intensely at the village level with the significant expansion of the central state after 1940.15 My argument is that these differences in the ways that the British and French built the formal administrative institutions of the colonial state stimulated variation in the informal institutions of reciprocity.

11

12

13

14 15

actually written by Africans themselves. Nevertheless, it is always important to problematize who is writing or collecting the documents, for what audience, and for what purpose. While I could probe endlessly backwards to locate the origins of these colonial norms in prior social structures or institutions in the metropole, for the purpose of this analysis, it is sufficient to say that these divergent normative frameworks were shaped by the particular social and political histories of each colonial power. See Young’s (1994: 99–100) brief comparison of differences between British and French (and other) state formation in Europe with their approach to colonial administration in Africa. While many contemporary scholars would acknowledge the social, economic, and political impact of colonialism on African societies, there is a range of views on how these processes unfolded, in particular with regard to the degree of African agency vis-a ` -vis the colonial state project. See Ekeh (1975); Berry (1993); Cooper (1996); and Young (1994). Crowder (1968: 171–2) acknowledges that Britain had a dominant tendency of indirect rule and France of assimilation (direct rule) but that there was no one official policy given the complexity of governing Africa. Nevertheless, he argues that the difference between British and French administration was not one of “degree” but much more profound (234). See Geschiere (1993) for a comparative analysis within Cameroun. See also Amon D’Aby (1988) and Piault (1987). See Firmin-Sellers (2000). See Cooper (2002).

The Legacies of the Colonial Administrative State

105

British Colonial Administration In general, in the colonial territory of the Gold Coast, the British authorities built a highly decentralized administrative organization that was based on the existing structure and authority of chieftaincy and predicated on a conception of subjects belonging to ethnic, tribal communities. I argue in this section that the distinctive British approach to colonial state-building reinforced the diversification of the informal institutions of reciprocity. Following the Portuguese who arrived first in 1471, the British established several slave forts along what was then known by Europeans as the Gold Coast. Throughout most of the nineteenth century, the Asante state attempted to expand its rule and promote its own trading interests, repeatedly battling with the British merchants and government.16 By 1821, the British government eliminated the African Company of Merchants who had operated there and formally established the British Gold Coast. Over the next several decades, control of the Gold Coast settlements would revert back and forth between the British merchants and the crown government. During this period, the British would take over the coastal territories controlled by the Danish (1850) and then the Dutch (1871).17 During the last quarter of the nineteenth century, the British turned their attention to conquering the interior, which led to four wars with the Asante state (1863–1864, 1873–1874, 1893–1894, and 1895–1896).18 The coastal areas became an official crown colony in 1874. Then, after defeating a final Asante rebellion and extending control over the Northern Territories, the British established the entire Gold Coast as a British colony under the authority of the British governor in 1902. With Asante resistance restrained and the Asantehene monarch in exile, British colonial administration was expanded and strengthened earlier than in neighboring Côte d’Ivoire.19 Since the mid-1800s, the British had established increasing judicial and legislative authority over the coastal areas but now the coast, Asante, and Northern Territories were unified under one central administration. Nevertheless, the character of British colonial administration continued to be highly decentralized. For example, since 1874, the Native Jurisdiction Ordinance had essentially established the minimal interference by the colonial officials in the political affairs of the Colony chiefs. According to this legislation, indigenous authorities could enact bylaws, organize tribunals, and administer justice in minor cases. 16 17

18 19

Ward (1958). On the rivalry between European powers to establish control and confirm boundaries in Africa, see Packenham (1991). Crowder (1968: 143–50). The British forced the Asantehene Prempeh I into exile in 1896.

106

Informal Institutions and Citizenship in Rural Africa

After the 1930s, the approach to colonial administration in the Gold Coast was heavily influenced by the writings of Lord Lugard, the high commissioner and later governor of Nigeria in the early 1900s and author of the British policy of “indirect rule.”20 The British goal was to establish a colonial system that was able to “adapt as far as possible the indigenous African institutions” (i.e., chieftaincy) for administrative goals.21 Lugard had argued that this system of indirect rule was efficient economically and politically. First, economically, fewer European personnel were required in the hinterlands as the British relied on indigenous political institutions to govern. Second, politically, the British system drew on the preexisting legitimacy of the chieftaincy structure to govern, thereby reducing the opposition to British rule. Several key legislative acts highlight how the British incorporated chieftaincy structures into their system of colonial administration in the Gold Coast as well as how they sometimes intervened to modify them.22 In 1925, provincial councils of chiefs were established in all three territories. In 1927, the British regulated and attempted to strengthen the authority of the chiefs and councils with the Native Administration Ordinance. In 1935, the Native Authorities Ordinance unified the central and native administrations, establishing that native authorities were appointed by the British governor, but giving them extensive power at the local level. In the Gold Coast, methods of taxation also remained highly decentralized and “customary” through the chief’s management of “native treasuries.” Although early on the chiefs were not required to collect taxes directly from their subjects to give to the Colony, they also had no local budgets allocated from the colonial government. Instead, they governed through the customary mechanisms of collecting tribute and selling stool (chiefly) lands. It was only in 1939 that the British expanded their regulatory control over the chiefs’ collection and management of tax revenue through the Native Treasuries Ordinance, and in 1944 that the British required chiefs to impose direct taxation through the Native Authority Ordinance.23 The implementation of direct taxation occurred significantly later than in neighboring Côte d’Ivoire. 20 21 22

23

Lugard (1926). Ghana National Archives (1938). Crowder (1968: 169) distinguishes between Lugard’s “interventionist indirect rule” which sought to improve indigenous institutions and his successors “non-interventionist rule” which allowed indigenous institutions to develop along their own lines. See also Boone (2003: 146–9). Crowder (1968: 224).

The Legacies of the Colonial Administrative State

107

The British approach to political administration was based on their view of African subjects as culturally distinct. In contrast with the French policy of assimilation, the British did not envision the residents of the Gold Coast as potential future British citizens. Rather, their African subjects were seen as members of separate cultural groups, with unique languages, customs, and political institutions. As a result, the British dedicated significant time and resources for colonial anthropologists and administrators to study, understand, and classify the different tribal groups in each colonial territory. Cooper highlights how the “ideological power of ‘tribe’” shaped other colonial policies, British migrant labor policies in particular, in the early twentieth century.24 So how was the British approach to political administration actually experienced at the local level in this region of Ghana? And how might this experience of state formation have influenced the informal institutions of reciprocity? Posner has shown that British colonial policies had long-term legacies that continue to shape the nature of political identities and conflict in contemporary Zambia. This book similarly analyzes the nature of colonial administration and traces the effects on political culture today. Oral histories that I collected reveal that many of the local people thought of their village chiefs during the colonial period as “powerful.”25 Another Ghanaian elder described chiefs as “leaders [and] spokesmen for the people . . . the chiefs were the local government.”26 In general, older respondents from the Ghanaian regions highlighted colonial-era chiefs as the dominant, primary, and exclusive authority figures; not subservient auxiliaries to the colonial officials, as articulated in the Ivoirian region across the border. Even when chiefs and commoners came into conflict over land rights or the use of native treasuries, Ghanaians in this region continued to look to the local political institutions of chieftaincy to resolve these issues. A crisis period that sheds light on the political dynamics between the colonial state, chieftaincy, and other societal groups is the cocoa holdup of 1936–1937.27 In the Gold Coast, the chiefs played a significant role in a farmers’ movement to refuse to sell cocoa to the European buyers who were perceived as colluding to suppress the local producer price.28 The chiefs used the Akan cymbal-like instrument called the “gong-gong” to spread the news of the holdup throughout their villages and enforce unified action among 24 25 26 27 28

Cooper (1996: 49–50). Posner (2005). Interview (anonymous) by author. Barima, Ghana, February 7, 1999. See Alence (1990); Austin (1988, 2005); and Miles (1978). Interview with elderly cocoa farmers by author. Makwan, Ghana, April 7, 1999.

108

Informal Institutions and Citizenship in Rural Africa

cocoa farmers. The chief’s use of the “gong-gong” was pivotal to the success of such a grassroots opposition to the colonial power. Today, elderly cocoa farmers still remember their fathers and uncles piling their cocoa beans in the village center and burning it in accordance with the chief’s edict.29 The British approach to political administration served to validate and strengthen the role of indigenous “tribal” political institutions, including participation in tribally based relations of reciprocity. The British central colonial state was not intervening directly to mediate risk for its subjects; rather, its minimal interventions clarified the importance of maintaining ties and membership in one’s ethnic or tribal community. In all likelihood, the lessnuanced understanding of tribes in the Gold Coast developed by the British reinforced the salience of affiliation with a broader ethnic community for reciprocal social support as opposed to the lineage or immediate nuclear family. Later in the chapter, I will show how these legacies of British colonial administration can be seen in subsequent attempts at political reform in the contemporary era. French Colonial Administration Overall, the French approach to colonial administration was highly centralized based on a gradual extension of central bureaucratic control and decision-making. The French style of direct administration was predicated on a conception of subjects as individuals who had the potential to evolve and assimilate as future French citizens. Here, I will argue that the French approach to colonial state formation stimulated an emerging trend toward concentrating the informal institutions of reciprocity on the nuclear family. Like in the Gold Coast, the Portuguese were the first to arrive in Côte d’Ivoire in the 1460s. Unlike the Gold Coast, however, Côte d’Ivoire’s harbors were not preferred by Europeans, so European contact was less intense, and the slave trade less devastating. Only in the middle of the nineteenth century did France take a more sustained interest in securing a commercial monopoly of trade along the coast. At that point, France began to negotiate agreements with local chiefs and then sought to conquer the interior. Nearly a decade after the Berlin Conference, Gabriel Angoulvant wrote that in 1893: “The Ivory Coast belonged to us, but only in the eyes of foreign powers.”30 Angoulvant later became governor in 1908 and solicited 29

30

Interviews (anonymous) by author. Barima, Ghana, February 7, February 13, February 19, and March 4, 1999, and Makwan, Ghana, April 7 and April 13, 1999. Gabriel Angoulvant, La Pacification de la Côte d’Ivoire, 1908–1915: Methodes et Resultats (Paris: 1916) as cited in Crowder (1968: 109).

The Legacies of the Colonial Administrative State

109

approval from Governor-General William Ponty to initiate a full-scale military conquest of Côte d’Ivoire. In the fieldsite region of Côte d’Ivoire, the French negotiated treaties with the Agni kingdoms early on.31 “Pacification” of the rest of Côte d’Ivoire was difficult and prolonged, however, with resistance continuing until 1917. Elders in this region of Côte d’Ivoire still spoke admiringly of Samory Toure’s legendary battles with the French. The French military subjugation of much of the interior of Côte d’Ivoire was achieved at a huge cost, including the loss of numerous lives, destruction and reorganization of many villages, displacement of populations, imposition of war fines on tribes, and imprisonment of rebels.32 As a result, the expansion of French colonial administration did not commence until much later than in the Gold Coast. Whereas in most of Côte d’Ivoire, the French adopted a policy of “direct rule” and appointed new agents of the central state to govern the rural areas, the French worked with the existing Agni chiefs in the fieldsite region.33 In a similar vein, in most of Côte d’Ivoire, the French ignored preexisting ethnic or linguistic boundaries and defined canton districts arbitrarily based on central administrative efficiencies rather than local cultural meanings. In contrast, in the fieldsite region, the French canton borders actually overlapped with meaningful precolonial Agni state boundaries.34 Despite these differences however, the way the French worked with the Agni chiefs differed profoundly from the British style of indirect rule. The Agni chiefs served primarily as a mouthpiece for the new central administration, not as functioning sovereigns with their own bases of authority. The objective of the colonial authorities was to use chiefs as “representatives of French sovereignty” or “indispensable auxiliaries” to spread the influence of the central government.35 Governor-General Van Vollenhoven articulates this perspective in his circular on indigenous chiefs on August 15, 1917: The Commandant de Cercle alone gives orders; only he is responsible. The native chief is only an auxiliary instrument . . . they have no power of their own of any kind, for there are not two authorities in the cercle: French authority and native authority; there is only one . . . The native chief never speaks, never acts in his own name, but 31

32 33

34 35

Boone (2003: 182) argues that colonial conquest was prolonged because a history of precolonial decentralization made it difficult for the French to find central rulers who could sign treaties and negotiate peaceful trading relationships. An exception was the treaties signed between the French and the Agni kingdoms of Sanwi, Moronou, and Indenie. Crowder (1968: 111). This point is highlighted by Boone (2003: 183–4) who demonstrates the necessity of looking at subnational experiences of state formation. Zolberg (1964: 53). French National Archives (1932–1947) and Alexandre (1970: 64).

110

Informal Institutions and Citizenship in Rural Africa

always in the name of the Commandant de Cercle and by formal or tacit delegation from the latter.36

Decision-making hence was not delegated to the local chiefs, but rather increasingly centralized after 1895 where the Governor of Côte d’Ivoire had to coordinate and seek approval for actions from the governor-general of the whole French federation of West Africa, based in Dakar.37 In fact, the use of Agni chiefs to promote such ignominious French policies as forced labor and direct taxation actually undermined the preexisting legitimacy of these rural elites.38 In stark contrast with the legislation passed within the Gold Coast giving chiefs greater autonomy, the relevant legislation for the colony of Côte d’Ivoire was usually made by decree by the minister of the colonies based in Paris, promulgated by the governor-general based in Dakar and lieutenant-governor in Abidjan, and then implemented by the Commandants de Cercle at the local level. At every level of the bureaucracy, the people making and carrying out decisions were French. Several decrees passed after 1902 increased the authority of the governor-general and gave him greater control over local administration throughout the West African Federation, including Côte d’Ivoire. In particular, the governor-general in Dakar had federal control over budgeting, economic development, justice, and military operations. Indeed, for most of this period, the colonial heads of government were titled “lieutenant-governors” and were redesignated as “governors” only in 1937.39 The only avenue for Ivoirian representation in decisionmaking was through the Superior Council of Colonies, again based in Paris, to which each colony sent one delegate.40 The French authorities also implemented a highly centralized system of direct taxation much more rapidly in Côte d’Ivoire than in the Gold Coast whereby individuals paid a personal tax directly to the central state representative. The objective of direct taxation was not only to generate revenue for colonial administration but also to force Ivoirian subjects to increase production of cash crops. The comparative impact of the different styles of colonial administration was not lost on the French, either. Indeed, French colonial authorities attributed the exodus of Akan people along the border of 36

37 38 39 40

Governor-General Joost van Vollenhoven, “Circulaire au sujet des chefs indigenes,” Dakar, August 15, 1917, reprinted in Une Ame de Chef: Le Gouverneur J. van Vollenhoven (Paris: 1920), p. 207 as cited in Crowder (1968: 187–8). Crowder (1968: 176–7). Firmin-Sellers (2000). Crowder (1968: 180). The colony of Senegal also had a Deputy of Senegal that was sent to the Chamber of Deputies in Paris since 1914.

The Legacies of the Colonial Administrative State

111

Côte d’Ivoire into the Gold Coast to the differences in their respective methods of taxation. The French approach to political administration in Côte d’Ivoire was based on a very different view of their African subjects and societies from the British in the Gold Coast. The French in Côte d’Ivoire emphasized the potential universal equality of men as individuals instead of the cultural differences between ethnic and cultural groups, as was done in the Gold Coast. On the one hand, the French could conceive of Africans as becoming full-fledged citizens in the future; on the other, this was predicated on the extermination of African language, culture, and law; and the inculcation of French language, culture, and law. As a consequence of this French policy of selective assimilation, French administrative institutions were built to encourage the evolution of individual citizens living in smaller, nuclear families. In the next chapter, I will show how this contrasting normative framework of the individual citizen, family, and community has created profound differences in social policy. The experience of this French approach to political administration in this region of Côte d’Ivoire is reflected in the oral histories I collected in the fieldsite villages. Many older inhabitants characterized the chief during the colonial era as “the representative of the state here,” who convened meetings to disseminate information from the central government.41 Many elderly Ivoirians also mentioned that the chief worked with the colonial authorities to satisfy their demands for forced labor, a brutal aspect of colonial state formation. In further contrast with the experience of the Gold Coast, in Côte d’Ivoire, when cocoa prices fell during the 1930s, no local movement was organized to resist.42 In several documents reporting on this period, colonial officials proudly note that Côte d’Ivoire had not been confronted with any of the serious difficulties plaguing their neighbor in the Gold Coast, attributing the difference to Ivoirian chiefs’ relative lack of power and prestige.43 Rather than protest visibly, Ivoirian farmers left their cocoa plantations untended and turned their attention toward food crops.44 According to one elderly farmer, not until “a paper came from France telling the farmers to take up

41 42

43 44

Interviews (anonymous) by author. Kyere, Côte d’Ivoire, July 14, 1999. One villager did remember that during the depression of the 1930s, the government did not buy any cocoa so it stayed in the fields for the next year. A few interviews stimulated memories of a time when the villagers burned the cocoa fields themselves. This seems to have occurred later however, at the insistence of the French colonial authorities during World War II when the state was unable to buy or export cocoa. Senegalese National Archives (1938). Ivoirian National Archives (1905– 1933).

112

Informal Institutions and Citizenship in Rural Africa

cocoa again” did they go and weed their plantations.45 Again, the farmer highlights how the French colonial state had centralized all decision-making and authority within the bureaucracy. The French colonial officials’ approach to political administration served to strengthen the role of the central state as the arbiter of order or insecurity. The French concentrated the power to make decisions and allocate resources in the urban capitals of Abidjan, Dakar, and Paris. At the same time, the use of chiefs to implement French policies of forced labor and direct taxation undermined more severely the precolonial legitimacy of similar indigenous political institutions in the Ivoirian region than in the Gold Coast. The highly interventionist French colonial state reinforced the importance of individuals maintaining or improving their status as individuals vis-a` -vis the central bureaucracy rather than their village chieftaincy. The salience of the village community or broader ethnic group was diminished as the relevance of the immediate nuclear family and lineage structure increased. In this way, the French patterns of political administration spurred the move toward concentrating the informal institutions of reciprocity on a narrower conception of the nuclear and proximate extended family, but linking outside of the village to the urban capitals. In the second half of this chapter, I turn to how these colonial legacies continue to influence political administration and reciprocity today.

postcolonial state-building: continued variation in these regions of ghana and coˆ te d’ivoire Since the colonial period, the political regimes in both Ghana and Côte d’Ivoire have changed radically several times, yet the very different patterns of state formation in these regions have remained relatively constant. Political institutions tend to remain more decentralized and socially mediated in Ghana, while remaining relatively centralized and bureaucratic in Côte d’Ivoire. To begin, even the process of decolonization differed dramatically. In informal conversations, elder respondents characterized the process of decolonization as less highly centralized in Ghana than in Côte d’Ivoire.46 Elders in the Ghanaian villages grounded Kwame Nkrumah’s leadership and his fight for independence more firmly within the political arena inside Ghana rather than in Britain. Often they highlighted how Ghana was the very first nation in Africa to obtain 45 46

Interview (anonymous) by author. Opanin, Côte d’Ivoire, September 28, 1999. These comments were made informally in everyday interactions that were unrecorded or in comments before/after official interviews.

The Legacies of the Colonial Administrative State

113

independence in 1957. In contrast, elders in the Ivoirian villages singularly highlighted Houphouët-Boigny’s leadership in his role as deputy in Paris. Frequently, they would speak admiringly of Houphouët’s status as a full citizen of France working in Paris as critical to his ability to challenge French colonial policies. With the exception of Guinea-Conakry, which famously refused to participate, Côte d’Ivoire was granted independence in 1960 at the same time as the rest of the federation of French West Africa. The first two decades of state-building after independence also differed along similar dimensions. Initially in Ghana, President Kwame Nkrumah attempted to recentralize politics with the declaration of a single-party state in 1964 and the concentration of power in the office of the president.47 The slogan of the time famously pronounced: “The CPP is Ghana; Ghana is the CPP.” Despite these efforts to centralize power and stabilize the country, Nkrumah’s Convention People’s Party (CPP) was more quickly challenged and overthrown by a military coup d’etat in 1966. Nkrumah’s regime was followed by a series of quickly overturned republics and juntas, with no regime lasting more than three years. Joseph Ayee has argued that successive political regimes throughout the postcolonial period have attempted to reform through further decentralization.48 Ayee cites eight different commissions and committees that proposed decentralization reform between 1967 and 1983.49 This era of rapid turnover ended with the longer period of authoritarian rule by Flight Lieutenant Jerry Rawlings from 1981 until 1992. During the early 1980s, Rawlings espoused a populist rhetoric and encouraged popular participation and self-government through local committees. Drawing on insights from both Crook and Ayee, I argue that these two decades of massive political instability and insecurity hindered Ghanaian state capacity overall, making it difficult to centralize or decentralize the formal institutions of the state. The British colonial legacy of state formation can be seen more in the repeated attempts to decentralize rather than in the actual formal institutional outcome. In Côte d’Ivoire, a one-party state was quickly established by President Houphouët-Boigny and his Parti Démocratique de Côte d’Ivoire (PDCI), and the French colonial legacy of strengthening a central bureaucratic administration continued unabated.50 President Houphouët-Boigny and the senior PDCI leadership designated party chairman at every level down to the smallest village and maintained power practically unchallenged throughout the

47 48 49 50

See Owusu (1970). Ayee (1994, 2004). Ayee (2004: 19–21). Zolberg (1964).

114

Informal Institutions and Citizenship in Rural Africa

1980s. Richard Crook has emphasized the Ivoirian bureaucracy’s distinctive level of state capacity and administrative effectiveness in implementing policies for taxation, infrastructural development, and agricultural production.51 Varied Implementation of Similar Decentralization Initiatives Since the 1980s As in the earlier postcolonial period, the formal state institutions for political administration continue to diverge in this region of Côte d’Ivoire and in Ghana in the contemporary period of the 1980s and 1990s. Ironically, in both Ghana and Côte d’Ivoire, the state has pursued an official policy of decentralization since the 1980s.52 They share a common rhetoric that espouses the creation of new institutions that more effectively link state and society, empowering people to make decisions and promote community planning and development at the local level. These reform initiatives coincided with donor pressure to adopt structural adjustment programs (SAPs) in order to improve governance and accountability. The simultaneous moves toward political liberalization begun in the late 1980s also served to reinforce this effort at decentralizing the state apparatus.53 Despite these common objectives and pressures, however, the design and implementation of decentralization varied significantly in Ghana and Côte d’Ivoire.54 Ghana: Decentralization to the Villages from the 1980s In Ghana, decentralization began in earnest in 1988–1989 with the creation of 110 district assemblies.55 The district assembly constituted the political body for each district and met two to three times a year to discuss local revenue collection, the allocation of the District Assembly Common Fund, and other projects. The local population elected 70 percent of the district assembly membership; the ruling government, in consultation with the traditional authorities and other interest groups, appointed 30 percent. Beginning in 1998, the state created an even more locally-oriented institution, the unit committee. Unit Committees were composed of village (or neighborhood in the urban areas) residents. Again, the ruling government appointed one-third of the representatives, while the rest were popularly 51 52 53

54 55

Crook (1989). On Ghana, see Crook and Manor (1998). These different types of authoritarian regimes both began to liberalize in the late 1980s due to a combination of domestic pressures and international events. See Widner (1994) and Bratton and Van de Walle (1992). See the comparative analysis of recent decentralization efforts in Crook and Manor (1998). Crook (1994).

The Legacies of the Colonial Administrative State

115

elected. The committee usually met twice a month to deliberate and decide local community development issues. Villagers could also request a meeting to review a particular case or urgent problem. One of the responsibilities given to the unit committee was to organize communal labor to maintain and develop community infrastructure. The mobilization of communal labor formerly was the sole privilege of the chief, but now some unit committees used the “gong-gong” to call groups to work early in the morning. These public announcements usually began by acknowledging the village chief – a way of recognizing his enduring authority in the community. The unit committee also mobilized the local people to pay their taxes. While village chiefs historically levied fines, the central government has experienced perennial difficulties with tax collection and thus aimed to increase compliance through the unit committee. Overall, despite some limitations and handicaps, Ghana’s decentralization program was much further advanced than that in Côte d’Ivoire, reflecting divergent colonial histories about how the state should be constructed. Côte d’Ivoire: Extending the Central Bureaucracy to Rural Areas Despite official rhetoric, “decentralization” in this region of Côte d’Ivoire was closer to what several Ivoirian officials termed “deconcentration,” consisting primarily of extending the central bureaucracy to new areas, with little change in the mechanisms for popular participation. The first part of the government’s two-pronged initiative was to create more sous-préfectures (administrative units equivalent to counties in the United States) to lessen the distance between each village and the administrative capital. Some villagers may spend all day at considerable expense traveling to and fro, and the sous-préfet can tour each village perhaps only once per year. The second part is a program of rural “communalization” in which a council would be elected to deliberate the concerns of all of the villages in the commune. The president of this council would not be elected, but rather appointed by the sous-préfet, himself an appointed civil servant. Furthermore, the rural commune would still encompass a sizable population and geographic area with only ten to fifteen members as elected representatives. In contrast with Ghana, in Côte d’Ivoire no electoral representation existed at the village level. The comments of villagers recently included in an expanded commune demonstrate local awareness that the new arrangements were still a far cry from village representation and participation.56 56

One of the fieldsite villages was included in the regional capital commune while the other was to be included under the authority of a new sous-préfecture within the next two years.

116

Informal Institutions and Citizenship in Rural Africa

To one villager in Côte d’Ivoire, decentralization meant that his village “now belongs to the ‘commune’ of the regional capital” and is thus eligible for infrastructural developments such as a youth center and a rural health clinic.57 Another older Akan man explained that the chief “was in agreement with the commune” in order to facilitate the construction of the above community centers.58 Neither of these comments indicated a genuine devolution of planning and participation at the local level, but instead emphasized the vertical and clientelistic nature of linkages between the village and central government (to be discussed in further detail in Chapter 7). Furthermore, both of these initiatives were moving rather slowly at the time of the fieldwork. Most authorities explained that the process was “too expensive” as each new sous-préfecture required office buildings, residences, telephone connections, and so on. Officials explained that it takes time to “create more civil servants.”59 Several government officials admitted that the decentralization program had not “descended to the villages” and that no big change had occurred yet.60 While decentralization did not bring new village institutions in Côte d’Ivoire as it did in Ghana, neither had the existing village authority of chieftaincy been revitalized. Following the French colonial legacies of the past, the chief today in Côte d’Ivoire was both less autonomous and less integral to central state government than in Ghana. Echoing the French colonial rhetoric seen earlier, state civil servants continued to consider village chiefs to be “the auxiliaries of the administration.”61 Villagers reported that because the current chief “doesn’t take decisions by himself” but deferred to the central government authorities in Abengourou, he “loses dignity.”62 In contrast, in Ghana, local and national government continued to consult chiefly authority before making decisions, and many disputes continued to be resolved in the traditional court of the village chief. In sum, the French colonial legacy of direct, top-down state intervention and the British colonial legacy of indirect, mediated state interaction continued to be reflected in the varied implementation of decentralization reform initiated in both countries in the 1980s and 1990s.

57 58 59 60 61 62

Interview (anonymous) by author. Kyere, Côte d’Ivoire, July 14, 1999. Interview (anonymous) by author. Kyere, Côte d’Ivoire, July 17, 1999. Interview (anonymous) by author. Abengourou, Côte d’Ivoire, August 5, 1999. Interviews by author. Ministry of Interior, DGCT, Abidjan, Côte d’Ivoire, June 18, 1999. Interview (anonymous) by author. Abengourou, Côte d’Ivoire, August 5, 1999. Interview (anonymous) by author. Kyere, Côte d’Ivoire, July 12, 1999.

The Legacies of the Colonial Administrative State

117

conclusion: different histories of political administration produce divergence in the structure of informal institutions of reciprocity This chapter has begun to reveal how crossnational differences in the approach to political administration were actually experienced differently on the ground over time in similar regions of Ghana and Côte d’Ivoire. In this way, I start to unpack the causal mechanisms underlying “country” differences highlighted in the first section of the book’s quantitative analysis. What had been similar precolonial experiences of state formation began to diverge in important ways during the colonial period in these two regions of Ghana and Côte d’Ivoire. In the Ghanaian region, the British colonial approach to political administration was much more decentralized. The experience of state action at the local level was mediated substantially through the preexisting indigenous institutions of chieftaincy. The legacies of the British colonial administration were evidenced in the numerous repeated attempts to adopt decentralization reforms since independence. In particular, since the late 1980s, a nationwide decentralization program has increased village participation in decision-making through the creation of a new system of village-level unit committees as well as district assemblies. In contrast, in the Ivoirian region, the French colonial approach to political administration was highly centralized. The state intervened more forcefully and directly at the village level, thereby undermining the legitimacy of the precolonial institutions of chieftaincy. Where the British colonial state was built on the particularistic notion of tribal ethnic communities, the French colonial state was constructed around a more universal concept of individual citizens living in nuclear families. The legacies of the French colonial administration were witnessed in the continued centralization of the single-party state since independence. Even when a program of decentralization reform was officially adopted in the 1980s, the results in this region of Côte d’Ivoire differed remarkably from what was observed in Ghana. Ivoirian decentralization was more accurately described as a slow extension of the central state bureaucracy out into the regional capitals and larger towns. Rural residents of the smaller fieldsite villages reported that nothing at all changed for them. These divergent patterns of political administration initiated during the colonial era and continuing today have shaped the informal institutions of reciprocity in two ways: directly and indirectly. First, the different patterns of political administration shaped the structure of reciprocity directly through peoples’ everyday interactions with the state. These different kinds of

118

Informal Institutions and Citizenship in Rural Africa

everyday interactions and experiences of the state in the fieldsite villages shaped the way people conceived of their own identities as well as the meanings they attached to citizenship, family, and community. In the Ghanaian region, the decentralized nature of formal state administrative institutions validated the continued authority of the preexisting village chieftaincy as well as the new unit committees to resolve disputes and mobilize communal labor at the village level. The nature of these everyday interactions with the state reinforced the salience of maintaining membership in broader lineage, ethnic, and village communities within the village itself. As a result, the informal institutions of reciprocity in the Ghanaian region were more diversified, meaning a greater allocation of help was given to a wider variety of social ties including the nuclear family, extended family, friends, and others. In the Ghanaian region, the central state was not perceived as the primary player in mediating risk in the village. And as a result, reciprocity appeared to be more locally-oriented in the Ghanaian region, again focused primarily on maintaining social relationships within the village. In the Ivoirian region, the centralized nature of formal state administrative institutions continued to undermine the role of village chiefs and did not facilitate any village-level decision-making or communal self-help. The character of the central state’s interventions spurred the village residents in the Ivoirian region to conceive of themselves more readily as individual members of smaller nuclear families or more narrowly delimited extended families. The informal institutions of reciprocity had become more concentrated on a narrower group of social ties, with the priority clearly given to the immediate nuclear family. To the residents of these rural villages in Côte d’Ivoire, the central state appeared to be the singularly dominant player in mediating risk. Thus, in the Ivoirian region, reciprocity was more externally-oriented and focused on accessing the political and economic resources of the central state itself, with village residents linking to “big men” in the regional and national capitals. The second way that these different histories of political administration have shaped the informal institutions of reciprocity is indirectly. The formal institutions of political administration were not only experienced on an everyday basis directly but also indirectly through the formulation of public policies and the provision of services in other realms. Thus, the formal state institutions for how to administer and maintain basic order in the political system shaped the way the state expanded and intervened in additional policy areas. This chapter’s comparative analysis of the underlying conceptual differences about the fundamental role of the state, the nature of the

The Legacies of the Colonial Administrative State

119

individuals or communities to be governed, and the organization of the family will serve as the foundation for the subsequent analysis of state formation in two additional policy areas. In the next two chapters, I show how these contrasting approaches to basic political administration shaped the history of state formation in the areas of social and agricultural policy with important implications for informal reciprocity.

chapter 5 The Construction and Retrenchment of State Social Service Provision and the Unintended Consequences for Reciprocity

Not until the basic challenges of political administration had been undertaken did the British and French colonial rulers consider expanding the state provision of social services in the Gold Coast and Côte d’Ivoire. State social service provision was minimal until the Great Depression, and really only intensified after World War II. The different histories of political administration established in the earlier part of the colonial era resulted in the subsequent development of contrasting social policies in the Gold Coast and Côte d’Ivoire that continue today. Since divergent political administrative configurations produced different sets of experiences among state and societal actors, state actors had differing views of local social welfare needs, as well as contrasting perspectives on the existing local capacity to fulfill those needs.1 These different assessments of need and capacity framed the development of divergent colonial and postcolonial social policy in each of the cases. During the colonial era, social policy in the Gold Coast was seen as supplementing the “traditional” systems of social welfare in the community while, in Côte d’Ivoire, administrators aimed to supplant these “inadequate” systems with more activist policies for individuals. Ironically, despite pressures from the World Bank and International Monetary Fund to adopt similar types of social service reforms during structural adjustment programs (SAPs), the formulation of policy and the local experience of state social service provision continued to differ in these two regions in the 1990s. In the Ghanaian region, the government was attempting to strengthen the 1

See Fraser (1989) on the politics of need interpretation.

120

State Social Service Provision

121

indigenous social welfare systems of the extended family and community through a more decentralized social policy, whereas in the Ivoirian region, the government aimed to replace these informal systems with the centralized arm of the bureaucracy. This fundamentally different approach to social policy was not just discussed abstractly in the ministry corridors of the capital cities, however, but was also actually experienced on a day-to-day basis in the four villages in these two regions. In the Ghanaian villages, only a small number of older Ghanaians had very vague memories of attending local schools and clinics, and those institutions were often still named and associated with the original sponsoring missionary religious organization rather than the central state. In contrast, older Ivoirian village residents more often spoke in French and always recalled “French” schools, and to a lesser extent, “French” clinics. The divergent experiences of state social policy was also evident in the different ways village residents described the role of midwives and other traditional healers in the 1990s. In the Ghanaian villages, many respondents proudly pointed to what the Ghanaian state termed “traditional birth attendants” and other traditional healers as community leaders. In contrast, in the Ivoirian villages, despite similar policies to provide support to midwives, they were rarely mentioned. Instead, village residents proudly emphasized the increasing number of births taking place in the clinic or hospital. In this chapter, I further unpack the legacies of state formation at the village level. I highlight how historical differences in the construction of state social service provision as well as varied paths of retrenchment during the contemporary era of neoliberal social policy reform have shaped the informal institutions of reciprocity in these regions of Ghana and Côte d’Ivoire. Where much of the literature on social policy looks at the differences in aggregate social spending, I find that the variation in the delivery of social services by the state through history is critical. I also move beyond typological descriptions of national models of social service delivery from the welfare state literature on advanced industrialized countries to explore how service delivery is actually experienced locally on the ground. The chapter’s first half shows how the construction and reform of social service provision has differed in the two regions since the colonial era. I link this varied historical experience in the orientation of state social service provision to the more diversified structure of informal reciprocity in the Ghanaian region in comparison with the Ivoirian one. Then, the chapter’s second half examines how the different historical levels of stability of state social service provision and the timing of its retrenchment explain the lower quantity of help given and the fewer number of people participating in the

122

Informal Institutions and Citizenship in Rural Africa

Ghanaian region. This chapter emphasizes how the informal institutions of reciprocity do not have a zero-sum relationship with the state. It is important not to look simply at how much risk is mediated by the state, but also how the interactions were structured and how consistent the relationships were over time.

contrasts in colonial social policy British efforts to supplement the informal system of social security In the Gold Coast, social welfare initiatives were begun much earlier than in Côte d’Ivoire and with a much larger role played by private, nonstate actors, such as merchant and subsequently missionary organizations. For example, as early as the mid-sixteenth century, European trading companies had created “Castle Schools” at the forts along the coasts.2 The company headquarters in the European metropole provided only limited financing, so local fund-raising was often used, a pattern that was repeated by the colonial state in later years. The missionary role in the construction of schools and clinics was also more pronounced and occurred earlier in the Gold Coast than in Côte d’Ivoire. For example, Wesleyan (Methodist) and Basel (Presbyterian) missionaries arrived in the Gold Coast as early as 1828 and had built several schools by the middle of the nineteenth century.3 By 1881, more than 136 mission schools had been established, and they enrolled about 5,000 students.4 These schools were primarily supported by the missions and received only limited assistance from and later regulation by the colonial government. At this point, only three schools were fully built, financed, and operated by the central government. All these efforts were concentrated on the southern coast and Akuapim ridge (not far north of Accra), distant from the fieldwork sites of this study. The first action by the colonial state to provide social services directly was in the early 1880s. Earlier in 1852, the governor had passed an education ordinance to develop education, but the poll tax initiated to generate revenue for the expansion was so unpopular that it was abolished by 1861.5 Later, the Educational Ordinance of 1882 succeeded in proposing direct public 2

3

4 5

These Castle Schools provided basic education to a small number of children of African merchants and local chiefs. George (1976: 23). By 1841, nine Wesleyan mission schools had been opened. By 1850, 885 students were enrolled in Wesleyan schools. Graham (1971). George (1976: 24). Berry (1995: 117).

State Social Service Provision

123

funding for government schools in addition to financial support for government-assisted schools. By 1900, there were about 135 government and government-assisted schools enrolling 12,000 students still almost exclusively in the coastal areas.6 By 1924, enrollment had nearly tripled to nearly 35,000 students in 236 government and government-assisted schools.7 By 1931, there were 347 schools, but the huge majority of them were mission-assisted, while only nineteen of them were government-run.8 Until this point, educational development was exclusively at the level of primary schooling. In 1927, the British governor of the Gold Coast opened one of the first post-elementary educational institutions in Africa, the Prince Wales College (now Achimota College) in Accra.9 While the demand for secondary school education increased in the 1930s and 1940s, the majority of the schools established were unaided by the government, and the absolute numbers of students enrolled was small and increased slowly, with only a very few girls able to attend.10 Despite these new efforts, the British colonial government continued to rely extensively on missionary organizations to provide health and education services in many areas throughout the colonial period. In several pieces of legislation and policy reports, the British colonial government focused on how to support and regulate private provision, not how to replace it. The British also emphasized the need to work closely with local indigenous institutions, thus the call for local advisory committees, vernacular textbooks, and local language instruction. The poverty and instability of the Great Depression spurred the British to reflect further on the welfare needs in their colonies, however. Often fielding British anthropologists, they expended considerable effort to characterize and understand what they termed the “traditional communal system.”11 They believed that “an effective social security system exists under the custom of the village group,” but that this system was breaking down due to the pressures of modernization. The British authorities saw their task as “(adapting) and (building) upon this communal machinery for social security . . . [thus] (supplementing) the tribal or village organization.”12 This decentralized approach to social services reflected the British goal of 6 7 8 9

10

11 12

George (1976: 24). Graham (1971). See Ghana National Archives (1931). Prince Wales College would become a very prestigious institution with such well-known alumnae as Kwame Nkrumah, J. J. Rawlings, and Robert Mugabe. In 1945, eighty-six students were enrolled at Achimota. See the GOG (1946) and George (1976: 27–32). Public Record Office (1943). See Ghana National Archives (1931).

124

Informal Institutions and Citizenship in Rural Africa

colonial self-sufficiency: The colonies had to pay their own social service expenditures from revenues they generated from native treasuries. My interviews with villagers confirmed that in this region during the colonial period there were few hospitals and schools, and few numbers of people attending either. Evidently, some minimal fees were charged at hospitals, clinics, and schools, but they were frequently waived for “paupers.”13 Only in 1939 did the British colonial authorities introduce the Colonial Development and Welfare Act and create a new Department of Social Services. Worried about the threat of labor unrest, primarily in the urban areas, the British developed legislation promising a wider role for the government in improving social welfare.14 The contentious process leading to approval of the policy highlights the reluctance of British colonial authorities to view social welfare as the dominion of their administration. The British tended to see only a few categories of “detribalized,” urban Africans, separated from the traditional safety net of the extended family in the villages, as possibly needing some type of social assistance. The Colonial Development and Welfare Act languished for several years during World War II due to a lack of resources for any substantive implementation. The Beveridge report in 1942 sparked further debate, however, and by the middle 1940s, British colonial discussions on social welfare had concluded that the government should not implement a centrally-organized social insurance system, but instead focus on making social services more available and affordable at the local level. Any direct financing of social services in the 1940s and 1950s often took the form of loans or matching grants. For the British, any future government activity with regard to social welfare would not be “stereotyped” but designed to fit into the particular local social and political structure.15 In 1945, the Colonial Social Welfare Advisory Committee recommended a policy of collaboration between public and private institutions and the creation of local committees to encourage community cooperation.16 Even the mass education program begun in 1952 emphasized local leadership, voluntary effort, and self-help.17 One British official dismissed the idea of a generalized social security plan saying, “The time has not come for old age pensions or the dole.”18 Colonial documents clearly indicate the view that social welfare should be promoted in collaboration with the private 13 14 15 16 17 18

See Ghana National Archives Kumasi (no date) and Ghana National Archives Kumasi (1931). Cooper (1996) argues that this was primarily to justify the continuation of imperialism. Ghana National Archives Kumasi (1948). Yimam (1990: 38). Asamoah and Nortey (1987). Ghana National Archives (1945).

State Social Service Provision

125

sector, including churches, voluntary societies, and members of the local communities. The British also concluded that since they lacked sufficient resources (and information) to provide social security for every individual, the government should additionally develop general economic policies to safeguard peasant producers against wide price fluctuations, particularly in the cocoa sector.19 Thus, the British strategy was to increase Africans’ standard of living to enable them to secure themselves and their extended families rather than to provide a centrally organized, public social safety net. French Colonial Authorities Take Activist Bureaucratic Role Meanwhile, in Côte d’Ivoire, the French colonial authorities started later (with a later date of pacification), but then took a more activist and secular role in providing social welfare.20 Even though French Catholic missionaries were the first Europeans to arrive in Côte d’Ivoire in 1637, their presence was always much more limited than the Protestant missionaries in the neighboring Gold Coast. While the British were more concerned about protecting the extended family, the French essentially applauded its dissolution, viewing the demise of the “traditional” collectivity as the liberation of the individual. One French colonial official wrote: The disaggregation of the traditional family corresponds to emancipation, a liberation of its members that one can’t complain about. The African man (and now the African woman) becomes, thanks to the action of the Administrative Power, a person existing by and for himself (and not for the Collectivity . . . ). It is not visionary to think that, in a time more or less close, the extended Collectivities (often tyrannical) of the past, founded on the primacy of their own existence, will be replaced by smaller communities (but more human) based on the respect of the individual. The extended family will be succeeded by the household (and even the monogamous household), established on reciprocal consent between spouses . . . And the African Collectivity, organized and protected by French public power, will be able, in its turn, to assure total and effective protection to the family and to the child.21

Since the French viewed the “traditional family” as inadequate, the central government used the revenues from direct taxation to provide a broader level of social infrastructure in Côte d’Ivoire. The French metropole also subsidized social service provision in the colonies directly through the Fonds d’Investissement pour le Developpement Economique et Sociale (Investment 19 20

21

Ibid. The French did not sign the first treaties with coastal chiefs until the mid-nineteenth century and were not able to pacify the colony until 1915. French National Archives (1951–1952).

126

Informal Institutions and Citizenship in Rural Africa

Funds for Economic and Social Development) as early as 1928.22 While school construction started later in Côte d’Ivoire, the central government dominated the missionary organizations from the beginning when the first schools were established in Grand-Bassam as early as 1887.23 As early as 1903, the French colonial government successfully passed an education ordinance to provide full financial support to all schools so that elementary education was given free of charge.24 This expansion of the central state role in education was comparatively uncontested and occurred much more rapidly than in the Gold Coast. Then in 1923, the West African governorgeneral issued a Circulaire unifying the educational system for all of the West African French colonies.25 During this period, this growth of the central state’s role in social service provision was not limited to schooling either. Public clinics were also constructed throughout various regions of the country, and most medical care was offered free.26 For example, the regional hospital was established in Abengourou in 1935.27 Civil servants and salaried workers were particularly well-provided-for by the state early on, receiving pensions, family allowances, workers’ compensation, sick leave, and pregnancy leave.28 Côte d’Ivoire was the first French colony to attempt to coordinate existing social services by organizing a special government department in 1950. But when the French colonial government described its mission of reinforcing family ties, the ideal family was conceptualized as a nuclear unit of parents and their children. As early as 1936, some Africans conceived the delivery of social services to be the responsibility of the French colonial government. In a letter to the minister for the colonies, an indigenous group living in the field research area demanded more public services. However, they complained of the overly centralized and European nature of institutions, requesting the accommodation of local African needs – for instance, changing the school day to allow children to help their families in the fields.29 Further evidence of a generalized

22 23

24 25 26

27 28

29

Yimam (1990: 37). Clignet and Foster (1966: 34) cite the first primary school teacher being sent to Côte d’Ivoire to open a primary school in 1887. Clignet and Foster (1966: 35). Ibid. All doctor visits and care provided at clinics and maternities were free. Only hospitalization required some payment except for indigent cases. FNA (1951–1952). Government of Côte d’Ivoire (1994). Pensions for higher-level civil servants were established as early as 1924 and extended to all civil servants in 1928 and 1946. Workmen’s compensation was introduced in the early 1930s. Family allowances were established in 1955. FNA (1951–1952). Ivoirian National Archives (1936).

State Social Service Provision

127

expectation by Africans for the central government to provide certain services at the local level is that by the early 1930s, the colonial authorities note a resistance by many Africans to engage in communal labor. When interrogated, the Africans would respond that they had already paid tax money, so they did not owe anything further.30 In addition to a more expansive provision of social services, the French government discussed the need for a broader social insurance system as early as 1910.31 One French colonial official from Dahomey (now Benin) suggested that because the indigenous “tontines,” or rotating credit groups, were often rife with cheating, the government should replace them with more formal savings accounts.32 A few years later, in 1916, the French government established Provident Societies in Côte d’Ivoire. The French funded these societies to provide tools to African farmers; encourage the storage of seed reserves for future planting; provide aid to members facing natural disaster or personal crises such as accident or illness; and disburse mutual loans to members for agricultural development. The manner of establishing these “voluntary” societies was perhaps more instructive than their success or failure. In 1910, the French decreed from Paris that these societies should be created in all of French West Africa. After several years of slow progress, the colonial authorities revised the policy slightly to promote the creation of these societies only in the regions where the local administrators deemed it feasible. Nevertheless, once a society was created in the administrative unit of the “cercle,” the French admonished that “all the indigenous farmers and herders living in that area will have to belong.”33 The French authorities note that “the societies do not establish themselves, and that they only prosper on the initiative and prompting of the Administrators.”34 The legislation outlined precisely how the societies are to be organized, and how French administrators could intervene to suspend, revoke, and exclude members who did not behave appropriately. Many Africans critiqued this top-down approach harshly, suggesting that this société des commandants de cercle should either be managed by Africans or, better yet, dissolved.35 This rare opportunity to hear the voices of African dissent reveals the struggle involved in the construction of state institutions for social welfare based on the hegemonic étatist framework of the French. 30 31 32 33 34 35

Ivoirian National Archives (1905–1933). French National Archives (1911). Ibid. Ivoirian National Archives (1915). Ibid. Ivoirian National Archives (1936).

128

Informal Institutions and Citizenship in Rural Africa

This section demonstrated how the colonial governments in Ghana and Côte d’Ivoire diverged in their approach to social policy. The next section explains how and why this variation continues in the postcolonial period.

colonial patterns reproduced: social policy in the postindependence period (1960–1980) While the nature of social policy in the contemporary period (1980–present) is the chapter’s primary focus, the next section demonstrates briefly how three key elements of British and French colonial legacies were reproduced in the first two decades after independence: (1) a tendency to target the needy in Ghana versus attempting to provide more generalized social coverage in Côte d’Ivoire; (2) a greater incorporation of private social service providers in Ghana than in Côte d’Ivoire; and (3) a more decentralized delivery system in Ghana than in Côte d’Ivoire. First, in Ghana, from 1957 until 1966, while Nkrumah made health and education services increasingly available and affordable to a broader section of the population, the focus remained on meeting the needs of the growing urban areas.36 The targeting of “detribalized” Africans, a policy initiated by the British, essentially continued with Nkrumah’s focus on the new urban masses who had left their rural homes and social networks to seek industrial jobs in the cities. Thus, the Nkrumah government identified one of its top priorities as the need for urban housing. During the early 1970s, the urban emphasis of Ghanaian social policy shifted to a more general targeting of the poor, including the rural poor. This shift in policy explains the Busia government’s (1969–1972) investment in rural feeder roads, rural electrification and water projects, rural housing, and an expansion in rural health centers. The Acheampong military regime that followed in 1972 tried to link urban and rural interests by attempting rather unsuccessfully to increase food production with Operation Feed Yourself and Operation Feed the Nation. By the late 1970s, the Five-Year Development Plan (1975–1980) again primarily focused social welfare activities on those groups believed to be suffering from the breakdown of traditional social systems in the growing urban areas. Meanwhile, the newly independent Ivoirian government reproduced earlier French colonial patterns in social policy. Led from independence for more than thirty years by President Houphouet Boigny, the government of Côte d’Ivoire was able to provide a more generalized public social service system that consistently functioned at a much higher level of quality and 36

Fadayomi (1991).

State Social Service Provision

129

accessibility than that in Ghana. Health care was provided essentially free of charge to all Ivoirians until the mid-1970s, at which point, the fees were kept low. The quality of the public hospital system was the envy of West Africa. Education was also heavily subsidized by the central government in Côte d’Ivoire with enrollments increasing dramatically between 1960 and 1980. The central government provided free lodging in addition to generous scholarships to many secondary school and university students. While health and education services continued to be more generalized in Côte d’Ivoire than in Ghana, most other social security measures were aimed at formal sector workers. Still, a more elaborate social insurance system was developed for the formal sector, and in particular, for government employees in Côte d’Ivoire than in Ghana. In 1964, the Mutual Fund of Government Employees was created to reduce the costs of pharmaceuticals, dental care, and eye care; in 1968, the Caisse Nationale de Prévoyance Sociale (CNPS) was established to coordinate social security initiatives; and in 1973, a mutual insurance company was established to help cover health care costs for government employees. Second, although public financing of the health and education sectors was dominant in both countries, the supplementary role of private service provision continued to be relatively more important in Ghana than in Côte d’Ivoire. Churches continued to play a larger role in providing both primary and secondary school education and health care facilities in Ghana than in Côte d’Ivoire. Government regulation of these private providers was relatively laissez-faire as well in Ghana. Even under the seven-year plan (1963–1970) when the Ghanaian government began to increase public expenditure on education, spending was focused on secondary and technical schools, thus, the central government continued to pressure local governments and voluntary organizations to support primary and middle schools. Throughout the 1970s as well, local communities in Ghana were frequently encouraged to support social services by supplying building materials, communal labor, and special financial collections. Also, as early as 1971, the Ghanaian government began to collect user fees for health services, although these were minimal until the adoption of structural adjustment in the 1980s. These types of self-help initiatives were never mentioned in descriptions of Ivoirian social service development. Instead, the emphasis was on the central government’s ability to finance the construction and operation of hospitals and schools. Another example of the greater role of private resources emerges from a comparison of social security in both countries. In Ghana, following the Social Security Act of 1965, a system of provident funds was established in 1972 whereby employers and employees contributed a certain percentage of

130

Informal Institutions and Citizenship in Rural Africa

the worker’s salary into the Social Security and National Insurance Trust (SSNIT) fund. In contrast, in Côte d’Ivoire, the central government followed the French example by providing social insurance pensions and other social security programs directly for those who qualified. Third, the Ghanaian government continued to pursue a relatively more decentralized approach to social service delivery than the Ivoirian government. Throughout the 1960s and 1970s, the Ghanaian government emphasized working with local communities to address their social needs more efficiently and effectively. Following the colonial notion of “selfsufficiency,” the continued emphasis on local participation in social service provision was similarly articulated in Nkrumah’s policy of “self-reliance” in the 1960s and in terms of self-help and community development in the 1970s. In contrast, in Côte d’Ivoire throughout this period of the 1960s and 1970s, the vast majority of social services were provided by the state through a highly centralized state system. The health care system was strongly biased toward curative rather than preventive care, with the majority of public expenditure allocated toward the government hospitals in the capital and a few other urban areas. In education, universities and teacher training institutions continued to be modeled on the French system and clustered in Abidjan, with one secondary university in Bouake and three “grandes ecoles” in Yamoussoukro.37 After a relatively brief “honeymoon” period under Nkrumah, state provision of social services began to decline seriously in Ghana. The economy was debilitated by the political instability of a rapid series of coup d’états and military governments and by the end of the cocoa boom in the mid-1970s. Between 1976 and 1983, government expenditures on education fell from 6.4 percent of GDP to 1.4 percent.38 A shortage of foreign exchange meant that most medical equipment, medicine, textbooks, and school supplies were simply unavailable. Hospitals and school buildings deteriorated, and a mass exodus of health and education professionals left the country poorly staffed. Although government health and educational facilities were officially free or of very low cost, in conditions of severe shortage and endemic political and economic crisis, most Ghanaians were pressured to make side payments to obtain public services, if at all. By the early 1980s in Ghana, the state social services system was in virtual collapse.

37

38

Bouake is practically the center of the country and is known as the capital of the Baoule, Houphouet-Boigny’s ethnic group. Yamoussoukro was Houphouet-Boigny’s hometown. World Bank (1996).

State Social Service Provision

131

The “good times” in Côte d’Ivoire, largely financed by almost two decades of high growth (averaging 7.5 percent in the 1970s) in the cocoa industry, did not last forever either and ended by the late 1970s. The steep drop in coffee and cocoa prices in 1978, the oil shocks, and the debt crisis all combined to produce a rapid financial decline in Côte d’Ivoire, and by 1981, the “Ivoirian miracle” was in recession.

continued colonial legacies on social policy: divergent orientation of structural adjustment from 1981–1999 Many of the social policy changes implemented in the 1980s and 1990s were part of the structural adjustment programs (SAPs) designed by the World Bank; thus the neoliberal reforms advocated in Ghana and Côte d’Ivoire were very similar. For example, in both countries, cost recovery became a major concern in the early 1980s. Ironically, however, both nominal and real government expenditures on health and education actually appeared to have increased during the period of structural adjustment in both countries.39 Furthermore, the share of total government expenditure allocated to these social service sectors appears to have grown.40 However, similar trends in expenditure data do not necessarily signify similar social service delivery. In fact, despite these similarities, significant differences in the orientation of social policy remain. The orientation of the Ghanaian government policies was to build on informal social support systems such as the network of the extended family, village, or other community. While government officials in Côte d’Ivoire admitted that “traditional” institutions such as the extended family played a role, their future goal was to create new formal bureaucratic structures, especially in the rural areas. It should be noted here that although the available statistics are limited, it does not appear that the Ghanaian government turned to the informal social welfare system simply because it was less able to provide wider formal sector coverage than the government of Côte d’Ivoire. This chapter demonstrates that the orientation of the Ghanaian state has differed historically since the colonial period. An analysis of aggregate social expenditures indicates that both the Ghanaian and Ivoirian governments have allocated approximately similar levels of expenditure to both health and education. (See Tables 5.1. and 5.2.) 39 40

Demery (1994) and Leechor (1994). While it is troubling that the share may be declining in more recent years from an earlier peak, it remains higher than prior to adjustment. Aryeetey and Goldstein (2000).

t a b l e 5 . 1 . Comparison of National-Level Health Expenditures and Infrastructure in Ghana and Côte d’Ivoire Ghana

Colonial Period Public health expenditure (% of total public expenditure) Hospital beds (per 1,000 people) Number of doctors (per 1,000 people) Number of nurses and midwives (per 1,000 people) Immunization for DPT (% of 12–23 mos.) Immunization for measles (% of 12–23 mos.)

Early Independence (1960)

Before SAP (1980)

*

7.1% 7% (1962/1963)

*

0.78

Côte d’Ivoire Most Late SAP Recent (2000) (2005)

Colonial Period

Early Independence (1960)

Before SAP (1980)

Most Late SAP Recent (2000) (2005)

10.9%

7%

*

*

3.9%

7.2%

0.90

*

1.35

*

*

4%

*

0.05

*

*

1.57 (1985) 0.07 (1984) *

*

*

7%

84%

84% (2006)

*

*

25% (1985)

72%

77% (2006)

16%

84%

85% (2006)

*

*

31% (1985)

73%

73% (2006)

*

0.06 0.15 (1996) (2004) 0.92 (2004)

*

0.03

0.07 (1985)

*

Sources: World Bank (WDI; 2007); and UNDP/World Bank (1989).

0.40 (2006) 0.09 0.12 (1996) (2004) 0.60 (2004)

t a b l e 5 . 2 . Comparison of National-Level Education Expenditures and Infrastructure in Ghana and Côte d’Ivoire Ghana

Colonial Period Public * education expenditure (% of total public expenditure) Primary school 5,000 enrollment (1881) 12,000 (1900) Secondary * school enrollment Tertiary school * enrollment

Côte d’Ivoire

Early Independence (1960)

Before SAP (1980)

12.9%

22%

483,425 (1959)

1,377,734 2,560,886 2,929,536

14 schools 220,028 in 1897 (1959)

1,024,585 1,943,501

2,111,975 (2006)

178,581 (1959)

113,157

223,597

340,164

*

9,698 (1959)

198,190

*

4,300 (1965)

7,926

9,251 (1989)

*

*

1,900 (1965)

19,633

Late SAP (2000)

Most Recent (2005) 22% (1994– 2004)

Colonial Period

Early Independence (1960)

Before SAP (1980)

Late SAP (2000)

*

27.9% (1970)

16.3%

21.5%

Sources: World Bank (1982, 1998, 2001b, 2007); GOCI (1991); George (1976); and Clignet and Foster (1966).

489,740 (1995) 167,055 (2000) 52,228 (1994)

Most Recent (2005) 21% (1994– 2004)

*

134

Informal Institutions and Citizenship in Rural Africa

Where there are differences, the Ghanaian government actually outpaced the Ivoirian one. Indeed, aggregate public health and education expenditures have increased over the period of structural adjustment reform. What this book emphasizes throughout, however, is that it is not simply the overall aggregate expenditure levels that matter to local people but instead how that money is spent. The different normative underpinnings in the orientation of social policy reform will be analyzed in each of the following seven subsections: the approach to the family, relations with nongovernmental and community groups, pauper policies in health care, educational reform, the exemption of vulnerable groups, poverty-alleviation programs, and national health insurance initiatives. The latter three subsections represent the most recent round of reforms where normative differences persisted in shared efforts to soften structural adjustment. Approach to Support of the Family Several officials in Ghana explained that social policy strategies “are all ways to keep the family together,” and that this was the first goal of state action. Officials emphasized that they intended social policy to build on “existing structures” and “strengthen existing institutions at the local level,” particularly the family. It is important to note that when the family was discussed in Ghana, the unit most often referred to was the extended family system. Several policies implemented in Ghana encouraged families to take care of their needy relatives. The government organized a “Celebration of the International Year of the Family” in May 1999 to promote reciprocity within the extended family. In education, the government was encouraging community-supported schools in which parents work with teachers to improve the quality of instruction; for example, privately supplementing teacher salaries in return for additional work. In health, the government provided tax credits to caretakers of elderly family members in order to strengthen families and make nursing homes unnecessary.41 In Côte d’Ivoire, a new ministry, created in 1994, was dedicated to women and the family; however, the focus of policies was clearly on the nuclear family. Government officials rarely spoke of strengthening the extended

41

While this tax exemption was not well publicized and only applied to the relatively small percentage of formal sector workers, the point remains that the Ghanaian government was actively seeking new ways to strengthen existing forms of private social support, in particular, the extended family.

State Social Service Provision

135

family system, but instead of rehabilitating the nuclear family to fight poverty, particularly in urban settings. The extended family system was more often referred to as a burden that hindered the proper economic and social development of nuclear families. Even the discourse among World Bank representatives based in Ghana and Côte d’Ivoire mirrored the previous differences in the conceptualizations of the family. World Bank representatives in Ghana acknowledged that although the support of the extended family was not as easily gained as in the past, this informal institution was the “basic social safety net that exists in Ghana.”42 In contrast, World Bank officials in Côte d’Ivoire discussed the fiscal limitations of the state and consequent need for reforms in an interview, but the informal social safety net was never mentioned. Relations with Nongovernmental and Community Groups The government of Ghana also collaborated more extensively with international and local nongovernmental organizations (NGOs), local clubs, and community associations to help the needy “indirectly.”43 Officials often noted that these groups frequently helped solve problems at the local level without relying on the central government bureaucracy. These local organizations actively supported community infrastructure and services; for example, the creation and maintenance of water sources, schools, and markets.44 In the health care field, NGOs financed 12 percent of all expenditures on health in Ghana, almost one-third of what the government spent.45 In addition to direct health spending, NGOs also worked on health education and outreach. One recently founded organization in the fieldsite region, Care-Elderly, actively recruited elderly members in order to improve their access to free health care under the exemption policy, and advocate for increased government support for the aged. The Ghanaian government has also recognized the important role of private individuals, such as traditional healers and traditional birth attendants. Table 5.1 shows the significantly higher numbers of officially recognized midwives in Ghana as compared with Côte d’Ivoire from the late 1960s/early 1970s until today. At the time 42 43

44

45

Interview by author. World Bank, Accra, October 29, 1998. Interviews by author. Department of Social Welfare, Accra and Bechem, October 1998 and February 16, 1999. In Tano District, Brong-Ahafo region in Ghana, where the field research was conducted, several local NGOs worked with district officials to support social infrastructure, for example, BRUM in water management and CEDEP in school improvement. Demery et.al. (1995).

136

Informal Institutions and Citizenship in Rural Africa

of fieldwork in the late 1990s, the Ghanaian government was trying to support (and to a certain extent, regulate) its services as alternatives to government-run facilities. In Côte d’Ivoire, government officials spoke of working with community groups in terms of “encadrement.”46 This conceptualization called for a more activist role by the state than the Ghanaian-style collaboration with existing local initiatives. For instance, the government created management committees for health clinics and schools. Most government officials admitted that these committees have had limited success, however. In the fieldsite region, many local residents and health officials reported that the management committee was suspected of embezzling funds so that community trust had dissipated and meetings were infrequent. The Ivoirian government has tried to follow the lead of Ghana in one area: increasing the training and support of traditional birth attendants. This policy has not yet been extended to traditional healers, even though they appear to be quite numerous and active in the fieldsite villages. Pauper Policies in Health Care The Ghanaian government recognized early on the need to help the indigent who simply did not have the resources to pay for certain services. When a needy patient required medical treatment, a social worker from the hospital initially tried to determine whether there were any resources available from the extended family. If the family member was reluctant despite entreaties from the government official, the patient was then treated for free, or released on the promise to pay in installments. Unfortunately, only 4.5 percent of Ghanaian respondents interviewed in the fieldsite villages were aware of any such policies to waive, reduce, or arrange payments for fees. In comparison, 33 percent of Ghanaian respondents reported that the government provided free care for certain exempted groups, in particular the elderly, while 9 percent described government educational scholarships, in particular for the children of cocoa farmers. While the pauper policy was not as widely publicized or implemented in Ghana as was perhaps necessary,47 a similar policy in Côte d’Ivoire remained almost exclusively on paper and was nearly nonexistent in practice. As in

46

47

In French, to “encadrer” is to frame, to surround or flank, to train and supervise, or to officer (in the military sense). The Tano District Department of Social Welfare recorded only thirteen cases in 1998. GOG (1998a).

State Social Service Provision

137

Ghana, when a person could not pay for health care, a social service officer from the health facility was supposed to assess other available resources. The Ivoirian social service officers not only sought the aid of family members but also sometimes sought the assistance of a mayor’s office or an embassy (for non-Ivoirian migrants). Unfortunately, due to budget cuts, many regional hospitals (CHRs) and clinics did not have any staff to carry out these policies. Furthermore, the larger urban hospitals (CHUs) did not receive an allocation for such cases from the national budget; thus, social workers felt pressure to limit the number of social cases in order to keep costs down. Educational Reform The Ghanaian government attempted in 1987 to implement reforms in the educational sector that would adapt further the British-style system to local needs. While the hasty implementation of the 1987 reforms has been frequently criticized, rarely were the stated objectives. The Ghanaian government intended to shorten the preuniversity education track and tailor the junior secondary school curriculum to local labor markets. Since many children could not continue education at higher levels, the reforms were intended to develop more practical vocational skills that would increase local employment rather than increasing a brain drain or, worse, a growing group of frustrated and unemployed school dropouts. In contrast with the health sector, government spending on education in Ghana was relatively equitable, particularly spending on primary education.48 Still, despite some positive trends, major disparities continued to exist in the quality of schooling in certain regions and in rural areas overall. These inequalities were further demonstrated by the growth of private schools, primarily in larger urban areas. World Bank data suggest that private primary school enrollment as a percent of the total primary school enrollment increased dramatically in Ghana from 7 percent in 1991 to 21 percent in 2005, as compared with an increase from 10 percent in 1991 to 12 percent in 2005 in Côte d’Ivoire.49 In Côte d’Ivoire, educational reform consisted primarily of changes in financing. Teacher salaries continued to constitute a disproportionately large percentage of the public sector wage bill, suggesting that rising education expenditures may have been focused on salaries and not necessarily other investments (i.e., building schools, providing books, etc.).50 The character of the school system continued to be highly centralized and hierarchical, based 48 49 50

Demery et.al. (1995). World Bank (2007). Demery (1994).

138

Informal Institutions and Citizenship in Rural Africa

directly on the French system initiated during the colonial period. For instance, not until the late 1980s did the government begin producing its own textbooks in order to incorporate local values instead of purchasing them from France. Exemption of Vulnerable Groups from Cost Recovery In response to growing domestic and international criticism after several years of structural adjustment, Ghanaian policymakers began to implement a second round of reforms. The Ghanaian government recognized that certain groups could not readily access social services due to the higher costs associated with cost recovery. The government has expended significant effort in trying to identify the vulnerable sectors of the population and develop appropriate policies for these targeted groups. For instance, over the past several years, several intersectoral commissions have studied the problems confronting children,51 the aged,52 and the urban poor. These commissions were composed of a diverse group of representatives from several government ministries, universities, and religious and other NGOs to facilitate information gathering and coordinate action. Based partially on the recommendations of these commissions, policies have been developed to exempt several vulnerable groups from health and education fees. Since October 1997, children under five years of age, preand postnatal women, and the elderly should have received free health care services when needed.53 The government allocated a certain section of the national ministry of health budget to provide care for these exempted categories. The problem with this targeting mechanism was that the truly poor may not even go to the hospital when sick, either due to the barrier of transportation costs or a lack of knowledge of these exemption policies. District officials reported a low rate of participation in these exemption programs so that budget allocations remained unspent. Other studies have also confirmed low use of the health system in Ghana as well as an increasingly regressive character to actual government spending on health.54

51 52

53

54

One example is the Ghana National Commission on Children. The National Aging Committee was formulating a framework for a comprehensive National Aging Policy. In addition, in Ghana, patients who are identified as suffering from tuberculosis, leprosy, or mental illness are supposed to be exempt from fees. See, for example, Demery et.al. (1995).

State Social Service Provision

139

In education, the Ghanaian government implemented the F-cube program in 1996 to address problems in the 1987 educational reforms that hindered access to basic education.55 The government also attempted to decentralize the financing and planning for education by creating district education planning teams. These policies reflect the historical tendency of the Ghanaian government to strive to identify, research, and fill gaps instead of providing universal coverage. While the Ghanaian government’s efforts to target the poor and vulnerable did not always yield successful results, this does not mean that targeting as a policy tool was useless; instead, the mechanisms of targeting needed to be more carefully considered and implemented. In contrast, little has been done in Côte d’Ivoire to recognize or address the needs of vulnerable groups. No targeting policy existed. There were only a few specific services that were provided free – for example, immunizations for children and childbirth kits for pregnant women. Instead, most of Ivoirian social welfare policy focused on formal sector workers such as teachers or civil servants who made up at most 10 percent of the total population. For these workers, Ivoirian social policy continued to be more extensive and substantial than the minimal benefits offered by the state in Ghana. The primary way that the Ivoirian government has attempted to increase access to health facilities was by building new rural health clinics. While this initiative did lower travel costs, it did not resolve the problem for the poor who could not pay anything at all. Other Poverty-Alleviation Programs In addition to the preceding exemptions to existing social policies in health and education, several new programs in each country focused explicitly on poverty alleviation. The orientation of these poverty-alleviation programs differed, even though the original program designers were often the same external donors. Ghana was one of the first African countries to implement a poverty-reduction scheme. In 1988, the Ghanaian government introduced the Programme of Actions to Mitigate the Social Costs of Adjustment (PAMSCAD). The intent of the program was to target financial assistance and employment opportunities to the poor and vulnerable throughout the country. Similarly, although the PAMSCAD program reproduced earlier patterns by attempting to work with existing private institutions and communities, the inputs elicited from beneficiaries were often overly 55

The goal of the F-Cube program is to ensure “free, compulsory, universal basic education.”

140

Informal Institutions and Citizenship in Rural Africa

burdensome, creating barriers to participation rather than poverty alleviation.56 Again, although the impact of the PAMSCAD program was less than anticipated, no such program was even attempted in Côte d’Ivoire. In more recent years in Ghana, even though the World Bank played a large role in designing and financing the Village Infrastructure Project, the program’s focus on supporting local capacity building to maintain the community’s social infrastructure paralleled the traditional orientation of government policy.57 Yet again, no such program existed in Côte d’Ivoire at the time of field research. While both countries have initiated several microcredit programs to promote income generation for the poor, program design has differed, resulting in a higher level of repayment problems in Côte d’Ivoire than in Ghana. Repayment problems were exacerbated when lending was made through international donors or central government agencies that were ignorant of local seasonality issues critical for successful loan disbursement.58 Interestingly, after an earlier unsuccessful experience with an International Fund for Agricultural Development (IFAD) program, the Ghanaian government was distributing new poverty alleviation fund microcredit loans through local rural banks in collaboration with district assemblymen. Introduced in 1998, this program also required each district to allocate the loans evenly across the entire administrative area. While the decentralization of disbursement did not eliminate the play of party politics or necessarily guarantee that the “deserving poor” received opportunity, it did appear to improve repayment. In contrast, the Ivoirian government continued to manage the credit schemes through the regional offices of various central government agencies. This appeared to bias the allocation of loans toward the urban centers and to perpetuate the notion that the loan was actually a “gift” from the government, thus hindering repayment. National Health Insurance Initiatives Finally, by the late 1990s, both the governments of Ghana and Côte d’Ivoire had been considering the possibility of a national health insurance program. Again, the legacies of past state institutions were evident in the varied 56 57 58

Kwadzo and Kumekpor (1994). This program had not yet been initiated at the time of research in 1999. For instance, many farmers complained that previous loans had been disbursed well after the planting season and their need for funds had passed. They were therefore unable to invest the money as planned to develop their farms so they did not reap the profits of expanded production and were later unable to pay back the loan with interest at the designated time.

State Social Service Provision

141

method of evaluation and the orientation of the proposed programs. In Ghana, the government was evaluating several pilot schemes at the time of fieldwork, all but one of which were previously existing local-level initiatives, such as the Nkoranza insurance scheme in the Brong-Ahafo region.59 Several government officials expressed the desire to identify existing local associations to organize the health insurance schemes, for example, the Ghana Private Road Transport Union (GPRTU), churches, or community associations. In contrast, the minister of health in Côte d’Ivoire appointed a highly centralized committee based in Abidjan to consider options for health insurance. Supported financially by the World Bank, the African Development Bank, and French Cooperation, pilot schemes were to be launched in two districts of the country. While the committee proposed basing the scheme on the idea of a mutual assistance organization, the central Ivoirian government appeared poised to play an active role in organizing and regulating these organizations. Furthermore, the organization would cover an entire administrative area, a relatively large unit for an “informal” organization. Despite the government’s rhetoric, which asserted that it “(does) not want a system that comes from Abidjan . . . [but rather] from the grassroots,”60 all indicators were that this initiative was being pursued in a highly centralized and top-down fashion to date.

impact of different orientation of social service provision on structure of reciprocity In summary, despite donor pressures to implement similar neoliberal SAPs, the orientation of state social service provisions varied in Ghana and Côte d’Ivoire, in ways that strikingly paralleled initial social policymaking during the colonial era. In Ghana, the government aimed to decentralize social service management and build on the existing informal social support systems of the extended family and community. In Côte d’Ivoire, the government organized new state institutions, where possible, through the arm of the central government. This chapter argues that the fundamental normative differences of British and French colonial officials in the conceptualization of the role of the state, family, and community led to the construction of 59

60

The one pilot program that was initiated by the Ghanaian government was disorganized and barely implemented due to bureaucratic tangles as of May 1999. In 2003, the Ghanaian parliament passed the National Health Insurance Scheme bill. The NHIS policy was officially launched in March 2005. Interview by author. Ministry of Health, June 16, 1999.

142

Informal Institutions and Citizenship in Rural Africa

different formal institutions of political administration during the colonial period. The interactions between citizens and the state within this institutional configuration subsequently framed the conceptualization of what types of social policy were both desirable and feasible. Thus, the Ghanaian government’s desire to reduce its role and promote an augmented responsibility for local government, NGOs and the private sector in providing social services was not simply a result of the hegemony of the neoliberal development paradigm in the 1990s, but rather a more long-term legacy of colonial administrative and social policy in the postindependence era. As citizens in this region of Ghana experienced the formal state institutions for social service delivery, the structure of their informal institutions of reciprocity was reshaped. Reinforcing earlier dynamics begun with the creation of the British political administrative institutions in the Gold Coast, this decentralized approach to state social service delivery strengthened the trend toward diversification of informal reciprocity. The orientation of the formal state institutions in the area of social policy increased the salience of social identities and reciprocal exchange among members of the broader extended family, lineage, and village community within a more localized geographic space. In Côte d’Ivoire, the preceding type of social policy was practically impossible to imagine. Colonial social policy was based on the norm of the individual receiving social welfare benefits from a provident central state. Local organizations of the extended family and community were either ignored or taken over by the central state. Even though the central government seemed to acknowledge that it could no longer provide all social services needed, the legacy of state dominance in the provision of social services obscured the possibility of working with “informal” actors at the local level in the current period. The local capacity to carry out this type of policy might have existed, but state officials were wholly unaware of the strengths or weaknesses of the informal social support system. As citizens in this region of Côte d’Ivoire experienced the more highly centralized formal state institutions for social service delivery, the structure of their informal institutions of reciprocity was also reorganized, but in a very different way than in the neighboring region in Ghana. Again, paralleling earlier interactions stimulated by the establishment of centralized political administrative institutions (discussed in Chapter 4), the more centralized and interventionist role of the state in the delivery of social services reinforced the concentration of informal reciprocity to focus more greatly on help given within the nuclear family. The orientation of the formal state social policy institutions strengthened the relevance of the social identities

State Social Service Provision

143

and reciprocal exchange among members of the nuclear and more immediate extended family. Importantly, the state’s emphasis on the central state as the key player in social service provision was reflected in a parallel shift toward the urban centers in the structure of informal reciprocity. Village residents focused more heavily on the social support exchanged with close family members based in the regional and national capitals of Côte d’Ivoire. Having now explored further the historical origins of the different structures of the informal institutions of reciprocity, it is now time to examine the surprising variation in the quantities of reciprocity exchanged in the two regions. Why did Ivoirians in this region typically give so much more help than Ghanaians?

history of stable state social service provision and timing of retrenchment affects quantity of informal institutions of reciprocity In this chapter’s second half, I investigate the origins of the puzzling differences in the quantity of reciprocity exchanged. I argue that the different historical levels of stability of state social service provision and the timing of its retrenchment explain the lower quantity of help given and the fewer number of people participating in the region in Ghana as compared with Côte d’Ivoire. History of More Stable State Social Service Delivery in Côte d’Ivoire than Ghana Not only was the overall orientation of state social service delivery very different in these two regions, but so was the stability of benefits over time. While social service delivery was initiated significantly earlier in this region of the Gold Coast, social service delivery was more consistent in this region of Côte d’Ivoire since the colonial era. The contrast in the higher stability of benefits in Côte d’Ivoire was especially stark by the late 1970s, a period acutely remembered by many respondents. These differences over time generated more expansive expectations of the central state role in Côte d’Ivoire and stimulated a higher quantity of informal reciprocity exchanged in the Ivoirian region when the state attempted to reduce its role. Starting with the French colonial authorities and continuing under the rule of Houphouet-Boigny, the Ivoirian state played a more activist role in mediating social risks. The objective was to supplant the “inadequate” communal systems of social welfare by delivering an expanding array of

144

Informal Institutions and Citizenship in Rural Africa

services through a growing, central state bureaucracy. In education, French colonial officials chose not to rely on nonstate missionary groups and decided to publicly finance and provide school services directly. In the first several decades of educational development, enrollments, particularly at the primary school level, lagged behind those in Ghana. But World Bank data suggest that by independence, Côte d’Ivoire had essentially caught up in primary school enrollments and far outpaced Ghana in enrollments at the secondary school and university level. (See Table 5.2.) After independence, two decades of sustained economic growth and political stability meant that the state delivery of services in both health and education was uninterrupted and generally expanding. In the fieldsite region, even the dormitory lodging for high school students in Abengourou, the regional capital, was provided gratis by the state until 1991 when it was finally deemed too expensive for the national budget.61 In health, the state provided office visits, hospitalization, and medicine all for free even as late as the 1980s.62 While this book focuses on the period immediately prior to the outbreak of civil conflict in 2002, it is important to acknowledge that today there is a lack of political and economic stability in Côte d’Ivoire, which has undermined severely the provision and uptake of health and education services in many parts of the country.63 In Ghana, on the other hand, from the early days during the colonial period, more emphasis was placed on a more decentralized delivery of services that supplemented and supported preexisting informal social security systems, in particular the extended family system.64 Social service delivery was initiated earlier due to the initiatives of missionaries who started many primary schools in the mid-nineteenth century, but service provision remained at a lower level and was more locally autonomous and targeted. Social service delivery was also interrupted frequently by political and economic instability at various junctures in Ghana’s history. Starting from the coup d’état that overthrew Nkrumah in 1966, a new regime was established in Ghana practically every three years. One sixty-five-year-old Ghanaian woman characterized the entire period “after independence” as difficult due to “the mismanagement of government.”65 Thus, Ghanaians have 61 62 63

64 65

Interview, teacher in Abengourou, August 2, 1999. Interview, CHR Abengourou, Abengourou, June 28, 1999. The poverty study completed by Côte d’Ivoire’s Institute for National Statistics (INS) in November 2008 as well as a UN paper on humanitarian needs are both cited in “Poverty Getting Worse,” UNIRIN, December 3, 2008, http://www.irinnews.org/www.irinnews.org. Ghana National Archives (1931). Survey interview (anonymous) by author. Makwan, Ghana.

State Social Service Provision

145

historically experienced a much more uneven (and generally lower) level of accessibility for their public social services than Ivoirians. One way to attest to the difference in the stability of social service delivery is to compare the actual outcomes revealed in the social indicators in Table 5.3. Unfortunately, it is impossible to obtain comparable statistics for the colonial period precisely because the Ivoirian administration was so centralized. Almost all statistical records are aggregated for French West Africa. Given these limitations, Table 5.3 focuses on how national-level social indicators have changed between 1960 and today in Ghana and Côte d’Ivoire. An analysis of health and education indicators shows that Côte d’Ivoire started from a lower point than in Ghana at independence, but then improved more rapidly (and most importantly, steadily) over the entire postindependence period. In contrast, indicators in Ghana improved more slowly and then stagnated earlier before the adoption of structural adjustment reforms in the early 1980s. Two factors explain these patterns. First, while the character of colonial social policy was less centralized and interventionist in Ghana, colonial administration was still implemented earlier in Ghana. This explains why literacy rates were higher in Ghana than in Côte d’Ivoire, for instance. Even though many mission schools were aided indirectly by the British colonial government, more children were attending primary school earlier than in Côte d’Ivoire. Despite Ghana’s earlier start, Côte d’Ivoire has closed most of this gap. Second, it is possible that the decentralized character of Ghanaian social policy was more efficient than the Ivoirian approach, resulting in improved health and educational outcomes despite a lower consistency and accessibility of services over time. For example, the persistent small gap in literacy rates today could reflect partially Ghana’s greater emphasis on a locally customized basic education system as opposed to Côte d’Ivoire’s heavier investment in secondary and tertiary education. Similarly, lower infant and under-5 mortality rates in Ghana might reflect a similar emphasis on locally accessible basic preventive health as opposed to the investment in big, urban hospitals for curative care in Côte d’Ivoire. An analysis of immunization rates is particularly insightful because it highlights how Ghanaian service delivery had been decimated by 1980 with only 7 percent of children twelve to twenty-four months receiving DPT immunization and only 16 percent receiving the measles vaccine. While these immunization rates in Ghana had rebounded to covering more than half of this population by 1990 (similar to the level found in Côte d’Ivoire), the earlier weakness in Ghanaian social service delivery could partially explain the slight regression in both infant and under-5 mortality during the 1990s and early 2000s.

t a b l e 5 . 3 . Comparison of Social Indicators Over Time in Ghana and Côte d’Ivoire Ghana

Life expectancy at birth (years) HIV/AIDS prevalence rate for adults (15–49 y.o.) Infant mortality (per 1,000) Mortality of children under 5 years (per 1,000) Adult literacy (%)

Côte d’Ivoire

Early Independence (1960)

Before SAP (1980)

Most Late SAP Recent (1999) (2005)

Early Independence (1960)

Before SAP (1980)

Most Late SAP Recent (1999) (2005)

46

53

44

53

*

*

58 60 (2006) (2000) 2.2– 1.9–2.3% 2.8%

*

*

47 48 (2006) (2000) 5.9– 4.0–5.0% 7.8%

126

92

212

150

27%

67% 73% (1984)

72 76 (2006) 193 (1965) (2000) 113 120 (2006) 289 (1965) (2000)

115

71%

43%

5%

* Statistics unavailable. Sources: World Bank (1982, 1998, 2001b, 2007); GOCI (1991); UNAIDS (2007); and UNESCO (2009).

168

95 95 (2006) (2000) 136 127 (2006) (2000) 61%

State Social Service Provision

147

The only health indicator that shows any regression at all in Côte d’Ivoire is life expectancy, which declined from the exact same level as Ghana in 1980 to a stunning low of forty-eight years in 2006, a full twelve years below the Ghanaian level. This difference in life expectancy is significant, but can be attributed almost certainly to the markedly higher prevalence of HIV/AIDS in Côte d’Ivoire. HIV/AIDS prevalence rates in Côte d’Ivoire might be as much as seven times higher than in Ghana.66 While not the subject of this book, Côte d’Ivoire’s higher rate of HIV/AIDS more likely reflects the higher rates of migrant labor in the country than a difference in the quality or accessibility of health care provision. Higher Expectations of the State Role in Côte d’Ivoire than Ghana As a result of these differences in the greater stability, and consequently more consistent level of state social service provision, Ivoirians in this region developed higher expectations of the state than Ghanaians. Early on in Côte d’Ivoire, colonial reports indicated that people expected the state to provide more comprehensive services in exchange for the taxation that was extracted. By the early 1930s, colonial authorities described resistance by Ivoirians to engage in communal labor for local development projects on the justification that their taxes were already paid.67 This evidence suggests that not only was the French colonial state more centralized but also that local people’s expectations of the central state role had been expanded accordingly. When the expected role was not fulfilled, Ivoirians expressed frustration. In contrast, taxation remained indirect in the Gold Coast, channeling political demands and protest primarily to local authorities rather than the central colonial government.68 In the archival materials reviewed, there were no accounts of local people in Ghana making claims on the British colonial state in exchange for the fines paid to these local treasuries as there were in Côte d’Ivoire. Buttressed by a nearly uninterrupted delivery of social services by the state in Côte d’Ivoire, the higher level of popular expectations of the state developed during the colonial era continued to evolve throughout the contemporary period. In interviews, many Ivoirian respondents described how the state had previously provided a higher level of social services in health and

66

67 68

HIV/AIDS prevalence rates for the adult population are notoriously unreliable because of the stigma associated with the disease and the lack of testing among a representative population. The tables draw from data compiled by UNAIDS. See INA (1905–1933). Crowder (1968: 222–3).

148

Informal Institutions and Citizenship in Rural Africa

education for free in comparison with today. One Kyere village resident described the current problems in the health sector: Now, a lot of people don’t go to the hospital. They die here. Before, it wasn’t like that. They gave you care and medicine. Now they are more interested in money, papers and bills before they treat you.69

Similar complaints from before and now were heard regarding the education sector. “Before the state helped the students. Now everything has to be paid for.”70 “The supplies are expensive. The fees for the students are too expensive. Before, this was free . . . ”71 Not only did Ivoirian respondents emphasize the more active role of the state in the past but they also revealed a continued expectation of a return to a greater state role in social service delivery. One thirty-five-year-old Akan woman in Côte d’Ivoire insisted, “The state should reduce (the price of) medicines in order for everyone to have access . . . The state should help the relatives like before . . . ”72 In comparison, in Ghana, while respondents complained of the high cost of health care and school fees, they rarely described a higher level of service provision provided by the state in the past, precisely because it was so erratic. Because Ghanaians described their current difficulties in gaining access to health and education services without reference to the past situation, popular expectations of the state appeared to be both lower and less relevant. In interviews, Ghanaians focused on the spectrum of strategies used to overcome present problems rather than how state delivery had changed. In general, Ivoirians also described the role of the state in much more expansive terms than Ghanaians. In reply to a question about what the state owes its citizens, one Ivoirian man began with a laundry list and then concluded with a vision of an omnipotent state that can and should provide for all social needs. Running water, to build a hospital, to fix our roads that lead to our plantations and everything that we have need of . . . We must inform the state what to give us. Everything that we need. Because only the state has the possibility of doing everything.73

Ironically, the Ivoirian government had actually built a new rural health clinic within the previous year in walking distance of one of the villages

69 70 71 72 73

Interview (anonymous) by author. Kyere, Côte d’Ivoire, July 8, 1999. Survey interview (anonymous) by author. Kyere, Côte d’Ivoire. Ibid. Survey interview (anonymous) by author. Opanin, Côte d’Ivoire. Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999.

State Social Service Provision

149

where medicine continued to be heavily subsidized. Nevertheless, respondents complained about the high costs of medicine and their lack of access to health care.74 The health clinic’s nurse explained: In the past, it [health care] was free; there are some who still have it in their head, and think everything is free. They are shocked that it all costs money. Often they return to their village . . . Also, in general, we see a lot of more serious cases now, because people delay [treatment].75

Many Ivoirian respondents also emphasized the connection between the state and the individual citizen. For instance, a sixty-five-year-old Ivoirian woman asserted, “The state should give me money to construct a house and care for me.”76 This view of state benevolence, which was both broad and individually-oriented, typified the Ivoirian response and is one indicator of how a legacy of more active public service provision fostered a higher level of expectations, and even entitlements. In comparison with the expansive lists of many Ivoirians, most Ghanaians cited one or two specific services that they expected from the state, such as farm loans or school construction. Then they usually proceeded to explain why that particular investment would be good for national development overall. Furthermore, most Ghanaian respondents more frequently invoked the linkage between the state and the community instead of the individual, as seen previously in Côte d’Ivoire.77 “Assistance from the state is to communities, school buildings, and so on. Bringing help to individuals . . . never! If an individual is in crisis, it’s up to them to figure their way out.”78 The implications of these contrary viewpoints for the development of indigenous conceptualizations of citizenship and national governance will be discussed later in Chapter 7. In addition to the previous direct articulations of higher popular expectations of the state, there was also indirect evidence of these expectations. A higher level of popular expectations of the state in Côte d’Ivoire resulted in a higher degree of frustration by Ivoirians at state cutbacks when that role was not played as it had always been in the past.79 This higher frustration was 74 75 76 77 78 79

Interview with nurse in health clinic near Kyere, Côte d’Ivoire, July 22, 1999. Ibid. Survey interview (anonymous) by author. Opanin, Côte d’Ivoire. Survey interviews (anonymous) by author. Barima, Ghana. Interview (anonymous) by author. Barima, Ghana, February 2, 1999. Prospect theory supports my argument. Kahneman and Tversky (1979) suggest that the perceptions of gains and losses affect the evaluation of utility to be received. Decisions are made based on changes relative to a prior state, not simply on an evaluation of the expected utility of an end state.

150

Informal Institutions and Citizenship in Rural Africa

evidenced by higher levels of student strikes, teacher protest, and more negative political attitudes by villagers in the Ivoirian region. First, cost recovery measures have been associated with an elevated incidence of student strikes in Côte d’Ivoire. Comments during survey interviews confirmed a connection between cost recovery and the rise of student protests. For instance, one woman explained, “Before, one didn’t spend money for school. Everything was almost free, but, now everything is expensive, and the students strike a lot.”80 Several respondents articulated their expectation that the state should intervene to resolve the “disorder” and lost time caused by the strikes in return for the taxation that is collected from the citizens. As one twenty-three-year-old woman who had only completed primary school asserted: The government should stop the strikes in the middle and high schools so that our children can succeed at school . . . I pay my “patente” and this is money for the state. And, I also pay taxes.81

The condition of the schools was objectively worse in Ghana, and although there were complaints from students, parents, and teachers, they were not as uniform or wholesale as in Côte d’Ivoire, in which the situation was notably superior. Ironically, it was the students in Ghana who not only paid for school fees and school supplies but also provided their own desks and chairs to attend a shorter day of school. The Ghanaian students in the fieldsite villages were even taken regularly from the classrooms to their teachers’ farms to work. The fact that there were complaints but not the same widespread student organization and protest reinforces the point that Ivoirians had higher expectations of the state than did Ghanaians. In addition to the student strikes, the teachers in Côte d’Ivoire were comparatively more organized and active in protesting their working conditions than the teachers in Ghana. The Ivoirian teachers had been on strike in 1999 to protest the fact that their salaries had been frozen for eighteen years and thus did not match the cost of living.82 While Ivoirian teachers’ salaries had not been increased, and in some cases had been significantly cut, Ivoirian teachers still received many more benefits than Ghanaian teachers. In most cases, Ivoirian teachers were provided with family housing and all 80

81

82

Survey interview with 52-year-old widowed woman with six children (anonymous) by author. Kyere, Côte d’Ivoire. A “patente” is a tax paid on a business or marketing activity. Survey interview (anonymous) by author. Opanin, Côte d’Ivoire. Interview, primary school teachers, Kyere, Côte d’Ivoire, July 14, 1999, and with secondary school teacher in Abengourou, August 2, 1999.

State Social Service Provision

151

the substantial benefits given to civil servants such as family allowances, limited health coverage, a pension, and so on. Again, this teacher protest in Côte d’Ivoire demonstrates indirectly the higher levels of popular expectations of the state by citizens in Côte d’Ivoire. A comparison of political attitudes toward reform from the survey research also provides evidence of a higher level of frustration in Côte d’Ivoire, which suggests again a higher level of popular expectations of state provision. More than 69 percent of Ivoirian respondents declared that the state does nothing to help villagers in need, whereas only 11 percent of Ghanaian respondents replied so negatively. In response to this question, several people emphasized how the lack of help from the state contrasts with previous higher levels of public social support. “The government does nothing for us (now). Except for talk to us of the PDCI-RDA. Even to obtain the ‘fonds sociaux’ is a problem . . . Before there were ‘cas sociaux,’ but now . . . no.”83 “Before the government was helping those who couldn’t pay for services, but now, if you have the means, they take care of you . . . .”84 In contrast with the overwhelmingly negative view of the current state role, the majority of respondents in Ghana were more positive, describing how the state gives individuals credit (37 percent) or supports villagers through the development of the community at large (29 percent). Unexpected Impact of Higher Expectations on Quantity of Reciprocity Exchanged When State Retrenched The impact of the higher level of expectations for the state role on the quantity of reciprocity exchanged in Côte d’Ivoire when the state retrenched was perhaps counterintuitive and unexpected. A functionalist institutional argument might predict that informal reciprocity networks would more or less automatically spring up in Ghana to fill the needs not being served consistently by the state. But, in fact, I find the reverse. Paradoxically, a history of higher levels of public social service provision in Côte d’Ivoire created the context initially for higher levels of giving through informal institutions of reciprocity in the contemporary period. In addition to political protest and verbal complaints, which were limited in their scope and ultimate effect, frustrated Ivoirians in this region turned to their informal reciprocity networks to provide the social services they missed. A history of more stable public service provision created higher expectations of the state in Côte d’Ivoire and stimulated Ivoirians’ attempt to replace the social services 83 84

Survey interview (anonymous) by author. Kyere, Côte d’Ivoire. Ibid.

152

Informal Institutions and Citizenship in Rural Africa

formerly delivered by the state at a much higher level than what was attempted by Ghanaians. One teacher and native of Opanin explained how the state itself created the higher level of expectations in Côte d’Ivoire: We have the mentality . . . Before, it was the “welfare state” (“etat providence”). The state itself helped individuals. With free health. With free schools. Even the lodging at the university was free. But the population grew a lot, and now this is impossible . . . We always wait for those in government to do something. We never try and manage. It’s the government who put this in our head.85

Thus, by historically providing a more stable and universal coverage of social services for free, the Ivoirian state created a higher level of popular expectations – both of the minimum set of social services and of the state’s role. Ivoirians actually did “try and manage” once the state reduced public provision, but not by sacrificing the level of services. Instead, Ivoirians attempted to replace the services previously provided by the state temporarily through their own resources; thus, informal relations of reciprocity were mobilized and utilized at a higher rate than in Ghana. This first factor combines with the next point about the timing of the reform to result in a higher level of giving and participation in the informal institutions of reciprocity in the Ivoirian region than in the Ghanaian one. Later Reform of a System in Less Serious Decline in Côte d’Ivoire Versus Earlier Reform of a Collapsed System in Ghana The second major point of difference in the way state social services were delivered over time in these two regions was the actual timing of retrenchment reforms. The two states reduced their role in the provision of social welfare at different points along a trajectory of state decline, and this occurred at different moments in chronological time. The quantity of informal reciprocity exchanged was heavily influenced by the Ivoirian state’s later reform of a system in less serious decline as compared with the Ghanaian state’s earlier reform of a collapsed system. In Côte d’Ivoire, problems were surfacing in the public sector by the early 1980s, but they were not as evident to the average person in the rural areas as in Ghana. Perhaps the hospitals and school dormitories were not as well maintained as in the past in Côte d’Ivoire, but the social sector was nowhere near the level of deprivation found in Ghana, in which school buildings were dilapidated, nearly half of the primary school teachers were completely

85

Interview (anonymous) by author. Opanin, Côte d’Ivoire, September 23, 1999.

State Social Service Provision

153

untrained, and pharmacy shelves were dangerously empty.86 The main challenge facing the Ivoirian government was how to cut costs in order to reduce budget deficits to maintain relations with donors and overseas investors. The problem in Ghana in the early 1980s was much more acute: how to rescue a social services sector that had essentially “broken down.”87 Thus, Ghana was forced to adopt neoliberal reforms by the early 1980s, whereas the Ivoirian state did not fully adopt them until the later 1980s. Devaluation, one of the cornerstones of the neoliberal reform program, began in incremental stages in Ghana as early as 1983, but was implemented all at once in Côte d’Ivoire, and considerably later: in 1994. In fact, World Bank documents often use 1994 as the benchmark for the beginning of adjustment reforms in Côte d’Ivoire. The rapid style and later date of devaluation in Côte d’Ivoire are significant because the change in relative prices was experienced almost immediately, even in the rural areas, and for certain medical and educational supplies in particular. Several respondents described how after the devaluation, “everything became expensive” and “even hospital costs, we couldn’t pay.”88 “It’s devaluation that changed things, that increased the prices of necessities. In the time of Houphouet (prior to 1994), there was no devaluation so everything was good.”89 Cost recovery measures in social services were also adopted in Ghana much earlier than in Côte d’Ivoire, in which they were first attempted piecemeal in 1991 and not generalized until 1994.90 For example, in Ghana, the “cash and carry” system in health care was introduced by the mid-1980s, while primary schools began to charge school and textbook fees in 1987. In contrast, the Ivoirian government started to require payment for medicines only in 1991, and this policy was not generalized to services in hospitals and clinics until around 1994. Overall, then, state retrenchment, popularly termed “the crisis,” had begun more recently in Côte d’Ivoire. Ivoirians recalled with pride the long history of expanding state infrastructure and could not seem to comprehend that things might not return to “normal” relatively soon. Thus, Ivoirians struggled to provide themselves with the same level of social services in what they assumed to be an interim period of “crisis.” In comparison, in Ghana, de facto state retrenchment had begun at least a decade prior to the adoption of structural adjustment, so economic hard times were practically a way of life by the late 86 87 88 89 90

See World Bank (1996: 2). Interview (anonymous) by author. Tano District Assembly. Survey interviews (anonymous) by author. Kyere and Opanin, Côte d’Ivoire. Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999. Interview (anonymous) by author. Ministry of Health, Abidjan, Côte d’Ivoire, October 12, 1999.

154

Informal Institutions and Citizenship in Rural Africa

1990s. One thirty-year-old Ghanaian woman described how she had only known hard times in her life: “Ghana has always been difficult. I have never seen good times . . . ”91 There was no commonly understood term or label for what seemed to be an enduring hardship in this region of Ghana. Ghanaians could not pinpoint a particular moment when these hard times hit like Ivoirians could recall where they were when the CFA was devalued. In general, unlike Ivoirians, Ghanaians did not make the distinctions of life before structural adjustment and after because both periods were difficult, albeit in different ways. When asked to identify an especially difficult period, several respondents responded “always” or as one sixty-year-old woman with ten children explained: “Hard times are always there. There is no money and there are no good jobs that we get money from, so it is always very hard.”92 More Negative Political Attitudes Toward Reforms in Ivoirian than Ghanaian Region This difference in the extent of decline before reforms were adopted resulted in different attitudes toward the reforms themselves in the two regions. As with the development of different expectations of the state role, these contrasting political attitudes toward reform in the two regions has led respondents in the two regions to mobilize the informal institutions of reciprocity in different ways, stimulating the variations in the quantity of informal reciprocity that are explored in the chapter’s final subsection. In general, although Ivoirians nearly all resented the tremendous increase in the cost of social services that they had consistently enjoyed for free, many Ghanaians actually noted some positive improvements. Again and again, Ivoirians asserted that everything became more costly “about five or six years ago” or “after the devaluation.”93 For example, one sixty-eight-year-old man complained about the huge increase in costs of health and education services wrought by the devaluation in 1994: Now, there isn’t one help (from the state). Education is now our burden. Now, we have to pay for the supplies . . . And, if you go to the hospital, they give you bills. You must pay. If not, you’ll just have to go back to the house to die.94

As one elderly non-Akan man explained: Before, one was taken care of for free. Now, everything is difficult. If your pocket is heavy, you are treated. Before, the state sent medicines and vaccinations for all the

91 92 93 94

Survey interview (anonymous) by author. Barima, Ghana. Ibid. Ibid. Interview (anonymous) by author. Kyere, Côte d’Ivoire, July 14, 1999.

State Social Service Provision

155

villagers. The state now . . . the means no longer exist. The state even is tired now. Now, it’s difficult . . . You must pay for the school supplies at school. If not, your child will come back to the house.95

Repeatedly, Ivoirians stressed that these rising costs have meant lowered access to state social services. In comparison, even though Ghanaians sometimes grumbled about the costs of state social services, they often appreciated the fact that medicines, medical equipment, and school supplies were even available. Many of the Ghanaian respondents remembered the economic crisis of the late 1970s that preceded the adoption of structural adjustment as a very difficult period because “goods were scarce,” even essential drugs.96 Several district government officials described how, in the past, when the government provided medicine and other goods for free, the goods were stolen and then resold on the black market.97 Thus, Ghanaians did not compare the new user fees to the past fees that were “officially” very minimal, but to the much higher costs of the side payments to public servants to access scarce public services, the exorbitant fees charged by private facilities in order to circumvent the public sector, or the opportunity cost of not being able to obtain any services at all. Some Ghanaian respondents even criticized themselves for “crying about hardships today” when “things are flowing on the shelves” in comparison with the past, when people had to stand in lines for basic foodstuffs such as sugar, milk, and soap, if they were even available.98 Results from a University of Ghana survey before the 1996 elections confirmed this trend of split opinion. Respondents were almost evenly divided as to whether the Ghanaian economy had improved (43.2 percent) or become worse (47.8 percent) between 1992 and 1996.99 These overall differences in attitudes toward adjustment reforms persisted in the more specific discussions of health care services. Village residents of this region of Côte d’Ivoire tended to stress the high cost of health care in general and of medicine in particular. More than 50 percent of Ivoirian respondents reported that their access to health care had been reduced because state services were unaffordable. This popular perception of declining access was validated by regional data that showed the occupation rate in the regional hospital in Abengourou had dropped significantly from 95 96 97 98

99

Interview (anonymous) by author. Kyere, Côte d’Ivoire, August 18, 1999. Survey interviews (anonymous) by author. Barima and Makwan, Ghana. Interviews (anonymous), Tano District Assembly, Bechem, Ghana, April 23, 1999. Interview (anonymous) by author. Barima, Ghana, February 7, 1999. Interview (anonymous) by author. Barima, Ghana, March 21, 1999. See UOG (1996b: 7).

156

Informal Institutions and Citizenship in Rural Africa

76 percent in 1989 to only 61 percent by 1993.100 More than 58 percent of the Ivoirian respondents complained in particular about the high cost of medicine. In comparison, only 10 percent of village residents in the Ghanaian region reported diminished access due to the high cost of health services, and only 2 percent of Ghanaians noted the high cost of medicine. One sixty-five-year-old Ivoirian man with twenty-seven children complained bitterly of cost recovery, emphasizing that urgent care was delivered freely in the past, but now the bills come first: Before, they gave the first medicines before you bought them, whereas now, when you arrive, they don’t treat you, and they give you bills. If you don’t have anything, you are going to die . . . There were three or four people from my house who died due to a lack of means.101

Confirming my survey research findings, a World Bank study found that the “very poor” in Côte d’Ivoire spent 70 percent of their total health expenditures on medicines.102 In contrast, one government official in Accra summarized the typical response that “the health care delivery system has improved immensely in terms of infrastructure.”103 Ministry of health officials in Ghana reported that the “cash and carry” system has enabled hospitals to have more drugs available and offer longer opening hours.104 At the village level, public opinion was split relatively evenly with regard to access to health care. While 18 percent of Ghanaian respondents replied that they had increased access to health care due to their own improved economic situation or better government policies, another 17 percent answered that they had access to health care only in an emergency. Respondents in Ghana were about evenly split with regard to the change in health care quality as well (29 percent saying quality had improved, and 26 percent reporting quality had worsened). As with health care, the overwhelming response in Côte d’Ivoire was that schooling was more expensive and thus more difficult to access (59 percent). For parents, the costs of educating a child were not only increased by the implementation of school fees but also by the increased cost of school

100

101 102 103 104

GOCI (1994). No comparable statistics were available for the district or regional hospitals in the Ghanaian region. Survey interview (anonymous) by author. Kyere, Côte d’Ivoire. Shepard (1996: 6). Interview (anonymous) by author. Accra, Ghana, April 21, 1999. Interview (anonymous) by author. Ministry of Health, Accra, October 27, 1998.

State Social Service Provision

157

supplies that those parents must now provide.105 One woman’s explanation typified many responses: “Before all of the supplies were free, even the school uniforms. But, now, everything costs money and is very expensive.”106 One study conducted by the Ivoirian government found that the cost of textbooks and school supplies in the Abengourou region totaled almost $30 per student by the last year of primary school and $208 by the last year of secondary school.107 The cost of these school supplies was clearly an enormous expenditure for individuals who reported a median agricultural income of $333 in a year. In the area of educational reform, attitudes among Ghanaians were also comparatively mixed. In comparison with the nearly two-thirds of Ivoirian respondents highlighted previously, only 9 percent of Ghanaian respondents described schooling as more expensive and thus less accessible after structural adjustment reform. In fact, nearly 22 percent of all Ghanaian respondents noted an increased access to schooling, compared with only 6 percent noting an increase in Côte d’Ivoire.108 Unfortunately, while some Ghanaians cited an increase in access to schooling, almost all the Ghanaian respondents noted a precipitous decline in the quality of education.109 One respondent explained, “Education is less expensive than ten years ago but unfortunately the children don’t know anything.”110 In particular, village residents in the Ghanaian region bemoaned the illiteracy of many junior high school graduates. “Standards have now fallen so low. Even JSS pupils can’t spell or even write a letter.”111 District-level data for Tano District, Ghana, showed that while 62 percent of district residents had completed “basic education,” only

105

106

107 108

109

110 111

Several respondents noted that that the cost of supplies is even higher because the books required for each grade level are changed so frequently that their younger children are unable to reuse the older siblings’ texts. Survey interviews (anonymous) by author. Kyere and Opanin, Côte d’Ivoire. Survey interview (anonymous) by author. Opanin, Côte d’Ivoire. Similar responses were transcribed from survey interviews in Kyere, Côte d’Ivoire. GOCI (1996). This perception was partially due to highly visible public expenditures to construct and rehabilitate school buildings, especially in rural areas. The poor quality is attributed to poor teaching due to the lack of decent salaries, housing, or teaching materials available for teachers; the attempt to teach too many subjects each year; a shorter school day; forced promotion policies; and the lack of effective evaluation. Men’s focus group (anonymous). Tape recording. Makwan, Ghana. This finding is confirmed by World Bank research (World Bank 1995: iv) that concludes that the poor quality of education was the “single most important concern” of interview respondents. Survey interview (anonymous) by author. Makwan, Ghana. Survey interview (anonymous) by author. Makwan, Ghana. A low level of literacy was confirmed by my informal discussions with junior secondary school children in both villages.

158

Informal Institutions and Citizenship in Rural Africa

37 percent were literate.112 Village residents in this region of Ghana shared stories of unemployed sons and daughters and admitted their unwillingness to sacrifice to school their children in the short term if the long-term payoffs were uncertain or minimal.113 This perception of lower quality in the public schools also propelled parents in one of the Ghanaian villages in 1993 to found a new private school, named the “International Preparatory School.”114 Despite charging fees that were eight to twelve times higher than the public primary schools, enrollment was growing, and parental satisfaction seemed high.115 In summary, although respondents in Ghana were more divided and unsure about the positives and negatives of the current round of adjustment policies, in Côte d’Ivoire respondents replied more unified and certain that social services had declined. As one youth leader in Côte d’Ivoire asserted emphatically about the educational system, “It’s clear; it’s gotten worse.” In contrast, the queenmother of one Ghanaian village expressed a guarded optimism saying, “It’s getting a little better now. It’s definitely not worse than the past twenty years.”116 Unexpected Impact of Timing of Retrenchment on Quantity of Reciprocity Exchanged The analysis of the timing of retrenchment on the quantity of reciprocity is key because people cope differently with what they perceive as a more recent, short-term “crisis” of a functioning system than they do with a long-term, chronic hardship due to a collapsed system. The higher levels of reciprocal giving in Côte d’Ivoire were thus partially due to the fact that the economic

112

113

114

115

116

GOG (1996: 52). These district-level results were confirmed in a national-level USAID study of P6 students that showed “appalling results” for pupil achievement. Cited in report by World Bank (1996: 5). District-level data confirmed a rise in unemployment from 5.6 percent in 1970 to 8.3 percent in 1984 to 19 percent in 1996. GOG (1996: 56). World Bank officials in Ghana described parallel results at the national-level finding that the perception by parents of low quality in state-run junior secondary and secondary schools had led to lower primary school enrollment. Interview by author. World Bank, October 29, 1998. Interviews (anonymous) by author. Makwan, Ghana, March 30, 1999 and April 23, 1999. A World Bank study finds that private primary and to a lesser extent junior secondary schools were spreading at a higher rate than public schools. World Bank (1996). While the other two primary schools charged 2,000 cedis ($0.95) per year for grades P1 through P6, the International Preparatory school charged 16,000 cedis ($7.62) per year for P1 through P3 and 24,000 cedis ($11.43) per year for P4 through P6. Interview (anonymous) by author. Makwan, Ghana, April 23, 1999. Interview (anonymous) by author. Makwan, Ghana, March 31, 1999.

State Social Service Provision

159

crisis was relatively new during the study period of the late 1990s. But the effects of the timing of retrenchment (in the previous section) combined in critical ways with the stability of social service delivery over time (discussed earlier). When all of a sudden faced with state cutbacks and the increased costs of social services, Ivoirians responded by trying to maintain their usual standard of living. They perceived the economic crisis as an aberration from the norm of Côte d’Ivoire’s long history of prosperity and more consistently accessible social services. As a consequence, Ivoirians in this region struggled to maintain their previously high level of social service provision by mobilizing their networks of reciprocity. Meanwhile, in Ghana, lower expectations of social service delivery combined with the rescue of a collapsed social service system to produce a perception that relatively more needs were already being met than before. As a result, Ghanaians did not make as many demands on their informal relations of reciprocity as Ivoirians to replace retrenched government services. These dynamics help explain a higher level of giving among a greater number of people in Côte d’Ivoire than in Ghana and are discussed in detail for each region in the following sections. Ivoirians Drawing Down Reciprocity Assets in Response to “Crisis” Even if the economic crisis was viewed as unquestionably short term, the struggle to adjust to these hard times was not easy for village residents in the Ivoirian region. Respondents frequently summed it all up with a heaving sigh, declaring: “It is hard.” At the time of this study, in the late 1990s, Ivoirians were still in the process of drawing down their wealth stores. Wealth stores in this context included cash savings (which was limited in the fieldsite regions), assets such as land or durable goods, and their relations of reciprocity. One teacher described this sale of assets due to the economic crisis by saying, “In the past one would see three or four teachers’ cars parked in front of the school; but today, there is not even a moto.”117 Ivoirians had cut other costs too – a quick walk around the villages made it clear that cars and homes that were once sources of deep pride had not been well maintained recently. Yet Ivoirians were unwilling to sacrifice the health or education of their families. One woman said that she would borrow money from someone to send her children to school: “You have to borrow, because the children are the hope of tomorrow, thus, you can’t let them trail behind.”118 So, some 117 118

Interview (anonymous) by author. Opanin, Côte d’Ivoire, September 25, 1999. Survey interview (anonymous) by author. Kyere, Côte d’Ivoire.

160

Informal Institutions and Citizenship in Rural Africa

have begun to sell land, and almost all have begun to “cash in” their personal and extended family obligations. While few Ivoirian respondents explicitly described relying on their social support networks more in the contemporary period than in the past, many implied this indirectly. For example, many Ivoirians explained that in the past they did not pay for anything (because the state provided it for free) and now they “are obliged to borrow” from family members in order to pay for social services. This explains the higher level of giving among Ivoirians. It also explains why Ivoirians gave a significantly higher amount to their nuclear families while still giving a lot to their extended families. The crisis had increased the urgent needs for informal reciprocity.119 Ghanaian Reciprocity Nearly Exhausted After Longer Period of Hardship In comparison, Ghanaians had already made downward adjustments to their lifestyle (perhaps long ago). Their expectations of state provision were also lower than in Côte d’Ivoire to begin with, and the changes caused by reforms were not all negative (as discussed earlier). Nevertheless, Ghanaians, too, probably struggled to maintain a more modest level of social services at the beginning, but their access had declined over time. They struggled for a long time to pay for certain services, often through private hospitals and schools, but over time they began to exhaust their reserves of money, durable goods, and favors. Ghanaians had already drawn down their wealth stores to the point that they were now minimally involved in the informal institutions of reciprocity. One Ghanaian man made the connections between the prolonged hardship and the “wickedness” of exclusion from the social support of the extended family: If there are four children and it is only one person’s child who is okay [in a position to help others], the other 3 are out. It is something that we came to meet. We are always in hardship here.120

The survey research results confirmed how the extended economic crisis has debilitated informal reciprocity in Ghana. Almost 14 percent of all Ghanaian

119

120

These general analyses and conclusions about the dynamics in Côte d’Ivoire do not mean that all Ivoirians were able to mobilize social networks and pay for the same level of social services through private financial means. There were Ivoirian respondents who reported withdrawing their children from school, being forced to seek out indigenous medicine or otherwise delaying treatment at the hospital until the situation was aggravated. Nevertheless, these accounts appeared to be in the minority. Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999.

State Social Service Provision

161

respondents reported that because they were in difficulty themselves, they could not give any help. One focus group participant articulated the implications of the long-term hardship in Ghana very clearly: This situation was not so in the past. It is these days that hatred has entered into the family such that one regards only those who are very close or direct brothers or sisters. These things are happening as a result of economic hardships that are facing people now.121

In contrast, only 4 percent of respondents in Côte d’Ivoire cited economic difficulty as a reason for not giving help. Another indicator of the diminishing frequency of social exchange was the fact that almost 19 percent of Ghanaian respondents reported that if someone needed help, they did not ask the respondent. So not only were these Ghanaian respondents not able to give help, but the requests were not even being made to them. No one reported similarly in Côte d’Ivoire. While reciprocity was theoretically a renewable resource, in the context of an extended economic hardship, many Ghanaians had not been able to maintain reciprocity with their social relations; thus participation in social support networks had diminished, and the amount given had dwindled.

conclusion In summary, because expectations of the state were higher, state delivery had not yet collapsed, and the crisis was relatively new, the relations of reciprocity were much deeper and more extensive in Côte d’Ivoire than in Ghana. Ironically, despite the Ghanaian government’s efforts to reinforce informal social security, these support systems were relatively weak in comparison with those in Côte d’Ivoire. This does not mean, however, that the state has had no effect on the informal institutions of reciprocity. In fact, contrary to the prevalent image of a “weak state” in Africa, I present evidence that state policies did matter. It is just that the consequences of state action were often unintended. The research design also strengthens the conclusion that the local variations in reciprocity were not due to differences in local custom or wealth accumulation. Instead, culture and wealth articulated with the way that state policy was actually implemented over time to shape these social relations. Second, I show that it is not the aggregate social spending levels that make a difference for equity and social exclusion, but rather, the way that social 121

Men’s focus group (anonymous). Tape recording. Barima, Côte d’Ivoire.

162

Informal Institutions and Citizenship in Rural Africa

policy was actually implemented, highlighting the implications of the delivery of social services for the poor. The historic variation in decentralized versus étatist service delivery appears to have been crucial in shaping public perceptions of the positive or negative impact of reforms. Thus, despite similar spending levels by the state, Ivoirians perceived reforms consistently more negatively than Ghanaians. Third, I reveal the unintended consequences of state formation. Ironically, in Côte d’Ivoire, where the state historically took a more active, étatist role in providing social services to individual citizens, the informal institutions of reciprocity were deeper and more extensive than in Ghana, in which the more decentralized state diminished its role. This is because the legacy of state provision created a higher level of expectations for social services in Côte d’Ivoire than in Ghana. In Ghana, because social services had never been provided consistently, Ghanaians did not mobilize their informal reciprocity relations as intensely to provide these services after the state retrenched. This finding does not necessarily mean that “strong states” create “strong informal networks,” and “weak states” produce “weak informal networks,” a hypothesis that would refute much of the literature on collapsed states and the informal political economy.122 If anything, the finding demonstrates the futility in such oversimplified dichotomies. These conclusions confirm the need to recognize the complexity of the everyday interactions between the state and social actors. Future research should analyze the actual experience of the state at the local level to understand the dynamic feedback between formal state and informal institutions. Finally, the analysis raises an interesting question to be answered: Is the variation in informal reciprocity institutions simply a difference in the point at which the study took its “snapshot”? Will the informal institutions in Côte d’Ivoire look exactly the same as those in Ghana if the “crisis” over time becomes “chronic hardship”? If the economic crisis were to continue in Côte d’Ivoire, the variation in the value and extent of participation in informal reciprocity likely would be diminished over time, but the process would be slower and on a different scale because of the other aspects of the legacies of social service delivery noted in this chapter. Ivoirians are starting from a higher level of expectations of social services, and those expectations were never crushed as abysmally as in Ghana by the near collapse of the state social sector. Thus, Ivoirians will most likely try harder to replace those services through their relations of reciprocity. It is likely that as the crisis lengthens in Côte d’Ivoire, the ratio of loans with interest will grow in 122

See Reno (1998) and Chazan (1988).

State Social Service Provision

163

proportion to those loans without interest as well as outright gifts. There is evidence to suggest that this was already happening in Côte d’Ivoire.123 Ultimately, over a longer period of time, their social networks will likely become exhausted and will look more like those in the Ghanaian region; however, the structure of the networks will still differ. Thus, although these institutional legacies of state social service provision explain much of the variation in informal reciprocity, they do not solve the entire puzzle. Chapters 4 and 5 demonstrated how the orientation of political administration and social service provision produced a higher level of diversification in the structure of reciprocity overall in the Ghanaian region, but a mystery remains. The lingering mystery is this: Why did Ghanaians place such a priority on friends, particularly if they were so hard pressed? In order to fit in this last piece of the puzzle, I turn now to the striking changes in cash crop production occurring in Ghana, but not in Côte d’Ivoire.

123

Men’s focus group, Kyere, Côte d’Ivoire, and women’s and men’s focus group, Opanin, Côte d’Ivoire, September 1999.

chapter 6 The Empire of the Young: Contrasting Legacies of State Agricultural Policy for Local Capitalism and Reciprocity

Different legacies of state formation have structured and restructured the informal institutions of reciprocity in diverse ways in these two regions of Ghana and Côte d’Ivoire. While it may be more apparent how the history of the state’s interventions in political administration and social service provision would influence patterns of informal reciprocity, it is important to look beyond these two policy areas to investigate a less-obvious one: the state’s role in agricultural policy over time. Village residents in these neighboring regions have experienced the state’s agricultural policies differently at the local level, and thus have responded with dissimilar production choices at various points in history. This chapter examines how a recent dramatic shift in local-level production strategies has influenced local social relations, spurring a redefinition of the boundaries of social and political exclusion.

recent divergence in local cash crop production in ghana and coˆ te d’ivoire In the late 1990s, farmers in these two regions were choosing different cash crops. Tomatoes were becoming the new cash crop of the young in Ghana, whereas cocoa remained the primary cash crop in Côte d’Ivoire. This was a recent divergence in local production strategies with important implications for the informal institutions of reciprocity. New Tribulations for the Cocoa Farmer Since at least the early 1930s, cocoa has been the main crop grown in the two fieldsite regions. Cocoa was first established in Côte d’Ivoire and Ghana in 164

Legacies of State Agricultural Policy

165

other nearby regions of the same umbrella Akan ethnic group, but quickly spread to the research areas.1 During this early era of cocoa production, the colonial governments in each country designed economic policies to encourage African farmers to produce more and higher-quality cocoa for the global market; in particular, for the chocolate manufacturers in each colonial metropole. Much like the colonial social policies examined earlier, colonial economic policies shared common objectives, but differed in important ways in the implementation of these broad economic goals in the British versus French colonies. By the 1950s and 1960s, cocoa was so dominant that both regions were considered relatively wealthy. Because cocoa is a permanent tree crop that requires long-term access to a sizable area of land,2 the indigenous Akan ethnic group, primarily males, dominated cocoa production. Women often worked on their husbands’ cocoa farms, but did not necessarily share or have control over the cocoa profits. They often took charge of various food crops such as plantain, cassava, cocoyam, tomatoes, peppers, and okra, which were used for household consumption and often sold in local or regional markets.3 Until the 1970s, cocoa plantations thrived, but in the early 1980s, several droughts sparked bushfires that destroyed a considerable amount of the forest, including many established cocoa farms. At the same time, neoliberal economic policies adopted in the 1980s in Ghana and Côte d’Ivoire made the export of timber extremely attractive, increasing the rate of logging in these regions and further reducing the semiforested land.4 Meanwhile, human populations continued to grow. Hence, Ghanaian and Ivoirian farmers in these regions found it increasingly difficult to find suitable semi-forested land to reestablish their lost farms or start new ones. One Ghanaian cocoa farmer’s complaints were echoed by farmers in Côte d’Ivoire: “We can’t get any 1

2

3

4

Hill’s (1963) classic study examines the first “pioneer” expansion of cocoa farming in Ghana around the turn of the twentieth century. Dunn and Robertson (1973: 46) describe the adoption of cocoa in the Ahafo region by 1906. See also Acquaah (1999). Cocoa trees require approximately five years before they can be harvested for the first time. After peaking between five and fifteen years, yields began to decline, and trees need to be replaced after thirty years. Food crops can be intercropped with cocoa trees for about five years, until the cocoa tree canopy provides too much shade for food cultivation below. At this point, food is grown on separate plots. The village chief often negotiates the deal with the logging companies without the input of the concerned farmers. Hence, a farmer might go to his plantation only to find loggers cutting down an enormous tree. Typically, the felled trees are so big that the trucks can only carry part of one. Many cocoa trees are destroyed immediately during the process of felling and transporting the timber tree. Then, others are hurt over the long-term from the lack of shade.

166

Informal Institutions and Citizenship in Rural Africa

land; there is no forest left at all.”5 Land pressure was hence a common problem facing farmers in both regions. The reduction of forested areas also generated a more permanent climactic change – the dry season is now longer and more severe, which has resulted in recurrent droughts and additional bushfires. Many Ghanaian and Ivoirian farmers agreed that “it doesn’t rain like it used to . . . ”6 Meanwhile, the surviving plantations were aging, soil fertility was diminishing, and yields were subsequently declining. An Ivoirian agricultural researcher estimated that where the big plantations formerly produced ten tons of cocoa annually, now, they might yield between one-half and one ton.7 Due to declining yields, migrant agricultural workers, typically from the Northern regions or Sahelian countries (e.g., Burkina Faso, Mali, or Niger), were moving southwest to the newer cocoa areas in the two countries, making it difficult to maintain and harvest the remaining cocoa farms. Dramatic Switch to Tomato Farming in Ghana, but not in Côte d’Ivoire Surprisingly, even though farmers in Ghana and Côte d’Ivoire have faced broadly similar ecological and economic challenges, they responded differently with divergent crop production strategies in the 1990s. In Ghana, many farmers were abandoning cocoa for tomato farming,8 while in Côte d’Ivoire, farmers continued to expand their existing cocoa plantations and establish new ones. The situation in Ghana constituted a dramatic shift in local-level production, with far-reaching implications for social relations. This recent divergence in cash crop production strategies was even more surprising given the similarities in transportation infrastructure and urban integration in Ghana and Côte d’Ivoire. Farmers in this region of Côte d’Ivoire theoretically could have adopted tomato production, too, and even earlier than in Ghana, but they chose to continue with cocoa. Transportation networks were actually consistently better in Côte d’Ivoire than in Ghana during the postcolonial period. At the time of fieldwork in the 1990s, Côte d’Ivoire’s infrastructure was not as well-maintained as it had been during the cocoa boom times, but the roads were still extensive and well-paved between the fieldsite villages and the regional and national capital. Even when the 5 6 7 8

Interview (anonymous) by author. Barima, Ghana, April 29, 1999. Interview (anonymous) by author. Kyere, Côte d’Ivoire, July 14, 1999. Interview with researcher at ANADER (anonymous) by author. Abengourou, July 22, 1999. Lyon (2000) finds a rise of tomato production in other districts in the same Brong-Ahafo region of Ghana. Berry (1997) finds a similar commercialization of vegetable production, in particular, that of tomatoes, in Kumawu, Ashanti region.

Legacies of State Agricultural Policy

167

Ivoirian economy began to struggle during the late 1970s, foreign exchange reserves were still adequate enough to continue importing fuel, vehicles, and spare parts. Finally, the distance between the villages in the Ivoirian region and the regional and national capitals was approximately equal to the distance traveled in the Ghanaian region. Of course, the market in Kumasi, Ghana, was much, much bigger in size than that in Abengourou, but this difference alone would not have driven the variation seen here in production choices. Tomatoes could have been adequately marketed in Abengourou or relatively easily been taken to Bouake (four hours) or Abidjan (five hours). Yet, farmers in this region of Côte d’Ivoire continued to produce cocoa while the farmers in the Ghanaian region were switching to tomatoes as the new cash crop. Cocoa versus Tomato Production Levels First, we need to compare the numbers of farmers, farm plots involved and production values in cocoa versus tomato production. This data reveal that cocoa production was declining in Ghana. In Ghana, only a handful of farmers (15 percent) grew cocoa on an average of 0.13 plots, while more than twice as many grew tomatoes (36 percent) on more than twice as much land (0.32 plots).9 The majority of the farmers in the Ghanaian villages no longer concentrated on any one particular cash crop, but grew a diverse mix of mostly food crops: some for subsistence, and some for sale to domestic markets. Again, these production changes are dramatic indeed for a region in which cocoa historically was dominant and had brought much prosperity. Elderly cocoa farmers confirmed that the cocoa farms in the contemporary period were “very, very small now” compared with the plantations of the past.10 The mean total cocoa production (U.S.$304)11 was significantly below the value of tomato production in Ghana ($465), and less than onethird of the mean value of cocoa production in Côte d’Ivoire ($932). One of the largest cocoa farmers in Barima, Ghana, reported that prior to the bushfires of 1983, he was harvesting fifty bags of cocoa per year; after 1983, he 9

10 11

The mean total number of agricultural plots was 1.87 in Ghana and Côte d’Ivoire. Since it was impossible to measure physically respondents’ plots, the survey asked for an estimation of the number of plots of land cultivated in the past year. Since most of these plots were intercropped, many crops were cultivated on less than one plot. A problem that remains unresolved in the data is that one farmer’s plot might be significantly bigger in size than another’s. Nevertheless, cross-sectional trends, in particular among crop mixes, remain valid. Interview (anonymous) by author. Makwan, Ghana, April 7, 1999. The mean total production of cocoa in Ghana was slightly higher than tomatoes, at $575, but when one large producer (earning over $7,000 annually from cocoa) is taken out of the sample, the mean drops to only $304, well below tomatoes.

168

Informal Institutions and Citizenship in Rural Africa

harvested only ten to fifteen bags per year.12 District-level data also confirmed the survey research findings from the two Ghanaian villages that cocoa production had declined during the early 1990s, dropping from more than 3,000 tons in 1990–1991 to only 1,163 tons in 1994–1995.13 Not only were more farmers growing tomatoes, tomatoes had become the primary income source for many more Ghanaian farmers than Ivoirian ones.14 The mean production of individual tomato farmers was also much higher in Ghana ($465) than in Côte d’Ivoire ($48), and the maximum value of tomato production was nearly ten times higher in Ghana ($7,200) than in Côte d’Ivoire ($720). The existence of many large tomato producers in Ghana reveals a qualitative change in the nature of tomato production there. In the past, a few tomato plants were typically intercropped among cocoa trees for subsistence purposes; now whole plots were dedicated entirely to tomatoes. Also, in Ghana, more than 70 percent of survey respondents reported major changes in and out of tomatoes, whereas in Côte d’Ivoire, there had been almost no change. The volatility in tomato production was due to the short growing season (three to four months), which facilitated experimentation, and the high risks associated with marketing such a perishable crop,15 which caused huge variation in prices. Farmers reported prices ranging from $38 all the way to more than $200 per box in the late 1990s. In contrast with these striking changes in Ghana, cocoa remained the single most important crop produced in Côte d’Ivoire. More than six times as many farmers (71 percent) produced cocoa in Côte d’Ivoire as in Ghana, on more land (0.35 plots). Some Ivoirian farmers were less confident about the capability of their cocoa plantations to provide them with security, due to the prevalence of bushfires and drought in recent years. Nonetheless, data from survey and in-depth interviews confirmed that not only were existing plantations being expanded but new ones were also being established. For

12 13 14

15

Interview (anonymous) by author. Barima, Ghana, April 29, 1999. GOG (1996: 24). 22 percent of farmers in the Ghanaian region derived their primary income from tomatoes compared to only 3 percent in the Ivoirian region. Many Ghanaian tomato farmers bemoaned the lack of storage for their produce. Apparently, there used to be a factory that canned tomatoes and provided storage but was closed. One Akan man described the risks in marketing tomatoes: “If you’re into tomato farming, and by three days, no one comes to buy, they’ll all go bad, so you lose all of your money.” Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Ghanaian farmers also described how “market women” would collude to “dictate the price” in the villages and together refuse to buy any tomatoes that had been transported privately by farmers to the markets in Kumasi and Accra. See also Clark (1994).

Legacies of State Agricultural Policy

169

instance, one Ivoirian woman declared that she was going to create a larger cocoa plantation to be secure from future hard times.16 A Comparison of the Demographic Profile of Tomato and Cocoa Farmers The demographic profile of the farmers producing tomatoes and cocoa was even more remarkable. In Ghana, tomato production was the empire of the young. One chief, when describing the agricultural activities of his village, began with tomatoes, saying that it was “the main work for the youth.”17 The queenmother of the second Ghanaian village confirmed this trend, attributing a rise in the village’s standard of living to the fact that “now, the young men are up and hard-working with tomatoes.”18 In informal interviews, older people, in defeated tones, also characterized tomatoes as “a young person’s crop.”19 One elderly cocoa farmer in Ghana explained how he would farm tomatoes, too, “if he had the strength and could.”20 After a couple of years of attempting to farm tomatoes, several Ghanaian farmers in their late forties and fifties described being exhausted and unable to continue. They seemed disappointed by their own physical limitations as they spoke of the need to shift to a less labor-intensive food crop such as cassava. Survey results indicated that most tomato producers in Ghana were between eighteen and twenty-five years old, with only three respondents more than forty years old. This strikingly young group was also earning much more from their production ($838) than any other age groups. Furthermore, the top three maximum values of production between $1,400 and $7,200 were from farmers aged eighteen to twenty-five. In Côte d’Ivoire, by contrast, tomato producers and their revenues were split relatively evenly between the ages of eighteen and fifty-five years, the same demographic pattern as was found in cocoa. Tomato production in Ghana challenged not only prior social hierarchies based on age but also those based on gender and ethnicity. Whereas indigenous males almost exclusively dominated cocoa production in the past, now a significant number of young women and non-Akans from the Northern regions and Sahel have started tomato farms in Ghana.21 Moreover, these 16 17 18 19 20 21

Survey interview (anonymous) by author. Kyere, Côte d’Ivoire. Interview with Nana Amankwah, Barima, Ghana, March 22, 1999. Interview (anonymous) by author. Makwan, Ghana, March 31, 1999. Interviews (anonymous) by author. Barima and Makwan, Ghana. Interview (anonymous) by author. Makwan, Ghana, April 7, 1999. Women accounted for 28 percent of all tomato producers whereas non-Akan accounted for 34 percent in Ghana.

170

Informal Institutions and Citizenship in Rural Africa

new cash crop farmers sometimes matched or outpaced their Akan male counterparts in the value of their tomato production.22 In comparison, Ivoirian cocoa farmers included the young and old, women and non-Akan.23 Whereas the mean age of Ghanaian cocoa farmers was sixty-two years, the mean age of Ivoirian cocoa farmers was much lower, at forty years. Indeed, the largest group of cocoa producers was twenty-six to-forty years old, relatively youthful compared with Ghana where the largest group was those who were more than seventy years old, including the single largest producer. Tellingly, in Côte d’Ivoire, the single largest producer ($12,500) was a forty-five-year-old Akan male, and the second largest producer ($9,025) was a twenty-year-old Akan male. There was also evidence that several young men were starting new farms, and thus cocoa did not at all appear to be on the wane among members of upcoming generations, as was so clearly the case in Ghana. While a few farmers were growing tomatoes in Côte d’Ivoire, this remained on a very small scale in comparison with farms in Ghana, and interestingly, the tomato farmers in Côte d’Ivoire were almost exclusively young non-Akans, and to a lesser extent women, indicating tomatoes’ lack of recognized value as a true cash crop there.24 Only 16 percent of non-Akan tomato farmers earned more than $100 in Côte d’Ivoire compared with more than 96 percent in Ghana; and in Côte d’Ivoire, only 4 percent of female tomato farmers earned more than $100 compared with more than 63 percent in Ghana.

why were people switching to tomatoes? So why were farmers in the Ghanaian region switching to tomatoes as a new cash crop while farmers in the Ivoirian region were choosing to remain with 22

23

24

Ghanaian females in the twenty-six- to forty-year-old age group actually earn more on average with their tomato production than their Akan male peers, and Ghanaian females in the forty-one to fifty-five-year-old age group match their mean value of production. NonAkan farmers in the youngest age group actually outstrip Akan farmers, and they either match or outpace Akans in the age groups of 41–55 and 56–69. It appears that more women and non-Akans have been able to establish relatively large cocoa plantations in Côte d’Ivoire in comparison to Ghana. The mean value of production for forty female cocoa farmers in Côte d’Ivoire was $646 (58 percent of the mean for men) and for ten female cocoa farmers in Ghana was $194 (only 16 percent of the mean for men). The mean value of production for fifty-three non-Akan cocoa farmers in Côte d’Ivoire was $684 (60 percent of the mean for Akans) and for 0 non-Akan farmers in Ghana was $0. The few Ivoirian women who appear to be trying to cultivate tomatoes seemed to view tomato production as injecting short-term cash into the household, whereas cocoa plantations constituted more long-term security despite the threat of bushfire.

Legacies of State Agricultural Policy

171

cocoa? This question is especially perplexing in light of the appreciably higher producer price for cocoa in Ghana in comparison with Côte d’Ivoire. This section highlights the effects of different state interventions over time in three areas: agricultural marketing, labor, and transportation. These state legacies stimulated different cash crop production choices in the late 1990s with implications for the informal institutions of reciprocity. The History of More Exploitative State Agricultural Policies in Ghana than in Côte d’Ivoire First, Ghanaians were switching to new cash crops, in particular annual vegetables instead of permanent tree crops, because of a long history of more exploitative agricultural policies against cocoa by the Ghanaian state.25 In 1947, the Cocoa Marketing Board (CMB) was established in the colonial Gold Coast as a legal monopoly to export Ghanaian cocoa and continues to operate today as the Ghana Cocoa Board (or Cocobod).26 The CMB fixed the local price of cocoa, normally below the world price, and saved the differential in theory to support a price increase when the world price was too low. In reality, when the economy deteriorated, the CMB profits were used to cover CMB costs or transferred directly to the central government coffers. These marketing policies, initiated during the colonial era but continued after independence right up until the time of this study’s fieldwork, implicitly taxed cocoa production at a much higher rate in Ghana than in Côte d’Ivoire. In Ghana, cocoa farmers captured only 15–40 percent of the world market price for cocoa, whereas Ivoirian farmers received at least 66 percent of the world price.27 Another study estimates that the producer price Ghanaian cocoa farmers received was even worse, less than 50 percent of the prices paid to farmers in Côte d’Ivoire and Togo.28 In the short term, these poor relative prices caused Ghanaian farmers to try and circumvent the statecontrolled marketing organization, spurring a substantial degree of crossborder smuggling to the neighboring region of Côte d’Ivoire.29 Over the long 25

26

27 28 29

Stryker (1990: 97). See also Boone (2003); Frimpong-Ansah (1991); Mikell (1989); Bates (1981); Beckman (1976) and Dunn and Robertson (1973). The CMB was reconstituted as the Ghana Cocoa Board during an earlier reform attempt in 1979. Leechor (1994: 156). Commander, Howell and Seini (1989: 108). Dunn and Robertson (1973: 46) highlight how smuggling was costly and so many Ahafo cocoa farmers accepted the lower producer price and became increasingly dependent on the Ghanaian state.

172

Informal Institutions and Citizenship in Rural Africa

term, the combination of such extractive marketing policies with severely weakened agricultural extension services eroded the farmer confidence necessary for making lasting investments in cocoa, where yields do not usually begin until year five and do not start to peak for another decade.30 This explains why Ghana dropped from its position as world cocoa production leader in the 1960s (producing more than 560,000 tons in 1965 or approximately 34 percent of the world market) to its historic low point, (producing only 160,000 tons by 1983). (See Table 6.1.) Whereas the Ghanaian state exploited the cocoa regions, fearful of their potential opposition threat, the Ivoirian state encouraged the expansion of cocoa as a fundamental pillar of its domestic power base with strongly supportive policies.31 A similar type of stabilization fund for cocoa was established in 1955, a few years after the CMB in Ghana. In 1964, a bigger state company, Caisse de Stabilisation et de Soutien des Prix des Productions Agricoles (CAISSTAB), was created to “stabilize and support agricultural producer prices.”32 However, there were two critical differences in Côte d’Ivoire’s approach: CAISSTAB administered a higher producer price, and numerous private agents were allowed to compete in the process of purchasing from cocoa farmers. Although in the Ghanaian region, farmers pointed out the “old” Cocoa Board (or Cocobod) purchasing warehouse, which plainly had previously been the only game in town, there was no single purchasing outlet in the Ivoirian villages. Several of the older cocoa farmers in the Ghanaian region recalled that in the past they were cheated on the already-lower price or delayed in actually receiving payment from the Cocobod purchasing agent. The Ivoirian cocoa farmers also complained about their lack of power vis-a` -vis the small number of buyers, but there was clearly more than one. Thus, the Ivoirian state historically supported the cocoa industry, guaranteeing a higher share of the world producer price, allowing private companies to market cocoa, and facilitating the immigration of agricultural workers from the northern regions and Sahelian countries into Côte d’Ivoire (discussed in the following subsection). As a consequence, Côte d’Ivoire virtually traded places with Ghana as the world

30

31 32

Political instability and the earlier and more severe fiscal collapse of the state severely weakened the accessibility and quality of agricultural extension services in Ghana as compared to Côte d’Ivoire. Weakened extension services meant that farmers had difficulty responding to the spread of diseases such as blackpod, swollen shoot, and capsid, which also contributed to production losses. Arhin (1985: 104). On the political origins of these different policies, see Boone (2003). McIntire and Varangis.

t a b l e 6 . 1 . Annual Cocoa Production in Tons in Ghana and Côte d’Ivoire 1965 Ghana

557,000

Côte d’Ivoire 122,000

1975 1983

1985

1988–1990 avg 1995

159,000 228,000 296,000 (1986–1987) 580,000 580,000 793,000

403,872

2000

2005

2006

389,772

740,458

614,532

1,061,600 1,249,000 1,408,000 1,229,000 1998–1900 avg

Sources: World Bank (Adjustment in Africa; Lessons from Ghana. 1996); LaVerle Berry, ed. Ghana: A Country Study; 1994); and IMF (Côte d’Ivoire Selected Issues and Statistical Appendixes; 2000).

174

Informal Institutions and Citizenship in Rural Africa

leader in cocoa production, increasing production from 122,000 during the period of 1962–1967 to 580,000 tons in 1985. The importance of these state policies on farmer production strategies is confirmed further by examining what happened to cocoa production in the two cases after these state policies were reformed during the era of structural adjustment.33 Starting in 1984, Ghana began to reform the cocoa sector, retrenching 90 percent of Cocobod’s staff, increasing the role of the private sector in crop transport, removing input subsidies, and privatizing state-owned plantations and processing plants. Ghanaian cocoa production began to rebound slowly, increasing to 296,000 tons by 1990.34 In the early 1990s, Cocobod raised the producer price and began to allow, and even encourage, private buyers. The state’s entire role was not eliminated during these neoliberal reforms, however. Importantly, the producer price of cocoa was never fully liberalized.35 The government worked to revitalize extension services by increasing the participation in government spray programs.36 By 1999, during the time of fieldwork, cocoa production had increased to 381,000.37 Teal argues that this increase is “very modest” because in absolute terms, it is still lower than the 1965 peak level of cocoa exports.38 He suggests that to outpace previous production in per capita terms, cocoa exports would have to reach two million tons. The most recent FAO statistics do show a marked increase in Ghanaian cocoa production to nearly 615,000 tons. While Teal, Zeitlin, and Maamah conclude this increase to be “a success for cocoa policy” in Ghana, cocoa production in 2006 was still barely above 1965 levels and only slightly more than half of what Côte d’Ivoire produced in the same year (1,229,000 million tons).39 Côte d’Ivoire’s continued place as the runaway world leader despite the past decade of coups and civil war is striking and further highlights the slower growth occurring in Ghana, which appears primarily driven by the establishment of new farms in pioneer areas outside of the Brong-Ahafo region. Based on this history of these extractive agricultural policies toward cocoa in Ghana, one would predict that Ghanaian farmers’ would resist adopting other permanent crops and be reluctant to make any long-term investments 33 34 35

36

37 38 39

Coleman, Takamasa, and Varangis (1993); and Deaton and Benjamin (1988). Gyimah-Boadi (1995) and Leechor (1994: 156). Cocobod has a Producer Price Review Committee made up of a diverse group of farmer, government, private sector, banking, and university representatives who fix the cocoa producer price and related fees in cocoa purchasing and marketing. These efforts were strongest in the Western region, the cocoa frontier. Teal, Andrew, and Haruna (2006). Teal (2002). Teal (2002: 1330). Teal et.al. (2006).

Legacies of State Agricultural Policy

175

in agriculture. Thus, this legacy of state exploitation explains why despite “the hard sell” and multiple incentives offered by agricultural extension agents, Ghanaians have refused to adopt another permanent crop such as coffee.40 Not one Ghanaian farmer reported growing coffee, whereas 43 percent of Ivoirian farmers polled in the survey cultivated coffee, with a mean annual value of production of $291. Ghanaian farmers also were not making any long-term investments in irrigation despite the fact that watering tomato plants involved enormous inputs of labor. Furthermore, even though water was becoming a relatively scarce commodity in these areas, there was no evidence of local-level initiatives to regulate the access and use of available water sources over the long-term. The Lower Price of Labor in Ghana Relative to Côte d’Ivoire While past antagonistic policies of the state elucidate why Ghanaians may have turned away from cocoa and other tree crops, it does not explain why they turned toward tomatoes. Another key part of the story is that the relative price of labor was much cheaper in Ghana than in Côte d’Ivoire. Unlike cocoa, tomatoes do not require large plots of land, but they do call for large inputs of labor. Tomato plants must be watered individually every day, a backbreaking task that usually begins at 5 A.M.41 Also, because tomatoes are not permanent plants, they must be replanted after each harvest, three times per year, or every three to four months. Cocoa trees, on the other hand, can live for between thirty and fifty years. Cocoa also requires minimal labor throughout the year to maintain the trees, with only two predictable peaks of labor during the major and minor harvests. Until recently, the Ivoirian government made it easy for foreigners from Mali or Burkina Faso to migrate and work in the cocoa regions. Historically, Ivoirians have depended on migrant labor to work their plantations much more exclusively than have farmers in Ghana, where the government did little to facilitate the influx of migrant labor. A variety of respondents frequently joked, “the Agnis (the local Akan in the fieldsite region of Côte d’Ivoire) are too lazy to work their own farms.” Now, however, because fires 40

41

Agricultural extension agents in Ghana reported serious resistance to coffee cultivation despite offers of free fire-resistant seedlings, technical assistance, and favorable producer prices. In contrast, agricultural extension agents in Côte d’Ivoire noted an increase in coffee production in recent years. Watering must be done early in the fieldsite villages for two reasons. One, the water source close to the field will be exhausted and dried up if you are the last one to arrive. Second, the plants must be soaked with water before the sun begins to scorch them.

176

Informal Institutions and Citizenship in Rural Africa

and other ecological changes have reduced cocoa yields over time, it has become increasingly difficult to attract and pay for foreign labor to work on Ivoirian farms in the fieldsite region. On the supply side, Burkinabe and other workers preferred to go to the newer pioneering cocoa regions in the West, where the same percentage of a larger yield earned more money. On the demand side, Ivoirian cocoa farmers could no longer afford to hire labor. As one Ivoirian cocoa farmer lamented: “Since the plantations are starting to spoil, we lack hired workers. It’s because of the market; the price has fallen so we can’t afford to hire.”42 The Ivoirian government aggravated this situation further by shifting tactics in the 1990s and making it much more difficult for foreign workers to immigrate and remain legally in the country. The price of identity cards skyrocketed, and anyone who looked “foreign” was routinely harassed for their papers.43 The Bédié government was also pursuing a land registration policy intending to give titles for land to the “original” indigenous owners, essentially appropriating land that had been sold or given to foreigners in the past. For the previous reasons, foreign labor was in short supply, and Ivoirians had a strong incentive to stick with the less labor-intensive cocoa production. The same dilemma with regard to the scarcity of migrant labor troubled Ghanaians, but they appeared to have more options.44 First, Ghanaians never relied as heavily on migrant labor; farmers and their family members always worked their own farms, even in the good times. While approximately half of all respondents replied affirmatively that they purchased some type of agricultural labor during the past year, the mean expenditure was much higher in Côte d’Ivoire ($67) than in Ghana ($6). This most likely reflects the higher proportion of permanent workers hired in Côte d’Ivoire, as well as the higher costs of labor for permanent and daily workers there.45 Second, the Ghanaian economy has experienced more downturns than the Ivoirian one, and so Ghanaians have been forced to be more flexible and willing to work in any sector. For example, many more Ghanaians reported working “by day” if necessary to earn additional income whereas this practice was rarely, if ever, acknowledged in Côte d’Ivoire. Third, the survey 42 43

44 45

Interview (anonymous) by author. Kyere, Côte d’Ivoire, July 14, 1999. This situation deteriorated gravely since the time of fieldwork due to the ethno-regional civil war which spurred anti-immigration violence and the subsequent flight of many non-Akans to their regions and countries of origin. See Austin (2005) for an in-depth history of labor changes in cocoa production in Asante. The daily wage in this region of Cote d’Ivoire ($1.20–3.33) was almost double that in the Ghanaian region ($0.71–1.66).

Legacies of State Agricultural Policy

177

research documented that Ghanaian respondents’ children lived closer to their parents’ homes than do the Ivoirians’ children. Thus, due to the greater availability of cheaper labor, Ghanaians were better able to switch to tomato farming, with the high daily labor inputs it demanded. Recent Improvement in Transportation Facilitates Tomato Marketing in the Contemporary Period in Ghana While the lower cost of labor explains why Ghanaians were more likely to adopt tomatoes than were Ivoirians, recent improvements in transportation in Ghana explain why Ghanaians have adopted tomatoes within the past decade, rather than earlier. Transportation improvements associated with the adoption of structural adjustment facilitated the transport of more perishable crops such as tomatoes to domestic markets. Throughout the 1970s and early 1980s, transportation in Ghana was extremely difficult. Roads were in serious disrepair, as were the few vehicles plying the routes. Cocoa farmers in general were minimally affected by these difficulties because cocoa can be stored for long periods of time and was usually sold at nearby purchasing centers.46 Furthermore, the Ghanaian state historically took responsibility for the transport of the cocoa from the state-operated purchasing centers. Beginning in the 1980s under Ghana’s SAP, the government began improving the road infrastructure throughout the country. The Ghanaian state shifted responsibility for cocoa feeder roads to the Department of Feeder Roads and embarked on a program to construct 3,000 km of new feeder roads in addition to upgrading existing roads. With an infusion of foreign exchange through new World Bank loans, the government was also able to increase gasoline imports, which was chronically in short supply. The lack of supply was only part of the reason why gasoline prices were historically so high in Ghana. The Ghanaian state also exercised a monopoly on distribution, facilitating monopoly rents and higher fuel costs. Ghana’s SAP liberalized distribution and thus brought prices down. Imported car parts also became more readily available, resulting in a marked increase in the number of vehicles circulating between markets. All these transportation improvements, combined with the availability of lower-cost labor and the

46

While Arhin (1985: 80) describes the impassability of many roads in Ghana in the 1970s, particularly in the rainy season, he also writes that over 80 percent of cocoa farmers interviewed stated that their farms were less than five miles to a buying center.

178

Informal Institutions and Citizenship in Rural Africa

history of exploitative policies against cocoa, encouraged farmers to drop cocoa in favor of tomato production in this region of Ghana while they have maintained or increased cocoa production in the Ivoirian region.

implications of dramatic cash crop changes for reciprocity Redefining Family and Community Reciprocity The dramatic switch away from cocoa to tomato production in Ghana has stimulated a redefinition of family and informal reciprocity at the local level. First, because so few Ghanaian farmers maintained, replanted or expanded their cocoa plantations in the fieldsite region after the bushfires in 1983, cocoa farms no longer promised as viable a future. The high levels of poverty in this region were unknown during the heyday of the cocoa boom. As one Ghanaian man noted: . . . Some people live the whole day and buy kenkey only. We have a big problem. All of our farms have been destroyed like the cocoa. At the time that we had our cocoa, such things were non-existent.47

Cocoa was no longer viewed as the way to become prosperous; now, tomatoes were. But this new type of cash crop production had dramatically different implications for social relations than did cocoa. As seen previously, a new group of younger tomato farmers has been able to accumulate wealth relatively rapidly. This substantial and rapid increase in the economic power of youth was already challenging traditional social, economic, and political hierarchies in which older Akan men with long-term investments in large cocoa plantations historically were the wealthy elders of the village.48 It is not only the new dynamics of wealth accumulation that were overturning previous power structures. Within this new production system, tomato farmers faced different sets of input requirements than cocoa farmers. Tomato farmers needed labor more than they needed land, the inverse of cocoa cultivation needs. Notably, in both of these regions of Ghana and Côte d’Ivoire, land pressure was increasing. Land for cocoa farming was not easy to locate. Despite similar pressures, however, farmers in the two regions made different choices. Figure 6.1 shows how these production choices spurred new input needs, which altered the incentives to maintain social 47 48

Men’s focus group (anonymous). Makwan, Ghana, April 1999. Okali (1983).

Legacies of State Agricultural Policy

179

Different Input Needs for Cash Crop Production Tomatoes in B/A Ghana

Cocoa in Abengourou Côte d’Ivoire

Need for Land

Low

High

Need for Labor

High

Low

Different Incentives for Social Ties

Maintain Ties to Elders in Nuclear and Extended Family Expand Ties to Many, Strong Workers

B/A Ghana

Abengourou Côte d’Ivoire

Low

High

High

Low

Structure of Informal Institutions of Reciprocity B/A Ghana Density of Ties Direction of Ties

Abengourou Côte d’Ivoire

Diversified

Concentrated

Horizontal Among Youth

Vertical Between Youth And Elders

fi g u r e 6 . 1 . Linking Economic Production to Informal Institutions of Reciprocity

ties with various social groups, thereby stimulating changes in the structure of the informal institutions of reciprocity.49 These new needs for little land but substantial amounts of labor created incentives to maintain highly diversified reciprocity networks in Ghana, in contrast with the more-concentrated networks found in Côte d’Ivoire. Furthermore, the relative costs and benefits of giving and receiving social support to an older parent or uncle versus a younger friend have changed in Ghana, but not in Côte d’Ivoire. The variation in the patterns of social relations between cocoa and tomato farmers in Ghana and Côte d’Ivoire is demonstrated by the survey data shown in Table 6.2. First, in both Ghana and Côte d’Ivoire, tomato farmers gave help to people with whom they shared a more diverse set of social ties, while cocoa farmers concentrated their help on nuclear family members.

49

This economistic way of conceptualizing reciprocity networks is based on local-level, indigenous perspectives. While notions of economic exchange were important components of reciprocity in both Ghana and Côte d’Ivoire, this does not mean that other less concrete and less tangible costs and benefits do not enter into an individual’s analysis, or that this analysis necessarily takes place consciously.

180

Informal Institutions and Citizenship in Rural Africa

t a b l e 6 . 2 . Comparison of Help Given from Cocoa versus Tomato Farmers B/A Ghana Primary Source of Mean value of help given ABENGOUROU Côte d’Ivoire Agricultural Income in U.S.$ Mean value of help given in U.S.$ Cocoa

Tomatoes

n=21 help to nuclear = 4.39 help to extended= 7.43 help to friends = 1.41 n=41 help to nuclear = 1.68 help to extended = 16.21 help to friends = 3.84

n=95 help to nuclear = 117.67 help to extended= 40.22 help to friends = 3.90 n=5 help to nuclear = 3.61 help to extended= 5.32 help to friends = 4.51

The survey data also show that giving to friends was not as important for cocoa farmers as for tomato farmers, in either Ghana or Côte d’Ivoire. The study found that tomato farmers in Ghana continued to give the most significant amount of social support to their extended family, but what is obscured is the ages and/or specific social relation to whom that help was offered. For example, the qualitative data suggest that the young tomato farmers in Ghana most likely gave more support to their younger siblings and cousins than their elderly uncles. A Diminished Need for Land in Ghana and Implications for Social Relations Because these younger Ghanaian tomato farmers did not need land as much as labor, inheriting the cocoa plantation of one’s father and/or uncle had become less material than maintaining a wide network of family, friends, and others. Since only a small plot of land was required for tomatoes,50 and for only a short period of time, there was no need to invest heavily in one or two particular relationships – a parent or uncle, for example. A farmer could borrow or rent the land from anyone rather than inheriting or obtaining the consent of elders for the long-term customary use of family land. Thus, because family land had become relatively less important economically, the incentives for maintaining particular social ties had also become less compelling. Then in a positive feedback loop, the fact that fewer young farmers were using family land in Ghana further weakened the extended family 50

Data collected at the district level in Ghana confirmed that the average cocoa farm was 3.6 acres, whereas the average farm size for vegetable crops was only 0.5 acres.

Legacies of State Agricultural Policy

181

component of informal reciprocity. When a person did use family land to establish a cocoa plantation, their incorporation into social networks became tighter and more obligatory. Family land was usually not given or sold, but instead entrusted to a family member to take care of for relatively long periods of time.51 Because cocoa production usually occurred on family land, the revenues were perceived as less autonomous and more closely tied to family welfare. Thus, a wealthy relative who refused to help a needy family member could be questioned and held accountable if his wealth was gained through family property, as was often the case with cocoa. “He can even be sacked for another person to be put in charge of the cocoa farm.”52 On the other hand, if his wealth was acquired through his own means, particularly on non-family land, as was usually the case with tomatoes, “nothing can be done to him.”53 In addition to land, tomato farming required a certain amount of inputs for pesticides, spraying machines, and other tools. Again, this financial assistance need not be provided by a family member, but could be given by anyone as a loan or as an investment in exchange for a share of the crop returns. In fact, it appeared that this sort of assistance was frequently given in conjunction with the needed land as part of a sharecropping arrangement. The survey research confirmed these predicted differences in land tenure patterns in tomato versus cocoa production in Ghana and Côte d’Ivoire. While cocoa farmers in Ghana exclusively used their own property (63 percent) or family land (37 percent), the majority of tomato farmers either made sharecropping arrangements (38 percent) or rented land (15 percent). While a substantial proportion of tomato farmers did use family land (38 percent), only a small minority used their own property (9 percent). In comparison, in Côte d’Ivoire, the overwhelming majority of farmers grew tomatoes on family land (48 percent) or on their own property (38 percent), and only 15 percent rented or made sharecropping arrangements in Côte d’Ivoire. These tenure patterns also confirmed the earlier finding that tomatoes were being cultivated as a cash crop in this region of Ghana, but not in Côte d’Ivoire. A comparative analysis of the land tenure patterns disaggregated by age group shows how the use of family land had declined in importance for younger farmers in Ghana, who were now primarily involved in tomato production. Younger farmers in Côte d’Ivoire, on the other hand, remained 51

52 53

The commonly held understanding that individuals have usufruct rights but not ownership rights was demonstrated by the bitter resentment of extended family members when a relative sold family property. Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Men’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999.

182

Informal Institutions and Citizenship in Rural Africa

actively engaged in cocoa production. In Ghana, at least as many, if not more, younger tomato farmers rented land or sharecropped, and older farmers did not grow tomatoes at all. In Ghana, older cocoa farmers tended to own their own plots whereas the use of family land was most important to the few, younger cocoa farmers. This general trend holds true in Côte d’Ivoire as well, although sharecropping was more important for younger cocoa farmers. This distinction reflects the greater role of migrant labor in cocoa production in Côte d’Ivoire.54 The changing relative importance of land as an input in tomato production was one reason why informal social relations in Ghana were more diversified than in Côte d’Ivoire. It also begins to reveal why social ties were more horizontally-oriented (linking similar age cohorts) in Ghana and more vertically-oriented (connecting youth to elders) in Côte d’Ivoire. If the need for land under the new system of tomato production was diminished and could be satisfied through a multiplicity of social ties, the former power of nuclear and extended family member elders over youth was eroded. The Role of Inheritance Laws While inheritance laws have also changed over time, the reorientation of reciprocity was primarily due to the shift in local-level production strategies, not to changes in official legal frameworks affecting inheritance. In both Ghana and Côte d’Ivoire, legislation has been passed to encourage a more patrilineal inheritance system, which potentially conflicts with the matrilineal system of the indigenous Akan in both fieldsite regions.55 In Côte d’Ivoire, legislation that designated the spouse and children as the sole inheritors was enacted much earlier, in 1964. Later, in 1985, the Ghanaian government introduced the Intestate Succession Law, which promoted the security of widows and children, but also specified that one-third of the estate be given to the extended family.56 54

55

56

Migrants represented 52 percent of those farmers who gained their primary income from cocoa production in Côte d’Ivoire. The dominant land tenure pattern for those non-Akan cocoa farmers was sharecropping (56 percent sharecrop; 31 percent own; 10 percent use family land; and 3 percent use at no charge). Historically, in Akan areas, when a man died, the man’s relatives on his mother’s side claimed rights to his property, often leaving his wife and children homeless and destitute. Interestingly, this policy development continues the Ghanaian legacy of maintaining a role for the extended family. Several respondents described how the Ghanaian law was a compromise and promoted peace within the family. As one Ghanaian man described: “It is now that the government has decreed that the children can inherit a certain percentage that has brought a little unity. Before this law, a married woman is sacked from the husband’s house and the keys collected from her when the husband dies.” Men’s focus group (anonymous), Makwan, Ghana, April 1999.

Legacies of State Agricultural Policy

183

One might predict, therefore, that these laws would encourage more nucleated social relations, particularly in Côte d’Ivoire where allocation to the extended family was not specified. While both laws did provide some support for the rights of the nuclear family, neither was consistently enforced at the local level. When asked if children inherit from their father, one Ivoirian woman replied skeptically, “Here, one talks about it like that, but really . . . those who inherit are one here, one there.”57 Families were seeking the application of the law and shaping their own customs on a case-by-case basis. One Ivoirian man described why the law was not always implemented in this primarily matrilineal region. In the following account, he explained that the 1964 law contradicted the customary inheritance rules among the Agni and highlighted the traditional legitimacy of the Akan system. He emphasized that the children who would inherit potentially from their fathers according to the 1964 law were “scared” to follow the formal rules because they did not want to die. His implication was that resentful extended family members have practiced witchcraft and killed those children who did inherit according to the 1964 law: It’s been since 1964 that this law was created. Let’s say that it’s the son who should inherit from his father. But, with us (the Agnis) here, the children are scared to inherit because the [extended] family members aren’t happy . . . [He tells a long story tracing the historical roots of the matrilineal system.] Voila, why it’s the nephew who inherits with us, the Akan, and that continues here with us. There are a lot of children who inherited from their father, but they are all dead. Therefore, they are fearful.58

He then went on to say how individuals dealt with the problem in an ad hoc manner trying to appease the demands of both the nuclear and extended family member: But, a survivor of the father, he could divide one part for the children and one part for the family. There wouldn’t be any problem following.59

Many other respondents similarly described how estates were divided unofficially between the nephews and the children to avoid problems. Others described the sometimes violent conflict that ensued when children attempted to wholly exclude the nephews from any inheritance:60

57 58 59 60

Women’s focus group (anonymous). Opanin, Côte d’Ivoire, September 1999. Men’s focus group (anonymous). Opanin, Côte d’Ivoire, September 1999. Ibid. The two-story house where our research team was accommodated in Kyere, Côte d’Ivoire, was left vacant because a son had apparently fallen through the first-floor ceiling to his death while sleeping. This was interpreted as retribution for his attempt to claim the house to the exclusion of the nephews and other extended family. The house was then labeled as haunted so no one was willing to stay there. What better place to house a visiting researcher!

184

Informal Institutions and Citizenship in Rural Africa

The law was voted but only in certain places is it followed to the letter. Here, that’s not the case. From the other side, it’s done, but here, they haven’t yet started. If the child inherits, the relatives will kill him, therefore, if he doesn’t inherit, he lives . . . 61

Since there was no commonly held expectation of a consistent enforcement of the statutory law to benefit the nuclear family, there was little evidence that these official legislative changes were the primary causal factor shaping variations in the relations of reciprocity. Two puzzles remain that were not adequately explained by the legal changes, but were better explained by production changes. First, why would the extended family system appear to be relatively vibrant in Côte d’Ivoire even forty years after the passage of a more nuclear-family-oriented law? The data reveal that the continued viability of cocoa production at the local level in Côte d’Ivoire strengthened the incentives of young Ivoirian farmers to maintain ties with their extended family elders, whereas in Ghana the system of tomato production eroded these incentives. Second, if the legislation potentially strengthened the role of the nuclear family in both cases, why were friends so important in Ghana but not in Côte d’Ivoire? The answer is that the need for less land and more labor to grow tomatoes in Ghana fostered a more diversified social support network so the role of friends was expanded.

a greater need for younger labor and the implications for reciprocity The shift to tomato farming in Ghana not only reduced the demand for larger plots of land but also increased the need for labor. The demand for labor was not generic, however. Because the work on tomato farms was back-breaking, those who could be of genuine help were younger, more active, and physically fit. One Ghanaian man in discussing the rise of tomato farming in the village asserted that “Strong people do it . . . very strong. It’s difficult work.”62 Thus, it is the younger and stronger men and women who helped each other out with watering and weeding. Younger men and women might lend a hand during the harvest or give money to help another young person plant. Respondents of all ages confirmed that the young tomato farmers in Ghana were now becoming an increasingly common source of loans for local people.63 This overturned traditional hierarchies in which elderly cocoa

61 62 63

Women’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Interview (anonymous) by author. Barima, Ghana, February 21, 1999. Survey interview (anonymous) by author. Barima, Ghana. Interview (anonymous) by author. Makwan, Ghana, April 9, 1999.

Legacies of State Agricultural Policy

185

farmers historically provided loans. One eighty-five-year-old cocoa farmer said despondently, “In the past, I could help anyone, even to the extent of borrowing on their behalf . . . Now, nobody would even come . . . I can’t help anyone now.”64 Increasingly, young tomato farmers were also the recipients of loans given by local people because they were perceived as capable of repaying the loans after their tomatoes were sold.65 We can hypothesize that in this region of Ghana reciprocity networks were based increasingly on horizontal ties among family members and friends of the same age cohort. Unfortunately, because the survey questionnaire did not ask directly the age of the donors and recipients of social support described by respondents, it is impossible to test definitively with this dataset whether reciprocity networks were denser among members of the same age cohort. Other quantitative indicators, however, suggest that relations were more horizontal in the Ghanaian region than in the Ivoirian one. Chapter 2 demonstrates that siblings were a particularly important social category for help in the Ghanaian villages, whereas giving to parents, grandparents, aunts, and uncles was not as substantial there as in the Ivoirian villages. Ghanaian respondents also allocated a greater percentage of their total help to friends and others than did Ivoirian respondents.66 Particularly when one takes account of the cultural emphasis placed on age hierarchies, it is more likely that a respondent’s friends would belong to approximately the same age group than not. This data reinforce the earlier findings from the regression analysis in Chapter 2. This analysis showed that in Ghana, the older the respondent was, the less help was given to anyone. In contrast, in Côte d’Ivoire, older respondents gave more help, in particular to the extended family.

intergenerational reciprocity under challenge Given the difficulties in proving indisputably with the survey data that reciprocity was denser among those of the same age cohort, I turn now to compare the predicted effects of different patterns of relations on intergenerational reciprocity. If informal reciprocity did diverge in the ways I have suggested, one would predict that reciprocity between the young and the old would be under more serious strain in Ghana than in Côte d’Ivoire. The 64 65 66

Survey interview (anonymous) by author. Makwan, Ghana. Women’s focus group (anonymous). Barima, Ghana, March 1999. Udry and Conley (2004) also note a higher number of connections between siblings and longterm friends than between husband and wife in a survey of informal networks in another region of Ghana.

186

Informal Institutions and Citizenship in Rural Africa

following four points provide qualitative evidence to support the prediction that treatment of the elderly has indeed worsened more gravely and rapidly in Ghana than in Côte d’Ivoire.67 The Characterization of Intergenerational Reciprocity by Youth First, let’s compare how the elderly and youth in these regions of Ghana and Côte d’Ivoire characterized the nature of intergenerational reciprocity. Perhaps not surprisingly, the elderly in both regions complained about a lack of respect and the inadequacy of care they received.68 The discontented comments from the elderly in both regions reflected their long-term expectations having been frustrated by the current economic crisis. They likely took care of their own parents during the boom times, and thus their expectations of their own futures developed accordingly. In contrast with the shared viewpoints of the elderly, the youth in these regions of Ghana and Côte d’Ivoire expressed quite different opinions from each other. Their comments regarding how the elderly were treated and why seemed particularly candid and insightful.69 While youth opinions clearly differed between the two regions, the characteristics and culture of the youth did not vary in any significant way. In Ghana, a young Akan man explained perceptively how the new system of tomato farming had inverted traditional hierarchies, resulting in a lack of respect for the elderly among self-sufficient youth: The town is even sinking . . . The youth do not respect the elders . . . [The interviewer queries: “Why don’t they respect?”] The youth say, “I’m caring for my own so they can’t tell me what to do.” It’s because the aged can’t afford for the young. The only job we do is tomatoes. The aged are sitting in the house and can’t go to farm. So the

67

68

69

Other ongoing processes (i.e., urbanization, demographic changes in the structure of the population, etc.) also influence relations between the youth and the elderly, but these appear to be relatively similar in both regions. This chapter argues that the dramatic changes in cash crop production produced a much more marked strain in intergenerational reciprocity in Ghana than in Côte d’Ivoire. For other scholarship on the aged in Africa, see Apt (1996) and Akuamoah-Boateng (2004), An elderly cocoa farmer in Ghana explained the lack of reciprocity: “People do not respect the elderly now, so they don’t care for them well.” Similarly, in Côte d’Ivoire, a 65-year-old Akan woman explained, “Now, children no longer respect the elderly. They don’t do anything for them. Even to go to the farm to look for food, some refuse.” Survey interviews (anonymous) by author. Barima, Ghana, and Opanin, Côte d’Ivoire. Perhaps talking to a young American outsider facilitated candor from the youth respondents. On many occasions, respondents described derogatorily how Westerners simply put their elderly in nursing homes. Since I was not presumed to understand the “traditional” norms of helping the elderly in Africa, I might be less apt to judge their statements about whether and how things had changed.

Legacies of State Agricultural Policy

187

young has money and the old do not. They can’t respect the old man because he has no money . . . This is a sickness to all Ghanaians . . . There has been a problem with respecting the elders for a long time, even since 1900, say, but now it is growing bigger and bigger. Now, maybe it’s 80% who don’t respect . . . The older ones have done something for the younger ones. They bathed us and did everything for us. We should be giving them things. It’s not a good thing that the young don’t respect them. This is how God said we should do (respect our elders).70

The comments of this young man and others suggested that this was not a totally new problem, yet the problem had been significantly aggravated by recent economic changes (in particular, those in the local system of cash crop production). The preceding comments also revealed that the changes in Ghana were not trouble-free for anyone: while the elderly were despairing or resentful, the youth often expressed guilt. These emotional expressions of resentment and guilt exposed how the informal institutions of reciprocity were under pressure in the Ghanaian region. At other times, Ghanaian youth seemed more matter-of-fact about the recent breakdown of long-term reciprocity. One young female trader explained simply: “The elders didn’t help us when they were in a position to do so, so we don’t mind them . . . ”71 Another Ghanaian man in his middle forties explained, “If the elderly were able to cater well and send their children to higher education, the children will take care of the parents well. If not . . . . It’s a give and take.”72 Even when parents did support their children well enough to complete secondary school, the children often found it difficult to get started in life. For instance, many young people reported struggling to even feed and sustain themselves in Kumasi or Accra during an apprenticeship. Opinion was relatively split as to whether the children’s lack of education resulted in a diminished ability to take care of the elderly or a lesser desire to do so even if they were able. In contrast, in Côte d’Ivoire, because cocoa production still promised the possibility of a prosperous future, young people had an incentive to maintain their relationships with their elders, and, in particular, to stay well-connected to their parents and uncles from whom they could inherit sizeable cocoa plantations. As one young Akan man boasted, “ . . . when I inherited from my uncle, ask anyone here, what didn’t I have as far as plantations go?”73

70

71

72 73

Interview with active leader of church and youth (anonymous), Barima, Ghana, February 24, 1999. Survey interview with 27-year-old married Akan female trader (anonymous) by author. Barima, Ghana. Interview (anonymous) by author. Makwan, Ghana, March 31, 1999. Men’s focus group (anonymous). Opanin, Côte d’Ivoire, September 1999.

188

Informal Institutions and Citizenship in Rural Africa

A young Akan woman in Côte d’Ivoire summarized this idea of the continued reciprocity between youth and elders: If the elderly person has well-off children, then these children will take good care of him and send him money at the end of every month. This happens only if you worked hard in your youth.74

Using a common but vivid expression, another sixty-year-old man exclaimed, “If you don’t have a well-off child, you are ‘foutou.’” (“Foutou” is a staple dish in Côte d’Ivoire that involves repeatedly pounding a long, heavy pestle into a wooden mortar to mash cassava or yam.) The idea here was not only that children who were well-supported felt morally obligated to reciprocate but also that they would most likely be better able to reciprocate due to the likelihood of higher education and access to regular salaries. This long-term reciprocity seemed to be considered “normal” to both the young and the elderly in Côte d’Ivoire.75 One elderly Akan woman used a proverb to explain the ongoing cycle of help: “I raised my children from when their teeth first came in. They also, can help me until my teeth fall out.”76 Because of the continuing relevance of the matrilineal system, discussed above with regard to inheritance, parents were not the only significant figures to the young – maternal uncles were important, too. Many more respondents in Côte d’Ivoire than in Ghana mentioned the uncle as important sources and outlets for help, often citing them before or even without reference to their fathers. Many respondents described fulfilling their duties as an uncle in terms of supporting or fostering their nephews and nieces. For example, one Ivoirian man said: It’s me that always takes care of the children of my sisters. When I inherited, I took care of the children like I should do. I did it without a second thought. We get along. But, if I weren’t taking care of them, they wouldn’t get along with me.77

Ironically, despite long-established legislation in Côte d’Ivoire that was more unambiguously supportive of a father’s children, uncles, and nephews remained more important there than in Ghana. This enduring relevance of uncles not only concentrated network ties within extended families, as discussed earlier, but it also strengthened intergenerational ties between youth and elders.

74

75 76 77

Survey interview with a 23-year-old married Akan woman who had one child (anonymous) by author. Opanin, Côte d’Ivoire. Ibid. Ibid. Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999.

Legacies of State Agricultural Policy

189

Caretaking of the Elderly in Practice: A Comparison Having examined how Ghanaians and Ivoirians self-reported relations between youth and elders, now I compare how the two groups actually behaved vis-a` -vis the elderly in their own family. First, fewer people in Ghana (27 percent) described taking care of an elderly person in comparison to Côte d’Ivoire (52 percent). One elderly mother of seven children in Ghana described how life was “always” hard because “No one is taking care of me. I have to take care of myself so I’m always in hardship.”78 Another indicator of the lack of care given by many children in Ghana was that many village parents, when surveyed, were totally unaware of their young adult children’s location or activities after they left school to seek work. Many respondents in Ghana were working under the assumption that the elderly would be self-reliant, but noted that this self-reliance had been undermined with the marginalization of the elders’ cocoa farms. One thirty-threeyear-old Akan man in Ghana explained, “Formerly, the elderly didn’t need anyone to care for them because they had their own cocoa farms. But with the destruction of all of these cocoa farms, they, just like the young ones, had to do their own petty farming to care for themselves.”79 One eighty-year-old widow who no longer had a cocoa farm complained, “These days . . . I have no means of getting enough money and things to live a good life.”80 Thus, the decline of cocoa production in this region of Ghana had in a sense leveled the playing field between the young and the old, but the old were less prepared to play the new game in town: tomato farming. Several respondents of all ages confirmed that if the elderly person could not sell his properties “to cater for himself,” he would be “forced to farm to feed himself,” “to take on some odd jobs to survive – like fetching firewood to sell or preparing brooms to sell,” or “to find another means of survival through begging or loans.”81 In comparison, respondents in Côte d’Ivoire did not emphasize the selfreliance of the elderly as Ghanaians frequently did. Children still articulated a responsibility to take care of their mothers and fathers, in collaboration with their siblings. Respondents frequently opined that children should care for their “own” parents and that the village would not take up this responsibility:82

78 79 80 81 82

Survey interview (anonymous) by author. Barima, Ghana. Ibid. Ibid. Survey interviews (anonymous) by author. Makwan, Ghana and Barima, Ghana. Men’s and women’s focus groups (anonymous). Tape recordings. Kyere, Côte d’Ivoire, August 1999.

190

Informal Institutions and Citizenship in Rural Africa

The old man who put you in the world, you can’t drop him and it’s not the neighbors either who are going to leave their elderly to take care of yours while you are there. It’s you or your brothers who can take care of the old man in your house.83

The continued concern for the elderly in Côte d’Ivoire was confirmed by focus group discussions in which participants described how one “must help” an elderly person. “If he needs help, it’s necessary to give him care, give him something to eat. If he can’t get up, one can even bathe him and wash him.”84 Although there were exceptions, in general, Ivoirians responded by describing how one cares for the elderly, not whether one should. Changes in Meal Preparation for the Elderly The preparation of meals for the elderly in these regions of Ghana and Côte d’Ivoire also differed substantially, a dissimilarity directly connected to the new system of tomato production found in Ghana but not in Côte d’Ivoire. In this region of Ghana, most of the elderly and children bought at least one of their meals from the many prepared food sellers that lined the main street and were spread throughout the village. Since everyone was needed on the tomato farms very early to water, no one stayed behind to cook for the family compound as was done in the past. Even the structure of the school day in the Ghanaian villages reflected the lack of home cooking; there were only two quick breaks for students to run and grab a snack at a nearby food seller. One elderly man described how household eating patterns had changed in Ghana and the negative impact this had on the elderly: In the past, the children of the elderly are there, and the nephews are around so they don’t allow the elderly to work. They give them good and everything. For example, my father was in Ohia and since they didn’t buy their food, if the old man wanted to eat 10 times a day, it was no problem. Foodstuffs were in abundance, bushmeat was all around, and livestock; only had to buy salt. Whatever the old man wanted, they could provide easily. Even oil, like palm oil, they made at home, so they always had plenty. Not this way today. The elderly are not looked after as before. Now, they go and take the bowl and buy food. The old man has to send for food. If you are elderly and you were irresponsible and didn’t take care of your children, then you will go hungry. But if you were responsible, they will buy for you . . . Now the women are also working so they buy something and go to work . . . We buy a lot more of our food now. We’ve always had farms, but now they’re doing it for the money. They’re cultivating tomatoes and you have to work hard to get much money, to get good yields.85 83 84 85

Men’s focus group (anonymous). Kyere, Côte d’Ivoire, August 1999. Women’s focus group (anonymous). Kyere, Côte d’Ivoire, August 1999. Interview with 74-year-old man (anonymous) by author. Barima, Ghana, March 4, 1999.

Legacies of State Agricultural Policy

191

Another Ghanaian man with a tomato farm said that he and his wife needed to go out early to water the tomatoes so their children were sent with 100 or 200 cedis each to buy kenkey or something else to eat. He rationalized that if his wife were to stay and cook for the whole family, she would be delayed until 10 A.M. or so. “In the olden days, we may have sent the kids to the extended family, but now, we can’t. They may have their own problems and burdens and not see them as deserving. They may see it as ‘stealing’ even!’”86 By comparison, in the Ivoirian region, fewer elderly and children purchased their morning meals. This was evidenced by the smaller number of prepared food sellers there. While a plethora of food sellers lined the main thoroughfare and dotted the courtyards all around the Ghanaian villages, in each of the Ivoirian villages, there were perhaps three “beignet” and other sellers at the village market, and one or at most two “chop” houses. The differences were not only quantitative in terms of the number of sellers – the Ghanaian sellers also tended to prepare more substantial “meals” (i.e., staple dishes such as kenkey, foufou, and banku with accompanying soups) rather than “snacks” (i.e., fried dough, plantain, or yam). Even the school day in Côte d’Ivoire was set up with a two-and-a-half-hour break at midday to allow students and teachers to return home and eat lunch at home. Those few who did buy food in Côte d’Ivoire appeared to do so because their parents or caretakers had to leave early to walk to more distant cocoa farms, not to water tomato plants. Changes in Accommodations for the Elderly Changes in the architectural design of homes suggested a more marked decline in the accommodation and daily supervision of the elderly in Ghana than in Côte d’Ivoire. In Ghana, a greater number of smaller rectangular homes without the more “traditional” courtyard were being constructed on the outskirts of the village center, primarily by younger Akan men with extended family in town. Interestingly, the changes in architecture toward smaller nuclear family houses were both an indicator of current changes in family relations and a catalyst for further change along the same lines. First, smaller houses did not facilitate cooking together, which hindered the pooling of resources and the monitoring of food intake by elderly family members. One elderly man described how the architectural changes

86

Interview with man in his late 50s (anonymous) by author. Barima, Ghana, February 21, 1999.

192

Informal Institutions and Citizenship in Rural Africa

intertwined with the preparation of food and care for young and old family members: Formerly, there were more people than this in one compound. There might be himself, children, grandchildren, and nephews. He knew so he built a family house in 1950. We used to have BIG pots that they cooked all the food for everyone. They may have had two mortars for foufou and all cooked together. Now each builds his own house for his smaller family. Now [we’re] separating into separate houses and so food is not done together. Everyone is separating . . . 87

Second, the more restricted allotment of space in the smaller houses did not as easily allow for extended overnight visits, as was usually the custom during ceremonies such as funerals. Lastly, the separate and usually lesscentral locations of the smaller homes impeded the supervision and attentive care of elderly family members.88 The social implications of these changes in architecture were not lost on Ghanaians. The construction of such a house by an individual was viewed both positively and negatively as an assertion of economic power and independence. Several people pointed proudly to half-completed walls, noting that when able to buy more materials, they would continue building their small house for the security of their wife and children. In contrast, older family members would note desolately that one of their younger relatives had abandoned the family compound to build their own smaller house. In Côte d’Ivoire, by contrast, there were fewer smaller houses, particularly among the indigenous Akan.89 The styles of large family houses in this region of Côte d’Ivoire were more varied and included three basic types: (1) the older style of large rectangular house with many rooms opening onto an open-air central courtyard (the type found more exclusively in Ghana); (2) a newer style of “villa” with several rooms opening onto an enclosed living room area; and (3) the most recent “motel-style” house with two or three long blocks of rooms, usually arranged in an “L” or horseshoe shape, looking out onto a larger open area. Based on respondent accounts of local history, it seems that the few “traditional” courtyards remaining were constructed many decades ago. During the cocoa boom times, most were replaced with villas, which were frequently in a state of disrepair due to the recent hard times. Houses built 87 88 89

Interview with 74-year-old man (anonymous) by author. Barima, Ghana, March 4, 1999. Many elderly Ghanaians complained of loneliness and reported symptoms of depression. In the sections of the village where northern migrant families lived, there were many smaller homes. This usually reflected the lack of extended family members for most recent migrants. Interestingly, the second- and third-generation migrant families often had constructed large family compounds in the local Akan style.

Legacies of State Agricultural Policy

193

within the past decade followed the third style. Since all of the rooms were not under the same square roof, this “motel-style” was a relatively simple and less-expensive model of construction. This style more readily facilitated piecemeal building as individuals were able to finance the necessary materials and need to accommodate growing families.90 All three types of houses accommodated relatively large or multiple households, and all three were referred to as family compounds. While all of the family house styles in Côte d’Ivoire shared one central, outdoor kitchen area, cooking was not necessarily as communal as in the past. Sometimes multiple households would pool resources and cook “in one big pot”; at other times, each household cooked with their own ingredients in their own pots.91 Still, the centralization of cooking facilitated a greater sharing of resources and information about each other’s food security than when people lived in completely separate, nucleated houses, as was more often the case in Ghana. We can see this evolution of architectural styles as representing adaptations to changing needs, and as allowing the family compound to persist in the face of mounting pressures. Before concluding this section, a few caveats must be presented in each case. First, in Ghana, while the shift in local-level cash crop production and social relations was dramatic, it remained in process. Reciprocities were being strained and renegotiated, but they were not completely reconstructed. Not all young tomato farmers were able to forfeit all social ties to their elders, particularly in light of the volatility of the market, and not all older cocoa farmers were utterly bereft of financial security. One story from a twenty-year-old Akan tomato farmer revealed the dual existence of the old and new hierarchies of economic power.92 After this young man earned nearly $4,000 in 1998–1999 from tomato farming, a tremendous amount of revenue given the region’s median agricultural income of $346, he was inundated with requests for monetary assistance. His revenues were so depleted that he found it impossible to start the new tomato season, so he had to turn back toward his father, a village elder and one of the few large cocoa farmers in the village, for a loan.

90

91

92

Frequently, families with older villas have added at least one “motel” block to house married children, nieces and nephews, or adult siblings. For instance, one wealthy cocoa farmer in Kyere gave his daughter “chop money” and she prepared for the whole extended family living in the compound, whereas another wealthy cocoa farmer in Opanin gave each of his four wives separate “chop” money to buy meat, fish, salt, and other purchased ingredients to prepare separately. Survey interview, Barima, Ghana.

194

Informal Institutions and Citizenship in Rural Africa

Second, in Côte d’Ivoire, even though cocoa was still relatively more prosperous, leaving vertical social linkages between the young and the old relatively strong, informal reciprocity in general was under strain as a result of economic crisis. Not every elderly parent or uncle in Côte d’Ivoire was well cared for by children or nephews. Even though incentives to look after elderly parents or uncles remained in place, young people often lacked the means to follow through. A middle-aged woman emphasized that it was not a lack of volition but rather a lack of means that hindered the care of the elderly: “Because it’s difficult, what one wants to give them, one doesn’t earn. Because the produce isn’t bought at a good price, the means are not enough to take care of the elderly.”93 Nevertheless, the fact remains that a greater number of Ivoirians than Ghanaians reported caring for an elderly person. Furthermore, in Côte d’Ivoire, this role was more often described in a nurturing and positive way than in Ghana. For instance, one Ivoirian man described in typical fashion how he takes care of his father: “Each morning, I visit my ‘old man.’ If he asks me to do something somewhere, I do it. If not, I am there next to him. If there is something to do for him, I do it.”94

conclusion Caveats now aside, this chapter presents evidence that the dramatic shift away from cocoa to tomatoes as a cash crop in Ghana has redefined the informal institutions of reciprocity. The declining relative importance of land and the growing need for labor has resulted in new sets of incentives, and consequently reshaped social relations in two important ways. First, Ghanaians in this region gave social support to people with whom they shared a wider range of social ties than did Ivoirians. Second, social ties in Ghana were denser among family, friends, and others of the same age cohort. We have seen how these more diversified, horizontally-oriented networks have weakened the basis for long-term reciprocity between the young and old. By contrast, with the continued emphasis on cocoa production in Côte d’Ivoire, intergenerational reciprocity between the young and the old remained relatively strong. Ivoirians gave help to a more concentrated group of social ties, and this support appeared to link rather than segregate different age groups. While the economic crisis obviously constrained the support given to the elderly in both countries, different systems of local 93 94

Survey interview (anonymous) by author. Opanin, Côte d’Ivoire. Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999.

Legacies of State Agricultural Policy

195

economic production altered the dynamics of elderly care, providing incentives for elderly care in Côte d’Ivoire and disincentives for the same in Ghana. So what are the broader implications of these findings? The informal institutions of reciprocity were mediated in different ways by the different histories of state formation in these two regions. Even state policies outside of political administration or social service provision, such as in agriculture, can have profound indirect impacts on the informal institutions of reciprocity by shaping the local systems of production. Africanist historian Sara Berry makes a valuable contribution by demonstrating the different ways that various types of farmers have gained access to productive resources through social networks, but she underestimates the potentially transformative effects of colonial and postcolonial state power. In the present case, I find that the history of more exploitative agricultural policies toward cocoa in Ghana facilitated a shift to tomato production, overturning prior local hierarchies of social, economic, and political power in Ghana and spurring a redefinition of the boundaries of community. Meanwhile, these boundaries had so far proved relatively “sticky” in Côte d’Ivoire, where cocoa production continued to prosper based on the continuation of more supportive state policies toward permanent tree crops. The renegotiation of social relationships was occurring in fits and starts, not in the fluid and ongoing manner that Berry describes. Second, renegotiating economic livelihoods involved risks and pains. In Ghana, older farmers were less able to take advantage of the new opportunities created by present state policies. And the risks were high for the youth who tried. While some young tomato farmers have rapidly become wealthy, others have lost their shirts. Consequently, as more young farmers in the area tried to win the same tomato “lottery,” prices fell further. Hence, the longterm sustainability of expanded tomato production is doubtful unless the tomatoes can be canned or transported “fresh” to international markets.95 To see these local shifts in economic production as simply responses by rational African farmers to changing prices prevents us from comprehending the real desperation felt by many farmers, both young and old, as they abandoned the relatively more stable, albeit less profitable production of cocoa.

95

Tomato canning has been attempted in the past but with little success to date. This is ironic given the number of imported tomato paste cans that are purchased and used to cook local dishes. One obstacle to cost-effective local canning is that only extremely small quantities are desired for purchase in the local market due to the liquidity constraints of most households; Berry (1985, 1989, 1993); (Gracia Clark, personal communication, November 2003).

196

Informal Institutions and Citizenship in Rural Africa

Such a perspective emphasizing the individual rationality of farmers also occludes the broader politics at work. Market prices and incentives were not simply generated by a “neutral” market, but by a market constructed, regulated and reregulated by the state in various ways in Ghana and Côte d’Ivoire. In this neoliberal era, the Ghanaian and Ivoirian states were integrating with the global economy in novel and different ways producing new sets of opportunities for local producers. Thus, neoliberalism did not automatically produce one uniform opportunity structure for local African farmers and hence stimulate one uniform production response. It is not simply the experience of state agricultural policies today that matters but also the varied local experience of earlier eras of state formation that shape farmers’ current decision-making in these two regions. Finally, this chapter raises questions concerning the political implications of the changing structure of informal reciprocity. It reveals that the more diversified and horizontally-oriented social ties in Ghana ceased to effectively link the young and the old, and have strained intergenerational reciprocity. While this is clearly bad news for the long-term social security of the elderly in Ghana, what are the implications for the practice of politics? The evidence presented from this region of Ghana suggests that youth could play a positive role in integrating political systems and strengthening democracy. In Ghana, reciprocity relations appeared to connect individuals from different extended families, ethnic backgrounds, genders, and classes; whereas in Côte d’Ivoire, ties were more concentrated within family systems of the same ethnicity, gender, and class. The more-diversified ties in Ghana might foster social ties across a broader range of groups, promoting broader political coalitions and supporting enhanced political participation in decision-making – a sort of grassroots democracy. In Côte d’Ivoire, by contrast, the more concentrated and vertical networks would most likely reinforce social and political exclusion, and hinder political participation and local democratic development. The following chapter explores these possibilities in more detail.

part iii INFORMAL INSTITUTIONS OF RECIPROCITY AND THE PROSPECTS FOR DEMOCRATIC CITIZENSHIP

chapter 7 Transformations of the Informal Institutions of Reciprocity and the Implications for Citizenship

Over the past 100 years, what were similar village communities in these Akan regions of Ghana and Côte d’Ivoire have become very different. The informal institutions of reciprocity have changed and continue to be transformed. These great transformations were not the same in the two regions, however. The different legacies of state formation in political administration, social service provision, and agricultural policy have stimulated variations in the quantity of help exchanged, extent of participation in reciprocal exchange, and structure of social ties involved in informal social reciprocity. The parameters of social exclusion and inclusion have been redrawn in different ways. And as a consequence, the local hierarchies of political power have also changed. These transformations of reciprocity are about connection and disconnection, inclusion and exclusion, with critical consequences for citizenship and democracy in rural Ghana and Côte d’Ivoire. In this chapter, I explore the contrasts in indigenous meanings of citizenship and everyday political practices at the village level in these two regions.1 Then, expanding the focus, I look at how these surprising differences in village political cultures and practices shape the ways village residents participate in politics at the national level. This analysis sheds light on why Ghana has become an economic and political success story, whereas the former “miracle” of Côte d’Ivoire has become mired in ethnoregional civil conflict. Before proceeding, it is important to qualify my claims. I am not asserting that differences in the informal institutions of reciprocity were the only 1

See Schaeffer (1999) on the indigenous meaning of democracy in Senegal.

199

200

Informal Institutions and Citizenship in Rural Africa

causal factors that explain puzzling contrasts between these Ghanaian and Ivoirian villages in political culture and everyday political practices. In fact, in a related article, I show that variation in the organization of local associations also had an important role to play in strengthening these differences with implications for the mediation of ethnic conflict at the grassroots.2 Nevertheless, because the research design controls for so many other rival explanations, the importance of the local experience of the state and the resultant informal institutions of reciprocity is revealed.

implications of divergent informal institutions of reciprocity for local conceptualizations of social exclusion Social exclusion is often used in the welfare state literature to refer to exclusion of certain individuals or groups from the benefits of the welfare state.3 Social exclusion is used here in a broader sense to signify the exclusion of certain individuals or groups from the social support of other individuals or the community at large. In this way, the concept includes access to benefits from the state as well as from informal institutions of reciprocity in its meaning. To flesh out these abstract conceptualizations, I draw on village residents’ responses to more generic, open-ended questions about the way social support operated – both formal and informal – rather than the specific, close-ended questions detailing respondents’ actual reciprocal exchanges over the past year that have been analyzed in earlier chapters. More Rapid Grasp of the Concept of Social Exclusion in Côte d’Ivoire than in Ghana The very concept of social exclusion, of some people being left out, was often unimaginable to respondents in the Ghanaian region, but was immediately understood in the Ivoirian region. When asked whether there were any categories of people who typically had more or less access to social support, Ghanaian respondents often looked baffled or unsure. After many lengthy pauses, the interviewers in Ghana found themselves repeatedly explaining the question to puzzled respondents. In contrast, the very same question, posed in an identical manner, elicited an almost visceral response from an overwhelming number of Ivoirians. Almost every single Ivoirian understood and answered the question within seconds, with no repetition or 2 3

MacLean (2004). Hills, Julian, and David (2002); Rodgers, Charles, and Jose (1995); and Silver (1994).

The Implications for Citizenship

201

explanation. Even after reposing the question and waiting patiently for the respondents to process it, a much lower percentage of Ghanaians acknowledged any type of social exclusion. Only 37 percent of Ghanaian respondents replied that there were people who had less access to social support, whereas between 94–95 percent of Ivoirian respondents answered positively. A More Homogeneous Concept of Rich versus Poor in Côte d’Ivoire than in Ghana In addition to grasping the concept of social exclusion more readily, Ivoirians as a group characterized the nature of exclusion more homogeneously. Whereas the large majority of Ivoirians perceived social exclusion in terms of rich versus poor, Ghanaians were much more split in their descriptions of who would have less access to help. In Côte d’Ivoire, the overwhelmingly dominant response (51 percent) was that “the poor” had less access to help.4 In comparison, only 4 percent of the Ghanaian respondents who recognized any social exclusion defined it in these socioeconomic terms as being “the poor.” Indeed, in Ghana, there were so many different responses that the data was extremely difficult to code.5 The only answers that Ghanaians repeated with any frequency were someone who cannot pay back assistance (21 percent); lazy people (17 percent); sick, weak, or mad people (11 percent); those without friends (10 percent); and “strangers” without a lot of family nearby (9 percent). A comparison of the types of people identified by Ghanaians and Ivoirians as having more access to help reveals similar patterns. Whereas in Côte d’Ivoire, the distinctly dominant response was “the rich or those with rich families,” only 9 percent of respondents in Ghana answered similarly. The only other significant responses mentioned in Côte d’Ivoire were “bureaucrats” and “someone who is well-known or well-connected outside of the village.” While these responses emphasized more directly ideas of social position and influence, both could overlap with the perception of wealth. Interestingly, when articulated in terms of position and social influence, the connections mentioned in Côte d’Ivoire were outside of the village. Many respondents explained that the poor get less help because they are “unknown,” “not seen,” or “not heard”; while the rich “are well-known,” “quickly seen,” or “well-regarded.”6 In contrast, the only frequently cited

4 5

6

There were no differences in response according to gender or ethnicity in Côte d’Ivoire. An indicator of the heterogeneity of responses is that “other” was the most frequent response (25 percent). Survey interviews (anonymous) by author. Kyere and Opanyin, Côte d’Ivoire.

202

Informal Institutions and Citizenship in Rural Africa

response in Ghana was “someone who is well-known in the village” (21 percent), a response that was rarely cited in Côte d’Ivoire (3 percent).7 This provides additional evidence that reciprocity was more community-based in this region of Ghana compared with this region of Côte d’Ivoire where reciprocity more readily extended outside of the village boundaries. Thus, in general, Ivoirians’ perception of social exclusion was almost wholly based on socioeconomic power whereas Ghanaians’ notions were more heterogeneous and included other attributes such as the presence or absence of local social ties, moral character, illness, and other types of physical vulnerability. Why the Structure of Reciprocity Fostered Less-Rigid Social Exclusion in Ghana than in Côte d’Ivoire The previous section reveals some intriguing patterns in the conceptualizations of social exclusion in these regions of Ghana and Côte d’Ivoire, but some puzzles remain. For instance, why did Ghanaians seem so unable to comprehend the very concept and boundaries of social exclusion when so few appeared to participate in reciprocity networks at all in comparison with Ivoirians? One might have predicted that with the smaller number of Ghanaians able to give any kind of social support, and with the average value of that support being comparatively small, that Ghanaians would be highly sensitive to the notion of social exclusion. Because of the impact of the extended economic crisis (analyzed in Chapter 5) and the changing structure of local-level production (analyzed in Chapter 6), Ghanaians had incentives to diversify their networks of reciprocity. No one particular set of social ties was sufficiently deep to provide social insurance. The data from another survey question confirmed our earlier findings. When asked to whom they might turn for help if they were sick, Ghanaian respondents mentioned a diversity of potential sources, whereas nearly two-thirds of all Ivoirians cited the nuclear family first. (See Table 3.1.) One Ghanaian woman explained how whole extended families were poor, so Ghanaians looked outside of their extended family networks for additional social support: So if you come from a family with no money, you suffer a lot because no one has any to help you since he too is struggling to get some for himself. So you too suffer. That’s how it is. That’s why we’re saying most of the help we get is outside the family.8 7

8

Again, there was such a diversity of infrequently cited responses in Ghana that “other” captured 23 percent of the total. Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999.

The Implications for Citizenship

203

Similarly, friends were mentioned more frequently than any other social tie in Ghana, but no one friend was the sole source of support. Ghanaian respondents also often listed churches and religious organizations, yet these organizations generally did not have very extensive resources. A fair number of Ghanaians (16 percent) cited rich people and moneylenders as additional sources of social support (Ivoirians rarely did so), but there were costs associated with accessing this type of help, making it the last resort. Since Ghanaians had shallow but wide relations of reciprocity, they tended to think of everyone as having some access to help, albeit limited. The horizontal orientation of the social ties that cut across family, ethnicity, and gender also gave the impression of a more inclusive system of reciprocity in Ghana. In contrast, Ivoirans had a few deeper options for social support. Since assistance was predicated primarily on the nuclear family, those with welloff nuclear families were secured, but those without were relatively excluded. As one Ivoirian man explained, “The poor help each other, and the rich also help each other. ‘Birds of the same feather . . . ’ ”9 The second option for most Ivoirians was the extended family, but frequently entire extended families were relatively rich or poor. Instead of viewing this as a reason to diversify social relationships, as was the case in Ghana, Ivoirians became more aware of and shared their dismay at the social exclusion that followed. “Well-off families are the ones who help each other. Then, they all succeed. It’s the poor families who have problems. Always, there are problems.”10 Churches, mosques, and producer associations that might have more readily incorporated the poor were relatively restricted by the onerous expectation of donations and/or membership dues. The more-concentrated reciprocity networks in Côte d’Ivoire were more exclusionary for the poor, and may also have been more precarious for the rich, because social insurance was normally based on spreading risk.

implications of different informal institutions of reciprocity for indigenous notions of citizenship Individuals’ experience of the informal institutions of reciprocity also shaped their reading of the boundaries of the political community, reinforcing certain types of identities over others and changing the contours of cleavage and coalition in a particular place and time. This was important for how 9 10

Survey interview (anonymous) by author. Opanin, Côte d’Ivoire. Interview with 32-year-old cocoa farmer (anonymous) by author. Kyere, Côte d’Ivoire, July 21, 1999.

204

Informal Institutions and Citizenship in Rural Africa

t a b l e 7 . 1 . Comparison of Local Conceptualization of State

Word choice Level of state/government described Characterization of state Nature of relationship between state and local people

B/A Region of Ghana

Abengourou Region of Côte d’Ivoire

“government” Administration as a whole

“state” Individual politicians

Generalized Personalized Paternalism of state but with More hierarchical reciprocity of local people paternalism without reciprocity of local people

individuals defined themselves as citizens in a particular community and/or vis-a` -vis the state. Here, I not only investigate the indigenous meaning or content of citizenship but also the nature of the political community (or communities) from which citizenship emerges, or the location of citizenship. Conceptualization of the State in Ghana and Côte d’Ivoire Ghanaians and Ivoirians conceptualized the state in very different terms. (See Table 7.1.) First, in Ghana, most respondents spoke about “the government,” whereas in Côte d’Ivoire, the overwhelming majority talked about “the state.” This variation in word choice was not simply a linguistic accident but reflected underlying conceptual differences in how Ghanaians and Ivoirians thought about the state. The Ghanaians’ use of the word government emphasized a civil organization that provided services, whereas the Ivoirians’ use of the state highlighted a political organization. Second, when describing the state, Ghanaians and Ivoirians mentioned different levels of administration with contrasting degrees of personalism. Ghanaians usually referred in general terms to the civil administration or bureaucracy as a whole. Very few people talked about the ruling party or then-President Rawlings by name or called the government “he” as they did so frequently in Côte d’Ivoire. In contrast, Ivoirians specified “state” almost immediately in terms of individual political authorities, not the civil administration. Only one person mentioned civil servants as the state; everyone else interviewed mentioned politically elected or appointed officials such as the mayor, préfet, ministers, president, and so forth. Frequently, Ivoirians simply referred to individual politicians by name, rather than by their offices. For example, one Ivoirian villagers attested, “Mayor Akon Yao comes to our

The Implications for Citizenship

205

aid.” Not the mayor’s office, but the individual himself. The state was also more frequently conflated with the ruling party in Côte d’Ivoire (the PDCI at that time). Finally, Ghanaians and Ivoirians in these regions differed in their descriptions of the relationship between the state and local people. While both Ghanaian and Ivoirian respondents described a paternalistic state, in Côte d’Ivoire, a more straightforward, top-down paternalism reinforced the perception of unequal levels of power between the state and local people. “When one speaks of the state, it’s all of us. But, the sous-préfet, the préfet, the gendarmes and police are our guardians. They watch over us.”11 One Ivoirian highlighted the tension between a commonly espoused ideal of “the state as all of us” and the reality of an abysmal lack of connection between the state and the villagers: The state is all of us. But, our elected officials do not come to the source for their information, let’s say, regarding our key problems . . . M. Brou Emile, President of the National Assembly, doesn’t even know here in Opanin. He takes decisions all by himself, and we don’t like that. Otherwise the state is all of us.12

In comparison, Ghanaians frequently articulated their notion of paternalism within a familial idiom of reciprocal duties between the “father” and his children. For instance, one Ghanaian woman described the state as a father helping his “child,” which represented the local people in the village. “Before anything is done, it is important the big people provide assistance so if the big people don’t offer help, or push you – you as a child, what can you do?”13 No such reciprocity was described in Côte d’Ivoire. Comparison of the Local Notions of the Rights of Citizenship Consistent with the preceding finding that Ghanaians and Ivoirians in these regions think of the state in different terms, they have also developed varying notions of citizenship. My goal was to explore these indigenous understandings of citizenship in the most unrestricted manner possible. To do this, I asked two open-ended questions at the very end of the survey about the rights and obligations that were owed between an individual and the state.14 The interviewers used the local language translation for “the

11 12 13 14

Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999. Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. The two questions were: “When people talk about citizenship, they often talk about rights and obligations. What do you expect the state to do for its citizens?”; and, “We have spoken just now about the rights of citizens. In your opinion, what do you owe to the state?”

206

Informal Institutions and Citizenship in Rural Africa

state.” They were careful not to use “Ghana” or “Côte d’Ivoire” as shorthand because of the way it might frame citizenship at the nation-state level and bias the respondents’ answers. Rather than taking the national political community as the given starting point, the survey questionnaire aimed to collect data about the nature or actual location of the political community for village residents. Another way to accomplish this objective was to place these two questions about rights and duties immediately after one that probed the respondents’ primary political identity. The prior question thus reminded respondents how they belong simultaneously to multiple different types of categories; that is, age, ethnicity, gender, occupational, regional, and national. Since much of the theoretical and public discourse about citizenship focuses on rights, and respondents might be expected to be more familiar with or primed about rights, the survey questionnaire asked about rights first. In fact, the survey question about duties was the very last one posed to respondents during the interview. This exploration of indigenous concepts of citizenship duties makes a valuable contribution to the theoretical literature, which has until now focused on rights. Again, I do not claim that differences in the informal institutions of reciprocity were the only factors influencing variations in conceptions of citizenship. Earlier in Chapter 4, Figure 4.1 illustrated how the different histories of the formation of political administrative institutions also have had an independent effect on notions of citizenship that was not mediated by informal reciprocity. Still, because the comparison of these two regions controls for so many other variables, the finding of a relationship between these patterns of informal social reciprocity and citizenship is a provocative starting point for further research. To start, the perceptions of the rights of citizenship differed in two important ways in these regions of Ghana and Côte d’Ivoire. (See Table 7.2.) First, Ghanaians responded with a relatively restricted list of rights that were frequently linked in the same sentence to corresponding duties. For example, one survey respondent replied, “I would do anything I am asked to do, be it payment of money (taxes), or whatever, if only that will help to improve education.”15 This finding strengthens the earlier evidence of a more explicit view of reciprocity between the state and citizens in Ghana than in Côte d’Ivoire. In contrast, Ivoirians had a much more expansive view of the state’s role in their lives, often presenting a long list of entitlements owed to them from an omniscient and omnipotent state. For example, one forty-year-old Akan man 15

Survey interview (anonymous) by author. Barima, Ghana.

The Implications for Citizenship

207

t a b l e 7 . 2 . Comparison of Local Conceptualization of Citizens’ Rights

B/A Region of Ghana

Abengourou Region of Côte d’Ivoire

Perception of rights

Restricted list of rights Expansive list of linked explicitly to duties entitlements

Nature of goods delivered by state

Public goods consumed by communities

Private goods consumed by individuals

Three most frequently mentioned rights

Social services (33%) Employment (24%) Roads/markets/electricity (21%)

Individual loans (28%) Cash grants to needy (20%) Housing (16%) Employment (16%)

Extremely low scores on Provision of physical citizenship rights security (2%) associated with Protection of private Western democracies property (1.5%) Protection of equality (1%) Guarantee of freedom (0.6%) Freedom to participate in political processes (0%)

Provision of physical security (2%) Protection of private property (0%) Protection of equality (0.4%) Guarantee of freedom (2%) Freedom to participate in political processes (0.4%)

ticked off several things that the state should give him and then summarized by saying that the state should provide “everything I need.”16 Second, the nature of the rights owed to citizens differed in Ghana and Côte d’Ivoire. Ghanaians conceptualized their rights as citizens in terms of public goods that the state provides to communities, for example, social services or the construction of roads. In Ghana, one-third of the survey respondents (33 percent) said that citizens had a right to the state provision of social services, often free or low-cost education or health care. For example, many mentioned that the state “should give good health” or “reduce fees at the hospital.”17 Only 10 percent replied similarly in Côte d’Ivoire. In contrast with the largely community orientation of the rights described by Ghanaians, every one of the most frequently mentioned rights by Ivoirians was individually-oriented. For example, the most frequently cited right in Côte d’Ivoire was that the state should provide loans to individuals (28 percent) for farm expansion or enterprise development. One typical Ivoirian response was, “The state should give a loan for my trading.”18 16 17 18

Survey interview (anonymous) by author. Kyere, Côte d’Ivoire. Survey interviews (anonymous) by author. Barima, Ghana. Women’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999.

208

Informal Institutions and Citizenship in Rural Africa

Several Ivoirians even mentioned credit as a way the state should help individual families school their children, a one-time consumption by an individual as opposed to the long-term, income-generating objectives of most farm loans. Consistent with the earlier point about more expansive entitlements, almost as many Ivoirians (20 percent) asserted that the state should “give me some money” as outright cash grants instead of credit. In addition to housing (16 percent) and employment (16 percent), Ivoirians frequently included food and even various individual consumer goods as something the state should give to citizens for free. For example, one Ivoirian declared in all seriousness, “The state should buy me a car, purchase me a refrigerator, and, build a house for me.”19 Perhaps unsurprisingly, after nearly two decades of efforts to liberalize the economy, a significant percentage of respondents in both Ghana and Côte d’Ivoire mentioned that the state should reregulate the prices of goods (11 percent in Ghana; 13 percent in Côte d’Ivoire). Ghanaians and Ivoirians both mentioned a wide range of goods, including crops produced in the field site regions for the international and national markets, other foodstuffs, consumer goods, and transportation. One Ghanaian typified many responses in both countries suggesting that the government should return to preliberalization policies where the government purchased goods and stabilized prices. “The government should buy our farm produce as it used to be in the past, to ensure ready market and fix prices for our farm produce.”20 Finally, the rights “classically” associated with citizenship in advanced industrialized democracies in the United States and Western Europe were almost never mentioned by either Ghanaian or Ivoirian respondents. Thus, 2 percent or fewer of respondents in both countries mentioned the right to freedom, equality, political participation, or the protection of private property. According to T. H. Marshall, the previous civil and then political rights should have evolved first but in these postcolonial contexts, social rights were more frequently cited.21 What differed between the cases was the way Ghanaians conceptualized these social rights as delivered primarily to reciprocating community groups, whereas Ivoirians conceived them as rights provided to entitled individuals. The previous points demonstrate how the conceptualization of rights was powerfully shaped by the political context as well as the particular moment in world historical time. Rights do not develop everywhere at every moment in 19 20 21

Survey interview (anonymous) by author. Kyere, Côte d’Ivoire. Men’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Marshall and Bottomore (1992).

The Implications for Citizenship

209

t a b l e 7 . 3 . Comparison of Local Conceptualization of Citizens’ Duties

B/A Region of Ghana

Abengourou Region of Côte d’Ivoire

Perception of duties

Explicit notion of reciprocity

Concept of obligations more difficult to grasp

Nature of duties owed to the state

Performed together as a community and/or to benefit the community

Performed as an individual

Three most frequently mentioned duties

Develop the community/ country (47%) Pay taxes (22%) Perform communal labor (21%)

Work on individual farm (58%) Participate in politics (10%) Develop community/ country (10%)

Extremely low scores on citizenship duties associated with western democracies

Participate in politics (7%) Obey laws (3%) Serve the military (0%)

Pay taxes (8%) Obey laws (7%) Serve the military (0%)

history in the same linear progression of stages from civil to political to social rights, as Marshall theorized. These postcolonial cases highlight how rights may be obtained in a different order, and furthermore that they can be won and later lost. Different political histories, including different everyday experiences of the state, shape how people develop their own concepts of the rights of citizenship.22 Ironically, the legacy of British laissez-faire liberalism in Ghana stimulated a more community-oriented notion of citizenship on the ground. The history of more decentralized state institutions in Ghana allowed space for local-level community institutions to thrive. Thus, the liberal emphasis on self-help was accomplished in the context of the community, not as separate individuals. In contrast, the French legacy of centralized étatism squashed community-based initiative and encouraged individual dependence on state largesse. Paradoxically, the highly centralized state in Côte d’Ivoire fostered a stronger liberal individualism at the local level. Comparison of the Local Notions of the Duties of Citizenship Paralleling the patterns found previously, Ghanaians and Ivoirians also thought about the duties of citizenship in distinct ways. (See Table 7.3.) 22

Tilly (1995) and Gore (1995).

210

Informal Institutions and Citizenship in Rural Africa

First, although Ghanaians linked rights and obligations explicitly in descriptions of reciprocity with the state, Ivoirians dwelled on entitlements and then struggled longer even imagining the concept of duties. For example, after a lengthy pause, one Ivoirian respondent finally responded, “I greet them,” referring to state officials who come visiting the village on rare occasion. As another Ivoirian exclaimed, “It’s us that don’t have anything. Who should help the state, and why?!”23 Second, the nature of the duties owed to the state continued to be more community-oriented in Ghana and individually oriented in Côte d’Ivoire. The most frequently cited response in Ghana was “to work to develop the community or country as a whole” (52 percent), whereas the most frequently cited reply in Côte d’Ivoire was “to work on the respondent’s individual farm” (58 percent). As one Ivoirian explained, “Every morning, I go to the farm. This is all that I can do for the state.”24 Almost all the Ivoirian respondents spoke in terms of how the state benefited from their individual cash crop production, but almost never in terms of how their individual food crop production increased the food security of the nation as a whole, as so many Ghanaians did. As one twenty-seven-year-old female tomato farmer in Ghana said: “I should work harder at my farming to produce more food to help feed Ghana.”25 Another indicator of the more community-oriented conceptualization of duties is that more than three times as many Ghanaian respondents (22 percent) mentioned the obligation to pay taxes as Ivoirians (7 percent). Ghanaians not only mentioned taxation more frequently but also almost always put it in the context of reciprocity for state-provided services. For example, several Ghanaians asserted that citizens should pay taxes in return for the state’s efforts to create jobs, and provide free medical care and free education. This is not to argue that all Ghanaians were eager to pay taxes; they were not. But in general, many more Ghanaians viewed the payment of taxes as an obligation they had to fulfill. In contrast, while several Ivoirians from this region indirectly referred to state taxation of their cash crop production, very few directly discussed paying taxes to the state as an obligation. They spoke of payment to the state only in highly personalized and ceremonial terms to acknowledge state representatives visiting the village. For example, several Ivoirians mentioned giving gifts to the state in the “traditional” idiom of honoring visitors with a sheep. These comments by Ivoirian respondents illuminated a popular sense 23 24 25

Survey interview (anonymous) by author. Opanin, Côte d’Ivoire. Ibid. Survey interview (anonymous) by author. Barima, Ghana.

The Implications for Citizenship

211

of submission vis-a` -vis a dominant state and lacked any notion of representation or reciprocity. Finally, the contrasting perceptions of communal labor in Ghana and Côte d’Ivoire revealed dramatically the nature of citizenship in the two cases. In Ghana, communal labor was mentioned explicitly by almost 21 percent of all respondents, whereas fewer than 1 percent of respondents in Côte d’Ivoire noted it. One long-time resident of Makwan, Ghana, emphasized the need for self-reliance by explaining, “The citizens are too many for the state to be able to help, so the citizen should rather do something to improve their lot instead of relying on the state.”26 During the time of fieldwork in both Ghanaian villages, the chief or local unit committee frequently organized workgroups to facilitate a government-sponsored rural electrification project, provide labor for the construction of a public school building, or prepare for a funeral. Opinion in both Ghanaian villages was that participation in communal labor was high, enforceable, and evenly distributed among residents.27 In comparison, communal labor in Côte d’Ivoire was viewed negatively as an indicator of state failure. When one Ivoirian respondent mentioned how the “cadres” had organized the village to construct several school classrooms, he was openly resentful. The message was that communal labor was something that should not have occurred if the state had been functioning properly.28 In Côte d’Ivoire, there were clearly community projects that needed to be accomplished (i.e., preparation for road resurfacing, clearing of brush for the youth hostel, etc.), but Ivoirians failed to organize any communal labor to facilitate them, as was so often done in Ghana. Again, as earlier, very few respondents in either Ghana or Côte d’Ivoire mentioned the “classic” obligations of citizenship historically associated with Western democracies. Very few respondents mentioned obeying the laws of the state, and not one person described military service as an obligation.29 The only exception was taxation, which was prominent only in Ghana. The duty to participate in the political process was also mentioned by a fair number of respondents in both countries (10 percent in Côte d’Ivoire and 7 percent in Ghana), but this was well behind the most significant responses. 26 27

28 29

Survey interview (anonymous) by author. Makwan, Ghana. One Ghanaian man explained the enforcement mechanism for communal labor: “Whoever refuses to take part is sent to where the law is.” Men’s focus group (anonymous). Tape recording. Makwan, Ghana, April 1999. Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. The history of the military is much different in Ghana and Côte d’Ivoire than in the United States or other Western democracies. The army is entirely professional. There is no requirement to register for military service at a particular age, and there has never been a need for a draft.

212

Informal Institutions and Citizenship in Rural Africa

To summarize, Ghanaians and Ivoirians had distinct ways of conceptualizing the state role as well as the specific rights and obligations they associated with citizenship. In Ghana, the more diversified and horizontally-oriented relations of reciprocity reinforced the legacies of state policies to strengthen the community orientation of citizenship. In contrast, in Côte d’Ivoire, the more concentrated and vertically-oriented relations of reciprocity added force to the effects of more statist policies to build up the individual orientation of citizenship. This analysis thus strengthens the earlier conclusions that we need to build new theories of citizenship to understand non-Western states. The conceptualization of citizenship in these non-Western cases was different from the West because it was a completely different moment in world historical time. We cannot simply abstract the process of moving from a more localized to a national sense of citizenship from a particular historical and political context. Thus, beginning to imagine a national community during the “Benedictine” moment at the end of the eighteenth century is not at all comparable with doing so in the twentieth or twenty-first century.30 To begin, the historical context is different. The imperial era of globalization of the eighteenth century was not the same as the rapid and neoliberal globalization of today. In many ways, the neoliberal era of globalization supports the continued value of local identities.31 Local identities are not being definitively crushed or progressively replaced with attachment to the nation-state, but instead being reconstructed in complicated and particular ways. The digital and communication technology of today’s era of globalization facilitates translocal linkages between individuals, organizations, and social movements. As a result, the evolution of citizenship today is not a simple, linear zero-sum competition between national and other competing identities. Second, the political context is different. Most of the non-Western countries were formerly colonies that have only recently achieved independence. Many are weak states, in the Weberian sense of having a monopoly of the use of force within their territory, and even more fragile democracies, with quite recent memories of authoritarian rule during the colonial and postindependence eras. As a result, civil and political rights have not been added

30

31

Jean Comaroff used this turn of phrase, “the Benedictine moment,” to great effect in a lecture on globalization and the commodification of ethnicity at Indiana University, November 9, 2007. Comaroff and Comaroff (2009).

The Implications for Citizenship

213

sequentially in a teleological evolution, but repeatedly been offered and then subsequently sacrificed. On the other hand, social rights have been emphasized as a cornerstone of national development and citizenship from the beginning, particularly because many of the colonies in Africa were obtaining independence during an era of great state reconstruction and welfare state expansion in Western Europe after WWII.

implications of divergent informal institutions of reciprocity for everyday local politics Now, this chapter moves beyond an analysis of the mindset of individuals in these regions of Ghana and Côte d’Ivoire to examine how these individuals actually participated in everyday local politics. I argue that differences in the informal institutions of reciprocity in the contemporary period framed how people viewed the long-term possibilities for participation in the local political community. Because individual Ghanaians and Ivoirians were connected in different types of relations of reciprocity, they organized along different political lines of coalition and cleavage. As a result, the Ghanaian and Ivoirian approaches to conflict resolution and decision-making varied at the local level. In general, Ghanaians and Ivoirians in these regions engaged with the local institutions of chieftaincy, the central state, and political parties in quite different ways.

Comparison of Village-Level Authority and Decision-Making In general, village-level state and chieftaincy institutions had greater authority and were more active in development initiatives and decision-making in the Ghanaian region than in the Ivoirian one. When asked who were the most important individuals or groups in making decisions or developing policies for the village, Ghanaians most frequently cited village-level political institutions, whereas Ivoirians noted, often by name, individuals who were based outside of the village. (See Table 7.4.) In Ghana, an overwhelming number of respondents (82 percent) cited the unit committee as among the three most important leaders of village development. In comparison, Ivoirians rarely mentioned the most local political authority, the sous-préfet (4 percent), or mayor’s (6 percent) office. Ghanaians in this region also evaluated the village chieftaincy as playing a more pivotal role than did Ivoirians. More than twice as many Ghanaians (55 percent) cited the chief as one of the three most important leaders for

214

Informal Institutions and Citizenship in Rural Africa

t a b l e 7 . 4 . Perceptions of Most Important Individual or Group Leader for Village Development and Policymaking Category of Leader or Group

Derivation of Authority

Most local state institutions

Based on elected or appointed political office

Most local “traditional” institutions

Local and nonlocal individuals

B/A Region of Ghana

High Unit committee (82.1%) District assembly (23.8%) Village chief (54.8%) Chief’s elders (12.9%) Low Based on personal Bureaucrats (0%) political and The rich (0%) economic Big farmers (0%) power

Abengourou Region of Côte d’Ivoire Low Sous-préfet (3.8%) Mayoral (5.8%)

Village chief (25%) Chief’s elders (4.5%);

High Bureaucrats (51.4%) The rich (16.4%) Big farmers (11.2%)

* Note: All statistics shown are cumulative amounts that combine results for first, second, and third mentioned responses.

development in Ghana as Ivoirians (25 percent).32 The chief’s elders were also rather frequently mentioned in Ghana (13 percent) and only rarely noted in Côte d’Ivoire (5 percent). Indeed, in Côte d’Ivoire, respondents were almost never able to name the elders of the village. One non-Akan cocoa farmer who had been an officer in the village cocoa producer cooperative admitted, “Me for example, I have been here since 1984 (fifteen years), but I don’t know who is an elder.”33 Others would tentatively put forward the names of one or two people, but their uncertainty was magnified by the lack of agreement among the other discussion participants. In comparison, in Ghana, almost everyone interviewed was able to name the elders in their village quickly, usually identifying them by their ceremonial subchief names as opposed to their common family name. Although the traditional 32

33

Other survey and interview data suggest that the institution of the chieftaincy was held in higher esteem than most individual chiefs in the Ghanaian region. Furthermore, many respondents distinguished in the Ghanaian region between the chief’s important role in community development as opposed to his less significant assistance to individuals. An argument could be made that since this man did not belong to the indigenous Akan ethnic group, that he would not have the same reasons to know the elders. Still, his highly public activity with the cocoa producer cooperative and his own role as an elder of his ethnic group in the village would have provided ample opportunity to come into contact with the Akan elders. Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999.

The Implications for Citizenship

215

queenmothers were more powerful and active in both Ghanaian villages than in either Ivoirian one, they, along with the rest of the royal family, were rarely cited in Ghana (3 percent) and never mentioned in Côte d’Ivoire (0 percent). A significant number of Ghanaian respondents (24 percent) mentioned the district assembly, the representative body based in the district capital, as being important. Still, the fact that nearly four times as many respondents mentioned the unit committee, the most local state institution based in the village, as mentioned the district assembly demonstrates the impact of the extension of local government to the villages in Ghana. Even though problems remain with Ghana’s decentralization initiative, local people did view the most local level of government, which was based on members elected from their village, as the most significant and influential in the decisions that affected their lives. This was obviously not the case in Côte d’Ivoire because no such elected village institution existed. In contrast, demonstrating a lack of confidence in local-level political institutions, most Ivoirians (51 percent) cited, often by name, wealthy civil servants (“cadres”) as being the most influential person or group for the development of their community. These individuals may have been born in the village and have family members living there whom they visited occasionally, but they were generally based in the regional or national capital. As one Ivoirian explained, “These are the ‘cadres’ who can develop the village, for example, Yao.”34 In Kyere, even the younger brother of the chief, who seemed to serve as a village elder, mentioned only this same Yao as the single most important leader for village development, not the chief. In Opanin, the minister of economy and finance was considered a “son of the village” and was mentioned frequently by name as being pivotal in various village development projects. In comparison, in Ghana, respondents never cited highlevel civil servants or particular “sons of the village” by name. The second most frequently identified leaders of village development and decision-making in Côte d’Ivoire were simply “the rich” (16 percent). Similarly, several big cocoa farmers were frequently mentioned (11 percent) by name as being important leaders for village development. Interestingly, in Opanin, these planters were not official elders of the chief, but served as de facto elders and interim leaders after the village chief’s grave illness and subsequent death. In contrast, Ghanaians never cited the rich or big farmers. The comparison of the frequency of these two responses demonstrates 34

The cadre’s name has been changed to protect his anonymity. Survey interview (anonymous) by author. Kyere, Côte d’Ivoire.

216

Informal Institutions and Citizenship in Rural Africa

dramatically the confluence of economic and political power that was rooted in personal wealth and achievement in Côte d’Ivoire instead of any elected or appointed office, as was the case in Ghana. An ethnographic comparison of the initial process of obtaining approval for the field research provides further evidence that the local state and chieftaincy institutions had more authority and decision-making power in Ghana than in Côte d’Ivoire, in which individuals with personal political and economic power seemed the most influential. To preview, in general, villagelevel state and/or chieftaincy institutions made the decision to approve my research stay in Ghana, whereas urban-based “cadres” and/or local economic elites did so in Côte d’Ivoire. In the first village in Ghana, the chief, in consultation with his elders, played a major role in personally encouraging the field research. The chief later consulted several members of the unit committee as well as the district assembly representative for the village, but the decision was practically taken. The research team was housed in the chief’s palace, and the chief ordered the “gong-gong” beater to announce our arrival to the village residents early the next morning. In the second Ghanaian village, the unit committee facilitated the official approval from the chief. The decision was formally announced at a meeting presided over by the chief but attended by all the unit committee members (as quasi-elders). The chief arranged accommodation in the compound of a village elder (in which one of the chief’s wives lived and cooked). In this village, the unit committee initiated the “gong-gong” announcement of our arrival to the village. In both cases, the village-level state and chieftaincy institutions played an active and often concerted role in approving of and accommodating the research team. In contrast, in the first village in Côte d’Ivoire, the start of the study was delayed because a village resident had to travel to Abidjan to seek the approval of an urban-based cadre. The cadre gave the keys for one of his village homes, then vacant and somewhat distant from the chief’s residence, and instructed the villager to accommodate the team there. The chief did hold a somewhat impromptu meeting in his courtyard to officially welcome the team one Sunday afternoon, but few village residents were aware or even around, and despite repeated requests, it was never clear whether the “gonggong” was beaten to announce the project. In the second village, one of the wealthiest farmers in the village played the primary role in approving the project, in consultation with several other local economic elites. The meeting to officially welcome the project was held indoors in the wealthy farmer’s living room, and the team was accommodated in his large compound.

The Implications for Citizenship

217

In both Ivoirian villages, the chieftaincy was included in more ceremonial and symbolic terms than as an actual decision-maker in the process.35 So what explains these puzzling differences in the authority and decisionmaking processes in similar villages in Ghana and Côte d’Ivoire? The divergent structure of reciprocity shaped how village residents evaluated their local political institutions. In the Ghanaian region, because reciprocity was more horizontally-oriented and widely diversified across a range of social categories, social ties frequently cut across ethnicity, gender, and class. These social linkages provided opportunities to gain knowledge of and develop confidence in individuals who belonged to other groups. Thus, the horizontal networks increased the likelihood of developing coalitions (social, economic, political, or otherwise) across these various categories at the community level. This more-heterogeneous local social experience also encouraged a greater willingness to come together to resolve community problems at the local level rather than trying to skip to higher levels in the political system. This inclination to resolve conflict and make decisions at the village level were strengthened because in Ghana, the village-level chieftaincy and unit committee were both available and actively pursuing opportunities to expand their local authority. Finally, more than two decades of authoritarian rule in Ghana reinforced the appeal of working with local government. The years under Rawlings’ PNDC regime (1981–1992) were described by village residents as particularly frightening, which had a localizing effect on political participation. In Ghana, then, authority and decision-making started and stayed in the village. In contrast, in the Ivoirian region, the networks of reciprocity were more vertically-oriented and concentrated on the nuclear family, so social ties rarely cut across ethnicity or class. The social exclusion of poor families versus rich ones was reinforced, and there were fewer opportunities for contact between individuals of different groups. The vertical networks increased a narrower type of social trust that was less likely to lead to coalition building across groups within the community. Since the village chieftaincy was relatively weak, and no equivalent to the Ghanaian unit committee existed in the Ivoirian villages, individuals had little incentive, or opportunity, to solve problems at the community level. Frequently, individuals and groups proceeded directly to personal contacts positioned in the

35

It is true that the chief was gravely ill at the time of the official start of the project in the second village, however, even earlier that year, and two years previous during a pre-dissertation trip, I was introduced to the chief more as a courtesy, or even an afterthought, instead of an essential first step in the usual protocol as was done in Ghana.

218

Informal Institutions and Citizenship in Rural Africa

higher echelons of the political system. Even though Ivoirians also lived under several decades of authoritarian rule since independence, the character of authoritarianism was quite different and fostered national linkages rather than local ones. While the single-party state of Houphouet-Boigny did not allow for competition, there was generally less-violent instability and fear of the ruling regime than in Ghana during the same period. In Côte d’Ivoire, then, more-homogeneous and extra-local social and political experience combined with the absence of effective local political institutions to undermine authority and decision-making in the village and shift it upward to the regional and national capital. Comparison of Village-Level Party Organization: Building Village Cells in Ghana versus Receiving Party Bosses in Côte d’Ivoire The next subsection shows that local people were also more active and continuously involved in political party organization at the village level in this region of Ghana than in Côte d’Ivoire. In each of the Ghanaian villages, several people were well-known as the village representatives for the major national political parties. At the time of the field research in 1998–1999, the opposition party representatives (the New Patriotic Party, or NPP) were the most energetic, organizing informal weekly meetings over a year and a half in advance of the 2000 elections (where the NPP won). These meetings included some limited discussion of programmatic issues, but focused largely on local party strategy to win voters, increase voter participation, and eliminate electoral manipulations and fraud. Local party leaders were also trying to institutionalize party organization at the village level by opening up party offices in the village. The NPP district office actively supported these developments by organizing events to inaugurate the new headquarters of the party’s village cell. Ironically, the ruling party (the National Democratic Congress, or NDC) was less lively at the village level. Party representatives in the village described extensive factionalization within the district branch of the NDC, paralleling the divisions at the national level and hindering their local organization. In contrast, in Côte d’Ivoire, political party meetings and organizational activities were much less frequent and institutionalized. Representatives of the ruling party acknowledged that they would begin to meet more regularly only immediately prior to elections. Even then, the party focused on organizing small festivals and rallies to receive visiting party bosses from Abidjan. These visits appeared to be largely initiated by the regional and national party officials, not the village representatives. The opposition parties were

The Implications for Citizenship

219

beginning to get organized for the 2000 elections, but much less energetically than in Ghana. The Rally for a Democratic Republic (RDR) was quietly establishing initial village cells, but there was no official representative, and there were no regular meetings. Opposition activity was also conducted in a tenser atmosphere of greater secrecy in Côte d’Ivoire than in Ghana. Opposition party followers claimed that there were actually more party members in the village, but that local people were fearful of openly declaring their opposition to the long-dominant PDCI. In the past, sous-préfet officials had difficulty finding even one official representative of each political party to monitor the elections at each polling station. The variation in reciprocity relations also helps explain the contrasting patterns of political party organization in these Ghanaian and Ivoirian villages. In Ghana, the more diversified and horizontally-oriented relations of reciprocity encouraged a more localized and openly heterogeneous political party organization there. In contrast, in Côte d’Ivoire, the more concentrated and vertically-oriented networks stimulated a more top-down, urbanactivated, and relatively secret party organization. For example, various members of a single extended family were often quite frank about belonging to opposing political parties in Ghana, a seemingly rare occurrence in Côte d’Ivoire where nuclear and extended families tended to pledge allegiance cohesively. The differences in the practice of local-level politics have important implications for political development and stability at the local and national level. First, the fact that village-level institutions existed, were active, and were considered to be so influential in Ghana facilitated local political integration and development. Although far from perfect, Ghanaian village residents reported that there were local avenues for participation and conflict resolution. In comparison, Ivoirian village residents relied on the personal power of local and urban-based economic elites to make decisions and organize politics. These patterns hindered broader political participation in decision-making and potentially threatened stability at the local level.

implications of divergent informal institutions of reciprocity for village residents’ everyday participation in national politics The different character of local reciprocity in Ghana and Côte d’Ivoire not only shaped how people viewed the everyday relevance of local political institutions in different ways but also how they participated in politics at the national level. While existing scholarship highlights the important role of

220

Informal Institutions and Citizenship in Rural Africa

clientelist networks for the state in Africa,36 rarely are the peculiar patterns of local social networks connected explicitly to the broader character of the national patron-client system. Not all patron-client systems are equal. While clientelism existed in both Ghana and Côte d’Ivoire, the character of the networks on the national scale differed in important ways. In general, in Ghana, clientelist networks were more localized, fluid, and multiethnic than in Côte d’Ivoire. This section traces these national differences to the variations seen in the structure of informal reciprocity developed earlier in the book. First, in Ghana, the more village-centered structure of reciprocity networks reinforced a more localized patronage system. Social ties were denser within the village or at most within the district in Ghana, and thus Ghanaian villagers were most active in mobilizing village, or at most district-level, patrons to help resolve a problem. Strengthening the localization of the broader patron-client system in Ghana were two factors: (1) the history of fear and intimidation of state authority, particularly during the early years of Rawlings’ rule; and (2) the lack of linkages that “scaled up” from the local to the national level.37 First, what was termed the “culture of silence” under the early years of the Rawlings regime was just beginning to erode in Ghana in the late 1990s.38 While Ghanaians generally discussed politics openly among friends, many Ghanaians spoke about their fear of voicing political discontent in a more public setting or through higher channels. “We, our voices are not heard in government . . . Truly, because of fear, we do not voice our worries.”39 Another woman’s comments reiterated how fear of Rawlings moved any political dissent out of the public realm to the private realm of prayer: If you go and stand somewhere and say he’s not governing this nation well, you’ll be killed. If you say he’s destroying the nation, it won’t go well with you. So the only thing you can do is pray for God to change him for everything to be okay.40

Several people mentioned praying for the national-level political leaders so that they would be “changed” or understand or think about local people. Another woman’s comment revealed how her fear elongated the distance between the village and the Ghanaian capital of Accra. “What can I do? I am in a village and they are in Accra. If you say something, and you are not careful, they will arrest you.”41 36 37 38 39 40 41

Hyden (1983); Price (1979); and Bayart (1993). Evans (1996: 1124). Gyimah-Boadi (1993). Women’s focus group (anonymous). Tape recording. Makwan, Ghana, March 1999. Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Women’s focus group (anonymous). Tape recording. Makwan, Ghana, March 1999.

The Implications for Citizenship

221

The second factor that further fostered a localization of clientelist networks in Ghana was the lack of linkages connecting the relatively vibrant local political institutions to the district, regional, and national capitals. While suggesting a certain level of fear, the following comment also implied that the necessary institutional connections simply do not exist: “If we could take these issues up there (Accra), we would have gone.”42 Both women and men expressed similar views that “If you talk, it does not reach anywhere . . . Our power does not reach anywhere.”43 Meanwhile, in Côte d’Ivoire, the greater dependence on reciprocity from cadres based in the regional and national capital strengthened a more nationalized patron-client system. When something was needed, village residents depended on one or two urban-based cadres from their own extended family, and thus ethnic group, and contacted them directly in Abengourou or Abidjan. As one woman declared succinctly, “If you have problems, those who work in Abidjan can help resolve the problem.”44 Many Ivoirian respondents described how personal connections were necessary in order to receive some kind of individual or community support from the state. “The government doesn’t intervene here (in the village). It’s necessary to have a person on high over there in order to have something.”45 In comparison, a few Ghanaians noted the influence of political party membership on whether the state helps individuals and communities, but individual patrons were never mentioned as frequently as they were in Côte d’Ivoire. Still, just because these clientelist networks were more personalized in Côte d’Ivoire, it does not mean that they were necessarily more open, inclusionary, or flexible. In Côte d’Ivoire, social connections clearly overlapped with economic and political power from the local all the way up to the national level. “We, the poor, who don’t have anyone in Abidjan, get less help and those who have contacts and, above all, those who have money get more help.”46 Thus, the structure of social relations at the local level was paralleled at the national level and reinforced the contours of social and political exclusion in society.

42 43 44

45 46

Women’s focus group (anonymous). Tape recording. Barima, Ghana, March 1999. Men’s focus group (anonymous). Tape recording. Makwan, Ghana, March 1999. The woman quoted was careful to specify that the help was not usually financial, but rather, pushing paperwork past bottlenecks along the proper route or bringing the issue to the right person’s attention. Women’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Survey interview (anonymous) by author. Kyere, Côte d’Ivoire. Survey interview (anonymous) by author. Opanin, Côte d’Ivoire.

222

Informal Institutions and Citizenship in Rural Africa

I can go find a well-connected person to help me. If you don’t know anyone (in the government), then that, truly . . . it’s not possible . . . In general, those who are wellconnected, the ‘big men’, get more help and those who are not known, the poor in general, get less help.47

Many Ivoirian respondents confirmed the preceding statement and spoke of being “unknown.” “The state does nothing for me; the state doesn’t know me.”48 One respondent expressed what it meant to “be known”: “Those who are in politics are known” or “those who have family in the government” or “who are in the party with those who are in power, up high.”49 Thus, if village residents enjoyed personal connections or social ties to someone in Abidjan, then they were “taken care of,”50 but if not, the exclusion experienced was so keen that one young man described himself as “blind and deaf.”51 For some in Côte d’Ivoire, the more vertically and nationally-oriented patron-client system meant that many levels of the political system and bureaucracy were skipped over in a usually efficacious way; for others, the protracted distance between themselves and political authority was unbridgeable. The second aspect of difference in the national patron-client networks in Ghana and Côte d’Ivoire was the degree of instability. Earlier chapters revealed that reciprocity in Ghanaian villages was more diversified among a wider range of social ties, partially in order to access labor and land on a short-term basis for tomato crop production. These social ties were also more temporary and shifting. For example, a nineteen-year-old tomato farmer may borrow a plot of land from an elder brother and ask one of his non-Akan friends to help him with weeding for the first harvest of the year, but during the second harvest, he might ask two other friends to help him with weeding and his elder sister to help him with a loan to buy medicine. The shifting nature of social support networks at the local level most likely rendered the larger patron-client system more fluid as well.52 The long-term reciprocity and loyalty between patrons and clients was hence destabilized. In comparison, in Côte d’Ivoire, social support networks were more concentrated on a narrower social base, primarily linking members of the nuclear and (to a lesser extent) the extended family. The structure of these social ties was earlier explained by the continued dominance of cocoa 47 48 49 50 51 52

Survey interview (anonymous) by author. Kyere, Côte d’Ivoire. Ibid. Ibid. Survey interviews (anonymous) by author. Opanin, Côte d’Ivoire. Interview (anonymous) by author. Kyere, Côte d’Ivoire, July 13, 1999. I am grateful to Goran Hyden for suggesting the implications of the temporary nature of social support networks on the larger patron-client system. Personal communication, September 2001.

The Implications for Citizenship

223

production, which required more long-term use of land and minimal labor inputs. In this case, the social support given was often of more significant value and repeatedly exchanged between the same people over an extended period of time. Thus, the local social support networks in Côte d’Ivoire were much more long-term and continuous, rendering the national patron-client system more stable in Côte d’Ivoire than in Ghana. One possible indicator of the greater stability of patron-client ties in Côte d’Ivoire was the more immediate perception of social exclusion mentioned earlier in this chapter. Since clientelist networks were relatively more entrenched and overlapped with economic and political power, Ivoirians more readily recognized who was “in” and who was “out” than did Ghanaians, whose networks were more constantly being renegotiated and redefined. Furthermore, this social and political exclusion in Côte d’Ivoire was usually characterized in terms of lack of access to the powerful outside of the village. In addition to a greater awareness of exclusion, a more stable clientelist system could also help explain the feeling of powerlessness vis-a` vis the state expressed by so many Ivoirians. “What they say (the ministers who form the state), we follow. We, for example, can’t make decisions.”53 The third way that variations in local reciprocity resulted in different clientelist networks on a national scale was that the existence of more horizontal linkages at the local level in Ghana might have facilitated parallel patterns of cross-class and cross-ethnic coalition building at the national level. This could potentially diminish the salience of ethnoregional politics at the national level in Ghana versus Côte d’Ivoire. Interestingly, although Ghanaian respondents identified primarily with their ethnic group in almost equal numbers as Ivoirians, ethnic identity was slightly less salient in national politics, or perhaps was mobilized in more subtle (and consequently less disruptive) ways. While ethnic politics clearly continued to play a role in recent elections in Ghana, politicians and party organizers usually managed it deftly. Both the membership and rhetoric of Ghanaian political parties were relatively more heterogeneous in terms of ethnic group identification and appeal. While many political parties were associated with a dominant ethnic group, Ghanaians frequently made distinctions between political parties in programmatic terms, not ethnic or regional ones. Certainly, the issue of ethnicity had not openly embroiled recent political campaigns to the extent it had in Côte d’Ivoire. And furthermore, ethnic violence had been more limited. To be sure, ethnic conflict was not unknown in Ghana, and lives had surely been lost. But 53

Men’s focus group (anonymous). Tape recording. Kyere, Côte d’Ivoire, August 1999.

224

Informal Institutions and Citizenship in Rural Africa

even the deadly ethnic conflict in the northern part of Ghana in 1995 did not spread throughout the country, infecting every vein of politics at the national level. In fact, at the national level, this Northern region event had been practically shelved. In contrast, in Côte d’Ivoire, the more vertically-oriented relations of reciprocity, which remain concentrated largely within families of one class and ethnic origin, did not foster this broader type of coalition building. In fact, the nature of local reciprocity reinforced the salience of ethnic identity from the local all the way up to the national level. There were several examples of the higher salience of ethnicity for national politics in Côte d’Ivoire. First, at the time of field research, ethnicity was being mobilized as a result of an Ivoirian state initiative to register and title all land. With the apparent approval of the state, mostly indigenous Akan landowners were demanding the return of land given or sold over the years to non-Akan migrant workers. Many of these second- or thirdgeneration non-Akan workers had used the land to establish their own cocoa plantations and resented not being able to pass it on to their heirs. Second, Ivoirian political parties were more obviously and exclusively associated with members of a particular ethnic group or region. Thus, the RDR was commonly referred to as the party of “Northerners” (including those immigrants from Sahelian countries north of Côte d’Ivoire), the PDCI was the party of the Akan, and the FPI was the party of the Bété. When Ivoirians were asked the differences between these political parties, they frequently responded in terms of ethnic categories and stereotypes in lieu of any actual programmatic differences. For example, one focus group participant exclaimed: Heh! The FPI! Isn’t that the party that has two fingers for its symbol?! Yes, I know that these are the Bétés, and they love to have “palabres”, and if they take power, we are “foutou.” It’s the same thing with the RDR.54

Another exclaimed about Outtara, the RDR candidate whose nationality was in dispute leading up to the aborted 2000 elections, “Alassane Outtara is a Mossi; a Mossi will come and lead our country . . . No!”55 Finally, former PDCI president Bédié raised nationality, or what he termed “Ivoirité,” as the dominant issue in the run-up to the 2000 elections. But nationality seemed to be essentially a cover for a regional politics that divided

54

55

Women’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999. Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire, September 1999.

The Implications for Citizenship

225

north against south. This regional cleavage overlapped with a religious one, creating tension between the Muslim-dominated North and the Christiandominated South. Ironically, the mirror image of these cleavages exists in Ghana too, yet, national politics was not organized along the same lines.56

conclusion The perceptions and everyday practice of social exclusion, citizenship, and politics from the local to the national level differed in dramatic ways in these regions of Ghana and Côte d’Ivoire. The variation in the informal institutions of reciprocity were a key, and heretofore understudied, causal factor in the creation of these important differences. First, the more horizontal and diversified character of reciprocity in Ghana elucidated the difficulty in comprehending the very notion of social exclusion by Ghanaian respondents. In comparison, the more vertical and concentrated character of reciprocity in Côte d’Ivoire explained the more immediate conceptualization of social exclusion in more homogeneous terms of “rich” versus “poor” in Côte d’Ivoire. Second, reciprocity that was more narrowly-concentrated on the nuclear family meant that Ivoirians were less connected to their larger community through social interaction and resulted in a more individuallyoriented notion of entitlements from the state in Côte d’Ivoire. In contrast, more diversified relations of reciprocity in Ghana fostered a wider basis of social interaction across the community, resulting in a more communityoriented notion of reciprocity with the state. Third, the character of reciprocity in Ghana and Côte d’Ivoire made authority, conflict resolution, and patronage more localized in Ghana than in Côte d’Ivoire. Finally, the ways that the preceding components combined – the indigenous concepts of social exclusion, notions of citizenship, and the everyday practice of local and national politics – in each of the country cases were perhaps counterintuitive. One might imagine that the strength of community-based citizenship and local organization in Ghana might threaten national unity and stability, but this does not appear to be the case. If anything, the national political outcomes are quite the reverse. Overall, the prospects for democratic consolidation continue to be more promising in Ghana than in Côte d’Ivoire.

56

This is not solely due to the articulation of local social support networks on the larger clientelist system, but also due to the history of a much more open policy of immigration in Côte d’Ivoire promoted by Houphouët-Boigny discussed earlier in Chapter 6.

226

Informal Institutions and Citizenship in Rural Africa

In Ghana, a less obvious and unyielding sense of social exclusion has combined with a more restricted and community-oriented notion of citizenship to produce a more optimistic readiness to participate and to compromise in politics. Even when one considers the potential for more unstable clientelist networks on the national scale in Ghana, the prospects for stability and good governance are greater than they are in Côte d’Ivoire due to the greater viability and legitimacy of local political institutions. Clientelism thus was not the only base of politics in Ghana, and it was arguably on the decline. These more locally-rooted institutions probably helped stabilize the Ghanaian political system during the elections in 2000, which gave power to the NPP opposition for the first time in twenty years, as well as in 2008, which returned power to the NDC after a razor-thin runoff victory. Nevertheless, the lack of linkage upward from the village to the national political system continues to be a problem that must be resolved for genuine long-term political stability in Ghana. In Côte d’Ivoire, a more rigid concept of social exclusion, in combination with a more expansive notion of individual entitlements from the state, produced a higher level of frustration and resentment than was found in Ghana. This frustration was exacerbated by the more personalized and “stable” type of clientelism that reinforced social exclusion in the political realm and hindered any type of accountability: The elected officials consider us to be animals. They don’t even respect us. All the time, they inform us that such and such country or NGO gave money to realize such and such thing, but we see nothing. They use it for their own needs. We are not happy.57

Basically, there was no more reason to buy into local or national-level politics in Côte d’Ivoire. This high level of frustration with no viable outlet for political participation provides a context for understanding the December 1999 coup d’état and subsequent instability in Côte d’Ivoire. The long-term resolution of the civil war in Côte d’Ivoire will require more than an elite-level political pact and cease-fire. Without workable local institutions that incorporate the heterogeneous populations that continue to reside throughout the country, no democratic peace will ever be sustained.

57

Men’s focus group (anonymous). Tape recording. Opanin, Côte d’Ivoire.

chapter 8 Conclusion

unexpected differences in informal institutions of reciprocity and citizenship in similar villages of ghana and coˆ te d’ivoire When I arrived in these villages in the Brong-Ahafo region of Ghana and the Abengourou region of Côte d’Ivoire in the late 1990s, I expected to see many similarities. I was interested in examining the informal institutions of reciprocity and indigenous concepts of citizenship so I had carefully chosen villages that were analogous to each other and representative of their similar regions on a number of critical theoretical criteria. To begin, I thought these four villages would be more similar than different because they shared the same precolonial Akan culture, history, and politics. Well before the establishment of British and French colonial rule in the late 1800s and early 1900s, these villages had all been settled by Akan peoples who resisted incorporation into an expanding Asante empire. The Akan continued to be the predominant ethnic group in each of the villages so I knew the villages still shared similar indigenous institutions of chieftaincy, justice, land tenure, and matrilineal inheritance. I anticipated that the local economies were alike, too, because Akan farmers had adopted cocoa as the main cash crop around the same time in the early 1900s. These farmers had since encouraged a minority settlement of migrant workers from the North and Sahel, and had recently been challenged by drought and bushfires in the exact same years in the early 1980s. In addition, it took about the same length of time to drive to the villages from the national and regional capital cities. The villages were integrated comparably into national and international markets, which were being liberalized 227

228

Informal Institutions and Citizenship in Rural Africa

nationally by the 1980s with the adoption of similar structural adjustment programs. Village residents also benefited from the same moderate amount of economic and social service infrastructure – a small daily market, a few kiosks, several-times-daily taxi service, one or two elementary schools, and a clinic within walking distance. Even the national political scene seemed to be converging in the two cases. After decades of instability in Ghana and single-party dominance in Côte d’Ivoire, both countries had adopted multiparty democratic political systems in the early 1990s. Opposition parties had competed, but lost in early rounds of elections in both cases. The opposition in both countries launched complaints about incumbent advantage and electoral fraud, but they accepted the results and proceeded to focus on winning future electoral battles. Finally, both countries were pursuing major decentralization initiatives in which the central state was supposed to devolve greater authority and decision-making to the local level. Differences in Informal Institutions of Reciprocity Despite all of these similarities, when I left at the end of eighteen months of village-level fieldwork, I was struck by the tremendous differences in the local cultures and politics between the villages in Ghana and those in Côte d’Ivoire. First, I found that the informal institutions of reciprocity differed dramatically in the two regions. These variations were explored in detail in Chapters 2 and 3 in the book’s first section. To summarize, informal reciprocity varied in four key ways: (1) the percentage of people participating at all, (2) the quantitative amount of social support exchanged, (3) the number and type of social ties involved in reciprocal exchanges, and (4) the character of the terms of exchange. Overall, my research indicated that a lower percentage of the population gave a much lower quantity of help but to a more diversified group of social categories and on less-stringent terms in Ghana than in Côte d’Ivoire. These differences are highlighted in the following section. To start, the survey data revealed that a surprisingly high number of village residents were not investing in any type of social exchange, neither as a donor nor a recipient, and that this trend was particularly pronounced in the Ghanaian region. More than 39 percent of respondents from the Ghanaian region reported that they did not give any type of help, whereas only 6 percent of Ivoirians did not invest in any type of social exchange. When only the help given for hospitals, medicine, school fees, school supplies, and clothing was analyzed, even more dismal rates of participation in reciprocity was revealed with nearly two-thirds of Ghanaians and one-fifth

Conclusion

229

of Ivoirians not participating at all. This finding of a substantial lack of participation in reciprocal exchange, which was particularly striking in Ghana, is important when one considers how ubiquitous the extended family system and village community are presumed to be across Africa. Second, the median amount of help given was significantly smaller in the Ghanaian region than in Côte d’Ivoire. Ghanaians only gave, on average, less than $29 per year for hospitals, medicine, school fees, school supplies, and clothing; while Ivoirians reported giving $117 per year, more than four times as much. It is important to place the level of giving in the context of the median income and cost of living in each region. While $29 was actually quite a lot of money to the giver in Ghana, approximately 8 percent of the median agricultural income in the two villages, the value of the help to the recipient was relatively low when one considers the cost of living. Thus, a Ghanaian might give all the help she can give for the entire year and barely be able to cover the cost of secondary school fees for one child. Needless to say, this amount would not cover room and board, transportation, or school supplies for that child; not to mention the possibility of giving help to any other children, family members, or friends. This research suggests that the amount of social support actually exchanged in Côte d’Ivoire and Ghana may not be enough to cover many health or educational needs, and that policymakers should not assume “Africa’s informal support systems to be working relatively well.”1 Third, while the largest share of help was allocated to the nuclear family in both regions, this was particularly concentrated in Côte d’Ivoire, where more than 69 percent of help was given to the nuclear family (conceptualized very minimally as the respondents’ children). When Ivoirians did give to the extended family, it seemed more likely transferred to parents and other elders than to siblings or peers. Friends and others in the village were also not mentioned nearly as frequently in the Ivoirian region. In comparison, village residents from the Ghanaian region gave help to a more diversified set of social categories overall, and noted the importance of friends and siblings in particular. Qualitative analysis of popular discourse confirmed the previous differences in the structure of reciprocity. Villagers in the Ivoirian region articulated a more closed and exclusionary norm of reciprocity, focusing on the help to and from the nuclear family, whereas villagers in the Ghanaian region expressed a more open and inclusionary view, emphasizing the role of friends. Finally, the terms of reciprocity varied as well in the two regions. Whereas in the past, help was often given with no concrete expectation of when and how that support might be returned, both Ghanaians and 1

World Bank (1994).

230

Informal Institutions and Citizenship in Rural Africa

Ivoirians described how frequently help was given now as a loan that must be more rapidly repaid in cash, and sometimes with interest. Informal reciprocity was becoming less generalized and long term overall, and particularly so in Côte d’Ivoire. Ivoirians reported in greater numbers that they were no longer able to reimburse loans in kind, by working or with a bottle of gin. Furthermore, village residents in the Ivoirian region described how it was no longer “obnoxious” but rather “expected” to charge interest on the loan today.2 In contrast, villagers in the Ghanaian region were noticeably more uncomfortable discussing the topic of interest on loans. Implications of Differences in Informal Institutions of Reciprocity for the Elderly, Women, and the Poor These empirical findings actually matter on the ground for the poverty and social welfare of particular individuals and groups. The findings of seriously fatigued networks in Ghana and relatively exclusionary ones in Côte d’Ivoire challenge many popular assumptions about the endurance of a cohesive village community and vibrant extended family system in Africa. Neither of these collectivities was providing a robust and unified social safety net for everyone in need. Reciprocity was a double-edged sword of inclusion and exclusion. An unexpected finding from the study was that the elderly appeared to be included and excluded in different ways in the two regions. The survey results suggested that Ghanaians were giving and receiving less help as they got older than Ivoirians. Since every elder generation might be expected to express disappointment with the values of the next, the book highlights the surprising elements of cross-regional variation in actual behavior in this particular time period. My observations during the course of our visits to sampled households confirmed the patterns reported by respondents; in the Ghanaian villages, I found a greater number of elders sitting all alone in a room or house. The qualitative data from in-depth interviews – in particular, with the youth – also suggest that there were important differences in the way Ghanaians and Ivoirians in these regions talked about the reciprocity owed to their elders. Again, the largely one-shot nature of the book’s evidence makes it difficult to assess the possible changes in reciprocity with elders over time. Nevertheless, there are important and unanticipated differences between these two regions that merit further exploration. 2

Women’s focus group (anonymous). Tape recording. Opanin, Cote d’Ivoire, September 1999.

Conclusion

231

Perhaps less surprisingly, women participated less than men in reciprocal exchange. Women were more likely than men to not participate in reciprocity at all, and they gave lower amounts of help when they did. In this case, the differences between men and women were more pronounced in the Ivoirian region than the Ghanaian one. Still, overall, gender differences within regions were not as striking as the differences between regions. So, even though women allocated a larger share of their overall help to their extended families than men in both regions, women diversified their help more significantly in the Ghanaian region than in the Ivoirian one. In particular, women in Ghana gave more to friends and others than did the women in Côte d’Ivoire. Poverty, however, was not a simple and direct determinant of an individual’s participation in reciprocal exchange. The qualitative evidence suggested that many of the most impoverished village residents were the least likely to give or receive help from anyone. Respondents in Côte d’Ivoire seemed particularly aware of how the presence or absence of wealth and economic power affected one’s ability to participate in reciprocity. This analysis was supported by the quantitative data showing that reciprocity was not affected by an increase in wealth on its own. Quite surprisingly, wealth was insignificant in the statistical model as an independent variable. Instead, help given was boosted by an increase in wealth in Côte d’Ivoire. This analysis revealed how the more-polarized distribution of wealth in the Ivoirian region was a critical context for the way wealth influenced the informal institutions of reciprocity. One of the principal contributions of this book was the disaggregation of informal reciprocity in order to reveal the different winners and losers of these transformations. Not everyone was losing. Certainly, some individuals were able to take advantage of new incentives and opportunities, and consequently reaped high rewards. But others became further excluded economically, socially, and politically. The point not to be lost here is that there were costs to both the wins and losses. This process of cultural change was not automatic, fluid, or seamless. The transformations involved bitter contestation over the new norms and informal institutions of reciprocity. Throughout the book, we hear in the words of Ghanaian and Ivoirian villagers how they experienced severe economic hardship, psychological suffering, and social strain. The increased frequency of witchcraft accusations in both countries revealed mounting conflict as individuals transgressed previously accepted informal institutions. Another indicator of many individuals’ search for normative certainty in a turbulent time was the tremendous rise of charismatic churches across Ghana and to a lesser extent in Côte

232

Informal Institutions and Citizenship in Rural Africa

d’Ivoire beginning in the late 1970s and early 1980s. Again, as reciprocity was reorganized in these regions of rural Ghana and Côte d’Ivoire, the social and political landscape was being transformed.

Differences in Indigenous Notions of Citizenship in Rural Ghana and Côte d’Ivoire The second broad set of remarkable differences found in the villages in these regions of Ghana and Côte d’Ivoire was in the conceptualization of citizenship and everyday practice of local politics. Village residents in these two regions differed astonishingly in how they conceptualized the state, their rights and duties as a citizen, as well as the salience of local government. These variations were explored in Chapter 7 of the book’s third section and are briefly summarized in the following sections. To begin, the Ghanaian villagers emphasized the civil bureaucracy as a whole when they spoke about “the government,” whereas the Ivoirians mentioned individual politicians by name when they talked about “the state.” Villagers in both regions described the state in paternal terms, but in the Ghanaian region, it was articulated in the context of reciprocity between a father and his children; in the Ivoirian region, residents framed the relationship as more one-way and top-down. Residents of these similar villages also differed in surprising ways in how they conceptualized the rights and duties of citizenship. Ghanaian villagers discussed a more restricted number of rights and more explicitly and immediately linked these rights to corresponding duties. In contrast, the village residents from the Ivoirian region quickly ticked off a more extensive list of entitlements and struggled to even comprehend the notion of reciprocal duties. Furthermore, the rights and duties mentioned by the villagers in the Ghanaian region tended to be public goods enjoyed or created by the community or nation at large; the Ivoirian villagers cited private goods that were consumed individually. Implications of Differences in Citizenship for Everyday Local Politics and Democracy These unanticipated differences in the ways that local people thought about citizenship had important implications for the practice of everyday local politics and the prospects for democracy at the national level. First, at the local level, the Ghanaian villagers viewed village-level authorities more frequently as the key decision-makers and players promoting development.

Conclusion

233

Ghanaians expressed more awareness of and confidence in the formal institutions of the village chieftaincy as well as the local organizations of the central state, the unit committee, and district assembly. In contrast, the Ivoirian villagers rarely cited the most local political figures and usually mentioned individual “big men” by name based in the urban capitals of Abengourou or Abidjan. These conceptual differences had important implications for the ways that village residents practiced politics at the local level on an everyday basis. In the Ghanaian region, village residents were much more openly and actively developing village-level party organization for the ruling as well as opposition parties. In contrast, in the Ivoirian region, with an election also on the horizon, political parties were not institutionalized at the village level, but were represented opaquely by individuals known to favor a political group. Village residents in the Ivoirian region described their activity primarily as waiting to receive the party bosses’ visits from Abidjan. Village residents in Ghana also demonstrated a greater willingness to come together within the village to resolve community problems at the local level. Conflicts that arose during the fieldwork stay were resolved quickly by the village chief and/or unit committee. Plainly, it would have been a mark of shame to have any important decisions made elsewhere so every effort was made to organize politically within the village in the Ghanaian region. This could not have been more different in the Ivoirian region. Relatively routine decisions were delayed interminably due to the perceived need to consult with the powerful in Abengourou or Abidjan. Conflicts, when they did arise, simmered and threatened to boil over in the village because any decision on how to proceed was delayed by transferring the issue upward to the central state authority in the regional capital. Already, we can begin to see how these differences in the practice of politics at the local level might have broader implications for the possibility of consolidating democracy at the national level. In the Ghanaian region, because local authority was validated, and conflict resolution was attempted first at the village level, the clientelist system was more localized. Ghanaian village residents sought assistance with their problems from patrons closer to home. Meanwhile, in the Ivoirian region, villagers did not have much experience or faith in resolving conflicts in the village. Assistance was mobilized from outside of the village in the regional and national capitals, and thus local tensions and cleavages became magnified as they were projected onto the larger political system. We can see how these differences actually mattered on the ground for poverty, social welfare, and the prospects for conflict resolution and

234

Informal Institutions and Citizenship in Rural Africa

democracy. So what explained all these differences in an area where village residents on either side of the border considered themselves part of “one family”? Now I turn to a review of the book’s argument.

how the history of the state role in mediating risk shaped the informal institutions of reciprocity and citizenship in rural ghana and coˆ te d’ivoire I argue that these Ghanaian and Ivoirian villagers have experienced the state in fundamentally different ways since the colonial era. This varied history of state formation has stimulated unexpectedly divergent transformations in the informal institutions of reciprocity and indigenous citizenship in these regions of rural Ghana and Côte d’Ivoire. Specifically, I analyzed the historical construction of the state role in mediating risk vis-a` -vis the individual, family, and village community. To make this argument, I drew on a multidisciplinary literature focused on institutional and agrarian change. Throughout the book, I highlighted how formal state policies and informal institutions of reciprocity do not simply interact in a zero-sum relationship, but how they deeply transform each other over time. I moved away from the typical top-down focus on the effects of formal state institutions at the national level and shifted to a more provocative bottom-up perspective on the local experience of the state and local participation in the rewriting of both the informal and formal rules. Hence, the informal institutions of reciprocity and citizenship were neither a natural component of a certain type of primitive economy, nor an automatic response to a market failure to provide social insurance. Instead, these informal institutions were powerfully shaped by politics and history. To thoroughly examine this institutional transformation, I analyzed the history of formal state interventions in political administration, social service delivery, and agriculture in Chapters 4, 5, and 6 in the book’s second section. By investigating the state role across multiple policy domains – and not just social policy – a more nuanced explanation for the differences in informal reciprocity and citizenship was illuminated. To begin, the formal state institutions for political administration were examined. Based on extensive archival research and oral histories, I showed that different state legacies in political administration shaped the structure of reciprocity directly through peoples’ everyday interactions with the state. The normative view of the appropriate role for the state and the nature of the family and community were reshaped through these repeated experiences. Thus, in the Ghanaian region, the more decentralized nature of formal

Conclusion

235

administrative institutions validated the authority and meaning of membership in the broader lineage, ethnic, and village community. The central state was not the exclusive, or even primary, actor in mediating risk in the village. This stimulated the greater diversification of informal reciprocity among a wider variety of social ties and localization of these social ties within the village. In contrast, in the Ivoirian region, the more centralized administrative institutions eroded the authority of the precolonial system of chieftaincy and did not facilitate village-level organization or self-help. To these village residents, the central state became the main player mediating the risk they faced as individual members of smaller, nuclear families. As a result, the informal institutions of reciprocity became more narrowly concentrated on a smaller group of social ties, with the priority allocated to the immediate nuclear family. In Chapter 5, the legacies of state social service provision are shown to have unintended consequences for the patterns of informal reciprocity in these regions of Ghana and Côte d’Ivoire. Here, I argue that the more historically active and centralized state role in the mediation of risk through social policy in the Ivoirian region actually stimulated a higher quantity of less-diversified reciprocal exchange in the late 1990s than in the Ghanaian region. Beginning in the colonial era, the British saw their role as merely supplementing the “traditional” systems of social welfare in the extended family and village community, whereas the French aimed to replace these “inadequate” systems with a more activist, central bureaucracy. Despite pressures from the World Bank/IMF to adopt similar types of social policy reform under the structural adjustment era of the 1980s and 1990s, social policy and service provision in the Ghanaian region was much more decentralized and incorporated more extensively nonstate actors of the family and community than in the Ivoirian region. Reinforcing earlier dynamics started with the establishment of political administrative institutions, the more decentralized approach to social service provision in Ghana reinforced the diversification of informal reciprocity in the Ghanaian villages, whereas the more centralized and activist social service interventions in Côte d’Ivoire reinforced the concentration of informal reciprocity in the Ivoirian villages. But why did Ivoirians give such a significantly higher quantity of help than did Ghanaians in these regions? To answer this question, the book focused on the different historical levels of stability of service provision and the timing of retrenchment. Thus, it is not only the varied historical orientation of social service provision that mattered for the patterns of informal reciprocity, but also the stability of benefits over time. While social service provision was started earlier in the British Gold Coast, service delivery was

236

Informal Institutions and Citizenship in Rural Africa

more consistent over time in this region of Côte d’Ivoire since the colonial era. The higher stability of benefits over time in the Ivoirian region further amplified the more expansive expectations of the state role and therefore stimulated a higher quantity of informal reciprocity exchanged when the state retrenched during structural adjustment. The second major difference in the delivery of benefits over time was that the two states reduced their role in social service provision at different points along a trajectory of decline, which occurred at different moments in chronological time. Thus, a higher level of reciprocity was being mobilized in the late 1990s in this region of Côte d’Ivoire because of the Ivoirian state’s later reform of a social service system in less severe decline. In comparison, the reserves of informal reciprocity were both less intensively mobilized and already more seriously exhausted by the late 1990s in the Ghanaian region. This was due to the Ghanaian state’s earlier reform of a critically collapsed social service system. The third aspect of the state’s role in mediating risk is in the area of economic policy, specifically agriculture. The history of the state role to support economic production, mediate market volatility, and tax productivity directly stimulated divergent local economic production decisions and then indirectly influenced the structure of informal reciprocity. I argued that the more extractive policies experienced in this region of Ghana caused farmers to choose less land-intensive crops with shorter-term yields, such as tomatoes, and subsequently led to a further diversification of their informal relations of reciprocity. In particular, because Ghanaian farmers needed less land and younger labor, the diversification of reciprocity was oriented toward peers of the same age group, especially friends and siblings. Meanwhile, in the Ivoirian region, the state had historically been more supportive of cocoa farmers. Facing lower levels of risk from state intervention in agriculture in the Ivoirian region, village residents developed more concentrated patterns of reciprocity. This reinforced earlier dynamics in Côte d’Ivoire, spurred by the state’s mediation of risk in other policy areas, where the nuclear family had become the more exclusive priority. Ivoirians did not spread risk as widely as did Ghanaians who emphasized more greatly the role of friends or others in the village. After the nuclear family, the next priority in the Ivoirian region was the extended family, in particular, intergenerational links between parents and children, uncles and nephews. Ghanaians also gave a significant share of help to their extended family, but intergenerational reciprocity was weakened by the rise of the new “empire of the young” growing tomatoes. Where these youthful tomato farmers were less motivated to bridge age groups through reciprocal exchange, they were linked more extensively with friends and others across

Conclusion

237

gender and even ethnicity. In contrast, although reciprocity in Côte d’Ivoire was more inclusionary of the aged, it hardened the lines of exclusion between the indigenous, largely Christian Akan and nonindigenous Muslim groups. Again, there were different sets of winners and losers involved in the transformations of each of these regions. Finally, I argue that this reorganization of the informal institutions of reciprocity then helped shape the varied patterns of citizenship seen earlier in the two regions. The changing patterns of reciprocity were not the only factor influencing the indigenous conceptualizations of the rights and duties of citizenship. Certainly, the differences I detailed in the construction of the state role in political administration, social service provision, and agriculture had an independent, direct effect on notions of citizenship. Further, the varied nature of participation in local voluntary organizations also had an independent influence on the way citizenship was conceptualized. The book revealed, however, an understudied indirect pathway of influence between formal state institutions and citizenship through the mechanism of the informal institutions of reciprocity.

implications for our theories of informal institutions and citizenship I turn now to the implications of my argument for our theories of informal institutions and citizenship. Where much of the institutional literature has focused almost exclusively on formal state institutions, an analysis of informal institutions is essential for our understanding of politics. But just as we should not focus exclusively on formal state institutions, we also cannot study informal institutions (or what some scholars have theorized as social capital) in isolation from state power. More of our institutional analyses should focus explicitly on the engagement and interactions between formal and informal institutions. Informal social institutions did not emerge endogenously as an almost automatic solution to market failure. This type of account is not sufficiently historical or political. Furthermore, the state did more than simply enforce the formal rules or formalize the informal ones. Indeed, this book demonstrated that even weak states in rural Africa can profoundly transform informal institutions over time. This theoretical conclusion will stimulate controversy because it deviates from much of the conventional wisdom on the African state and informal institutional change. Most scholarship on the African state has emphasized the institutional weakness or incompleteness of the state’s extension to the rural peripheries. For example, Herbst used the term “rural hinterlands” to

238

Informal Institutions and Citizenship in Rural Africa

emphasize the colonial and postcolonial state’s inability to project power across large and sparsely populated African geographies.3 Goran Hyden has famously portrayed the African state as a “balloon suspended in mid-air,” unable to capture and subordinate the peasants in Tanzania.4 Indeed, Naomi Chazan and others argued that as the African state progressively weakened and collapsed starting in the late 1970s, informal networks and voluntary associations grew and thrived.5 But this notion of a zero-sum relationship between an expanding and retrenching state and society was overly simplistic and obscured the complex interactions over time highlighted in this book. Helmke and Levitsky’s typology of formal-informal institutional interaction ends up reinforcing this zero-sum conceptualization.6 This book emphasizes instead the historical process of mutual transformation. Other scholars have given up this blunt dichotomous categorization of state versus society and appropriately highlighted the blurred boundaries between the state and informal institutions. Bayart showed that local networks of power and reciprocity became entangled and inseparable from state power in the “politics of the belly.”7 Similarly, in Reno’s work, local warlords used informal reciprocity in order to weaken central states such as Sierra Leone and Liberia from below until they could no longer govern and provide security at all.8 But the power to penetrate and undermine the state from below has perhaps been overstated and certainly has caused scholars to underestimate the state’s capability to transform informal institutions through everyday interactions at the local level. Yet, this is not simply a call to “bring the state back in” to our study of informal institutional change. Political scientists definitely have a positive contribution to make, but this book’s analysis of institutional transformation required a holistic and genuinely multidisciplinary approach. The approach was holistic because it examined multiple policy areas in which the state and local communities interacted on an everyday basis. Even though the book focused on informal social reciprocity, it was not limited to the effects of formal state social policies and social service provision alone. The book also extensively evaluated the history of state interventions in political administration and agricultural policy. Rarely 3 4 5

6 7 8

Herbst (2000). Hyden (1983: 19). See, for example, Chazan (1983); Pellow and Chazan (1986); Azarya and Chazan (1987); Dei (1992); Bratton (1994); MacGaffey (1991); Cheru (1997); and Tripp (1997). Helmke and Levitsky (2006). Bayart (1993) and Bayart, Ellis, and Hibou (1999). Reno (1998).

Conclusion

239

have the effects of all three of these policy areas been investigated simultaneously.9 Drawing from the literature on agrarian change in anthropology and geography, the book highlighted the importance of examining the local experience of the state in particular places. The study combined a careful comparative research design, which aimed to reduce some of the complexity in the world, with intensive village-level fieldwork and ethnographic observation, which sought to investigate the meaning of big abstract concepts such as citizenship to particular people in particular places. I have resisted the temptation to theorize about commonalities across the African continent in order to investigate the origins of the puzzling differences I found in similar regions across the national border of Ghana and Côte d’Ivoire. The book highlighted how national policies may be similar with parallel objectives in the capital city, but then diverge in their local implementation and experience on the ground. The local experience of the state varied substantially across these two similar regions, but these local experiences were also mediated by history. The effects of the state’s interventions in administration, social welfare, and agriculture in the contemporary era were shaped by the history of state formation during earlier periods. Historians, political scientists, and anthropologists have long debated the local impacts of the colonial state, particularly in rural areas.10 Acemoglu and Robinson have made bold claims about the persistence of institutions and the importance of history, but this book showed in rich empirical detail how colonial legacies were transmitted over time and indeed mattered for culture and politics in relatively remote parts of rural Ghana and Côte d’Ivoire today. These colonial legacies were not simply a result of one set of formal state institutions that persisted unchanged over time, but actually a consequence of the interaction between colonial state institutions with local normative frameworks over time that influenced later stages of state formation under different regimes. To capture these changes over time, the study combined one-shot survey research and interviews about contemporary behavior and attitudes with the collection of oral histories and an in-depth analysis of the colonial and postcolonial archives. I worked to untangle the historical construction and local experience of the state role in mediating risk over time. For example, Chapter 6 demonstrated how broadly similar efforts to liberalize the cocoa market and encourage cocoa production during structural adjustment have met with quite varied responses by Ghanaian and Ivoirian farmers. These varied responses had less 9 10

See Bates (1989) and Ensminger (1992) on the rise of capitalist markets in agriculture. See, Hanson’s (2003) analysis of Uganda. See also Cooper (2002); Crowder (1968); Young (1994, 2004); Herbst (2000); Colson (1974); and Moore (1986).

240

Informal Institutions and Citizenship in Rural Africa

to do with the minimal cross-national differences in agricultural policies today and more to do with how farmers have interpreted these policies given their very different past experiences with the state. This locally-grounded analysis over a long period of time revealed three important points about informal institutional change that has challenged existing theories. First, informal institutional change occurred relatively quickly. Certainly, these institutions had changed significantly in each of the regions within peoples’ lifetimes and likely within the past decade or two. This notion of rather rapid informal institutional change contrasted markedly with the conventional conceptualizations of informal institutions (and ultimately culture) changing extremely slowly, even over thousands of years, and as a consequence, lagging far behind formal institutional change.11 Second, even if informal institutional change was relatively quick, it was not fluid. Instead, informal institutional change occurred in fits and starts of normative conflict and contestation. Informal institutions did not evolve continuously as a result of constant negotiation and renegotiation as Berry has theorized.12 Change was not this smooth, but instead jerky, with periods of disjuncture and tension, followed by spurts of reorganization. Finally, the dynamism of informal institutional change should not be mistaken for inclusion. This book disaggregated the changing institutions of informal reciprocity and revealed different sets of winners and losers over time in the two regions. These informal institutional changes were not necessarily efficient or equitable. In fact, overall, informal reciprocity was narrowing in both places. Fewer people were participating in reciprocal exchange, less help was being given and received, and the terms of exchange had become tougher of late. In his recent book, Goran Hyden characterized informal reciprocity as weaker and “more fickle” than in the past because people were increasingly seeking benefits from more distant sources outside of the immediate kin group.13 On the one hand, the diversification of reciprocity spread risk across a wider group of social connection; on the other, it reduced an individual’s ability to cope with risk as they were forced to seek help from more distant and less reliable contacts. These negative changes suggest that much of the earlier Africanist literature emphasizing the resiliency of informal social networks was too optimistic. The same could be said of the Ghanaian government’s assessment of local needs and capacity. Thus, the persistence of formal state institutions that devolved authority for social service provision to “the informal” was 11 12 13

North (1990); Knight (1992); and Putnam (1993). Berry (1993). Hyden (2006: 80).

Conclusion

241

also not technically efficient but instead could be interpreted politically as enabling the state to reduce its responsibilities without direct accountability. Having considered some of the major implications for institutional theory, I now turn to our understanding of citizenship. The policy feedback literature suggests that the experience of state policies shapes political attitudes, political participation, civic engagement, and notions of citizenship.14 Almost all this literature, however, has focused on how welfare state policies have influenced citizens in advanced industrialized democracies over time, in particular, explaining social policy and American political development. Yet, similar sorts of policy feedbacks occur in non-Western contexts, where weaker states may have lower levels of bureaucratic capacity and shape less literate, multiethnic populations. Demonstrating the existence of a policy feedback in Ghana and Côte d’Ivoire is likely a stronger test of this theory. The book further contributed to this literature by illuminating a specific causal mechanism for a policy feedback effect: Policies shape individual conceptualizations of citizenship and political participation indirectly via the changes in the local patterns of informal reciprocity. The direct links between formal welfare state policies and citizenship have been explored in much more depth already. Here, the study revealed an indirect pathway of influence, where state policies first transformed the informal institutions of reciprocity, which then shaped local notions of citizenship. This historical analysis of citizenship in these regions of Ghana and Côte d’Ivoire, also calls into question Marshall’s evolutionary expansion of citizenship. Citizenship rights were not continuously expanded in a linear progression over time in either case. Citizenship rights were not fixed and could be retrenched.15 A lot of instability and uncertainty existed in all three areas of civil, political, and social rights. Furthermore, social rights were not the ultimate expansion of rights accorded by the state but instead some of the first rights to be declared for all citizens at the time of independence from the colonial rulers. These social services were frequently delivered, however, via highly complex, public-private partnerships. As a result, an unambiguous connection between state and citizen was never solidified through the state’s guarantee of social rights.16 With the increasing role of the private sector in

14 15

16

Mettler and Soss (2004); Mettler (2005); and Campbell (2003). Ndegwa (1998) argues that citizenship is a social process that is constantly being contested and revised. Smith (1993) argues that citizenship is transformed when states contract with nonprofit agencies in hybrid systems of social welfare.

242

Informal Institutions and Citizenship in Rural Africa

service delivery during structural adjustment, social rights may be even less integrative of the national political community than Marshall theorized.17 Finally, this book’s examination of citizenship in these regions of Ghana and Côte d’Ivoire demonstrated the importance of theorizing about obligations as well as rights.18 An emphasis on obligations highlights citizens as active participants in their political communities, not simply passive recipients of rights. Logan and Bratton draw on the Afrobarometer public opinion data and suggest that Africans have not yet become “active citizens.”19 They may register and vote in elections periodically, but they are not demanding accountability from their leaders on a regular basis. My exploration of the indigenous meaning of citizenship also suggests that the ways Africans think about duties has important implications for the future of democracy.

informal reciprocity, citizenship, and democracy elsewhere in africa This book’s analysis has been grounded intentionally in a thorough examination of local-level variation in a rigorously designed comparative study. The claims have been qualified carefully. These same outcomes would not be replicated across every other region of Ghana or throughout Côte d’Ivoire. Indeed, the book’s strength is the ability to hold constant several of the key variables that might be presumed to influence reciprocity and citizenship – for example, precolonial culture, geography, and economic history. What do we learn when we move beyond this tightly controlled comparison of these two particular regions? First, we should consider what we might gain if we compared a different set of regions within the borders of one nation-state, either Ghana or Côte d’Ivoire. If the Brong-Ahafo region of Ghana were compared with the northern region of Ghana, for example, all three of the previously controlled rival explanatory factors would then differ. On top of that, based on the subnational comparisons highlighted by Catherine Boone, we can anticipate that this book’s explanation – the history of the local experience of state formation – would also vary across these two regions within Ghana.20 There would probably be a number of similarities in the state experience, but so many differences that it would be extremely difficult to parse out a viable explanation. A more promising set of comparisons would be to try and locate another similar precolonial community that had been divided by the same Ghana–Côte 17 18 19 20

Bantig (1999) makes a similar point. Janoski (1998). Logan and Bratton (2006). Boone (2003).

Conclusion

243

d’Ivoire border, perhaps in the northern part of each country. In this way, we could again control for a different set of precolonial cultures, geographies, and economic histories. The comparison would again highlight the effect of the construction of national state institutions at the local level in these similar regions. A slightly different way to design a future study would be to compare one colonial legacy as it operated in two different countries. Thus, if one precolonial community straddled the Côte d’Ivoire/Burkina Faso border, for example, one could tease out the local experience of the French colonial state in these two different territories. This could also be done looking at one precolonial community on either side of two formerly British colonies, perhaps more easily done in eastern or southern Africa. Again, careful attention would have to be dedicated to the selection of the cases because of all the potential other factors that might be important; for example, whether it was a settler colony or not. Based on this book’s analysis, we might anticipate that there would be more variation in the experience of state policies among the British colonies than the more centralized federations of French colonies, but that would remain to be validated empirically. Most of the existing contemporary work on state policies, informal reciprocity and citizenship in the rest of sub-Saharan Africa is not comparative, however, and focuses on only one of the previous issues, but does not make connections between all three, like this book.21 The limited research on the nature of state social policies suggests that across Africa, the official formal social security regime covers a very small minority of people.22 These works often obliquely mention the continued importance of the “traditional” social security system for the majority of Africans. Surveys conducted on poverty since the 1990s also indicate that although macroeconomic indicators have improved, poverty persists, and economic inequality is growing for many Africans.23 Thus, despite growth overall, hard times and the need for informal reciprocity continue.

21

22

23

Cammett and MacLean (n.d.) do highlight the connections between state social policy, nonstate providers of social welfare, and citizenship in their edited volume emerging from a Harvard Academy conference in May 2009. ILO (2006); Devereux (2001); Kaseke (1998); Asante (1987); Leliveld (1994); Ntsama (1997); and Ahmad et. al. (1991). See Gough et. al. (2004) for their theorization of informal security regimes. Haggard and Kaufman (2008) do not include Africa in their recent analysis of welfare states in the developing world. On Ghana, see Vanderpuye-Orgle (2006) and Amoako-Tuffour and Armah (2008). On Côte d’Ivoire, see also the government of Côte d’Ivoire’s National Statistics Institute (INS) released a study in November 2008 documenting an increase in poverty from 38 percent in 2000 to nearly half in 2008, the highest rate in twenty years. See “Cote d’Ivoire: Poverty Getting Worse-Study,” IRIN, December 3, 2008 (http://www.irinnews.org/www.irinnews.org.). On poverty in Africa, see Chen and Ravallion (2008).

244

Informal Institutions and Citizenship in Rural Africa

Most of the scholarship on informal reciprocity in other parts of Africa supports this book’s argument that these informal institutions are neither vibrant nor inclusive. Based on studies from other regions of Ghana, Côte d’Ivoire, and other countries as wide-ranging as Burkina Faso, Nigeria, Tanzania, and Zambia, scholars have highlighted the weakness or incompleteness of the informal institutions of reciprocity.24 Evidence from other studies elsewhere in Africa also reinforces this book’s conclusion that the structure of informal reciprocity institutions is changing over time, and perhaps relatively rapidly.25 These scholars disagree on how informal reciprocity is changing in these very different parts of Africa, but that finding actually supports the book’s argument about the importance of the local experience of the state role over time. The Afrobarometer project has conducted multiple rounds of survey research focused on political attitudes, including individuals’ notions of citizenship. Fortunately, Ghana is one of the eighteen countries in subSahara Africa included in the sample, but Côte d’Ivoire is not. Not surprisingly, Bratton, Mattes, and Gyimah-Boadi find that “country effects” are one of the most powerful determinants of political attitudes toward democracy and economic reform.26 Interpreting the meaning of these country differences is not straightforward, however, with the enormous variation between such cases as Botswana, Madagascar, and Zimbabwe. Statistical averages for “Africa” thus obscure the meaning of exactly what this book tries to illuminate: the historical experience of the state in particular places.

conclusion The representations of rural Africa as “one” homogenous continental crisis or as idyllic and unchanged since time immemorial did not match the reality of change and difference on the ground in these regions of Ghana and Côte d’Ivoire. Village residents described how the informal institutions of reciprocity and indigenous notions of citizenship were being transformed in different ways in the two regions. Ghanaians and Ivoirians had experienced the construction of national state institutions in very different ways since the colonial era. These varied histories of everyday interactions with the state 24

25 26

Udry and Conley (2004) and Kumado and Gockel (2003: 1, 6). See also Gockel (1996); Duflo and Udry (2004); Kazianga and Udry (2006); Isamah and Okunola in Guyer (2002); and Agbaje in Guyer (2002). Pellow (2002); Kaseke (1998: viii); and Hoon (2005). Bratton, Mattes, and Gyimah-Boadi (2005: 316–20).

Conclusion

245

profoundly shaped these local transformations. This micro-level political history of relatively small, remote villages in Ghana and Côte d’Ivoire can help us understand politics in the rest of postcolonial Africa. The analysis sheds light on some of the roadblocks and potential pathways to democratic consolidation from the grassroots all the way up to the national level.

appendix Additional Notes on Survey Sampling and Statistical Data Analysis

Recruitment of Research Assistants

Because I was unable to find research assistants who were fluent in English, French, and the Akan dialects of both fieldsite regions, it was necessary to use two different sets of research assistants in Ghana versus Côte d’Ivoire. In Ghana, my two research assistants were recruited in Accra and traveled with me to the fieldsite region. In Côte d’Ivoire, my two research assistants were recruited in the regional capital and only traveled out to the villages from there. I found many advantages to the latter recruitment strategy. While the Ivoirian assistants had less prior academic and research experience and required more intensive training and supervision, the job was a greater relative opportunity for them – both in terms of financial remuneration and work experience – so they maintained a high level of enthusiasm throughout the fieldwork. Their morale was further bolstered by the fact that they could easily go home on the weekends to their families. Finally, while the Ivoirian assistants did not originate from the fieldsite villages, their surnames were more familiar and so they were not considered to be as foreign as my Ghanaian assistants. Sampling Strategy

I limited the sample size to 100 adults in each village for two reasons. First, with n=100 in each village, I was able to use statistical analysis to explore intravillage variation as well as the variation between the two regions in Ghana and Côte d’Ivoire. Second, I restricted the survey interviews to 100 in each village due to time and budget constraints. The survey interview was relatively long, taking as little as forty-five minutes, but usually an hour and a 247

248

Additional Notes on Survey Sampling and Statistical Data Analysis

half to complete. In addition, it often took us a significant amount of time to locate sampled respondents and arrange a convenient time to conduct the interview. Usually we interviewed village residents either very early in the morning or in the late afternoon or early evening after they returned from their farms. Frequently for women, we would sometimes split the interview in two parts to allow them to cook or take care of pressing chores. I used a multistage stratified random sampling technique in order to select the 100 individuals in each village. To begin, in the first stage, we conducted a census of all the housing units in the village. I took a random sample from this list using a random numbers chart, and we returned to those units to obtain a list of all adults who slept in the house. It was vital to ask specifically for everyone who slept there because often people ate at more households than they slept. Also frequently non-indigenous people were not mentioned as members in a household until we probed: “is there anyone else who sleeps here?” Similarly, domestic servants were often not initially remembered or cited as living in the housing unit. I then put these names into four lists: Akan males, Akan females, nonAkan males, and non-Akan females. Based on information I collected from the village and district capital, I knew the approximate size each category represented in the village population so I randomly selected for the village sample a number from each category that was proportionate to the actual population. The goal of this stratification by gender and ethnicity was to ensure that the sample included enough women and nonindigenous ethnic members. Before beginning data analysis, I weighted the cases in each category by calculating their probability of selection. Since I did not draw my sample from individual housing units but from these four lists of individuals, it was possible and did occur that multiple individuals from the same housing unit were sampled. This was fine because I assumed that housing units might contain more than one household. Furthermore, I theorized that within households, usually multiple patterns of reciprocity existed that were not necessarily pooled or perfectly cohesive. The participation rate for the survey interviews was 98 percent. This is astoundingly high, especially when compared to survey research conducted in advanced industrialized countries such as the United States. Ironically, our biggest problem was finding a polite way of turning interested village residents away rather than cajoling them to participate. In the beginning of our village stays, the high participation rate might have been due to an expectation of some concrete material benefit or future development assistance. Despite our efforts to communicate immediately and openly, the message about the purpose of our activity in the village was sometimes

Additional Notes on Survey Sampling and Statistical Data Analysis

249

misinterpreted in the early days. Throughout our stay, we did offer as a gesture of gratitude an opportunity to take a Polaroid picture after the interview was completed. This was perceived as an attractive incentive to village residents but was muted intentionally by our willingness to take and share additional photographs with the “ordinary” thirty-five millimeter camera we toted on all our interview visits. More likely, the high participation rate was due to our willingness to accommodate village resident’s schedules as well as their eagerness to share their personal lived experiences and problems with an interested stranger. Techniques to Improve Data Reliability and Respondent Recall

Several decisions were made during the design, conduct, and analysis of the survey to improve data reliability and respondents’ ability to recall accurately the details surrounding these informal exchanges. To begin, the statistical analysis uses the data on help given rather than help received. The recall of values should be more reliable for help given for three reasons. First, particularly for gifts that were given in kind, the giver would be more knowledgeable about the value of the gift than the recipient. Second, even for gifts that were given in cash, the giver is probably making something of a sacrifice to share very limited resources, thus the gift is more memorable than to the recipient, who is appreciative but not “hurt” by the exchange. Third, the recipient might be receiving more than one gift simultaneously, particularly if the gift is for a health or school crisis, whereas the giver most likely gives one gift for one crisis. Second, to stimulate their memory of exchanges over the past year, respondents were reminded of the various seasonal events, such as the primary cocoa harvest or the Christmas holidays, beginning twelve months prior to the interview. Arguably, the recall would have been better if the time period had been compressed to only the past month or two, but the data would have been extremely uneven across village fieldsites due to the seasonality bias. Third, after asking an open-ended question about help given or received, research assistants then enumerated all seventeen types of help that might be possible to stimulate recall. The types of help given and received included help to pay for hospital or clinic fees; help to pay for medicine; help to receive care from a traditional healer or herbalist; help with the costs of a funeral; help to get engaged or married; help to pay school fees; help to provide school books or supplies; help to provide accommodation; help with food or prepared meals; help with cloth, clothing, or shoes; help with tools; help by working

250

Additional Notes on Survey Sampling and Statistical Data Analysis

for someone without getting paid; help with land; help to pay a debt; help to start a business; and help with anything else that was not already mentioned. This list was vetted with non-African and local scholars and then revised after a pretest in an Akan village close to Accra, Ghana. The research assistants also listed all the potential social relationships to prevent respondents from forgetting to include those ties that were so close they were “taken for granted.” The range of social ties included spouse; children; parents; grandparents; siblings; cousin; aunt or uncle; other extended family member; friend; association; church; government; village chief; agricultural worker; landlord; “stranger”; and other. Frequently a respondent would act surprised when asked about the help given to their spouse or to the schoolage children seen returning home to the courtyard in uniform. Several respondents explained that they had not considered this type of support “help”; it was what was expected normally – nothing extra. Respondents were also probed early on about whether they had any children who lived overseas so they would be reminded about remittances. Surprisingly, remittances from overseas were not common in any of the villages. Fourth, the research assistants were trained how to help respondents give monetary estimates for “in kind” exchanges that were at times significant over the long term. For instance, a Ghanaian woman in her fifties might reply that she gave cooked meals to her elder brother on a daily basis. Despite the study team’s efforts, these figures are inherently less reliable. Not all respondents conceptualized these exchanges as help nor took the time during the interview to estimate accurately the value of this help. For this reason, the statistical analysis draws on what are considered to be the most reliable figures in the data: help for hospital fees, medicine, school fees, school supplies, and clothing. Strategies to Address Sensitivity of Survey Questions

Several strategies were also devised to improve the reliability of the data given the potential sensitivity of questions about the intimate relationships within the family, village, and state. For the study to be successful, it was vital to generate trust within multiple communities at several levels in the political system. First, the project was approved by all relevant local authorities and then carefully introduced according to local custom. An ongoing awareness of and willingness to incorporate the indigenous culture of the village in our daily lives went a long way toward opening doors and undoubtedly improved the quality of the data collection. Second, the questionnaire was structured in such a way as to build trust and reduce sensitivity. Respondents were asked relatively innocuous and

Additional Notes on Survey Sampling and Statistical Data Analysis

251

“traditionally” acceptable (if not required) questions about their length of residence in the village and their children before moving on to more sensitive questions regarding income and transfers. The section that documented the help given and received was broken down into many smaller questions so the overall picture was obscured. Even though receiving help does not appear to be as stigmatizing in the fieldsite villages as it is in American culture, giving help does carry a certain prestige value and honor, so it seemed wise to begin with help given and then move on to help received. Finally, several of the most sensitive questions were posed in general terms or as hypothetical scenarios to diminish the focus on the individual respondent and their particular situation or replies. Third, the survey was administered in a place where the respondent was comfortable and could answer without being overheard by anyone else. Usually, we would ask respondents to point out a place that they felt was best for a quiet conversation. Almost always, interviews were conducted outside, a bit away from the house or cooking areas. Frequently, several stools or chairs would be taken to sit under a tree, but sometimes we sat under a front porch or cooking shed. Curious friends or family members usually gave the respondent privacy to conduct the interview, and if they insisted on interloping, we politely addressed them, answered their questions, and then asked delicately if they could return later after we had finished. Despite the many efforts to improve the accuracy of the data collection, there are still errors and omissions in the data. Instead of pretending otherwise, which is simply unrealistic, I try to uncover the strengths and weaknesses of the data collection to facilitate a more accurate interpretation of the results. The survey instrument was effective at stimulating respondents’ recall of relatively significant exchanges, but smaller transfers were more difficult to document. There was also the possibility that respondents lowered their estimates of resources given and received in order to appear needy to an apparently wealthy outsider such as myself. The results, therefore, are most likely moderately underestimated. This underreporting should be constant across all the fieldsite villages, however, so even if the actual numbers are slightly low, the trends are consistent. Building the Statistical Model

To build the statistical model, I drew on the relevant theory to think through what to include and how best to operationalize the most theoretically salient variables and controls. The dependent variable was the help given in the past year to others for hospital fees, medicine, school fees, school supplies, and clothing. My primary hypothesis was the historical role of state formation,

252

Additional Notes on Survey Sampling and Statistical Data Analysis

which was represented simply as “country.” I conceptualized the rival explanatory variables as those that influenced an individual’s ability to give help and those that shaped an individual’s opportunity to give help. These variables are discussed in turn in the following sections. Variables that Influence the Ability to Give Help

One of the most compelling rival explanations to examine is the role of “individual wealth.” The most basic economistic logic might simply predict that as individuals’ wealth increases, their ability to give increases. Perhaps Ivoirians continued to enjoy greater wealth as a result of the earlier boom period known as the “Ivoirian Miracle” and were able to continue to help others more readily in the late 1990s. Modernization theory might further predict a nuclearization of the structure of reciprocity. According to this logic, as wealth increased, individuals would become increasingly attached to their “modern” nuclear family and detached from the particularistic social insurance obligations of their “traditional” extended family. Because of the very different way that wealth is distributed in these regions of Ghana and Côte d’Ivoire, I included wealth on its own as well as an interaction term for wealth and country. The measurement of wealth in largely farming communities with low levels of formal sector employment was challenging. Instead of using national-level GDP per capita for all of my cases, I developed a wealth index from the dataset. I could not use agricultural income as a proxy because not all respondents were involved in agricultural production. Income data would have been problematic anyway because of recall and trust issues. Instead, I developed an index that summed three housing characteristics (roof type, wall type, and floor type) and three asset characteristics (radio, television, and bike ownership). The reliability analysis indicates that this index has an alpha of 0.6221. While this alpha is not spectacularly high, it is considered sufficient, particularly given the previous difficulties involved in measuring wealth in a rural area of the developing world.1 Next, “education level” was tested. Very simply, an increasing level of education might mean increasing levels of income generated, for instance, through agricultural improvements in production and marketing, or access to off-farm employment. In this case, higher levels of available income might lead to both greater demands for help and an increased ability to fulfill those 1

Although my summed index is developed using a different technique, I benefited greatly from discussions with Gisele Kamanou.

Additional Notes on Survey Sampling and Statistical Data Analysis

253

demands, thus, a higher value of help given. Like wealth, modernization theory would predict that increasing levels of education would lead individuals to give more of their help to their nuclear families. Here again, education is seen as a mechanism of value changes that would decrease respondents’ attachment to the clan or ethnic group, resulting in a lower value of help given to the extended family. Third, several variables tested whether children served as an asset to village respondents. The idea here is that certain types of children might make more money and actually send remittances back to their parents in their village home, enabling their parents to give help to others. The model therefore included the number of children who completed secondary or higher education, or earned a salary as explanatory variables. The “number of children who live or work in a larger urban area” was originally included but later removed. It seemed to overlap and confound the role of the other previously mentioned variables related to children. Finally, age was included. Age could have two possible relationships with the level and structure of support given to others. The simplest relationship would be positive and linear such that with increasing age, the respondent had increasing numbers of dependents and/or access to a greater depth of resources, thus the respondent gave an increasing amount of help. But, based on the qualitative data, I suspected that the relationship might actually be curvilinear. Instead of rising directly with age, the value of help given might first increase and then decrease. For instance, a person might not give very much help to others when they were young and just starting out; then their assistance might rise as they got older and produced more; then their help would fall off during old age when their resources were drawn down. The project tested for curvilinearity in the data and found that age was not curvilinear in any of the equations. Variables that Shape the Opportunity to Give Help

The next set of explanatory variables influenced the number and kinds of opportunities an individual might have to give help. First, a final variable related to the respondent’s children was included. This variable was simply the respondent’s “number of children.” Theoretically, the value of help given would increase with a higher number of dependents. Furthermore, I would predict that with a greater number of children, respondents would tend to give more of their total help to their nuclear families than to other social ties. Second, “gender” was included as a rival explanatory variable. I predicted that the level of giving would be lower for females and that their giving might be more concentrated on the nuclear family or friends. This is first because

254

Additional Notes on Survey Sampling and Statistical Data Analysis

many married women (particularly the Akan) are living in their husbands’ home village, away from their own extended families. And second, because women’s lower rates of participation or control of the profits from cash crop production would decrease their available income to give help.2 Third, “ethnicity” was tested in the model. One might anticipate that the Akan indigenous ethnic group in the fieldsites would have stronger and more extensive informal social relations because they would be living closer to more members of their extended family for a greater period of time than the nonindigenous ethnicities that had migrated for work and were now living far from most of their families. Fourth, “the activity level of the lineage” combined several pieces of data in an index that measured the initiative and organizational activity of a respondent’s lineage system.3 Perhaps respondents with more active lineage systems gave more help to their extended families because their lineage system met frequently to organize and publicize the needs of various family members in crisis. Fifth, a “participation in local voluntary organizations” index was created.4 This index included involvement in a variety of organizations

2

3

4

See Berry (1993: 168) on the consequences of diminished female control over cash crop production for lower levels of investment in social relations. The “activity level of the lineage” index combines the following variables: (1) whether or not the lineage meets to discuss and resolve problems; (2) frequency of lineage meetings (regardless of respondent’s attendance); (3) frequency of respondent’s attendance at lineage meetings; and (4) annual financial contribution to the lineage organization by the respondent. This index measures not solely the participation of the respondent, but also the activity level of the organization itself. For instance, the first variable reflects whether or not the lineage meets at all to discuss and resolve problems, and the second variable reflects the frequency of lineage meetings irrespective of the respondent’s own personal record of attendance. The next two variables reveal the participation of the respondent in the lineage organization, first measuring the frequency of respondent attendance and then his or her annual financial contribution to the lineage. Although the fourth variable, “annual financial contribution to the lineage organization” did not score that high on the correlation matrix, it was retained in the index on theoretical grounds. Factor analysis demonstrates that the index explains 66.9 percent of the variance in the data. A reliability analysis shows an alpha of 0.8157. The index can thus be interpreted as both reliable (producing the same results each time it is administered to the same person in the same setting) and valid (actually measuring the construct that it is attempting to measure). Since there was no significant difference between the alpha coefficients when the items were standardized, the raw coding was used in the index. The “participation in local voluntary organizations” index combines the following variables: (1) number of associations to which the respondent belongs; (2) highest frequency of attendance at meetings of any organization by respondent; (3) whether or not the respondent holds any leadership position in any organization; and (4) highest annual financial contribution to any organization. Each respondent was asked for detailed information on his or her participation in up to four associations. Since very few respondents belonged to more than two

Additional Notes on Survey Sampling and Statistical Data Analysis

255

such as local agricultural cooperatives, rotating savings clubs, youth groups, ethnic associations, sports clubs, religious organizations, school groups, business associations, social groups, funeral associations, and so forth. Much of the civil society literature would predict that respondents who were more active in such local voluntary organizations would have denser social networks outside of their family and thus greater opportunities to help friends and neighbors.5 Of course, this voluntary participation might equally increase the respondent’s ability to give help. Sixth, marital status6 was included as a control variable. The hypothesis that married respondents might report a higher and/or more extensive level of giving because they probably have more dependents and essentially two extended families was nullified. Marital status did not seem to have any effect on giving. Finally, length of residence in the fieldsite village was included, but also had surprisingly little effect on the extent or value of giving. One might imagine that the longer a respondent was living in the village, the denser the social networks would become, but it does not appear to be the case. This means that it is not simply by virtue of residence in the village that one participates in informal institutions of reciprocity. The density of social networks seems to require more active participation, as will be demonstrated in the following section. Controls for Village and Interviewer Effects

Since the book hypothesizes that state institutions were highly significant, it was essential to control for the effects of particular villages or interviewers. To investigate the research questions posed, it was necessary to spend more time in a fewer number of carefully selected villages rather than accomplish very little in many randomly selected villages. Although the villages were selected only after extensive research of the regions, districts, and villages, involving an in-depth review of all available demographic, social, and

5

6

associations, it was impossible to use all of the available information without having many missing cases. The above variables pick up the highest values from any of the associations to which the respondent belongs to capture the overall participation level of the respondent rather than the participation within a particular organization. Factor analysis indicates that the index explains over 67.9 percent of the variance in the data. A reliability analysis shows a standardized item alpha of 0.8376 thus, the index can be interpreted as both reliable and valid. Since the standardized alpha was higher than the unstandardized one, the z-score index was retained. Putnam (1993) continues the argument another step by saying that the density of social networks contributes to democratic governance and economic development. Marital status was included in the model as a set of dummy variables for the following: married; divorced; and widowed. The referent category was: single.

256

Additional Notes on Survey Sampling and Statistical Data Analysis

economic statistics and consultation with national, regional, and local government officials and residents, the possibility remains that the two villages chosen in each case were unique in some way and therefore confounded the effects of country.7 Interviewer effects could also shape the results. Due to the different language requirements in each country, I needed to use different sets of research assistants to help conduct the fieldwork. In order to prevent interviewer bias, assistants were selected who shared similar social and educational backgrounds, and they were given similar training, including group discussion of pilot tests of the survey questionnaire. Again, despite these preventive efforts, different interviewers might still skew the results, making it difficult to interpret the effect of country differences. In order to control for the possibility of village and interviewer effects, the preliminary round of analysis of the multiple regression model was run as three separate equations. All the other theorized independent variables were run together with village dummy variables in the first set, interviewer dummy variables in the second set, and a country dummy variable in the third set. In this way, the results indicated the predicted impact of village, interviewer, or country when one controls for all of the other variables. Then, the coefficients of the three alternative sets of variables were compared in tandem. When the results from the village and interviewer equations were rank ordered, they clearly clustered together by country. (See Table A.1.) Furthermore, the distance across the country divide was bigger than the distance within the country. The Ghanaian villages and interviewers differed more from their Ivoirian counterparts than each other. Since country does appear to be driving the results rather than village or interviewer effects, country was retained in subsequent rounds of analyses, and the village and interviewer variables were dropped. Thus, although the later versions of the model do not control explicitly for village or interviewer effects, the first round of analysis demonstrated clearly that these effects were not significant. See Tables A.2., A.3., A.4., and A.5. for the final regression results discussed in Chapter 2.

7

For example, Ghana village A might be more similar to Côte d’Ivoire village C than to Ghana village B, but since Ghana village B and Côte d’Ivoire village D are so different, there appears to be a country difference.

Additional Notes on Survey Sampling and Statistical Data Analysis t a b l e a . 1 . Comparative Analysis of Country, Village, and Interviewer Effects on the Total Value of Help Given to All Social Ties Country

Village

−100 −50 0 Ghana 0 50 100 109.58 Côte d’Ivoire

Interviewer −114.04 Faustie (GH) −101.48 Kweku (GH)

−6.70 Barima (GH) 0 Mankwan (GH) 100.70 Opanin (CI) 111.52 Kyere (CI)

0 Célestin (CI) 5.56 Fulgence (CI)

257

258

Additional Notes on Survey Sampling and Statistical Data Analysis

t a b l e a . 2 . Regression Model Results for Value of Help Given to All Social Ties Dependent Variable: Value of help given to all social ties for hosp, rx, school fees, school supplies, and clothing in U.S.$ Independent Variables

Unstandardized Coefficients

Standard Errors

Country (Côte d’Ivoire)

131.39***

19.33

Ability to Give Help Wealth index Country X wealth interaction Education level Number of children with SSS education or higher Number of children who earn a salary Age

−1.82 14.68*** 21.14*** 70.85*** −94.63*** −1.06

4.10 5.24 5.80 12.94 17.10 0.878

22.54*** −41.47** −17.31 −0.90 2.06*

2.81 18.56 23.34 0.80 1.24

9.19 0.77 −70.52*

18.53 0.65 39.12

Opportunity to Give Help Number of children Female gender Akan ethnicity Activity level of lineage Index of participation in local voluntary organizations Married Length of residence in village Constant Overall Model R² N Notes: Significance level: *** p < = 0.01;

0.46 346

Additional Notes on Survey Sampling and Statistical Data Analysis

259

t a b l e a . 3 . Regression Model Results for Value of Help Given to the Nuclear Family Only Dependent Variable: Value of help given to the nuclear family for hosp, rx, school fees, school supplies, and clothing in U.S.$ Independent Variables

Unstandardized Coefficients

Standard Errors

Country (Côte d’Ivoire)

106.84***

17.41

Ability to Give Help Wealth index Country X wealth interaction Education level Number of children with SSS education or higher Number of children who earn a salary Age

−3.14 12.45*** 17.89*** 66.36*** −84.71*** −1.45*

3.70 4.72 5.23 11.65 15.40 0.79

21.77*** −37.02** −25.73 −0.99 1.53

2.53 16.72 21.02 0.723 1.12

11.45 0.92 −55.90

16.68 0.58 35.23

Opportunity to Give Help Number of children Female gender Akan ethnicity Activity level of lineage Index of participation in local voluntary organizations Married Length of residence in village Constant Overall Model R² N Notes: Significance level: *** p < = 0.01; ** p < = 0.05; * p < = 0.10.

0.45 346

260

Additional Notes on Survey Sampling and Statistical Data Analysis

t a b l e a . 4 . Regression Model Results for Value of Help Given to the Extended Family Only Dependent Variable: Value of help given to the extended family for hosp, rx, school fees, school supplies, and clothing in U.S.$ Independent Variables

Unstandardized Coefficients

Standard Errors

Country (Côte d’Ivoire)

21.69***

6.14

Ability to Give Help Wealth index Country X wealth interaction Education level Number of children with SSS education or higher Number of children who earn a salary Age

1.02 1.36 1.32 0.13 −5.21 0.31

1.30 1.66 1.84 4.12 5.43 0.30

0.107 −5.81 11.17 −0.06 0.60

0.89 5.90 7.41 0.26 0.40

Opportunity to Give Help Number of children Female gender Akan ethnicity Activity level of lineage Index of participation in local voluntary organizations Married Length of residence in village Constant

0.87 0.05 −15.54

Overall Model R² N

0.11 346

Notes: Significance level: *** p < = 0.01; ** p < = 0.05; * p < = 0.10.

5.88 0.21 12.43

Additional Notes on Survey Sampling and Statistical Data Analysis

261

t a b l e a . 5 . Regression Model Results for Value of Help Given to All Social Ties Split by Country Dependent Variable: Value of help given to all social ties for hosp, rx, school fees, school supplies and clothing in U.S.$ Ghanaian Region Independent Variables Ability to Give Help Wealth index Education level Number of children with SSS education or higher Number of children who earn a salary Age Opportunity to Give Help Number of children Female gender Akan ethnicity Activity level of lineage Index of participation in local voluntary organizations Married Length of residence in village Constant Overall Model R² N Notes: Significance level: *** p < = 0.01; ** p < = 0.05; * p < = 0.10.

Ivoirian Region

Unstandardized Standard Unstandardized Coefficients Errors Coefficients

Standard Errors

−1.48 12.45** 80.44***

2.74 5.21 15.23

6.50 32.73*** 58.25***

4.88 10.55 21.76

−64.19*

38.34

−112.89***

23.02

−1.82**

0.82

0.12

1.52

15.22*** −15.98 −5.70 −0.49 0.96

2.79 16.88 21.61 0.71 1.04

29.45*** −65.17* −18.75 −0.20 2.25

4.60 33.23 40.09 1.46 2.52

−15.94 −0.41

16.99 0.593

35.95 2.51**

32.11 1.15

26.92

33.89

−54.95

58.11

0.39 181

0.52 164

Bibliography

Archival Documents British Public Records Office (PRO). 1943. Social Insurance, West Africa, CO 859/ 78/13, London. French National Archives (FNA). 1911. Questionnaire-Programme du 3e Congrès de lamutualité coloniale et des pays de protectorat, Serie Q/60, 200 Mi 1558, C.A. R.A.N., Paris. 1932–1947. Succession de la chefferie de l’Indénié, 5G/38, 200 Mi 2124, C.A.R.A. N., Paris. 1951–1952. Rapport Bi-annuel sur la protection de la collectivité, de la famille, et de l’enfance, Serie 2 G 52/3, 200 Mi 1945, C.A.R.A.N., Paris. Ghana National Archives (GNA). 1931. Annual Report on the Social and Economic Progress of the People of the Gold Coast, ADM 5/1/108, Accra. 1938. Report of the Commission on the Marketing of West African Cocoa, September 5, 1938, ADM 5/3/33, Accra. 1945. Correspondence Relating to Paper No. 5-Social Security in the Colonial Territories, ADM 5/51, Accra. Plan for Mass Literacy and Mass Education. Department of Social Welfare, Gold Coast. Ghana National Archives Kumasi (GNAK). n.d. Village Dispensaries, ARG 1/14/ 116. 1927–1939. Cocoa Farms in Ashanti, ARG 1/12/1/31, Kumasi. 1931. School Fees, ARG 1/13/22. 1943. “Secondary Education for Girls,” Report of the Commission Appointed by the SYNOD of the Methodist Church in the Gold Coast, January 1943. 1946. Report of the Committee on Higher Education. Accra, Ghana: Government Printing Department. 1948. Social Welfare in the Colonies. Ivoirian National Archives (INA). 1905–1933. Rapports sur la situation économique agricole et zootechnique. Cercle de l’Indénié, 1RR38, Abidjan.

263

264

Bibliography

1915. Arrêté promulguant en AOF le décret réorganisant les Sociétés indigènes de prévoyance de secours et de prêts mutuels agricoles en Afrique Occidentale française, Abidjan. 1936. Lettre du comité indigène du cercle de l’Indénié au Ministre des Colonies, Côte d’Ivoire, cited in appendix of Jean-Noel Loucou, Ph.D. Thesis, Abidjan. Senegalese National Archives (SNA). 1938. Evenements de Gold Coast. Lettre du Ministre des Colonies a Monsieur le Gouverneur General de l’AOF, February 17, 1938, 7G 249 (108), Dakar. Other Publications Acemoglu, Daron and James A. Robinson. 2006. “De Facto Political Power and Institutional Persistence.” American Economic Review 96 (2): 326–30. Acemoglu, Daron, Simon Johnson and James A. Robinson. 2001. “The Colonial Origins of Comparative Development: An Empirical Investigation.” The American Economic Review 91 (5): 1369–401. Ackah, C. A. 1988. Akan Ethics: A Study of the Moral Ideas and the Moral Behaviour of the Akan Tribes of Ghana. Accra, Ghana: Ghana Universities Press. Acquaah, Benjamin. 1999. Cocoa Development in West Africa: The Early Period with Particular Reference to Ghana. Accra, Ghana: Ghana Universities Press. Adams, Glenn and Victoria Plaut. 2003. “The Cultural Grounding of Personal Relationship: Friendship in North American and West African Worlds.” Personal Relationships 10: 333–47. Adams, Richard H. 1991a. “The Economic Uses and Impact of International Remittances in Rural Egypt.” Economic Development and Cultural Change 39 (4): 695–722. 1991b. The Effects of International Remittances on Poverty, Inequality, and Development in Rural Egypt. Research Report No. 86. Washington, DC: International Food Policy Research Institute. Adcock, Robert and David Collier. 2001. “Measurement Validity: A Shared Standard for Qualitative and Quantitative Research.” American Political Science Review 95 (3): 529–47. Addison, Ernest K. Y. 2006. “The Macroeconomic Impact of Remittances.” In At Home in the World?: International Migration and Development in Contemporary Ghana and West Africa, ed., Takyiwaa Manuh. Legon, Ghana: Sub-Saharan Publishers, 118–38. Ahmad, Ehtisham, Jean Dreze, John Hills and Amartya Sen, eds. 1991. Social Security in Developing Countries. Oxford: Clarendon Press. Akuamoah-Boateng, R. 2004. “Retirement and Poverty Among the Aged in Africa– The Ghanaian Experience.” In Africa Through Ghanaian Lenses, ed., Samuel E. Quainoo. Binghamton, NY: Vestal International Press. Albarran, P. and O. Attanasio. 2002. Do Public Transfers Crowd Out Private Transfers? Evidence from a Randomized Experiment in Mexico. Discussion Paper No. 2002/6. Helsinki: World Institute for Development Economics Research. Alderman, Harold and Christina H. Paxson. 1992. “Do the Poor Insure?: A Synthesis of the Literature on Risk and Consumption in Developing Countries.” Policy

Bibliography

265

Research Working Papers No. 1008. Agriculture and Rural Development Department, World Bank. Alence, Rod. 1990. “The 1937–38 Gold Coast Cocoa Crisis: The Political Economy of Commercial Stalemate.” African Economic History 19: 77–104. Alexandre, Pierre. 1970. “Chieftancy in French West Africa.” In West African Chiefs, ed., Michael Crowder. London: Hutchinson & Co. Amoako-Tuffour, Joe and Bartholomew Armah. 2008. Poverty Reduction Strategies in Action: Perspectives and Lessons from Ghana. New York, NY: Lexington Books. Amon D’Aby, Francois-Joseph. 1988. Le Probleme des Chefferies Traditionnelles en Cote d’Ivoire. Abidjan: Les Nouvelles Editions Africaines. Anderson, Benedict. 1983. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso. Anderson, Jock R. 2001. “Risk Management in Rural Development: A Review.” Rural Strategy Background Paper No. 7. World Bank. Annan-Yao, Elizabeth. 1997. “Reproduction Economique–Sociale et Democratization: Les Effets du Programme d’Ajustement Structurale (PAS) et de la Devaluation Monetaire sur les Femmes en Cote d’Ivoire.” Unpublished paper. Apt, Nana Araba. 1996. Coping with Old Age in a Changing Africa: Social Change and the Elderly Ghanaian. Surrey, United Kingdom: Ashgate Publishing. Apter, David. 1965. The Politics of Modernization. Chicago, IL: University of Chicago Press. Archibugi, Daniele and David Held. 1995. Cosmopolitan Democracy: An Agenda for a New World Order. New York, NY: Polity Press. Ardayfio-Schandorf, E., ed. 1996. The Changing Family in Ghana. Accra, Ghana: Ghana Universities Press. Arhin, Kwame. 1985. The Expansion of Cocoa Production: The Working Conditions of Migrant Cocoa Farmers in the Central and Western Regions. Unpublished report. 1976. “The Pressure of Cash and its Political Consequences in Asante in the Colonial Period.” Journal of African Studies 3 (4): 253–368. Arrow, Kenneth J. 1971. Essay in Theory of Risk-Bearing. Amsterdam, The Netherlands: North Holland. Aryeetey, Ernest and Markus Goldstein. 2000. “Ghana: Social Policy Reform in Africa.” In Reforming Social Policy: Changing Perspectives on Sustainable Human Development, eds., A. Daniel, Morales-Gomez, Necla Tschirgi and Jennifer L. Moher. Ottawa, ON: International Development Research Centre, 9–44. Asamoah, Yvonne and D. N. A. Nortey. 1987. “Ghana.” In Social Welfare in Africa, ed., John Dixon. New York, NY: Croom Helm, 22–68. Asante, Ben, ed. 1987. The Growth of Social Security in Liberia. London: PanAfrican Communications Services. Assimeng, Max. 1989. Religion and Social Change in West Africa: An Introduction to the Sociology of Religion. Accra, Ghana: Ghana Universities Press. Associates in Rural Development. 1991. The Experience in Ivory Coast with Decentralized Approaches to Local Delivery of Primary Education and Primary Health Services. Washington, DC: United States Agency for International Development.

266

Bibliography

Attanasio, Orazio and Jose-Victor Rios-Rull. 2000. “Consumption Smoothing in Island Economies: Can Public Insurance Reduce Welfare?” European Economic Review 44: 1225–58. Austin, Gareth. 1988. “Capitalists and Chiefs in the Cocoa Hold-Ups in South Asante, 1927–38.” International Journal of African Historical Studies 21: 63– 95. 2005. Labour, Land and Capital in Ghana: From Slavery to Free Labour in Asante, 1807–1956. Rochester, NY: University of Rochester Press. Ayee, Joseph R. A. 1994. An Anatomy of Public Policy Implementation: The Case of Decentralization Policies in Ghana. Avebury: Aldershot Press. 2004. “An Overview of Local Government Experience in Ghana.” In Africa Through Ghanaian Lenses, ed., Samuel E. Quainoo. Binghamton, NY: Vestal International Press. Azarya, Victor and Naomi Chazan. 1987. “Disengagement from the State in Africa: Reflections on the Experience of Ghana and Guinea.” Comparative Studies in Society and History 29 (1): 106–31. Bakary, T. 1997. “Political Polarization Over Governance in Côte d’Ivoire.” In Governance as Conflict Management: Politics and Violence in West Africa, ed., I. William Zartman. Washington, DC: Brookings Institution Press, 9–48. Bakayoko, Adama and Sylvestre Ehouman. 1987. “Ivory Coast.” In Social Welfare in Africa, ed., John Dixon. New York, NY: Croom Helm, 69–99. Bannerman-Richter, Gabriel. 1982. The Practice of Witchcraft in Ghana. Winona, MN: Apollo Books. Bantig, Keith. 1999. “Social Citizenship and the Multicultural Welfare State.” In Citizenship, Diversity and Pluralism: Canadian and Comparative Perspectives, eds., Alan C. Cairns, John C. Courtney, Peter MacKinnon, Hans J. Michelmann and David Smith. Montreal, QC: McGill-Queen’s University Press. Bardhan, Pranab. 1989. “The New Institutional Economics and Development Theory: A Brief Critical Assessment.” World Development 17 (9): 1333–47. 2004. “History, Institutions and Underdevelopment.” In Scarcity, Conflicts and Cooperation: Essays in the Political and Institutional Economics of Development. Cambridge, MA: MIT Press. Barker, Jonathan. 1989. Rural Communities Under Stress: Peasant Farmers and the State in Africa. New York, NY: Cambridge University Press. Bates, Robert. 1981. Markets and States in Tropical Africa: The Political Basis of Agricultural Policies. Berkeley, CA: University of California Press. 1989. Beyond the Miracle of the Market: The Political Economy of Agrarian Development in Kenya. New York, NY: Cambridge University Press. 1990. “Capital, Kinship and Conflict.” Canadian Journal of African Studies 24: 151–64. 1997. “Social Dilemmas and Rational Individuals: An Assessment of the New Institutionalism.” In New Institutional Economics and Third World Development, eds., J. Harriss, J. Hunter and C. M. Lewis. New York, NY: Routledge, 27–48. Bates, Robert and Amy Farmer Curry. 1992. “Community Versus Market: A Note on Corporate Villages.” American Political Science Review 86 (2): 457–63.

Bibliography

267

Bayart, Jean-François. 1993. The State in Africa: The Politics of the Belly. New York, NY: Longman Group Publishers. Bayart, Jean François, Stephen Ellis and Beatrice Hibou. 1999. The Criminalization of the State in Africa. Oxford and Bloomington: James Currey Press and Indiana University Press. Beckett, W. H. 1944. Akokoaso: A Survey of a Gold Coast Village. London: London School of Economics and Political Science. Beckman, Bjorn. 1976. Organising the Farmer: Cocoa Politics and National Development in Ghana. Uppsala: Scandinavian Institute of African Studies. Benda-Beckmann, F. von, K. von Benda-Beckmann, E. Casino, F. Hirtz, G. Woodman and H. F. Zacher, eds. 1988. Between Kinship and the State: Social Security and Law in Developing Countries. Dordrecht, The Netherlands: Foris Publications. Berger, Suzanne and Ronald Dore, eds. 1996. National Diversity and Global Capitalism. Ithaca, NY: Cornell University Press. Berman, Sheri. 2001. “Civil Society and Political Institutionalization.” In Beyond Tocqueville: Civil Society and the Social Capital Debate in Comparative Perspective, eds., Bob Edwards, Michael W. Foley and Mario Diani. Hanover, NH: University Press of New England, 32–42. Berry, L. 1995. Ghana: A Country Study. Washington, DC: Federal Research Division. Berry, Sara. 1985. Fathers Work for their Sons: Accumulation, Mobility, and Class Formation in an Extended Yoruba Community. Berkeley, CA: University of California Press. 1989. “Social institutions and Access to Resources.” Africa 59 (1): 41–55. 1993. No Condition is Permanent: The Social Dynamics of Agrarian Change in Sub-Saharan Africa. Madison, WI: The University of Wisconsin Press. 1997. “Tomatoes, Land and Hearsay: Property and History in Asante in the Time of Structural Adjustment.” World Development 25 (8): 1225–41. Blau, Peter M. 1964. Exchange and Power in Social Life. New York, NY: J. Wiley. Boafo-Arthur, K. 1998. “Party Organization, Finance and the Democratic Process: The Case of the Opposition Parties.” In The 1996 General Elections and Democratic Consolidation in Ghana, ed., Joseph R. A. Ayee. Legon, Ghana: University of Ghana. Boone, Catherine. 1995. “The Social Origins of Ivoirian Exceptionalism: Rural Society and State Formation.” Comparative Politics 27 (4): 445–63. 2003. Political Topographies of the African State. New York, NY: Cambridge University Press. Bortei-Doku Aryeetey, Ellen and Ernest Aryeetey. 1996. Operation, Utilization and Change in Rotating Susu Savings in Ghana. Technical Publication No. 59. Accra, Ghana: ISSER. Bourdieu, P. 1986. “The Forms of Capital.” In Handbook of Theory and Research for the Sociology of Education, ed., J. Richardson. New York, NY: Greenwood Press. Bratton, Michael. 1989. “Beyond the State: Civil Society and Associational Life in Africa.” World Politics 31 (3): 407–30.

268

Bibliography

1994. “Peasant-State Relations in Post-Colonial Africa.” In State Power and Social Forces: Domination and Transformation in the Third World, eds., Joel Migdal, A. Kohli and V. Shu. New York, NY: Cambridge University Press, 231–54. 2008. “Formal Versus Informal Institutions in Africa.” In How People View Democracy, eds., Larry Diamond and Marc F. Plattner. Baltimore, MD: The Johns Hopkins University Press. Bratton, Michael and Nicolas van de Walle. 1992. “Popular Protest and Political Transition in Africa.” Comparative Politics 24 (4): 419–22. Bratton, Michael, Robert Mattes and E. Gyimah Boadi. 2005. Public Opinion, Market Reform and Democracy in Africa. New York, NY: Cambridge University Press. Browne, J. Orde. 1935. “Witchcraft and British Colonial Law.” Journal of the International African Institute 8 (4): 481–7. Brunet-Jailly, Joseph. 1997. Innover dans les systèmes de santé: Expériences d’Afrique de l’Ouest. Paris: Karthala. Busia, K. A. 1951. The Position of the Chief in the Modern Political System of Ashanti. London: Oxford University Press. Cammett, Melani and Lauren M. MacLean, eds. n.d. The Politics of Non-State Social Welfare Provision in the Global South. Campbell, Andrea. 2003. How Policies Make Citizens: Senior Citizen Activism and the American Welfare State. Princeton, NJ: Princeton University Press. Carey, John M. 2000. “Parchment, Equilibria, and Institutions.” Comparative Political Studies 33 (6/7): 735–61. Chabal, Patrick and Jean-Pascal Daloz. 1999. Africa Works: Disorder as Political Instrument. Oxford: James Currey Press and Indiana University Press. Chaudhry, Kiren. 1993. “The Myths of the Market and the Common History of Late Developers.” Politics and Society 21 (3): 245–74. Chauveau, Jean-Pierre and Jean-Pierre Dozon. 1988. “Ethnies et Etat en Cote d’Ivoire.” Revue Francaise de Science Politique 38 (5) (October): 732–47. Chazan, Naomi. 1983. An Anatomy of Ghanaian Politics: Managing Political Recession, 1969–1982. Boulder, CO: Westview Press. 1988. “Patterns of State-Society Incorporation and Disengagement.” In The Precarious Balance: State and Society in Africa, eds., Donald Rothchild and Naomi Chazan. Boulder, CO: Westview Press. 1994. “Engaging the State: Associational Life in Sub-Saharan Africa. In State Power and Social Forces: Domination and Transformation in the Third World, eds., Joel Migdal, Atul Kohli and Vivienne Shue. New York, NY: Cambridge University Press. Chen, Shaohua and Martin Ravallion. 2008. The Developing World is Poorer Than We Thought, But No Less Successful in the Fight Against Poverty. Policy Research Working Paper No. 4703. Washington, DC: Development Research Group, The World Bank. Cheru, Fantu. 1997. “The Silent Revolution and the Weapons of the Weak: Transformation and Innovation from Below.” In Innovation and Transformation in International Studies, eds., Stephen Gill and James H. Mittelman. New York, NY: Cambridge University Press, 153–69.

Bibliography

269

Clark, Gracia. 1994. Onions are My Husband: Survival and Accumulation by West African Market Women. Chicago, IL: University of Chicago Press. Clignet, Remi and Philip Foster. 1966. The Fortunate Few: A Study of Secondary Schools and Students in the Ivory Coast. Evanston, IL: Northwestern University Press. Coleman, James. 1988. “Social Capital in the Creation of Human Capital.” American Journal of Sociology 94: 95. 1990. The Foundations of Social Theory. Cambridge, MA: Harvard University Press. Coleman, Jonathan R., Takamasa Akiyama and Panos Varangis. 1993. How Policy Changes Affected Cocoa Sectors in Sub-Saharan African Countries. Working Paper No. 1129. Washington, DC: World Bank, Policy Research on International Trade. Colson, Elizabeth. 1974. Tradition and Contract. Chicago, IL: Aldine Publishing Co. Collier, Paul and Ashish Garg. 1999. “On Kin Groups and Wages in the Ghanaian Labour Market.” Oxford Bulletin of Economics and Statistics 61 (2): 133–51. Collier, Ruth Berins and David Collier. 1991. Shaping the Political Arena: Critical Junctures, the Labor Movement, and Regime Dynamics in Latin America. Princeton, NJ: Princeton University Press. Comaroff, J. and Comaroff, J. L. 1999. “Occult Economies and the Violence of Abstraction: Notes from the South African Postcolony.” American Ethnologist 26 (2): 279–303. 2009. Ethnicity, Inc. Chicago, IL: University of Chicago Press. Commander, Simon, John Howell and Wayo Seini. 1989. “Ghana: 1983–7.” In Structural Adjustment and Agriculture: Theory and Practice in Africa and Latin America, ed., Simon Commander. London: Overseas Development Institute. Cooper, Frederick. 1996. Decolonization and African Society: The Labor Question in French and British Africa. New York, NY: Cambridge University Press. 2002. Africa Since 1940: The Past of the Present. New York, NY: Cambridge University Press. Coulibaly, Loucoumane and David Lewis. 2009. “Ivory Coast Election Chief Confirms Poll Delay.” Reuters, November 11. http://www.reuters.com/article/ latestCrisis/idUSLB279965 Couret, Dominique, Jonas Ibo and Marc Le Pape, eds. 1994. Crises, Ajustements et Recompositions en Côte d’Ivoire: La Remise en Cause d’Un Modèle. International Colloquium, November 28–December 2, 1994. Abidjan: ORSTOM. Crawford, Sue E. and Elinor Ostrom. 1995. “A Grammar of Institutions.” American Political Science Review 89 (3): 582–600. Crook, Richard. 1989. “Patrimonialism, Administrative Effectiveness and Economic Development in Cote d’Ivoire.” African Affairs 88 (351): 205–28. 1990. “State, Society and Political Institutions in Côte d’Ivoire and Ghana.” IDS Bulletin 21 (4): 24–3. 1994. “Four Years of the Ghana District Assemblies in Operation: Decentralization, Democratization and Administrative Performance.” Public Administration and Development 14 (4): 339–64.

270

Bibliography

Crook, Richard C. and James Manor. 1998. Democracy and Decentralization in South Asia and West Africa: Participation, Accountability and Performance. New York, NY: Cambridge University Press. Crowder, Michael. 1968. West Africa Under Colonial Rule. London: Hutchinson & Co. Crowder, Michael and Obaro Ikeme, eds. 1970. West African Chiefs: Their Changing Status Under Colonial Rule and Independence. London: Hutchinson & Co. Danquah, J. B. 1928. Gold Coast: Akan Laws and Customs and the Akim Abuakwa Constitution. London: Routledge. Obligation in Akan Society (West African Pamphlet Series No. 8). Dasgupta, Partha and Ismail Serageldin. 2000. Social Capital: A Multifaceted Perspective. Washington, DC: World Bank. Deaton, Angus. 1992. “Household Saving in LDCs: Credit Markets, Insurance and Welfare.” The Scandinavian Journal of Economics 94 (2): 253–73. Deaton, Angus and Dwayne Benjamin. 1988. The Living Standard Survey and Price Policy Reform: A Study of Cocoa and Coffee Production in Côte d’Ivoire. Washington, DC: World Bank. Dei, George J. Sefa. 1992. Hardships and Survival in Rural West Africa: A Case Study of a Ghanaian Community. Dakar: CODESRIA. Demery, Lionel. 1994. “Côte d’Ivoire: Fettered Adjustment.” In Adjustment in Africa: Lessons from Country Case Studies, eds., Ishrat Husain and Rashid Faruqee. Washington, DC: World Bank, 72–152. Demery, L., S. Chao, R. Bernier and K. Mehra. 1995. The Incidence of Social Spending in Ghana. Poverty and Social Policy Discussion Paper No. 82. Washington, DC: World Bank. Dercon, S. 2002. “Income Risk, Coping Strategies and Safety Nets.” World Bank Research Observer 17: 141–66. 2003. “Poverty Traps and Development: The Equity-Efficiency Trade-Off Revisited.” Unpublished paper. Dercon, Stefan and Pramila Krishnan. 1996. “Income Portfolios in Rural Ethiopia and Tanzania: Choices and Constraints.” The Journal of Development Studies 32 (6): 850–75. 1997. “In Sickness and in Health . . . Risk-Sharing Within Households in Rural Ethiopia.” Working Paper Series No. 97–12, April. Devereux, Stephen. 2001. Social Pensions in Namibia and South Africa. IDS Discussion Paper No. 379, February 2001. Brighton, England: Institute of Development Studies. DiMaggio, Paul J. 1998. “The New Institutionalisms: Avenues of Collaboration.” Journal of Institutional and Theoretical Economics 154 (4): 696–705. Dixon, John, ed. 1987. Social Welfare in Africa. New York, NY: Croom Helm. 1999. Social Security in Global Perspective. Westport, CT: Praeger. Douglas, Mary. 1985. Risk Acceptability According to the Social Sciences. New York, NY: Russel Sage Foundation. Duflo, Esther and Christopher Udry. 2004. “Intrahousehold Resource Allocation in Cote d’Ivoire: Social Norms, Separate Accounts and Consumption Choices.” Unpublished paper. Dunn, John and A. F. Robertson. 1973. Dependence and Opportunity: Political Change in Ahafo. New York, NY: Cambridge University Press.

Bibliography

271

Dunning, Thad. 2008. “Improving Causal Inference: Strengths and Limitations of Natural Experiments.” Political Research Quarterly 61: 282–93. Ekeh, Peter. 1975. “Colonialism and the Two Publics in Africa: A Theoretical Statement.” Comparative Studies in Society and History 17 (1): 91–112. Engerman, Stanley L. and Kenneth L. Sokoloff. 2002. “Factor Endowments, Inequality, and Paths of Development Among New World Economies.” Economia 3 (1): 41–109. Ensminger, Jean. 1992. Making a Market. Cambridge, MA: Cambridge University Press. Evans, Peter. 1996. “Government Action, Social Capital and Development Reviewing the Evidence on Synergy.” World Development 24 (6): 1033–7. Evans-Pritchard, E. E. 1937. Witchcraft, Oracles and Magic Among the Azande. Oxford: Clarendon Press. Eyoh, Dickson. 1998. “Conflicting Narratives of Anglophone Protest and the Politics of Identity in Cameroon.” Journal of Contemporary African Studies 16 (2): 249– 76. Fadayomi, T. O. 1991. “The History of Social Development in Gambia, Ghana, Nigeria.” In Social Development in Africa: Strategies, Policies, and Programmes After the Lagos Plan, ed., Duri Mohammed. New York, NY: Hans Zell Publishers. Fafchamps, Marcel. 1992. “Solidarity Networks in Preindustrial Societies: Rational Peasants with a Moral Economy.” Economic Development and Cultural Change 41 (1): 147–76. Ferguson, James. 2006. Global Shadows: Africa in the Neoliberal World Order. Durham, NC: Duke University Press. Fine, Ben. 2001. Social Capital Versus Social Theory: Political Economy and Social Science at the Turn of the Millenium. New York, NY: Routledge. 2003. “Social Capital: The World Bank’s Fungible Friend.” Journal of Agrarian Change 3 (4): 586–603. Firmin-Sellers, Katherine. 1996. The Transformation of Property Rights in the Gold Coast: An Empirical Analysis Applying Rational Choice Theory. New York, NY: Cambridge University Press. 2000. “Institutions, Context, and Outcomes: Explaining French and British Rule in West Africa.” Comparative Politics 32 (3): 253–72. Fortes, Meyer. 1950. “Kinship and Marriage Among the Ashanti.” In African Systems of Kinship and Marriage, eds., A. R. Radcliffe-Brown and D. Fordepages. London: Oxford University Press. 1969. Kinship and the Social Order: The Legacy of Lewis Henry Morgan. London: Aldine Publishing Company. Foster, Philip and Aristide R. Zolberg. 1971. Ghana and the Ivory Coast: Perspectives on Modernization. Chicago, IL: The University of Chicago Press. Fraser, Nancy. 1989. “Women, Welfare and the Politics of Need Interpretation.” In Unruly Practices: Power, Discourse, and Gender in Contemporary Social Theory, ed., Nancy Fraser. Minneapolis, MN: University of Minnesota Press, 144–60. Frimpong-Ansah, A. 1991. The Vampire State in Africa: The Political Economy of Decline in Ghana. London: James Currey.

272

Bibliography

Fukuyama, Francis. 1995. Trust: The Social Virtues and the Creation of Prosperity. New York, NY: The Free Press. Galvan, Dennis. 2004. The State Must Be Our Master of Fire: How Peasants Craft Culturally Sustainable Development in Senegal. Berkeley, CA: University of California Press. George, Betty Stein. 1976. Education in Ghana. Washington, DC: U.S. Department of Health, Education and Welfare. Geschiere, Peter. 1993. “Chiefs and Colonial Rule in Cameroon: Investing in Chieftancy, French and British Style.” Africa 63 (2): 151–75. 1997. The Modernity of Witchcraft: Politics and the Occult in Postcolonial Africa. Charlottesville, VA: University of Virginia Press. 2009. The Perils of Belonging: Autochtony, Citizenship and Exclusion in Africa and Europe. Chicago, IL: University of Chicago Press. Geschiere, Peter and Francis Nyamnjoh. 2001. “Capitalism and Autochthony: The Seesaw of Mobility and Belonging.” In Millenial Capitalism and the Culture of Neoliberalism, eds., Jean Comaroff and John L. Comaroff. Durham, NC: Duke University Press, 159–90. Ginneken, Wouter van, ed. 1999. Social Security for the Excluded Majority: Case Studies of Developing Countries. Geneva, Switzerland: International Labour Office. Glewwe, Paul and Dennis de Tray. 1988. The Poor During Adjustment: A Case Study of Côte d’Ivoire. Washington, DC: World Bank. 1999. “Chop Time, No Friends: Intrahousehold and Individual Insurance Mechanisms in Southern Ghana.” Unpublished Article based on Ph.D. Dissertation, University of California at Berkeley. Gockel, Augustine Fritz. 1996. The Formal Social Security System in Ghana. Accra, Ghana: Friedrich Ebert Foundation. Goldstein, Markus, Alain de Janvry and Elisabeth Sadoulet. 2000. “You Can’t Always Get What You Want: Inclusion and Exclusion in Mutual Insurance Networks in Southern Ghana.” Unpublished paper. Gore, Charles with contributions from Jose B. Figueiredo and Gerry Rodgers. 1995. “Introduction: Markets, Citizenship and Social Exclusion,” In Social Exclusion: Rhetoric, Reality and Responses, eds., G. Rodgers, C. Gore and J. Figueiredo. Geneva, Switzerland: International Institute for Labour Studies, International Labour Organization. Gough, Ian, et.al. 2004. Insecurity and Welfare Regimes in Asia, Africa and Latin America: Social Policy in Development Contexts. New York, NY: Cambridge University Press. Government of Côte d’Ivoire (GOCI). 1993. Agissons Tous Ensemble Pour La Sauvegarde de la Famille. Abidjan: Ministere de la Famille et de la Promotion de la Femme. 1994. Rapport de Gestion 1994: CHR Abengourou. Abidjan: Ministere de la Sante Publique et des Affaires Sociales. 1996. Cout Estimatif d’un Eleve a la Rentree Scolaire, 1995–1996. Abidjan: Ministere de l’Education National et de la Formation de Base. 1997. Enquete Prioritaire Sur les Dimensions Sociales de l’Ajustement Structurel: Depenses des Menages, vol. 1. Abidjan: Institut National de la Statistique.

Bibliography

273

1999a. Profil de Pauvrete: 1998. Abidjan: Ministere de la Planification et de la Programmation du Developpement et Institut National de la Statistique. 1999b. Partageaons Ensemble le Risqué Maladie. Abidjan: Ministere de la Sante Publique. Government of Ghana (GOG). 1946. Report of the Committee on Higher Education. Accra, Ghana: Government Printing Department. 1993. Rural Communities in Ghana: Report of a National Rural Community Survey Carried Out as Part of the Third Round of the Ghana Living Standards Survey 1991/92. Accra, Ghana: Ghana Statistical Service. 1995. The Pattern of Poverty in Ghana, 1988–1992: A Study Based on the Ghana Living Standards Survey. Accra, Ghana: Ghana Statistical Service. 1996. Development Plan for Tano District (1996–2000). Bechem: Tano District Assembly, Ministry of Local Government and Rural Development. 1998a. Annual Evaluation Report, Tano District, January–December 1998. Bechem: Government of Ghana. 1998b. Unit Committees. Accra, Ghana: National Commission for Civic Education. Graeber, David. 2001. Toward an Anthropological Theory of Value: The False Coin of Our Own Dreams. New York, NY: Palgrave. Graham, C. 1971. The History of Education in Ghana. London: Frank Cass and Co., Ltd. Graham, Carol. 1994. Safety Nets, Politics and the Poor: Transitions to Market Economies. Washington, DC: The Brookings Institution. Granovetter, Mark. 1985. “Economic Action and Social Structure: The Problem of Embeddedness.” American Journal of Sociology 91 (3): 481–510. Gray, Natasha Adriene. 2000. The Legal History of Witchcraft in Colonial Ghana: Akyem Abuakwa, 1913–1943. PhD diss., Department of History, Columbia University. 2005. “The Politics of Policing Anti-Witchcraft Movements in Colonial Ghana, 1908–1927.” Journal of Religion in Africa 35 (2): 139–58. Greenwald, Bruce C. and J. Sitglitz. 1986. “Externalities in Economies with Imperfect Information and Incomplete Markets.” Quarterly Journal of Economics 101 (2): 229–64. Gregory, C. A. 1982. Gifts and Commodities. New York, NY: Academic Press. Greif, Avner. 1997. “Cultural Beliefs as a Common Resource in an Integrating World: An Example from the Theory and History of Collectivist and Individualist Societies.” In The Economics of Transnational Commons, eds., P. Dasgupta, K.-G. Mäler and A. Vercelli. Oxford: Clarendon Press. 1992. “Institutions and International Trade: Lessons from the Commercial Revolution.” American Economic Review 82 (2): 128–33. Grischow, Jeff. 1998. “Corruptions of Development in the Countryside of the Northern Territories of the Gold Coast, 1927–57.” The Journal of Peasant Studies 26 (1): 139–58. Grootaert, C. and T. van Bastelaer, eds. 2002a. The Role of Social Capital in Development: An Empirical Assessment. Cambridge, MA: Harvard University Press. 2002b. Understanding and Measuring Social Capital: A Multidisciplinary Tool for Practitioners. Washington, DC: World Bank.

274

Bibliography

Grootaert C., Deepa Narayan, Veronica Nyhan Jones and Michael Woolcock. 2004. Measuring Social Capital: An Integrated Questionnaire. World Bank Working Paper No. 18. Washington, DC: World Bank. Gruat, J. V. 1990. “Social Security Schemes in Africa: Current Trends and Problems.” International Labour Review 129: 405–21. Gupta, Akhil. 1995. “Blurred Boundaries: The Discourse of Corruption, the Culture of Politics, and the Imagined State.” American Ethnologist 22 (2): 375–402. Guyer, Jane I. 1995. Money Matters: Instability, Values and Social Payments in the Modern History of West African Communities. Portsmouth, NH: Heinemann. Guyer, Jane, LaRay Denzer and Adigun Agbaje, eds. 2002. Money Struggles and City Life. Portsmouth, NH: Heinemann. Gyimah-Boadi, E. 1993. Ghana Under PNDC Rule. Dakar: CODESRIA. 1995. “Ghana: Adjustment, State Rehabilitation and Democratization.” In Between Liberalisation and Oppression: The Politics of Structural Adjustment in Africa, eds., Thandika Mkandawire and Adebayo Olukoshi. Dakar: CODESRIA. Habyarimana, James, Macartan Humphreys, Daniel Posner and Jeremy Weinstein. 2007. “Why Does Ethnic Diversity Undermine Public Goods Provision?” American Political Science Review 101 (4): 709–25. Haggard, Stephan and Robert R. Kaufman. 2008. Development, Democracy and Welfare States: Latin America, East Asia and Eastern Europe. Princeton, NJ: Princeton University Press. Hailey, Lord. 1957. An African Survey: A Study of Problems Arising in Africa South of the Sahara. London: Oxford University Press. Hall, Peter A. and Rosemary Taylor. 1996. “Political Science and the Three New Institutionalisms.” Political Studies 44 (5): 936–57. Handloff, Robert, ed. 1991. Côte d’Ivoire: A Country Study. Washington, DC: Library of Congress. Hansen, Karen. 2004. Informality Reconsidered: Perspectives from Urban Africa. Uppsala: Nordic Africa Institute. Hanson, Holly Elizabeth. 2003. Landed Obligation: The Practice of Power in Buganda. Portsmouth, NH: Heineman. Hecht, R. M. 1983. “Ivory Coast Economic ‘Miracle’: What Benefits for Peasant Farmers?” Journal of Modern African Studies 21: 25–53. Helmke, Gretchen and Steven Levitsky. 2004. “Informal Institutions and Comparative Politics: A Research Agenda.” Perspectives on Politics 2 (4): 725–40. eds. 2006. Informal Institutions and Democracy: Lessons from Latin America. Baltimore, MD: Johns Hopkins University Press. Henry, Alain, Guy-Honoré Tchente and Philippe Guillerme-Dieumegard. 1991. Tontines et banques au Cameroun: Les principes de la Société des amis. Paris: Karthala. Herbst, Jeffrey. 2000. States and Power in Africa: Comparative Lessons in Authority and Control. Princeton, NJ: Princeton University Press. Heyman, Richard D., Robert F. Lawson and Robert M. Stamp. 1972. “The British Model of Education in Her African Territories.” Studies in Educational Change. Minneapolis, MN: Winston Press.

Bibliography

275

Hill, Polly. 1963. The Migrant Cocoa Farmers of Southern Ghana: A Study in Rural Capitalism. London: Cambridge University Press. Hills, John, Julian Le Grand and David Piachaud, eds. 2002. Understanding Social Exclusion. New York, NY: Oxford University Press. Hirtz, Frank. 1995. Managing Insecurity: State Social Policy and Family Networks in the Rural Philippines. Saarbrucken: Verlag fur Enwicklungspolitik Breitenbach GmbH. Hochschild, Adam. 1998. King Leopold’s Ghost: A Story of Greed, Terror and Heroism in Colonial Africa. New York, NY: Houghton Mifflin Co. Hoon, Parakh. 2005. Between Exchange and Reciprocity?: The Politics of Wildlife Conservation in Botswana and Agricultural Diversification in Zambia. Ph.D. Dissertation, University of Florida. 2009. “Working is Celebrating: Institutional Creativity among the Ngonis in Eastern Zambia.” Paper presented at the annual meeting of the African Studies Association, New Orleans, LA, November 2009. Hukansson, Thomas. 1989. “Family Structure, Bridewealth, and Environment in Eastern Africa: A Comparative Study of House-Property Systems.” Ethnology 28: 117–33. Hyden, Goran. 1980. Beyond Ujaama in Tanzania: Underdevelopment and an Uncaptured Peasantry. Berkeley, CA: University of California Press. 1983. No Shortcuts to Progress, African Development Management in Perspective. Berkeley, CA: University of California Press. 2006. African Politics in Comparative Perspective. New York, NY: Cambridge University Press. Ihonvbere, Julius. 1994. “The ‘Irrelevant’ State, Ethnicity, and the Quest for Nationhood in Africa.” Ethnic and Racial Studies 17 (1): 42–60. Immergut, Ellen. 1992. “The Rules of the Game: The Logic of Health Policy-Making in France, Switzerland, and Sweden.” In Structuring Politics: Historical Institutionalism in Comparative Analysis, eds., Sven Steinmo, Kathleen Thelen and Frank Longstreth. Cambridge, MA: Cambridge University Press. International Labor Office (ILO). 2006. La Contribution des Systemes de Protection Sociale Dans La Lutte Contre La Pauvrete: Exemples de la Guinee, Du Mali, de la Mauritanie et du Senegal. Dakar: Bureau International du Travail. Institute of Statistical, Social and Economic Research (ISSER). 1996. National Health Insurance in Ghana: A Study of the Informal Sector. Accra, Ghana: ISSER. Jacob, J. P. and Ph. Lavigne Delville. 1994. Les associations paysannes en Afrique: Organisation et dynamiques. Paris: Karthala. Janin, Pierre. 1995. Immuable, Le Changeant et L’Imprevu: Les Economies de Plantation Bamilieke et Beti du Cameroun Confrontees aux Chocs Exterieurs. These de geographie, Universite de Paris IV, Sorbonne, Mai. Janoski, Thomas. 1998. Citizenship and Civil Society: A Framework of Rights and Obligations in Liberal, Traditional and Social Democratic Regimes. New York, NY: Cambridge University Press. Kahneman, Daniel and A. Tversky. 1979. “Prospect Theory: An Analysis of Decisions Under Risk.” Econometrica 47 (2): 263–91.

276

Bibliography

Kaseke, Edwin, ed. 1998. Social Security Systems in Rural Zimbabwe. Harare: Friedrich Ebert Stiftun. Kaufman, Robert and Joan M. Nelson, eds. 2004. Crucial Needs, Weak Incentives: Social Sector Reform, Globalization and Democratization in Latin America. Baltimore, MD: Woodrow Wilson International Center for Scholars and Johns Hopkins University Press. Kaufman, Robert and Alex Segura-Ubiergo. 2001. “Globalization, Domestic Politics, and Social Spending in Latin America: A Time-Series Cross Section Analysis, 1973–1997.” World Politics 53: 553–87. Kazianga, Harounan and Christopher Udry. 2006. “Consumption Smoothing?: Livestock, Insurance and Drought in Rural Burkina Faso.” Journal of Development Economics 79 (2): 413–46. Keck, Margaret and Kathryn Sikkink. 1998. Activists Beyond Borders: Advocacy Networks in International Politics. Ithaca, NY: Cornell University Press. Kipre, Pierre. 1987. La Cote d’Ivoire Coloniale. Tome Second. Abidjan: Edition Ami. Knight, Jack. 1992. Institutions and Social Conflict. Cambridge, MA: Cambridge University Press. Kochler, Hans. 1995. “The Concept of the Nation and the Question of Nationalism: The Traditional ‘Nation-State’ Versus a Multicultural ‘Community State.’” In Citizenship and Rights in Multicultural Societies, eds., Michael Dunne and Tiziano Bonazzi. Keele, Staffordshire: Keele University Press, 43–52. Konate, Hamadou. 1997. Politique Sociale en Afrique de l’Ouest et du Centre. Ottawa, ON: IDRC. Krishna, Anirudh. 2002. Active Social Capital: Tracing the Roots of Development and Democracy. New York, NY: Columbia University Press. Kumado, Kofi and Augustine Fritz Gockel. 2003. Social Security in Ghana. Accra, Ghana: Friedrich Ebert Foundation. Kwadzo, G. T. M. and M. L. Kumekpor. 1994. PAMSCAD Evaluation (CIDAsupported programs). Ottawa, ON: Canadian International Development Agency. Kymlicka, Will. 1995. Multicultural Citizenship: A Liberal Theory of Minority Rights. Oxford: Oxford University Press. Kymlicka, Will and Wayne Norman. 2000. Citizenship in Diverse Societies. Oxford: Oxford University Press. La Ferrara, Eliana. 2003. “Kin Groups and Reciprocity: A Model of Credit Transactions in Ghana.” The American Economic Review 93 (5): 1730–51. Landy, Marc. 1993. “Public Policy and Citizenship.” In Public Policy for Democracy, eds., Helen Ingram and Steven Rathgeb Smith. Washington, DC: Brookings Institution Press, 19–45. Lauth, Hans-Joachim. 2000. “Informal Institutions and Democracy.” Democratization 7 (4): 21–50. Leechor, Chad. 1994. “Ghana: Frontrunner in Adjustment.” In Adjustment in Africa: Lessons from Country Case Studies, eds., Ishrat Husain and Rashid Faruqee. Washington, DC: World Bank, 153–92.

Bibliography

277

Leliveld, Andre. 1994. Social Security in Developing Countries: Operation and Dynamics of Social Security Mechanisms in Rural Swaziland. Amsterdam, The Netherlands: Thesis Publishers. Leonard, David K., ed. 2000. Africa’s Changing Markets for Health and Veterinary Services: The New Institutional Issues. London: Macmillan. Leonard, David. 1982. “Analyzing the Organizational Requirements for Serving the Rural Poor.” In Institutions of Rural Development for the Poor: Decentralization and Organizational Linkages, eds., David Leonard and Dale Marshall. Berkeley, CA: Institute of International Studies. Lerner, Daniel. 1958. The Passing of Traditional Society. New York, NY: MacMillan Press. Levi, Margaret. 1996. “Asocial and Unsocial Capital.” Politics and Society 24: 45– 55. Levi-Strauss, Claude. 1969. The Elementary Structures of Kinship. Boston, MA: Beacon Press. Levy, Jonah. 2006. The State After Statism: New State Activities in the Age of Liberalization. Cambridge, MA: Harvard University Press. Lin, Nan. 2001. Social Capital: A Theory of Social Structure and Action. New York, NY: Cambridge University Press. Locoh, Therèse. 1993. “Les familles africaines fâce a` la crise.” Afrique contemporaine 166: 3–14. Logan, Carolyn and Michael Bratton. 2006. “Voters but Not Yet Citizens: The Weak Demand for Political Accountability in Africa’s Unclaimed Democracies.” Afrobarometer Working Paper No. 63, September 2006. Lugard, Frederic. 1926. The Dual Mandate in British Tropical Africa. Lyon, Fergus. 2000. “Trust, Networks and Norms: The Creation of Social Capital in Agricultural Economies in Ghana.” World Development 28 (4): 663–81. 2003. “Community Groups and Livelihoods in Remote Rural Areas of Ghana: How Small-Scale Farmers Sustain Collective Action.” Community Development Journal 38 (4): 323–31. MacGaffey, Janet. 1991. The Real Economy of Zaire: The Contribution of Smuggling and Other Unofficial Activities to the National Wealth. Philadelphia, PA: University of Philadelphia Press. MacLean. 1999. “The Politics of Poverty Alleviation in Ghana.” Briefing Paper, Center for Democracy and Development, 1 (2) (July), Accra, Ghana. 2002. “Constructing a Social Safety Net in Africa: An Institutionalist Analysis of Colonial Rule and State Social Policies in Ghana and Cote d’Ivoire.” Studies in Comparative International Development 37 (3): 64–90. 2003. “State Social Policies and Social Insurance Networks: The Unintended Consequences of State Policymaking on Informal Networks in Ghana and Cote d’Ivoire.” International Journal of Public Administration 26 (6): 665–91. 2004. “Empire of the Young: The Legacies of State Agricultural Policy on Local Capitalism and Social Support Networks in Ghana and Cote d’Ivoire.” Comparative Studies in Society and History 46 (3): 469–96. 2004. “Mediating Ethnic Conflict at the Grassroots: The Role of Local Associational Life in Shaping Political Values in Cote d’Ivoire and Ghana.” Journal of Modern African Studies 42 (4): 589–617.

278

Bibliography

Mahieu, François. 1990. Les Fondements de la Crise Economique en Afrique. Paris: l’Harmattan. 1993. “L’opposition villes-campagnes est-elle fatale?” Afrique contemporaine, no. speciale (4): 108–21. Malinowski, Bronislaw. 1950 [orig. 1922]. Argonauts of the Western Pacific: An Account of Native Enterprise and Adventure in the Archipelagoes of Melanesian New Guinea. New York, NY: E.P. Dutton & Co., Inc. Mamdani, Mahmood. 1996. Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism. Princeton, NJ: Princeton University Press. Manuh, Takyiwaa. 1994. “Ghana: Women in the Public and Informal Sectors Under the Economic Recovery Programme.” In Mortgaging Women’s Lives: Feminist Critiques of Structural Adjustment, ed., Pamela Sparr. London: Zed Books. 2006. At Home in the World?: International Migration and Development in Contemporary Ghana and West Africa. Legon, Ghana: SubSaharan Publishers. Marie, Alain. 1995. “ ‘Y a pas l’argent’: l’endette insolvable et le creancies floue, deux figures complementaires de la pauvrete abidjanaise.” Revue Tiers-Monde 36 (142): 303–24. 1997. L’Afrique des individus: Itinéraires citadins dans l’Afrique contemporaine (Abidjan, Bamako, Dakar, Niamey). Paris: Karthala. Marshall, T. H. and Tom Bottomore. 1992 [orig. 1950]. Citizenship and Social Class. Concord, CA: Pluto Press. Marx, Karl. 1978. [orig. 1843]. “On the Jewish Question.” In The Marx-Engels Reader, ed., Robert Tucker. New York, NY: W.W. Norton & Co, 26–52. 1978. [orig. 1852].“The Eighteenth Brumaire of Louis Bonaparte.” In The Marx-Engels Reader, ed., Robert Tucker. New York, NY: W.W. Norton & Co, 594–617. 1978. [orig. 1887]. “Capital.” In The Marx-Engels Reader, ed., Robert Tucker. New York, NY: W.W. Norton & Co, 294–442. 1978. [orig. 1845–1846]. “The German Ideology.” In The Marx-Engels Reader, ed., Robert Tucker. New York, NY: W.W. Norton & Co, 146–202. Mauss, M. 1954. [orig. 1925]. The Gift: Forms and Functions of Exchange in Archaic Societies. Glencoe, NM: The Free Press. Mazzucato, Valentina, Bart van den Boom and Nicholas N. N. Nsowah-Nuamah. 2006. “Origin and Destination of Remittances in Ghana.” In At Home in the World?: International Migration and Development in Contemporary Ghana and West Africa, ed., Takyiwaa Manuh. Legon, Ghana: Sub-Saharan Publishers, 139–52. McCaskie, T. C. 1983. “Accumulation, Wealth and Belief in Asante History: To the Close of the Nineteenth Century.” Africa 53 (1): 23–43. 1995a. “Konnurokusem: Kinship and Family in the History of the Oyoko KOkOO Dynasty of Kumase.” The Journal of African History 36 (3): 357–89. 1995b. State and Society in Pre-Colonial Asante. New York, NY: Cambridge University Press. 2001. Asante Identities: History and Modernity in an African Village, 1850–1950. Bloomington, IN: Indiana University Press. 2007. “Denkyira in the Making of Asante, 1660–1720.” Journal of African History 48: 1–25.

Bibliography

279

McIntire, John and Panos Varangis. “Reforming Cote d’Ivoire’s Cocoa Marketing and Pricing System.” Washington, DC: World Bank. Menocal, Alina Rocha. 2004. “And If There was No State?: Critical Reflections on Bates, Polanyi and Evans on the Role of the State in Promoting Development.” Third World Quarterly 25 (4): 765–77. Merry, Sally Engle. 1988. “Legal Pluralism,” Law and Society Review 22: 869–96. Mershon, Carol. 1994. “Expectations and Informal Rules in Coalition Formation.” Comparative Political Studies 27 (1): 40–79. Mesa-Lago, C. 1992. “Comparative Analysis of Asian and Latin American Social Security Systems.” In Rethinking Social Security: Reaching Out to the Poor, eds., I. P. Getubig and S. Schmidt. Kuala Lumpur: Asian & Pacific Development Center. Mettler, Suzanne. 2005. Soldiers to Citizens: The G.I. Bill and the Making of the Greatest Generation. New York, NY: Oxford University Press. Mettler, Suzanne and Joe Soss. 2004. “The Consequences of Public Policy for Democratic Citizenship: Bridging Policy Studies and Mass Politics.” Perspectives on Politics 2 (1): 55–73. Meyer, Birgit. 1995. “‘Delivered from the Powers of Darkness’: Confessions of Satanic Riches in Christian Ghana.” Africa 65 (2): 236–55. Midgley, James. 1994. “Social Security Policy in Developing Countries: Integrating State and Traditional Systems.” Focaal 22/23 (1): 219–30. Midgley, James and Edwin Kaseke. 1996. “Challenges to Social Security in Developing Countries: Coverage and Poverty in Zimbabwe.” In Challenges to Social Security: An International Exploration, eds., James Midgley and Martin Tracy. Westport, CT: Auburn House. Migdal, Joel. 1988. Strong Societies and Weak States: State-Society Relations and State Capabilities in the Third World. Princeton, NJ: Princeton University Press. 2001. State in Society: Studying How States and Societies Transform and Constitute One Another. New York, NY: Cambridge University Press. Migdal, Joel, A. Kohli and V. Shue, eds. 1994. State Power and Social Forces: Domination and Transformation in the Third World. New York, NY: Cambridge University Press. Mikell, Gwendoline. 1989. Cocoa and Chaos in Ghana. New York, NY: Paragon. Miles, John. 1978. “Rural Protest in the Gold Coast: The Cocoa Hold-Ups, 1907– 38.” In The Imperial Impact, eds., C. Dewey and A. G. Hopkins. London: Athlone Press. Montiel, Peter, P. R. Agenor and N. Haque. 1993. Informal Financial Markets in Developing Countries: A Macroeconomic Analysis. Oxford: Basil Blackwell. Moore, Henrietta. 1992.“Households and Gender Relations.” In Understanding Economic Processes, ed., Sutti Ortiz. Lanham, MD: Society for Economic Anthropology. Moore, Henrietta and Todd Sanders. 2001. Magical Interpretations, Material Reality: Modernity, Witchcraft and the Occult in Post-Colonial Africa. London: Routledge. Moore, Sally Falk. 2001.“Certainties Undone: Fifty Turbulent Years of Legal Anthropology, 1949–99.” Journal of Royal Anthropological Institute 7: 95– 116.

280

Bibliography

1986. Social Facts and Fabrications: “Customary Law” on Kilimanjaro, 1880–1980. New York, NY: Cambridge University Press. Morduch, J. 1999. “Between the State and the Market: Can Informal Insurance Patch the Safety Net.” World Bank Research Observer 14 (2): 187–207. Munro, William A. 1996. “Power, Peasants and Political Development: Reconsidering State Construction in Africa.” Society for Comparative Study of Society and History 38 (1): 112–48. Naryan, Deepa, Michael Woolcock, Christiaan Grootaert and Veronica NyhanJones. 2003. Measuring Social Capital: An Integrated Questionnaire. Working Paper No. 18. Washington, DC: World Bank. Ndegwa, Stephen. 1998. “Citizenship Amid Economic and Political Change in Kenya.” Africa Today 45 (3): 351–67. Nelson, Joan. 2007. “Elections, Democracy and Social Services.” Studies in Comparative International Development 41 (4): 79–97. North, Douglass C. 1990. Institutions, Institutional Change and Economic Performance. New York, NY: Cambridge University Press. Ntsama, Etienne. 1997. La Securite Sociale au Cameroun: Enjeux et Perspectives. Yaounde: Editions Saagraph. Nukunya, G. K. 1992. Tradition and Change: The Case of the Family. Accra, Ghana: Ghana Universities Press. Okali, Christine. 1983. Cocoa and Kinship in Ghana: The Matrilineal Akan of Ghana. London: Kegan Paul for the International African Institute. Olowu, Dele and James Wunsch, eds. 1990. The Failure of the Centralized State: Institutions and Self-Governance in Africa. Boulder, CO: Westview Press. Oppong, Christine. 1974. Marriage Among a Matrilineal Elite: A Family Study of Ghanaian Senior Civil Servants. Cambridge, MA: Cambridge University Press. Osseo-Asare, Francislee Syversen. 1991. We are Magicians: Chronic Stress and the Economy of Affection in Africa. Ph.D. Dissertation, Pennsylvania State University. Ostrom Elinor. 2000. “Collective Action and the Evolution of Social Norms.” Journal of Economic Perspectives 13 (4): 137–58. 2005. Understanding Institutional Diversity. Princeton, NJ: Princeton University Press. Owusu, Maxwell. 1970. Uses and Abuses of Political Power: A Case Study of Continuity and Change in the Politics of Ghana. Chicago, IL: University of Chicago Press. Packenham, Thomas. 1991. The Scramble for Africa: White Man’s Conquest of the Dark Continent from 1876–1912. New York, NY: Avon Books. Parish, Jane. 2001. “Black Market, Free Market: Anti-Witchcraft Shrines and Fetishes Among the Akan.” In Magical Interpretations, Material Realities: Modernity, Witchcraft, and the Occult in Postcolonial Africa, eds., Henrietta L. Moore and Todd Sanders. New York, NY: Routledge, 118–35. Patterson, K. David. 1981. Health in Colonial Ghana: Disease, Medicine, and SocioEconomic Change, 1900–1955. Waltham, MA: Crossroads Press. Pellow, Deborah. 2002. Landlords and Lodgers: Socio-Spatial Organization in an Accra Community. Westport, CT: Praeger Publishers.

Bibliography

281

Pellow, Deborah and Naomi Chazan. 1986. Ghana: Coping with Uncertainty. Boulder, CO: Westview Press. Penrose, Perran. 1997. “Cost-Sharing in Education.” Unpublished Draft Commissioned by Overseas Development Association (ODA). Cambridge, MA: ODA. Perrot, Claude. 1982. Les Anyi-Indenye et le pouvoir aux 18è et 19è siècles. Abidjan: CEDA. Piault, Marc H. 1987. La Colonisation: Rupture ou Parenthese? Paris: L’Harmattan. Pierson, Paul. 2000. “Increasing Returns, Path Dependence and the Study of Politics.” American Political Science Review 94 (2): 251–67. 1993. “When Effect Becomes Cause: Policy Feedback and Political Change.” World Politics 45 (4): 595–628. Platteau, Jean-Philipe. 1997. “Mutual Insurance as an Elusive Concept in Traditional Rural Communities.” The Journal of Development Studies 33 (6): 764–96. 1994. “Behind the Market Stage Where Real Societies Exist–Part I: The Role of Public and Private Order Institutions.” The Journal of Development Studies 30 (3): 533–77. 1994. “Behind the Market Stage Where Real Societies Exist–Part II: The Role of Moral Norms.” The Journal of Development Studies 30 (3): 753–817. Polanyi, Karl. 1944. The Great Transformation: The Political and Economic Origins of Our Time. Boston, MA: Beacon Press. Popkin, Samuel. 1979. The Rational Peasant: The Political Economy of Rural Society in Vietnam. Berkeley, CA: University of California Press. Posner, Daniel. 2005. Institutions and Ethnic Politics in Africa. New York, NY: Cambridge University Press. Price, Robert. 1979. Society and Bureaucracy in Contemporary Ghana. Berkeley, CA: University of California Press. Putnam, Robert D. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press. 2000. Bowling Alone: The Collapse and Revival of American Community. New York, NY: Simon & Schuster. Rattray, R. S. 1923. Ashanti. London: Oxford University Press. 1929. Ashanti Law and Constitution. London: Oxford University Press. Ray, Donald I. 1996. “Divided Sovereignty: Traditional Authority and the State in Ghana.” Journal of Legal Pluralism 37/38: 181–202. Reno, William. 1998. Warlord Politics and African States. Boulder, CO: Lynne Rienner Publishers. Rhodie, Sam. 1968. “The Gold Coast Cocoa Hold-Up of 1930–31.” Transaction of the Historical Society of Ghana 9: 105–18. Rodgers, Gerry, Charles Gore and Jose B. Figueiredo, eds. 1995. Social Exclusion: Rhetoric, Reality, Responses. Geneva, Switzerland: International Labor Organization. Roncoli, Maria Carla. 1994. Managing on the Margins: Agricultural Production and Household Reproduction in Northeastern Ghana. Ph.D. Dissertation, State University of New York at Binghamton. Rothchild, Donald. 1997. Managing Ethnic Conflict in Africa: Pressures and Incentives for Cooperation. Washington, DC: Brookings Institution Press.

282

Bibliography

Rothchild, Donald and Naomi Chazan, eds. 1988. The Precarious Balance: State and Society in Africa. Boulder, CO: Westview Press. Rudoph, Lloyd and Susan Rudoph. 1967. The Modernity of Tradition. Chicago, IL: University of Chicago Press. Schaeffer, Frederic. 1999. Democracy in Translation: Undestanding Politics in an Unfamiliar Culture. Ithaca, NY: Cornell University Press. Scott, James C. 1976. The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia. New Haven, CT: Yale University Press. 1985. Weapons of the Weak: Everyday Forms of Peasant Resistance. New Haven, CT: Yale University Press. 1998. Seeing Like a State: How Certain Schemes to Improve the Human Conditions Have Failed. New Haven, CT: Yale University Press. Sen, Amartya. 2000. Development as Freedom. New York, NY: Anchor Books. Senah, Kodjo. 1989. “Problems of the Health Care Delivery System.” In The State, Development and Politics in Ghana, eds., Emmanuel Hansen and Kwame A. Ninsin. London: Codesria. Sengupta, Somini. 2004.“Land Quarrels Unsettle Ivory Coast’s Cocoa Belt.” New York Times, May 26, p. A3. Sery, Jean-Pierre. 1998. Les Principaux Axes de la Reforme du Financement de la Santé en Côte d’Ivoire: Le Recouvrement des Couts des Actes de Santé et L’AssuranceMaladie Auto-Financée. Paper presented at International Seminar on the PrePayment of Healthcare and Health Mutualism in West Africa, Paris, France. Shehata, Samer. 2006. “Ethnography, Identity and the Production of Knowledge.” In Interpretation and Method: Empirical Research Methods and the Interpretive Turn, eds., Dvora Yanow and Peregrine Schwartz-Shea. Armonk, NY: M.E. Sharpe, 244–63. Shepard, Donald S. 1996. Financement de la Securite Sociale et de la Sante en Côte d’Ivoire. Study Conducted for the World Bank. Waltham: Institute for Health Policy, Brandeis University. Shipton, Parker. 2008. The Nature of Entrustment. New Haven, CT: Yale University Press. Silver, Hilary. 1994. “Social Exclusion and Social Solidarity: Three Paradigms.” International Labour Review 133 (5–6): 531–78. Smith, Steven Rathgeb. 1993. “The New Politics of Contracting: Citizenship and the Nonprofit Role.” In Public Policy for Democracy, eds., Helen Ingram and Steven Rathgeb Smith. Washington, DC: Brookings Institution Press. Soss, J. and Schram, S. F. 2007. “A Public Transformed? Welfare Reform as Policy Feedback.” American Political Science Review 101 (1): 111–27. Sparr, Pamela, ed. 1994. “Feminist Critiques of Structural Adjustment.” In Mortgaging Women’s Lives: Feminist Critiques of Structural Adjustment. London: Zed Books Ltd. Steger, Manfred B. 2002. “Robert Putnam, Social Capital and a Suspect Named Globalization.” In Social Capital: Critical Perspectives on Community and “Bowling Alone”, eds., Scott P. McLean, David A. Schultz and Manfred Steger. New York, NY: New York University Press, 260–80. Stiglitz, Joseph E. 2005. “Rational Peasants, Efficient Institutions and a Theory of Rural Organization: Methodological Remarks for Development Economics.”

Bibliography

283

In New Institutions for Agrarian Change, ed., Pranab Bardhan. Cambridge, MA: MIT Press. 1986. “The New Development Economics.” World Development 14 (2): 257–65. Strathern, Marilyn. 1988. The Gender of the Gift: Problems with Women and Problems with Society in Melanesia. Berkeley, CA: University of California Press. Stryker, J. Dirck. 1990. Trade, Exchange Rate and Agricultural Pricing Policies in Ghana. Washington, DC: World Bank. Tarrow, S. 1996. “Making Social Science Work Across Space and Time: A Critical Reflection on Robert Putnam’s Making Democracy Work.” American Political Science Review 90 (2): 389–97. Teal, Francis. 2002. “Export Growth and Trade Policy in Ghana in the Twentieth Century.” The World Economy 25 (9): 1319–37. Teal, Francis, Andrew Zeitlin and Haruna Maamah. 2006. “Ghana Cocoa Farmers Survey 2004: Report to Ghana Cocoa Board,” March 11, 2006. Terray, Emmanuel. 1986. “L’Etat, le hazard et la necessite: Reflexions sur une Histoire.” L’Homme 26 (1–2): 213–24. Thelen, Kathleen. 2004. How Institutions Evolve: The Political Economy of Skills in Germany, Britain, the United States and Japan. New York, NY: Cambridge University Press. Thompson, E. P. 1993a. “The Moral Economy of the English Crowd.” Customs in Common: Studies in Traditional Popular Culture. New York, NY: The New Press. 1993b. “The Moral Economy Reviewed.” Customs in Common: Studies in Traditional Popular Culture. New York, NY: The New Press. Tilly, Charles, ed. 1995. Citizenship, Identity and Social History. New York, NY: International Review of Social History. Toqueville, Alexis de. 1945. [orig. 1835]. Democracy in America, vol. 1. Edited by Philipps Bradley. New York, NY: Vintage Books. 1990. [orig. 1840]. Democracy in America, vol. 2. Edited by Philipps Bradley. New York, NY: Vintage Books. Tripp, Aili Mari. 1997. Changing the Rules: The Politics of Liberalization and the Informal Economy in Tanzania. Berkeley, CA: University of California Press. Tsai, Lilly. 2007. Accountability Without Democracy: Solidary Groups and Public Goods Provision in Rural China. New York, NY: Cambridge University Press. Udry, Chris. 1995. “Risk and Saving in Northern Nigeria.” American Economic Review 85 (5): 1287–300. 1996. “Gender, Agricultural Production and the Theory of the Household.” Journal of Political Economy 104 (5): 1010–46. Udry, Chris and Tim Conley. 2004. “Social Networks in Ghana.” Unpublished paper. UNESCO. 2007. Institute for Statistics. http://stats.uis.unesco.org/unesco/TableViewer/ document.aspx?ReportId=143&IF_Language=eng UNIRIN. 2008. “Cote d’Ivoire: Shaky Peace Leading Into Elections.” United Nations Integrated Regional Information Network (UNIRIN), October, 1. http://www. irinnews.org/Report.aspx?ReportId=80702. University of Ghana at Legon (UOG). 1996a. Ghana Post-Election Survey, 1996: Variables, Coding and Marginals. Legon, Ghana: Department of Political Science.

284

Bibliography

University of Ghana at Legon (UOG). 1996b. Ghana Pre-Election Survey, 1996: Variables, Coding and Marginals. Legon, Ghana: Department of Political Science. Van de Walle, Nicolas. 2001. African Economies and the Politics of Permanent Crisis, 1979–1999. New York, NY: Cambridge University Press. 2000. “Africa’s Range of Regimes: Elections Without Democracy.” Journal of Democracy 13 (2): 66–80. Vanderpuye-Orgle, Jacqueline. 2006. “Economy and Poverty in Ghana in the 1990s: A Review.” Discussion Paper No. 29, Institute of Statistical, Social & Economic Research, University of Ghana Legon. Van Gijsegem, Veerle. 2006. “Criminal Law of French Origin and Criminal Behavior of African Occult Origin: The Modernity of Witchcraft Trials in the West of the Republic of the Ivory Coast.” Comparative Studies of South Asia, Africa and the Middle East 26 (2): 191–202. Varshney, Ashutosh. 2002. Ethnic Conflict and Civic Life: Hindus and Muslims in India. New Haven, CT: Yale University Press. Verstralen, Karin. 1995. On Whose Account . . . ?: A Study on the Impact of Structural Adjustment Policies on Financial Self-Help Activities Among Women in Mamobi, a Low-Income Area in Accra, Ghana. Master’s Thesis, Wageningen Agricultural University. Vimard, Patrice. 1987. Diversité des Structures Familiales en Côte d’Ivoire: Une Approche a` Partir d’E´ tudes de Cas en Milieu Rural Akan. Abidjan: ORSTOM. 1993. “Modernité et pluralité familial en afrique de l’ouest.” Revue Tiers Monde 34 (133): 89–115. Ward, W. E. F. 1958. A History of Ghana, 2nd ed. London: George Allen and Unwin. Watts, Michael. 1983. Silent Violence: Food, Famine and Peasantry in Northern Nigeria. Berkeley, CA: University of California Press. Weiner, Annette. 1980. “Reproduction: A Replacement for Reciprocity?” American Ethnologist 7: 71–85. 1992. Inalienable Possessions: The Paradox of Keeping-While-Giving. Berkeley, CA: University of California Press. Weyland, Kurt. 2002. “Limitations of Rational-Choice Institutionalism for the Study of Latin American Politics.” Studies in Comparative International Development 37 (1): 57–85. Widner, Jennifer A. 1993a. “The Discovery of ‘Politics’: Smallholder Reactions to the Cocoa Crisis of 1988–1990 in Côte d’Ivoire.” In Hemmed in: Responses to Africa’s Economic Decline, eds., Thomas M. Callaghy and John Ravenhill. New York, NY: Columbia University Press. 1993b. “The Origins of Agricultural Policy in Ivory Coast, 1960–86.” Journal of Development Studies 29 (4): 25–59. 1994. “Political Reform in Anglophone and Francophone African Countries.” In Economic Change and Political Liberalization in Sub-Saharan Africa, ed., Jennifer Widner. Baltimore, MD: Johns Hopkins University Press, 49–79. Wilks, Ivor. 1993. Forests of Gold: Essays on the Akan and the Kingdom of Asante. Columbus: Ohio University Press. 1975. Asante in the Nineteenth Century: The Structure and Evolution of a Political Order. New York, NY: Cambridge University Press.

Bibliography

285

Wolf, Joyce and Martina Odonkor. 1997. How Educating a Girl Changes the Woman She Becomes: An Intergenerational Study in Northern Ghana. Washington, DC: United States Agency for International Development. World Bank. 1981. Accelerated Development in Sub-Saharan Africa. Washington, DC: World Bank. 1989. Sub-Saharan Africa: From Crisis to Sustainable Growth. Washington, DC: World Bank. 1990. World Development Report 1990: Poverty. Washington, DC: World Bank. 1994. Averting the Old Age Crisis: Policies to Protect the Old and Promote Growth. New York, NY: Oxford University Press. 1995. Ghana: Poverty Past, Present and Future. Washington, DC: Population and Human Resources Division, World Bank. 1996. Staff Appraisal Report: Republic of Ghana, Basic Education Sector Improvement Program. Washington, DC: Population and Human Resources Division, World Bank. 2001a. African Development Indicators: 2001. Washington, DC: World Bank. 2001b. World Development Indicators: 2001. Washington, DC: World Bank. 2005. World Development Indicators: 2005. Washington, DC: World Bank. 2007. World Development Indicators: 2007. Washington, DC: World Bank. Woronoff, Jon. 1972. West African Wager: Houphouet Versus Nkrumah. Metuchen, NJ: The Scarecrow Press, Inc. Yanow, Dvora and Perigrine Schwartz-Shea, eds. 2006. Interpretation and Method: Empirical Research Methods and the Interpretive Turn. Armonk, NY: M. E. Sharpe. Yimam, Arega. 1990. Social Development in Africa, 1950–1985: Methodological Perspectives and Future Prospects. Brookfield, IL: Avebury. Young, Crawford. 1994. The African Colonial State in Comparative Perspective. New Haven, CT: Yale University Press. 2004. “The End of the Post-Colonial State?: Reflections on Changing African Political Dynamics.” African Affairs 103: 23–49. Young, Iris Marion. 1989. “Polity and Group Differences: A Critique of the Ideal of Universal Citizenship.” Ethics 99: 250–74. Zolberg, Aristide. 1964. One-Party Government in the Ivory Coast. Princeton, NJ: Princeton University Press. 1966. Creating Political Order: The Party-States of West Africa. Chicago, IL: Rand McNally and Co.

Index

Abengourou region 26, 27, 28, 116, 157, 167 distance from the capital 34 economic production 34, 180 geography 34 infrastructure 28 level of infrastructure 34, 126, 144, 155 administration 100 British colonial 105--108 centralization 113--114 decentralization reform in the 1990s 114--116 deconcentration 115 differences between British and French 104 French colonial 108--112 Afrobarometer 242, 244 age 17, 44, 48, 59, 61, 169, 170, 181--182, 185. See also elderly, youth aged 135. See elderly Agni chiefs 109. See chiefs agricultural labor 66. See labor agricultural marketing board. See Caisstab, agricultural policies, Cocoa Marketing Board agricultural policies 58, 171--174. See also agricultural marketing boards price liberalization 174 Akan ethnic group 3, 102, 165. See also inheritance matrilineal inheritance 44 Asante kingdom 3, 102, 227 and British colonial rule 105--108 precolonial history 101--103 assimilation 111 British contrast to 107

Bates, Robert 14, 15, 21, 22 Bayart, Jean-Francois 238 Bédié, Henri 176, 224. See also Ivoirite Berlin Conference 7, 108 Berry, Sara 195, 240 Boone, Catherine 26, 242 Brong-Ahafo region 26 distance from capital 34 economic production 34 geography 34 level of infrastructure 34 CAISSTAB 172 case selection 42. See research design Chazan, Naomi 23, 238 chiefs 85, 214. See also administration, colonial rule, non-Akan and British colonial rule 105--108 and chiefly authority 109--111, 112, 217 and collective action 107--108 and collective action in the 1930s 111--112 and French colonial rule 108--112 approaching for help 25 in Côte d’Ivoire 99, 116, 118, 217, 235 in Ghana 99, 115, 118, 211, 217, 233 children 43. See also education, level of children; elderly, respect by youth and support of family 60, 189 help given to 67, 72--73, 86 number of 43 with salaries 59--60, 73 churches 46, 203. See also missionary and British colonial rule 87

287

288 churches (cont.) and French colonial rule 87 charismatic and pentecostal growth 46, 231 in Côte d’Ivoire 46, 85 in Ghana 46, 85, 129 citizenship 10, 232. See also administration differences between British and French, Marshall, T. H. and reciprocity 32, 237 British and French views of 100 community-based orienatation of 10 community-based orientation of 6, 212, 226 duties of 209--211 implications for theories of 241--242 implications of for democracy 232--234 individualized orientation of 6, 7, 212 rights of 205--209 civil society 15, 62 class and citizenship 31 and reciprocity 217--218, 223--224 clientelism 220. See also District assembly (Ghana) nature of in Côte d’Ivoire 221--223, 226 nature of in Ghana 220--221, 222, 226 climactic change 166 and drought 166 effects on yields 166 cocoa and agricultural labor costs 175--177 and bushfires in the 1980s 80, 165 and bushfires of the 1980s 178 and effect on coffee production 174--175 and intergenerational reciprocity 187--188 boom 130, 131 cocoa producer organizations 85 demographics of 169--170 expansion of in Côte d’Ivoire 172--174 exploitation of in Ghana 171--172 history of production 45, 164--165 hold ups in the 1930s 107--108 land tenure of 181--182 landholding size 167, 168 levels of production 165, 167--168, 172--174 migrant labor and 77, 166, 175--177, 182 prices 112, 125, 131 yields 166, 176 Cocoa Marketing Board (CMB) 171, 172 coffee production in Côte d’Ivoire 28, 175 in Ghana 175

Index colonial rule 104--112. See also administration, direct rule, indirect rule, Ghana, C^ ote d’Ivoire Convention People’s Party (CPP) 113 Cooper, Frederick 107 cost recovery 131, 150, 153 exemptions from 138--139 Côte d’Ivoire. See also forced labor; inheritance laws; missionary; Houphouet-Boigny colonial pacification of 108--109 identity cards 81, 176 independence 113 land registration process 176 multi-partyism 228 northern 6, 44, 166, 172 pacification 125 PDCI 12, 114, 205, 219 single-party state 114, 117, 218 CPP. See also Nkrumah, Kwame credit markets 70. See loans, moneylenders de Tocqueville Alexander 30 decentralization 113, 114--116, 117, 215, 228. See also administration deforestation 165--166 democracy. See also citizenship; associational life, political parties consolidation of 13, 225 grassroots 196 direct rule. See colonial rule District Assembly (Ghana) 215, 216 and community leadership 215, 233 decentralization 140 economy of affection 17. See also Hyden, Goran economy of affection 17, 23 education 59 cost of 148, 157--158 during colonial period 122--123, 124, 126 level of 59, 61 level of children 187 levels of 35 quality of 129, 130, 134, 139, 144, 145, 157--158 reforms 137--138 student strikes 150 teacher strikes 150, 151

Index elderly 230 care of 82--83, 189--190 food preparation for 190--191 housing for 191--194 intergenerational reciprocity 186--188 loneliness 13 ethnic conflict 12 nature of in Côte d’Ivoire 12, 14, 174, 224--225 nature of in Ghana 13, 224 resolution of 6, 223, 224 expectations effects on reciprocity 151--152 frustration with cutbacks 154--158 of state role 28 of the state 147--151 extended family 4. See family family. See also children, elderly, inheritance, land tenure, youth family 83--86, 112, 134--135, 160, 202--203. See also children, elderly, inheritance, land tenure, youth extended 4, 13, 17, 28, 44, 61, 121, 125, 131, 141, 180, 229 giving to nuclear vs. extended 52--54 meaning of 48, 71--74 nuclear 4, 6, 7, 17, 49, 60, 62, 108, 118, 179, 229 field research methodology 35. See also case selection entry into villages 40--42 language used in 36 use of research assistants 37 forced labor 99, 110, 111, 112 foreign labor 176. See migrant labor friendship in Côte d’Ivoire 75--76 in Ghana 74--75 gender 37, 55, 60, 169, 231 Ghana. See also CPP, District assembly (Ghana), ethnic conflict, forced labor, inheritance, missionary, Nkrumah, NPP, Rawlings colonial pacification of 105--107 decentralization reforms 113, 114--115 independence of 34, 113 multipartyism 13 northern 6, 44, 166, 224 single-party state 113 Great Depression 38, 99, 120, 123

289 health care. See also cost recovery and NGOs 136 and pauper policies 137 and vulnerable groups 138, 139 decline in 130, 145 during colonial period 126 during postindependence period 128, 129, 130 during the postindependence period 144 expenditures 131, 134 life expectancy 147 national health insurance 141 reforms 153 responses to reforms 156 Herbst, Jeffrey 237 historical institutionalism 10, 22 Houphouet-Boigny, Felix 34, 113, 128, 143, 153, 218 household pooling of resources 44, 55, 74, 193 Hyden, Goran 17, 23, 238, 240 IMF. See World Bank in kind exchange 66, 67, 69, 230 indirect rule. See colonial rule, Lugard, Lord inheritance. See also Akan ethnic group, matrilineal inheritance 1964 law (Côte d’Ivoire) 182 Akan customary system 44 Intestate Succession Law (Ghana) 182 role of laws governing 182--184 institutions 47--48, 234--237. See also historical institutionalism, norms change of 11, 232 formal 11, 22, 233 implications for theories of 237--242 informal 4--6, 15--20, 228 interaction of formal and informal 24--26 interactions of formal and informal 20--24 role of state formation 26--29 interpretive methods 65 Ivoirité 224 Knight, Jack 20 Kymlicka, Will 30 labor 13. See also forced labor, migrant labor, tomatoes labor communal labor 13, 79, 83, 115, 118, 127, 129, 147, 211

Index

290 labor (cont.) daily 45, 177 daily labor 176 family labor 176 permanent labor 176 price 175--177 land tenure. See also inheritance, C^ ote d’Ivoire, Akan ethnic group in crop production 181--182 loans 67--71. See also moneylenders and collateral 69 and interest 67--70, 94, 163, 230 rural banks 140 local government 30, 129, 142, 217, 232. See also administration, chiefs, C^ ote d’Ivoire, District assembly (Ghana), unit committee leadership of 213--215 Lugard, Lord 106 Marshall, T. H. 30, 208, 209, 241, 242 Marx, Karl 30, 31 matrilineal descent 4. See Akan McCaskie, T. C. 102, 103 methods 19. See research methods migrant labor 44--45, 77, 107, 147, 166, 175, 182, 224, 227. See also cocoa government policies 107, 175, 176 migrant workers 47. See migrant labor missionary 121. See also churches Basel Presbyterian (Ghana) 122 Catholic 125 presence in Côte d’Ivoire 125, 126, 144 presence in Ghana 122, 123, 144 Wesleyan (Ghana) 122 moneylenders 70, 85. See also structural adjustment programs (SAPs), loans collateral 69 interest rates 69 moral economy 24, 81. See also Scott, James multiculturalism 30, 31--32. See also Kymlicka, Will neoliberal reforms 131, 153--154, 174. See also state retrenchment; structural adjustment programs (SAPs) devaluation 153 new institutionalists 14, 20--21. See also North, Douglas, Knight, Jack New Patriotic Party (NPP) 218, 226

Nkrumah, Kwame 34, 112, 113, 128, 130, 144 non-Akan 36, 57. See also migrant labor and cocoa farming 170, 224 and reciprocity 55 and tomato farming 169, 170 housing 45 leadership 45 percent of population in villages 45 sampling of 44--45 non-governmental organizations (NGOs) 135--136, 138, 142. See also health care, civil society, voluntary associations norms 16, 19--20, 32, 49, 94, 101, 102, 103, 231 North, Douglass 20 nuclear family. See family PAMSCAD. See poverty alleviation programs patron-client systems. See clientelism pauper policies. See cost recovery PDCI. See Côte d’Ivoire pensions 126, 130, 151 PNDC 217. See also Rawlings, Jerry political parties 233. See also CPP, NPP, PDCI and ethnicity 223, 224--225 village cells 218--219 poverty 135, 178, 243. See also poverty alleviation programs and patterns of reciprocity 58, 230, 231 distribution of wealth 57 measurement of in wealth index 56 poverty alleviation programs 139--140 microcredit 140 PAMSCAD 140 pre-colonial. See Akan, Asante kingdom Provident Societies 127 Putnam, Robert 15, 16 qualitative methods 32, 35, 48, 65 quantitative methods 32, 35 Rawlings, Jerry 113, 204, 217, 220 reciprocity conceptualization of 15--18 measurement of 18--20 of village community 78--83 participation levels 51--53 quantity of 158--161 stringency of 69

Index remittances from overseas 17 Reno, Will 238 research assistants 36. See field work methodology research design 32--35 national level case selection 32--34 regional level case selection 34 sampling at individual level 36, 47 time period covered by study 136, 159 village level case selection 34 research methods archival research 37 ethnography 11, 36 focus groups 19, 46 in-depth interviews 19, 36, 37 oral histories 36, 38 risk and economic production 22 diversification of 21 ecological 22 political construction of 20--23 social construction of 20--23 state mediation of 9 Scott, James 23 sharecropping 181--182 smuggling, crossborder 171 social capital 11, 14, 15--17, 237. See also Putnam, Robert social exclusion 161, 199, 217, 223, 225, 226 and reciprocity 202--203 local conceptualizations of 200--202 social service delivery. See also social service spending, education, missionary, health care during colonial period 120--121, 122--128 during postindependence period 128--131 orientation of 121, 131, 141--143 stability of 121, 143--147 timing of retrenchment 121, 151--154 social service spending. See also health care, education aggregate 121, 134, 161 sous-préfecture 116 state. See also administration, Bayart, Chazan, Herbst, Hyden, institutions, Reno, state retrenchment, Tripp

291 capacity of 11, 21, 23, 32, 238 formation of 7--11 state retrenchment 27, 58. See also neoliberalism, social service provision, structural adjustment programs political attitudes toward 154--158 timing of 152--154, 158--159 strangers 41, 46, 77, 90, 201 structural adjustment programs (SAPs) 114, 120, 131, 134, 141, 145, 177, 228, 235. See also neoliberal reforms, cost recovery, health care, social service delivery, state retrenchment in Ghana 178 subnational analysis 26, 242 taxation during colonial rule in Côte d’Ivoire 111, 112, 125, 147 during colonial rule in Ghana 106, 110, 147 in postcolonial Ghana 115 public opinion toward 122, 211 teachers 46, 134, 137, 139, 150, 152 tomatoes demographics of 169--170 labor required 175 landholding size 175, 180 perishability 168, 177 price volatility 168, 193 transnationalism 30, 31 transporation effects on agricultural marketing 178 transportation 43, 177--178 condition in the 1970s 177 improvements 177 similarities between Ghana and Côte d’Ivoire 167 Tripp, Aili Mari 23 unit committee 25, 45, 99, 114, 115, 117, 118, 211, 213, 215, 216, 217, 233 vegetable production. See tomatoes village chieftaincy. See chiefs village elders 216 and community leadership 214--215 public knowledge of 214

292 voluntary associations 23, 27, 30, 46, 238. See also churches, de Tocqueville, civil society producer associations 46, 203 rotating savings accounts or susu or tontines 127 wealth. See poverty

Index witchcraft 19, 42, 65, 76--78, 89--90, 91, 183, 231 World Bank 15, 16, 33, 120, 131, 140, 141, 235 youth and intergenerational reciprocity 186--188 and tomato farming 169

Other Books in the Series (Continued from page iii) Michael Bratton, Robert Mattes, and E. Gyimah-Boadi, Public Opinion, Democracy, and Market Reform in Africa Michael Bratton and Nicolas van de Walle, Democratic Experiments in Africa: Regime Transitions in Comparative Perspective Valerie Bunce, Leaving Socialism and Leaving the State: The End of Yugoslavia, the Soviet Union, and Czechoslovakia Daniele Caramani, The Nationalization of Politics: The Formation of National Electorates and Party Systems in Europe John M. Carey, Legislative Voting and Accountability Kanchan Chandra, Why Ethnic Parties Succeed: Patronage and Ethnic Headcounts in India José Antonio Cheibub, Presidentialism, Parliamentarism, and Democracy Ruth Berins Collier, Paths toward Democracy: The Working Class and Elites in Western Europe and South America Christian Davenport, State Repression and the Domestic Democratic Peace Donatella della Porta, Social Movements, Political Violence, and the State Alberto Diaz-Cayeros, Federalism, Fiscal Authority, and Centralization in Latin America Thad Dunning, Crude Democracy: Natural Resource Wealth and Political Regimes Gerald Easter, Reconstructing the State: Personal Networks and Elite Identity Margarita Estevez-Abe, Welfare and Capitalism in Postwar Japan: Party, Bureaucracy, and Business M. Steven Fish, Democracy Derailed in Russia: The Failure of Open Politics Robert F. Franzese, Macroeconomic Policies of Developed Democracies Roberto Franzosi, The Puzzle of Strikes: Class and State Strategies in Postwar Italy Geoffrey Garrett, Partisan Politics in the Global Economy Scott Gehlbach, Representation through Taxation: Revenue, Politics, and Development in Postcommunist States Miriam Golden, Heroic Defeats: The Politics of Job Loss Jeff Goodwin, No Other Way Out: States and Revolutionary Movements Merilee Serrill Grindle, Changing the State Anna Grzymala-Busse, Rebuilding Leviathan: Party Competition and State Exploitation in Post-Communist Democracies Anna Grzymala-Busse, Redeeming the Communist Past: The Regeneration of Communist Parties in East Central Europe Frances Hagopian, Traditional Politics and Regime Change in Brazil Henry E. Hale, The Foundations of Ethnic Politics: Separatism of States and Nations in Eurasia and the World

Mark Hallerberg, Rolf Ranier Strauch, and Jürgen von Hagen, Fiscal Governance in Europe Gretchen Helmke, Courts under Constraints: Judges, Generals, and Presidents in Argentina Yoshiko Herrera, Imagined Economies: The Sources of Russian Regionalism J. Rogers Hollingsworth and Robert Boyer, eds., Contemporary Capitalism: The Embeddedness of Institutions John D. Huber and Charles R. Shipan, Deliberate Discretion? The Institutional Foundations of Bureaucratic Autonomy Ellen Immergut, Health Politics: Interests and Institutions in Western Europe Torben Iversen, Capitalism, Democracy, and Welfare Contested Economic Institutions Torben Iversen, Jonas Pontussen, and David Soskice, eds., Unions, Employers, and Central Banks: Macroeconomic Coordination and Institutional Change in Social Market Economies Thomas Janoski and Alexander M. Hicks, eds., The Comparative Political Economy of the Welfare State Joseph Jupille, Procedural Politics: Issues, Influence, and Institutional Choice in the European Union Stathis Kalyvas, The Logic of Violence in Civil War David C. Kang, Crony Capitalism: Corruption and Capitalism in South Korea and the Philippines Junko Kato, Regressive Taxation and the Welfare State Robert O. Keohane and Helen B. Milner, eds., Internationalization and Domestic Politics Herbert Kitschelt, The Transformation of European Social Democracy Herbert Kitschelt, Peter Lange, Gary Marks, and John D. Stephens, eds., Continuity and Change in Contemporary Capitalism Herbert Kitschelt, Zdenka Mansfeldova, Radek Markowski, and Gabor Toka, Post-Communist Party Systems David Knoke, Franz Urban Pappi, Jeffrey Broadbent, and Yutaka Tsujinaka, eds., Comparing Policy Networks Allan Kornberg and Harold D. Clarke, Citizens and Community: Political Support in a Representative Democracy Amie Kreppel, The European Parliament and the Supranational Party System David D. Laitin, Language Repertoires and State Construction in Africa Fabrice E. Lehoucq and Ivan Molina, Stuffing the Ballot Box: Fraud, Electoral Reform, and Democratization in Costa Rica Mark Irving Lichbach and Alan S. Zuckerman, eds., Comparative Politics: Rationality, Culture, and Structure, second edition Evan Lieberman, Race and Regionalism in the Politics of Taxation in Brazil and South Africa Julia Lynch, Age in the Welfare State: The Origins of Social Spending on Pensioners, Workers, and Children

Pauline Jones Luong, Institutional Change and Political Continuity in PostSoviet Central Asia Doug McAdam, John McCarthy, and Mayer Zald, eds., Comparative Perspectives on Social Movements Beatriz Magaloni, Voting for Autocracy: Hegemonic Party Survival and Its Demise in Mexico James Mahoney and Dietrich Rueschemeyer, eds., Historical Analysis and the Social Sciences Scott Mainwaring and Matthew Soberg Shugart, eds., Presidentialism and Democracy in Latin America Isabela Mares, The Politics of Social Risk: Business and Welfare State Development Taxation, Wage Bargaining, and Unemployment Anthony W. Marx, Making Race, Making Nations: A Comparison of South Africa, the United States, and Brazil Bonnie Meguid, Competition between Unequals: The Role of Mainstream Parties in Late-Century Africa Joel S. Migdal, State in Society: Studying How States and Societies Constitute One Another Joel S. Migdal, Atul Kohli, and Vivienne Shue, eds., State Power and Social Forces: Domination and Transformation in the Third World Scott Morgenstern and Benito Nacif, eds., Legislative Politics in Latin America Layna Mosley, Global Capital and National Governments Wolfgang C. Müller and Kaare Strøm, Policy, Office, or Votes? Maria Victoria Murillo, Labor Unions, Partisan Coalitions, and Market Reforms in Latin America Ton Notermans, Money, Markets, and the State: Social Democratic Economic Policies since 1918 Aníbal Pérez-Liñán, Presidential Impeachment and the New Political Instability in Latin America Roger Petersen, Understanding Ethnic Violence: Fear, Hatred, and Resentment in Twentieth-Century Eastern Europe Simona Piattoni, ed., Clientelism, Interests, and Democratic Representation Paul Pierson, Dismantling the Welfare State? Reagan, Thatcher, and the Politics of Retrenchment Marino Regini, Uncertain Boundaries: The Social and Political Construction of European Economies Marc Howard Ross, Cultural Contestation in Ethnic Conflict Lyle Scruggs, Sustaining Abundance: Environmental Performance in Industrial Democracies Jefferey M. Sellers, Governing from Below: Urban Regions and the Global Economy Yossi Shain and Juan Linz, eds., Interim Governments and Democratic Transitions Beverly Silver, Forces of Labor: Workers’ Movements and Globalization since 1870 Theda Skocpol, Social Revolutions in the Modern World

Regina Smyth, Candidate Strategies and Electoral Competition in the Russian Federation: Democracy without Foundation Richard Snyder, Politics after Neoliberalism: Reregulation in Mexico David Stark and László Bruszt, Postsocialist Pathways: Transforming Politics and Property in East Central Europe Sven Steinmo, Kathleen Thelen, and Frank Longstreth, eds., Structuring Politics: Historical Institutionalism in Comparative Analysis Susan C. Stokes, Mandates and Democracy: Neoliberalism by Surprise in Latin America ed., Public Support for Market Reforms in New Democracies

Duane Swank, Global Capital, Political Institutions, and Policy Change in Developed Welfare States Sidney Tarrow, Power in Movement: Social Movements and Contentious Politics Kathleen Thelen, How Institutions Evolve: The Political Economy of Skills in Germany, Britain, the United States, and Japan Charles Tilly, Trust and Rule Daniel Treisman, The Architecture of Government: Rethinking Political Decentralization Lily Lee Tsai, Accountability without Democracy: How Solidary Groups Provide Public Goods in Rural China Joshua Tucker, Regional Economic Voting: Russia, Poland, Hungary, Slovakia, and the Czech Republic, 1990–1999 Ashutosh Varshney, Democracy, Development, and the Countryside Jeremy M. Weinstein, Inside Rebellion: The Politics of Insurgent Violence Stephen I. Wilkinson, Votes and Violence: Electoral Competition and Ethnic Riots in India Jason Wittenberg, Crucibles of Political Loyalty: Church Institutions and Electoral Continuity in Hungary Elisabeth J. Wood, Forging Democracy from Below: Insurgent Transitions in South Africa and El Salvador Elisabeth J. Wood, Insurgent Collective Action and Civil War in El Salvador