Democratic Participation in Rural Tanzania and Zambia: The Impact of Civil Education 9781626370845

Citation preview

Democratic Participation in Rural Tanzania and Zambia

Democratic Participation in Rural Tanzania and Zambia THE IMPACT OF CIVIC EDUCATION

Satu Riutta

Published in the United States of America in 2009 by FirstForumPress A division of Lynne Rienner Publishers, Inc. 1800 30th Street, Boulder, Colorado 80301 www.firstforumpress.com and in the United Kingdom by FirstForumPress A division of Lynne Rienner Publishers, Inc. 3 Henrietta Street, Covent Garden, London WC2E 8LU © 2009 by FirstForumPress. All rights reserved Library of Congress Cataloging-in-Publication Data A Cataloging-in-Publication record for this book is available from the Library of Congress. ISBN: 978-1-935049-14-2 British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library. This book was produced from digital files prepared by the author using the FirstForumComposer. Printed and bound in the United States of America The paper used in this publication meets the requirements of the American National Standard for Permanence of Paper for Printed Library Materials Z39.48-1992. 5 4 3 2 1

This study is dedicated to Zhulea Mohamed of Mbae village and the future generations of Africans whom she represents. This Tanzanian girl who at the time of field research was 10 years old came to sit next to our chairs every day, always smiling and eagerly observing every interview. One day she pointed to a water bottle less than half full, and was very happy when it was given to her. It is hoped that the “bottle” will one day be full for all Africans and people everywhere who live in poverty, and that this study will in a small way contribute toward a better future for them.

Contents List of Tables List of Figures Acknowledgments

ix xi xiii

1

Why Civic Education Matters

1

2

Understanding the Context: Tanzania and Zambia

21

3

The Effects of Civic Education on Cognition

63

4

Civic Education and Participation

99

5

Conclusion

135

Appendixes A: The Survey Instrument B: Details on the Survey Items C: Selection of Respondents D: Variables Used in Regression Analyses List of Acronyms Personal Interviews Bibliography Index

vii

153 169 193 197 203 205 211 219

Tables 2.1 Poverty in Tanzania and Zambia

25

2.2 Democratic Participation in Tanzania and Zambia

34

2.3 Demographic and Other Characteristics of Mtwara Region and Luapula Province

40

2.4 Characteristics of, and Civic Education Provided in, the Tanzanian Research Sites

52

2.5 Characteristics of, and Civic Education Provided in, the Zambian Research Sites

55

3.1 Explaining Aggregate Civic Knowledge: Civic Education and Other Factors Compared

70

3.2 Explaining “First Generation” Civic Knowledge: Civic Education and Other Factors Compared

79

3.3 Explaining Democratic Attitudes in Zambia: Civic Education and Other Factors Compared

90

4.1 Explaining Aggregate Participation: Civic Education and Other Factors Compared

105

4.2 Explaining Individual Participation: Contacting the Ward Councilor

116

4.3 Explaining Individual Participation: Active Participation at Community Meetings

119

4.4 Explaining Collective Participation: Memberships in Community Groups

127

4.5 Explaining Collective Participation: Raising Development Issues with Others

129

5.1 Results of Testing the Hypotheses

139

ix

Figures 1.1 Expected Paths from Civic Education to Participation, via Effects on Civic Knowledge and Attitudinal Factors

15

3.1 Self-Reported Exposure to Civic Education (I): Overall Exposure, Formal Sources, and Three Types of Formal Sources

65

3.2 Self-Reported Exposure to Civic Education (II): Informal Sources, Five Types of Informal Sources, and Political Representatives/Party as a Source

65

3.3 Self-Reported Exposure to Formal Civic Education by Sex

66

3.4 Aggregate Civic Knowledge: Tanzania

68

3.5 Aggregate Civic Knowledge: Zambia

68

3.6 Aggregate Civic Knowledge by Self-Reported Exposure to Formal Civic Education

69

3.7 Civic Knowledge by Type: Government Policies, Rights, and Responsibilities

72

3.8 Identification of Children’s Rights by Self-Reported Exposure to Formal Civic Education

73

3.9 Civic Knowledge by Type: Civil, Human, and Political Rights, and Two Subcategories of Rights and Responsibilities

75

3.10 Civic Knowledge by Type: Socioeconomic Rights and Two Subcategories

73

3.11 “First Generation” Civic Knowledge by Self-Reported Exposure to Formal Civic Education

77

3.12 Knowledge of Socioeconomic Rights by Self-Reported Exposure to Formal Civic Education

77

3.13 Trust in Local Politicians: Does the Leadership of the District Care about People’s Questions and Concerns?

85

xi

xii

Figures

3.14 Interest in Politics

87

3.15 Political Discussion

87

4.1 Communing and Contacting in Tanzania

101

4.2 Communing and Contacting in Zambia

101

4.3 Communing and Contacting by Sex

103

4.4 Communing and Contacting by Self-Reported Exposure to Formal Civic Education

104

4.5 Tanzanians’ Level of Active Participation at Community Meetings

109

4.6 Zambians’ Level of Active Participation at Community Meetings

110

4.7 Active Participation at Community Meetings by Sex

111

4.8 Active Participation at Community Meetings by Self-Reported Exposure to Formal Civic Education

111

4.9 Contacting the Ward Councilor by Sex

113

4.10 Contacting the Ward Councilor by Self-Reported Exposure to Formal Civic Education

114

4.11 Total Group Affiliations: Tanzania

121

4.12 Total Group Affiliations: Zambia

121

4.13 Group Affiliations by Sex

122

4.14 Affiliation with Political Party, Community Government, Religious Group, and Another Community Group

123

4.15 Group Affiliations by Self-Reported Exposure to Formal Civic Education

124

4.16 Raising Development Issues with Others by Sex

125

Acknowledgments This book is a result of two years of research on civic education in Africa. It would not have been possible without the opportunities and support given to me by the Department of Political Science at Georgia State University, and GSU as a whole. More than anybody else, thanks go to Dr. William Downs, the chair of the department, who reviewed and critiqued my manuscript. And while I was in graduate school, he played the biggest role in my development as a researcher and writer. I am also grateful to Dr. Michael Herb and Dr. Carrie Manning for teaching me about democratization and African politics. This research would also not have been possible without the generous funding of a number of sponsors. I was happy to learn that Finnish foundations so readily support research done by Finns at foreign universities. In my case these included Helsingin Sanomat Centennial Foundation and Emil Aaltonen Foundation—both financing field research—Ella and Georg Ehrnrooth Foundation, which funded the last stages of write-up, and Finnish Konkordia Fund, which contributed to my travel to academic conferences. In the U.S., the final months of research were funded by the Dan E. Sweat Fellowship at GSU’s Andrew Young School of Policy Studies. I am deeply thankful for each source of funding, and especially for those individuals and families whose names the foundations and fellowships bear. Research in Africa was facilitated by many kind and resourceful individuals. Professor Marja-Liisa Swantz provided invaluable information based on her 50 years of living in Tanzania. I appreciate her taking the time to invite me to her family’s home and giving suggestions on points of contact in that country. Thanks also to Professors Jeremy Gould and Juhani Koponen at the University of Helsinki for sharing their expertise on Zambia and Tanzania, respectively. In Finland, I wish I had had the chance to thank family friend Dr. Juhani Lindgren before his untimely passing, for his advice concerning my stay in Tanzania, which was based on his own experience there. Of the persons in Africa, I would first like to thank the 280 individuals who participated in the survey, and all others who gave their time to be interviewed. Without them, this study would have nothing to report. After data analysis, I went back to Africa and shared the lessons learnt with the villagers in Mbae, Mtawanya, Mabumba, Chamalawa, and Makasa and the key development agents in the research sites.

xiii

xiv

Acknowledgments

Thanks also to village leadership and others who facilitated the interviews. These include two talented research assistants, Community Development Officer Anthony Nyange in Tanzania, and teacher Jonathan Yamba in Zambia. To me, they both were a gift from heaven. They gave crucial advice on how to improve the survey questions and interpret respondents’ answers. Interviews were also facilitated by Ms. Sylvia Mwanache in Tanzania, and in Zambia by Mr. Tembah, Mr. Ayson Simbeye, and Ms. Everlyn Nambeye, the whole family included. In Tanzania missionaries Riitta and Olavi Heino translated the entire questionnaire into Swahili and spent countless hours describing and explaining the culture in Southeastern Tanzania to me. In addition to them, Twisa Mwambona, Allan Mkopoka, Bright Msalya, and Julie Adkins were especially helpful in aiding my understanding of the Tanzanian context. What Ms. Adkins said during our first meeting on September 3rd, 2005, in Dar es Salaam was revolutionary: it led me to the discovery of this study’s research question. To me, this experience reinforced the truth that sometimes major processes and turning points in life are facilitated by seemingly small events. In Zambia, too, many individuals’ contributions were crucial for the success of the research. Dr. Gun Mickels-Kokwe provided feedback on survey questions and gave recommendations on whom to contact in Mansa. Chiman Patel, Raphael Kauseni, and Ray, Terry, and Ted Barham all helped in many practical ways in Northern Zambia. A big thanks also to the staff of the Anti-Voter Apathy Project (AVAP) in Mansa for welcoming me to learn about civic education through their activities at the Democracy Centre, to Paavo and Annikki Erelä in Lusaka, and Professor Fred Mutesa at the University of Zambia. I want to thank Lynne Rienner, Claire Vlcek, and FirstForumPress for their support and help with editing the manuscript, and my closest family members and everyone who prayed for me. My Mom’s love and encouragement helped sustain me during this project. During field research we frequently discussed our respective writing tasks, scientific writing in my case, and biographic narrative in hers, as she was completing one of her book projects concurrently with mine. I also want to thank my sister, Marianna, and her family for their love and practical help. And last, I want to thank my Heavenly Father for giving me the perseverance, wisdom, courage, and finances necessary to complete this project and for connecting me with the right people. Satu Riutta

1 Why Civic Education Matters Many transitional countries today are in need of a “jump start”1 to restore stalled democratization processes—similar to the shock therapy that is sometimes prescribed for struggling economies. Although the “third wave” was heralded by its inventor2 as the most important political phenomenon of the late twentieth century, welcomed (understandably) by many people, and deciphered with enthusiasm in scholarly circles, the political development that took place during that era has not continued in much of the developing world. In February 2007 Freedom House reported that although “freedom was on the march” for much of the past thirty years, democracy’s expansion has come to a standstill, with the share of countries identified as free remaining stagnant for the past nine years.3 In fact already by 2002 most of the “‘transitional countries’ . . . . ha[d] entered a political gray zone,” in which one characteristic is “low levels of political participation beyond voting” (Carothers 2002, 9). For example, Thomas Carothers, an analyst of democratization and democracy aid, has spoken of the former Soviet Union as a “democratic wasteland,” concluded that South America was experiencing a “crisis of democracy”; and observed that “[d]ozens of African countries have seen once-promising democratic openings deliver only weak pluralism at best” (2004, 412). Why didn’t the third wave last? Though reasons are many, clearly one of the missing components in many transitional countries’ attempts at democratic consolidation is the existence of an active citizenry participating in the conduct of public affairs. Except for brief interludes of popular uprising and the “people power” associated with transitions, in few places have people really gained the power and position to be regularly—that is, not only at election time—included as meaningful participants in their societies with a real say in public affairs. So a crucial question for the prospect of consolidation seems, How could the masses—the poor—be better included and empowered? Many past studies of democratization have focused on macro-level processes including international influences, domino effects, and transitions in general; on the other hand their focus has largely been on the elites. But these approaches alone do not promote a wholesome

1

2

Democratic Participation in Rural Tanzania and Zambia

understanding of consolidation or well-rounded scholarship, as can be concluded from Geddes’ persuasive remarks: Among the methodological practices that most impede the development of a body of theoretical knowledge in comparative politics, I argue, is our standard approach to explaining these big, complicated outcomes [such as democratization, economic development, ethnic conflict]. I suggest an alternative approach. When trying to get some theoretical leverage on compound outcomes (otherwise known as big questions), it is often more useful to divide the big question into the multiple processes that contribute to it and propose explanations for the separate processes rather than the compound outcome as a whole. [In other words:] Outcomes such as democratization, the collapse of empires, and revolution result from the convergence of a number of different processes, some of which may occur independently of others (2003, 23, 37).

One implication of this is that the compound outcomes such as democratization and democratic consolidation should be examined on the micro level. Geddes continues: In order to unpack these mechanisms [in processes contributing to the big phenomena], we need to focus on the fundamental unit of politics, in most cases individuals. We need to break up the traditional big questions into more precisely defined questions about what individuals do in specific situations that recur often enough to support generalizations about them . . . . A carefully constructed explanatory argument built up from fundamentals usually has multiple implications, at least some of which are testable (2003, 38).

In the research on democratic consolidation, this question of “what individuals do” is most neglected when it comes to the role of the poor. What do individuals on the grassroots level do—or what could they do—to contribute to democracy? This study tackles the question of how to enhance rural citizens’ democratic participation, especially with regard to the role that civic education (CE) might play in it. How, if in any way, does this little-researched tool of democracy promotion boost citizen participation in new democracies? To what extent does educating citizens of their democratic rights and obligations actually empower them as participants in democracy? Does civic education increase civic awareness and/or elicit a change in democratic attitudes and patterns of behavior? These questions are asked and answered at the local, community, level—the only arena in which, it is probably safe to say, the majority of

Why Civic Education Matters 3

the rural poor will ever really be involved. It is on this level that a foundation for a civic culture is built. By examining the effects of rights education on local level participation in five villages in Tanzania and Zambia, this study contributes to the “emerging literature on the effects of civic education in new democracies” (Bratton et al. 2005, 40). The two countries are prototypes of the “hybrid”4 regimes that occupy the gray area between authoritarianism and democracy. But in a major difference to previous studies, this study examines civic education’s effects among the rural poor, a population which most extant studies have neglected, even as a great majority of people in many developing countries still lives in rural areas.5 The democratic orientations (or disorientations) of the (rural) poor really do matter for lasting democracy and stability. Also, in contrast to most other studies, the present one considers what kind of knowledge civic education promotes—and who specifically among the rural poor benefit the most from civic education. It is important to know how civic education, as any act of democracy promotion, affects different groups of people so as to know whether the consequences are likely to equalize or reinforce existing inequalities among them. In dissecting these questions the study thus represents a contribution to our understanding about the level of awareness among the poor, and the cognitive and attitudinal changes that are necessary for democratic consolidation. Democracy requires a critical mass of cognitively aware citizens capable of critically evaluating policies and political representatives and holding the government to account. But we can expect such awareness to be limited among unschooled populations in rural areas. Ignorance and the exclusion of certain groups are often perpetuated by strong cultural norms and traditions, affecting women in particular. This is demonstrated by the following account of an NGO employee in southeastern Tanzania of the kinds of problems that women face in participating in public: The village chairman arranged the meeting to be held at the market. . . Everyone sitting in the chairs were men so we asked, “Where are the women?” Then we saw women coming to listen, but they went to hide… I went around, and found some of them, and when I went back to the meeting… [Interviewer:] You brought the women with you? I didn’t bring them, because . . . first time, you must not bring them. You have to talk to the men. So I went to the men; I said: “You know, my husband, he loves me very much. But he [lets] me come here, to

4

Democratic Participation in Rural Tanzania and Zambia

talk with you, work with you. Why don’t you let your women sit with you here? Or at least sit somewhere else…[e.g., under] a tree nearby, so that the women can sit there”…. [Then] they [the women] came. So in the end the women did come? There were not very many, like men, but those who were eager to come, they came. And I had to apologize, “ Please – for these women who have come here, Mr. Chairman, please protect them, so that they are not beaten [by] their husbands because they came.” . . . So you have to joke and say [something like] this…6

A Question for Democracy

To the extent that civic education helps change these kinds of mindsets and broaden participation, it is relevant to democracy. Broadening participation beyond elites is a prerequisite for democracy’s survival. As Bratton et al. stress, “[a]bsent mass participation, the door is open for autocrats or embezzlers to seize power or, at best, for nonelected technocrats to assume responsibility for governance and economic management” (2005, 130). And in the words of Chaligha et al., “Democracy can only survive and mature where citizens take an active role in the governance of their country, for example by voting, contacting representatives, and taking part in community affairs” (2002, 29). The inclusion of the poor is especially challenging in Africa, due to a combination of widespread corruption and poverty. Though the wave of democratization swept over several countries on the continent in the early 1990s—beginning in Zambia—by the mid-1990s the wave had come to a standstill, including in Zambia. Since then, many African countries have experienced retrenchment.7 Yet lack of inclusion, or “departicipation,” has been a problem in African countries ever since independence (Hyden 1983; Weiner 1987), with some countries suffering from it more than others. In a recent Afrobarometer report the authors note, “half of the Africans . . . interviewed were psychologically disengaged from politics” (Mattes and Bratton 2003, 25), and in Mali, Zimbabwe, Nigeria, and Zambia “the electorate is seriously disengaged from politics” (Bratton et al. 2005, 144). This does not apply only to participation between elections; even “voter turnout has declined across sub-Saharan Africa between founding and subsequent elections” (ibid, 144). Thus, while participation has understandably been less meaningful in single-party regimes, its absence continues to plague the continent’s electoral democracies today. There is thus considerable ground for

Why Civic Education Matters 5

democracy promoters, whether indigenous or foreign, to strive to deepen democracy in Africa. Solidifying citizen participation would be vital for the future of democracy, so as to ensure democracy’s survival in economic conditions even as dismal as those of Zambia, one of the very few countries whose Human Development Index (HDI) today is lower than it was in the 1970s.8 To be sure, there are no easy or quick solutions to empower the poor so as to increase their participation in the democratic process, especially in the absence of economic prerequisites for democracy’s survival. But certainly, any attempts at consolidation need to be accompanied by efforts to ensure the basic level of understanding among the population about citizens’ rights and responsibilities—or else other attempts will likely not bear much fruit. Without such basic awareness, individuals cannot express their preferences, and thus participate meaningfully. The CE scholarship is still too young to have created the kind of understanding about effects that is needed for effective utilization of civic education by practitioners. Indeed, the empirical evidence accumulated through research on the consequences of civic education on participation and other democratic attributes in emerging democracies, particularly in Africa, is very limited. In particular, while school-based civic education has been researched more, the effects of adult education are vastly under-researched—not to mention those on the rural poor. And even the processes involved in civics taught at school are not understood: “while there is abundant evidence for the existence of a strong positive relationship between educational attainment and a variety of civic orientations and behaviors . . . how schooling does it remains an enigma” (Ichilov 2002, 82). Therefore, “while we can point to a number of excellent studies on civic education and civic engagement over the past 4 decades,” note Dudley and Gitelson, “we still know relatively little about what knowledge, both qualitatively and quantitatively, is necessary and desirable for an informed and active citizenry” (2002, 180). Thus, there is an overall lack of comparative, international data on how CE is connected to democratic participation and the related cognitive abilities. A Question for Donors

This lack of understanding also affects the prospects of international development aid, of which civic education is often a part. In fact most funds for the activity come from donors, who, especially since the early 1990s, have funded civic education as part of democracy aid.9 But, due to the lack of research based evidence, we have little understanding of

6

Democratic Participation in Rural Tanzania and Zambia

whether the investments made by international donors are bearing fruit. This study seeks to bridge this gap. Should donors continue to fund civic education? The lack of understanding of the effects of CE programs is symptomatic of the lack of understanding of what democracy aid, in general, achieves (Blair 2004). Though it may be true that foreign aid “can only claim to contribute marginally to the longer term impact of civil society building interventions” (Biekart 1999, 300), we need to understand what kind of contribution this is. If the effects of the processes which international actors fund are not understood, “donors will continue to apply discredited ideas likely to undercut their purposes” (Kasfir 1998, 138). Such points serve as the overall justification for the present study. A Question for Domestic Actors in Developing Countries

Although the provision of civic education is conditioned by resources from, and even agendas of, developed countries—civic education, like the supply of democracy in general—is first and foremost a question for the domestic actors in developing countries. What role are the state and civil society playing, and what role should they play, in the provision of civic education? The state, while having the resources and being usually more able than others to reach all corners of a country, may not necessarily have the motivation to provide civic education. Because civic education has the potential to threaten some power holders’ positions and agendas, providing it may not be a popular activity for those in office. And so, although international aid is increasingly given in the form of budget support —in which donors pool their resources to support the government’s own plans—there may continue to be a role for donors to at least fund civic education in the future. Indeed, it is the civil society that has clearly been more active in providing civic education, with non-governmental organizations (NGOs) often delivering CE messages in practice. Local NGOs in rural areas are normally in the closest contact with communities. According to the literature, the kinds of groups that are more likely to foster genuine participation are development groups rather than the more political groups involved in, for example, advocacy. This is because the latter tend to be narrowly based elite groups with sometimes little connection to the grassroots (Carothers and Ottaway 2000). These points about the involvement of donors, NGOs, and to a lesser extent, the state, serve to demonstrate that a better understanding of CE’s effects would affect several actors’ contributions to the lives of the rural poor.

Why Civic Education Matters 7

Civic Education and Participation Defined

How is civic education defined? According to the literature it is that education which promotes recipients’ understanding of the political system, their own interests, and options to contribute to government (Niemi and Junn 1998), or citizens’ rights and obligations (Kanaev 2000). Here, civic education is defined as the latter: awareness raising about rights and responsibilities. While understanding the political system and knowing certain facts about it (such as who one’s representatives are) are vital for effective participation, being aware of one’s rights and responsibilities is more personal, and, it can be argued, a first step to taking action. What, then, are the most important sources of awareness? In rural parts of the developing world most citizens have not been formally schooled nor have they necessarily attended any CE program outside school. Yet this does not mean that they have not been exposed to “rights messages.” In fact, according to a Tanzanian government employee interviewed for this study, “every organization is trying to participate in civic education.” And according to a Tanzanian academic, “In societies in transition people are constantly bombarded with information on what is and is not expected of them.” Among other things, people are targets of health campaigns and “how-to-avoidcorruption” messages (ibid). Even community meetings can be arenas for learning civic knowledge. There, villagers discuss their rights, responsibilities, and topics of importance to them. In these contexts, what is shared and learnt is application oriented civic knowledge, as opposed to the more abstract information taught at school. It is “created by and through groups rather than by isolated individuals; its origin lies in collective attempts to solve problems, and its meaning is only realized through application in an organizational setting” (Field 2005, 4). This is a good description of what learning and civic education may often be in practice in developing countries. Another source of civic information for many people is the media, and particularly the radio; also, by raising awareness of it, even research on civic education can be civic education. The ubiquity and elusiveness of civic education in developing country contexts means it is better to adopt a subjective definition of CE exposure. Therefore, in contrast to past studies, this study does not examine specific (donor funded) programs, but—so as to capture all the relevant sources of CE— defines respondents exposure by each person’s self-assessment. There also exists a standard operationalization of participation found in the literature. It includes voting and voter registration, party and campaign work, community activity, contacting officials, protesting, and

8

Democratic Participation in Rural Tanzania and Zambia

communicating (Milbrath and Goel 1977). Of these, this study focuses on interelection activity at the local level: involvement in community groups, participation at community meetings, joining others to raise development issues, and contacting the ward councilor, the local elected representative. Thus both communing and contacting are included. Though voting activity is surveyed, it falls outside the main forms of participation investigated because most people vote anyway—especially in rural Africa (Bratton et al. 2005)—hence there is little variation to explain. Also, electoral timetable sets constraints on how often a person can participate (Milbrath and Goel 1977); and, voter turnout may better reflect organizational capacity of political groups than citizen interest in politics (Dalton 1996). In turn, protest activity is excluded because most rural respondents do not have any experience in it. Lessons from Previous Studies Civic Education Literature

Extant literature gives some clues about how civic education might influence the cognition and behavior of the rural poor; however, the absence of a specific body of research on this group makes formation of hypotheses challenging. Indeed it has only been since the 1990s that scholars have turned their attention to developing countries in civic education studies. Before then, from the late 1950s until the 1990s, CE studies were almost exclusively limited to school settings in the industrialized world. They examined the extent to which civic education might influence students’ cognition—knowledge, values, and attitudes—being more pessimistic than the more recent studies have been about CE’s potential value. The recent tide of optimism was ushered in by Niemi and Junn’s seminal findings—for example, that recent civics courses alone increases political knowledge by four percent, making it a “significant part of political learning” (1998, 145). Those CE studies that have examined developing countries have focused on out-of-school experiences, due to generally lower levels of formal schooling in these contexts. They have been conducted in the former Soviet bloc, Africa, and the Central America/Caribbean region, reporting findings not only on cognition but also behavior. Though scholars disagree about effects—reaffirming that a better understanding of CE in the developing world remains wanting—there seems to be agreement among most that knowledge is the easiest realm to influence. A study in Zambia found that civic education has “consistently greater

Why Civic Education Matters 9

impact” on knowledge and values than on political behavior.10 The same cannot be said of democratic attitudes, in which there is less agreement about CE’s impact. Whereas a study on Senegal11 found adult education to support the formation of democratic attitudes, such as self-esteem, research on South Africa,12 the Dominican Republic, and Poland13 have found CE to exert little effect on efficacy, political tolerance, and trust in political institutions. In fact, some studies14 have found evidence of a negative impact on institutional trust. Beyond these, another important attitudinal factor, interest in politics, has received little attention in the literature, despite the fact that it is a prerequisite for “effective citizenship” (Bratton et al. 2005, 41). There simply is no clear understanding of the role that civic education plays in the formation of democratic attitudes. Scholars are similarly in disagreement about whether CE promotes participation, or more fundamentally, whether knowledge promotes participation. On the one hand, there are those who argue that “accountable governance requires an educated and well-informed citizenry” (Bratton and Liatto-Katundu 1994, 545). This could be through the fact that well-informed citizens “take the trouble to express their views so that government is directed to do what the well-informed citizenry want” (Halpern 2005, 188-189). But does civic education help in the formation of such citizenry, and if so, how, if in any way, does knowledge translate into effective participation? The literature suggests that any causal chain from civic education to participation would seem to be affected by, first, the kind of group that conducts the training, and the methods it uses, with hands-on, applied training having larger effects than more theoretical, classroom based instruction (Finkel 2002; Finkel and Ernst 2005). Second, effects hinge on certain recipient characteristics like education and status in society. But there is no consensus on whether the well-to-do or the relatively disadvantaged benefit more. The same applies to how sex and age mediate CE’s impact: some suggest that men are better able and positioned to “translate [CE] messages into actual behavior” (Finkel 2002, 1013), while others would maintain that women are more likely to benefit from learning new things and becoming aware of opportunities. The surroundings in which a CE recipient lives are also significant: CE messages are more likely to induce a change in behavior if also the recipient’s family and community reinforce the messages learnt, and do not contradict them (Kanaev 2000; Levinson 2004).

10

Democratic Participation in Rural Tanzania and Zambia

Literature on Democratic Participation

In attempting to understand the effects of civic education on participation, one needs, beyond reviewing studies on CE, to be aware of how civic education ties into the primary explanations of democratic participation: socio-economic status, institutions, and culture. Unless one places his/her investigation within this broader explanatory framework, claims about the contribution of civic education will likely not remain realistic. Of the three major explanations, socio-economic status is the one with the least amount of variation among this study’s participants. Thus the study does not contrast this explanation with the other two. Instead, since a person’s socio-economic status often largely determines his/her level of participation, what causes variation in levels of participation among the rural poor? The literature strongly suggests that people with low education and income are not expected to participate. What, then, causes some of them to participate? Does civic education play a role? In contrast, one can expect to find variation in the poor’s institutional affiliations, the second explanation of participation. This is true when institutions are defined in a micro sense, as they have to be when explaining different levels of participation within a nation-state. Institutional affiliations refer to citizens’ connections to “organized bodies of formal rules”—often, political parties and voluntary associations (Bratton 1999, 554). The affiliations that are especially meaningful in promoting other forms of participation are those in “[g]roups organized around community, workplace, or religion [which] provide opportunities for individuals to sharpen citizenship skills including public speaking, running meetings, and communicating with outside agencies (Brady et al., 1995)” (Bratton 1999, 554). In turn, by aggregating individual preferences and mobilizing citizens, political parties link citizens with the state, thus promoting interaction with the government (Bratton 1999, Milbrath and Goel 1977; Verba et al. 1978). While this study does not analyze how these institutional affiliations come to exist per se, it does examine what role civic education has in propelling people to join parties and associations. Finally, when explaining individual level differences in participation within single countries, culture must similarly be defined in a micro sense, as individuals’ “psychological orientations” (Bratton 1999, 553), or, differently put, democratic attitudes. The most widely discussed in the literature seem to be political interest and efficacy. Though political interest and political participation seem to be almost tautological, Verba

Why Civic Education Matters 11

et al. (1978) point out that it is possible to be interested, yet inactive, and conversely, not interested and still participating. But most of the time, interest coincides with participation. Does civic education contribute to increased participation by raising political interest? Second, does it enhance recipients’ efficacy, and does efficacy then promote participation? Thus, with institutional affiliations and democratic attitudes appearing as the likely major explanations of participation by the poor, it seems that one would need to target these domains when promoting this group’s participation. The theoretical framework for this study therefore consists of the institutions-culture debate: which of these exerts a stronger influence on participation, and in particular, which of them— institutional affiliations or cultural dispositions—is civic education likely to affect more? While at the outset, culture (attitudes) would seem more amenable to manipulation than would institutional affiliations (behavior), we also know, based on extant studies, that attitudes are not necessarily always (positively) affected by CE. Again, the current scholarship cannot explain the role of civic education in the formation of democratic attitudes and patterns of behavior. The literature’s inconclusiveness about both the magnitude and direction of, especially, attitudinal impact, and its lack of attention to some other aspects of impact form the backdrop for the study’s hypotheses. Hypotheses, Methods, and Approach Hypotheses Advanced

Because CE’s effects on participation are likely connected to (its effects on) knowledge and attitudes, this study hypothesizes in each of these areas. Its arguments are divided into the immediate effects—that is, on awareness and attitudes—and into the (indirect) ones on participation. This division does not constitute a formal proposition that effects are always mediated by awareness and attitudes, but it does symbolize the logical order or path along which they are likely to proceed in promoting democratic dispositions. While agreeing with past studies that the immediate effects are likely highest in awareness, this study points to a general shortcoming in them. That is, by failing to distinguish between different categories of civic knowledge, extant studies are not getting at the whole picture. It is not enough, or necessary, that citizens can correctly identify certain factual information about office holders. Rather, for knowledge to be translated into action, citizens first need to

12

Democratic Participation in Rural Tanzania and Zambia

understand the various types of rights to which they are entitled. Do they know that they have a right to express their opinions? Does CE contribute to the awareness of these kinds of “first generation” rights? It is anticipated that it does—more so than “second generation” rights. Therefore it is hypothesized: H1. Civic education promotes the knowledge of civil, human, and political rights and responsibilities more than it promotes the knowledge of socioeconomic rights. Gathering data required to test this hypothesis will allow one to know the extent to which poor people understand their rights to go beyond their immediate physical needs. Though not hypothesized, it is anticipated that “first generation” rights are more clearly linked to participation. People first need to know that they have a right to come together (freedom of assembly), voice their opinions (freedom of expression), and evaluate government performance (right to hold the government accountable), before they will be in a position to pursue specific socioeconomic rights. The study also hypothesizes that though more difficult to achieve, attitudinal change is possible, and most likely to manifest in efficacy. It is intuitive that participation in a CE program would uplift and empower people, while it will not necessarily elevate institutional trust or raise interest in politics. Participants in these programs, which can only accommodate a limited number of people, may feel privileged and excited about the opportunity to participate. Also, learning about such important issues as rights, responsibilities, the political system, and/or government policies should boost people’s confidence in being able to tackle the challenges they face. Therefore it is hypothesized: H2. Civic education increases efficacy. If civic education does boost one’s sense of efficacy, this is likely to have implications for participation, especially among the disadvantaged and discriminated segments of the population. In contrast, trust is expected to be impacted less, and possibly negatively. The reason is clear: civic education often reveals unflattering information about the government and its representatives and “rais[es] the standards to which citizens hold public institutions” (Bratton et al. 1999, 813).15 Along these lines, it is expected: H3. Civic education decreases trust in politicians.

Why Civic Education Matters 13

An important question is what kinds of consequences this lack of trust has. If civic education imparts distrust in politicians, does it do more harm than good by deepening the gulf between the poor and the elites? Or, “Will this mistrust lead to enhanced motivation to participate in, monitor or improve government?” (Torney-Purta et al. 2001, 96).16 Last of the cognitive elements, interest in politics is perhaps the one in which CE’s effects are the most uncertain. In the absence of clues from the literature, it is conceivable that civic education could be as likely to reduce one’s political interest as it is to raise it. But here, the scale tips in favor of expecting civic education to increase this interest because CE can help recipients understand better the options they have in resolving their problems. Therefore: H4. Civic education increases interest in politics. On whom are these effects the largest? When one wants to understand the participatory patterns of the rural poor, most of whom are relatively uneducated, it is not enough to know whether civic education has a greater effect on the educated or the less educated. This study looks into this overlooked aspect of impact: who among the mostly uneducated poor are likely to benefit the most from civic education? It argues that these are the underprivileged—here, women. They have more to gain from what civic education has to offer, and participation in civic education is often a unique experience to them. It is hypothesized: H5. Civic education has the greatest positive effect on the cognition and behavior of the relatively disadvantaged. This hypothesis applies not only to awareness and attitudes but also to participation. Therefore Hypothesis 5 differs from the others in that it will be tested both with regard to cognition, and in assessing CE’s effects on participation. Finally, there is another previously overlooked aspect of CE’s impact on participation: what kind of participation is most likely to be affected? It is not at all clear why civic education should affect all participatory acts to the same extent: not all participation requires the same cognitive skills and dispositions. Benefits should be larger on individual forms of participation relying on individual initiative, that is, active participation at community meetings and contacting the ward councilor, while having less impact on collective forms of participation—that is, participation in community groups and raising

14

Democratic Participation in Rural Tanzania and Zambia

issues with others—or mobilized acts such as voting. Because CE’s sphere of influence is individuals’ cognitive skills and dispositions, civic education is expected to have a greater effect on participatory acts requiring these cognitive skills and dispositions than on those not requiring them. It is hypothesized: H6. Civic education boosts individual forms of participation more than it boosts mobilized or collective participation. If corroborated, this hypothesis would speak to the relevance that civic education has for the formation of a democratic orientation and identity based on individualism. Below, Figure 1.1. presents the rough paths through which civic education is anticipated to affect democratic participation. Notice that it is only intended to give an overall picture of the main linkages between civic education, the various elements of cognition, and participation. Although the study does not hypothesize about these paths per se, 17 Figure 1.1. suggests some routes by which civic education likely connects with participation, through democratic attitudes and institutional affiliations. It suggests, for example, that the primary determinants of participation are associational memberships, interest in politics, and efficacy.18 With the thickness of the arrows, it also suggests that behavioral change is most likely to result from a change in attitudes than from an increase in knowledge. The thickest arrow denotes the expected strong connection between associational membership and (other forms of) participation. Thus important questions for the study are: what is the relationship between civic education and memberships in associations? And: does civic education promote relatively more the associational memberships of the disadvantaged?

p15.qu

3/9/09

5:00 PM

Page 1

15

16

Democratic Participation in Rural Tanzania and Zambia

Methods and Approach As its main instrument in gathering individual-level data on CE exposure and participation, the study used a semi-structured questionnaire administered to 280 respondents in their native language as a one-on-one oral interview lasting 25-45 minutes.19 There were 140 respondents per country. The survey questions and how they operationalized the dependent, independent, and control variables are described in Appendix A, with details on the distribution and mean values of key survey items given in Appendix B. Respondent selection is outlined in Appendix C. Interviews were oral because both countries have an oral communication culture and because many respondents were illiterate. They took place in respondents’ home village or town, in an outdoor public place to which respondents had been asked to gather. 20 The interviewer posed the questions orally, marking the answers on the questionnaire. Several of the questions were open-ended. Data were gathered at one point in time during election year: October-November 2005 in Tanzania and February-March 2006 in Zambia. Because data collection coincided with voter education in both countries, results are expected to represent the most that civic education can achieve. As has been a standard procedure in CE studies, interviews were conducted among treatment and control groups. However, as suggested, an important difference with most other studies is the target group: whereas others have analyzed data on students,21 or focused on elites, 22 this study is restricted to the rural poor. In fact, in Bratton et al.’s (1999) study on Zambia—which used “quasi-experimental” interview methodology—most respondents were educated and 80 percent lived in urban areas. The authors found that “none of the civic education messages germinated” among those with low education and no media exposure—that is, the bulk of respondents in this study (817)!23 In their analysis, all these studies, including the present one, utilize OLS (ordinary least squares) multiple regression methodology.24 More specifically, this study uses hierarchical regression, in which groups of variables are entered into the model in stages, determined on theoretical grounds. This enables one to assess the unique contribution of various explanatory groups: social structure, cognitive awareness, institutional influences, and democratic attitudes. It thereby facilitates answering the question whether institutions matter more than culture. A summary of all variables considered for regression models can be found in Appendix D.25 An additional method was to conduct in-depth interviews of key observers so as to enhance understanding of the context and enable a more accurate interpretation of results. This was done among NGO staff,

Why Civic Education Matters 17

local government officials, donor representatives, church and community leaders, and others in district and national capitals.26 By understanding civic education contextually, and defining exposure to it by respondents’ self-assessment, this study aims to capture all the relevant CE activities in the regions in which data were gathered.27 This concurs with the “decentred” or “arena” model of evaluating development aid (Seppälä 2000). In it, “the existing social processes in the given location are placed in focus, and aid is analysed only in relation to these on-going processes” (ibid, 17). The starting point should be the context, not the aid intervention. But when one examines civic education with a commitment to context sensitivity, it is all the more difficult to tell apart the role of various agents and identify real causal factors. For example, when CE is given as part of participatory development aid, it is very difficult to distinguish the impact of the aid intervention from the roles of indigenous actors and institutions. Context sensitivity also means that opportunities for citizen participation in each locale are taken into consideration. After all, if community meetings are not held, participation does not have the same meaning as it does when meetings are held regularly. Self-assessment is used to determine not only CE exposure but also respondents’ level of participation. That is, while in Tanzania each person’s exposure to civic education is cross-checked against village government records,28 the primary means of determining whether respondents had received civic education and to what extent they participate was by asking them. Though not perfect, self-assessment is important and arguably the only way to really get at a comprehensive picture of CE exposure and how respondents view their participation in the community. Yet employing this approach needs to be accompanied by the awareness that often, those interviewed tend toward positive comments to questions posed by foreigners (Carothers 1999). Parameters and Organization of the Book

This study is located at an intersection of multiple disciplines. Though civic education is inherently political, that is, a question of power, it is obviously a subject matter in the field of education, but also intersects with sociologists’ interest in social inequalities. It is furthermore related to anthropology, the discipline that houses most studies on development interventions. The book is organized the following way. Chapter 2 describes in detail the types of contexts that Tanzania and Zambia represent for the testing of hypotheses. It shows that although at first glance, these

18

Democratic Participation in Rural Tanzania and Zambia

countries are probably more, or as, similar as any two countries in subSaharan Africa, important differences remain in variables crucial for participation. These have to do with: (1) historical patterns of participation (with, for example, community meetings being more institutionalized in Tanzania), and the prevalent (2) political orientations and (3) religious affiliations in the research sites. The Tanzanian site is a government stronghold and largely Muslim, while most people in the Zambian site are critical of the government and profess Christianity. Chapter 2 also outlines CE given in each location. It presents aggregate demographic and other data from the villages, giving the reader an idea of the kinds of communities in which participation is assessed. Chapters 3 and 4 present research findings, with the former discussing CE’s immediate effects on cognition and the latter explaining effects on participation. A comparison of data from the two countries reveals how very different levels of cognizance and participation can be among the rural poor. This supports the expectation that civic education has a varying impact on different individuals within the same socioeconomic stratum. Findings strongly suggest that CE does indeed promote democratic participation and cognitive skills conducive to it, though not always in ways expected. Results are most encouraging in that rights education seems to boost relatively more the participation of those that stand to gain the most from it, the disadvantaged. This suggests that civic education can help level the disparities within a population—and is good news for those seeking to broaden democratic participation and help jump-start consolidation where it has stagnated. Notes 1

The expression is borrowed from Blair (2003). Samuel Huntington (1991). 3 Article originally appearing in National Post (Canada), February 20, 2007 (and referenced at www.freedomhouse.org), by Jennifer Windsor and Arch Puddington. 4 Larry Diamond (2002). 5 In Tanzania this share is about 75 percent, while in Zambia it is about 65 percent (2005; Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat, World Population Prospects: The 2006 Revision and World Urbanization Prospects: The 2007 Revision, http://esa.un.org/unup). 6 Personal interview of a Tanzanian NGO employee in Mtwara (September 20, 2005). 7 According to Freedom House, in 1995-2000 the percentage of democracies in Africa declined from 19 to 17 (cited in Schraeder 2002). 2

Why Civic Education Matters 19

Nevertheless, in general Tripp prefers not to speak of “reversals” of democratic gains in Africa: “it appears premature to talk about ‘reversal’ in many African countries when it is not clear that substantial gains were ever made beyond the holding of multiparty elections.” (2000, 212). 8 Whereas in 1975, Zambia’s HDI was 0.470, in 2005 it was 0.434 (http://hdrstats.undp.org/countries/data_sheets/cty_ds_ZMB.html). 9 CE has also been a part of support for decentralization, a process underway in many developing countries, including Tanzania and Zambia. With most support coming from the U.S. and Europe, “the total investment in civic education activities in the 1990s reached over $230 million (USAID Office of Budget, 2000)” (Finkel and Ernst 2005, footnote 1). 10 Bratton et al. (1999). 11 Kuenzi (2005). 12 USAID (2002), Finkel (2002, 2003), Blair (2003), and Finkel and Ernst (2005). 13 Finkel (2002) and USAID (2002). 14 Bratton et al. (2005) and USAID (2002). 15 Bratton et al. found that civic education tends to impart “healthy skepticism” toward leaders (2005, 250). 16 It should be noted that although Putnam (1993) found that interpersonal trust and participation in civic organizations go together, this is not evidence of causation (Peters 1998). Also, it is less clear how institutional trust (or trust in politicians) is related to participation. 17 That would be a topic for another study. 18 Though the figure only depicts associational memberships, institutional affiliations also include membership in political parties. 19 The languages were Kiswahili in Tanzania and Bemba in Zambia. 20 The interviewer in most cases was a male, but about 15 interviews were conducted in the Zambian village of Mabumba with the help of a female interpreter (an educated, English speaking resident of the village). 21 For example, Finkel and Ernst (2005); Levinson (2004). 22 Blair (2003); Bratton et al. (1999). 23 To be sure, there is one medium to which many respondents in this study reported having access: the radio (see media exposure in Appendix B). 24 Data were analyzed with the Statistical Package for Social Sciences (SPSS). 25 The choice of the independent and control variables was made based on evaluating those used by others—especially Bratton (1999), Bratton et al. (1999), Bratton et al. (2005), Finkel and Ernst (2005), and Niemi and Junn (1998). 26 See list of interviewees at the back of the book. Furthermore, in Zambia two cases of direct observation were part of the author’s learning about the context, including a CE lesson given by the NGO Anti-Voter Apathy Project (AVAP) at an elementary school, and a brief session of voter education by the same organization in a village from which data were otherwise not gathered. 27 The downside of the contextual approach is that one cannot analyze CE interventions in the same level of detail as studies analyzing specific programs. 28 In the Zambian research sites such records are not really kept.

2 Understanding the Context: Tanzania and Zambia What are the circumstances in which linkages between democratic participation and civic education are examined? This chapter provides a detailed account of the national and local level contexts—including the historical, political, and economic conditions that affect citizen dispositions and patterns of participation. The discussion is organized according to the similarities that unite, and the differences that differentiate the two countries. The purpose is to provide the reader with enough background information for contextualizing the findings. As will be elaborated below, the choice of Tanzania and Zambia is based in the first place on them being typical examples of “hybrid” 1 countries in transition—those located between authoritarianism and consolidated democracy. This means there is sufficient freedom for citizens to participate, yet the polities suffer from low levels of participation endemic to hybrid regimes. Thus there is both the opportunity and need to increase democratic participation. The countries therefore represent potentially fruitful ground for strengthening participation by the poor; their commonalities also make it possible to compare findings cross-nationally. Yet, despite a similar need to broaden and deepen participation, and despite very similar historical experiences, political systems, economic and educational challenges, and cultural contexts, the two countries also have important differences. These have to do with citizens’ attitudes, opportunities for and patterns of participation, (local) government’s involvement in civic education, and some cultural features. For example, Tanzanians are generally more interested in politics and in some ways seem to have a higher sense of efficacy. Their government has been more active in promoting participation and decentralization, and more involved in providing civic education. However, the Zambian civil society has been more vibrant and influential. Hence it may not be surprising that Afrobarometer survey data reveal typical participation in Tanzania to be mobilized and politicized, while Zambians are somewhat more likely to participate individually such as to contact their representatives. Therefore this study investigates how CE’s impact may

21

22

Democratic Participation in Rural Tanzania and Zambia

differ based on the country context. After describing the national and local level contexts, the chapter outlines the most important CE programs conducted in the study sites, and briefly discusses the implications of the timing of data gathering. Similarities between Tanzania and Zambia 1. Regime Type and Political History

No longer liberalized autocracies but not yet electoral democracies either, Tanzania and Zambia are typical “hybrid regimes.”2 They belong to the large number of countries in the “political gray zone” whose transitions are more or less stalled (Carothers 2002, 9). Despite more than 13 years since the first multiparty elections,3 they have a strong dominant party, with the opposition, especially in Tanzania, having difficulty garnering support. Their political systems are also characterized by strong presidency, weak parliament, and weak courts, with the executive branch interfering heavily in civil society, especially in Tanzania. The courts’ weakness in both countries is manifested by the fact that, “[d]espite notable recent developments . . . Tanzanian courts have not been able to significantly limit executive dominance or the ‘hyper-presidential’ nature of Tanzanian politics,” while in Zambia, “[d]uring the 38 years of independence, Zambian courts have rarely delivered decisions that significantly inconvenience the sitting government” (Gloppen 2003, 118). Perhaps most important for ordinary citizens, the lower courts—branches of the judiciary with which citizens mostly interact—are poorly equipped and highly corrupt (ibid; Bauer and Taylor 2005). The weakness and corruption of courts likely deters citizens from participating and pursuing their rights. Both countries are former British colonies that gained their independence in the same year: 1964. The concurrence of this is not a coincidence: Zambians “drew support for their independence from Tanzania” (Msabaha 1995, 164). In some other aspects, too, Zambia imitated Tanzania’s choices, not the least because the founding fathers, Kenneth Kaunda and Julius Nyerere were good friends.4 Following independence, both countries enjoyed a few years of multipartyism before instituting a one-party system, which happened earlier in Tanzania (1965) than Zambia (1972).5 Indeed, Tanzania’s one-party constitution of 1965 was “the first systematic one-party constitution in the Commonwealth . . . . provid[ing] a model for other African states including Zambia” (Read 1995, 131).

Understanding the Context 23

Since independence, Nyerere has significantly shaped the Tanzanian nation, with equality and the African version of socialism, ujamaa,6 as bedrocks of his philosophy. Encouraging citizen participation, ujamaa— literally, “familyhood”—urged “the involvement of all members of a family unit (the family being the whole village community) in cooperative work as well as co-operative living” (Legum 1995, 190). Though it did not always manifest as such, Nyerere’s concept of equality pronounced that all Tanzanians have equal rights. In fact, his “villagization” policy (1973-1976) deterred individualized participation in that while it purported to promote development, it was also aimed “to prevent the spread of independent initiatives . . . regarded as a possible source of local resistance against state power” (Seppälä 1998, 16). Also, “villagization seriously disrupted the social fabric in the villages” (ibid, 17). But one area in which Nyerere’s impact was clearly positive is education. Under him, primary school enrollment almost tripled from 25 percent of the age group in 1960 to 72 percent in 1985. And most importantly for this study, Nyerere placed high value on adult or civic education, especially in the 1970s.7 Mmari elaborates on Nyerere’s philosophy: “The whole nation was turned into a large class of seminars; radio broadcasts and public meetings were all used to inform, educate, and involve the population. Critics have described this as a case of indoctrination and have not been slow to point out that despite such efforts incidents of disloyalty and disaffection occurred” (1995, 181-2). While Zambia’s founding father is not as well known for his contributions to education or for emphasizing civic education, one thing that Kaunda had in common with Nyerere—significant for the prospects of participation—was his successful promotion of peaceful interethnic relations. Therefore, the generally low levels of participation in Tanzania and Zambia are likely not explainable by ethnic discrimination. Nyerere promoted interethnic amity by practicing the policy that no group is given too much power, and that all groups be included.8 As a result, Tanzanians have not resorted to mobilization on the basis of ethnic or religious lines.9 Nyerere promoted unity also by instituting a common language, Swahili, for the country’s approximately 120 ethnic groups; therefore, “[e]ven though Tanzania is a very diverse society, its citizens exhibit high levels of national identity and low levels of ethnic consciousness” (Chaligha et al. 2002, 2). On his part, Kaunda “discouraged ethnic discourse [by] develop[ing] a system of governance which included ‘tribal balancing’ as one of its cardinal principles” (Duncan et al. 2003, 21-22). And after Kaunda was forced to step down during the transition in 1991, Presidents Chiluba and Mwanawasa “for

24

Democratic Participation in Rural Tanzania and Zambia

the most part . . . followed Kaunda’s practice of ethnic balancing” (Bauer and Taylor 2005, 53). In both countries, low levels of ethnic consciousness have contributed to stability and to the fact that neither country has experienced a civil war—a rare situation in Africa. This is positive and helps create an environment in which all ethnic groups can feel free and relatively safe to participate. 2. Transition and State-Society Relations: Exceptions to the Similar Historical Experiences

Despite similar historical experiences during the one-party era, Tanzanian and Zambian transitions to multiparty democracy were very different. Zambia’s (1991)—the first of them all in sub-Saharan Africa—represents the “bottom-up” kind typical of Africa, which followed mass protests by civil society against the ruling elites. Protests were led by a constellation of civic groups, students, churches, trade unions, and business groups, with Frederick Chiluba’s Movement for Multiparty Democracy (MMD) emerging from this era as the new protagonist of popular will and multipartyism. So strong was popular pressure that Kaunda was “forced to accept a popular call for multiparty elections” (Bratton and van de Walle 1992, 40), and later, his successor President Chiluba was brought face to face with the power of civil society when his bid for a third term was shot down by civic groups. The bottom-up character of Zambia’s transition sets a strong precedent for citizen participation and mobilization in that country. In contrast, Tanzania’s transition (1992-) was relatively smooth and managed from above. It was initiated by the ruling party (CCM) as a response to the economic crisis that had worsened in particular in the 1980s, the collapse of communism, and donors’ requirements about democracy and human rights (Hossain et al. 2003). One may wonder why the two countries’ modes of transition were so different, especially as it can be argued that the ACP-EU Courier’s remark that “Tanzanians are unaccustomed to demonstrating in order to achieve change” (1999, 10) applies equally well to most Zambians. Although Zambian civil society has subsequently been more vibrant—while Tanzania’s has been more constrained by restrictions on NGOs’ registration and on NGOs’ ability to hold meetings, rallies, and celebrations (Tripp 2000)—it should be noted that both civil societies have come under attack from the government. Also, the state and donors determine much of policy in both countries. As one analyst of Zambian society commented, there has been a lot of government intervention and manipulation of NGOs in Zambia, especially in the political processes of civic education and

Understanding the Context 25

election monitoring. This is significant, as constraints on organizational life limit the extent to which civic education can promote citizens’ participation in associations. 3. Low Levels of Economic Development, Education, and Civic Awareness

Both Tanzania and Zambia are heavily indebted poor countries (HIPC) with a low ranking in the Human Development Index10 and a low percapita income.11 Though Zambia has a somewhat higher per-capita income, its residents seem to suffer from concrete manifestations of poverty at least as much as Tanzanians do. For example, as shown in Table 2.1., Zambians lack access to almost everything more often than Tanzanians. However, despite the recent economic upswing in Tanzania, it seems that poverty has also increased in Tanzania (REPOA 2006).12 Table 2.1. Poverty in Tanzania and Zambia

___________________________________________________ 2001

Tanzania 2003 2005

1999

Zambia 2002 2005

_____________________________________________________________ Food Drinking water Medical treatm. Cash Electricity/fuel* Income**

47 52 48 77 87 $270

45 49 54 73 30

53 62 55 84 32 $340

61 50 69 80 64 $300

78 46 73 87 50

71 71 75 89 57 $490

___________________________________________________ “In the past year, how often, if ever, have you gone without ______? (percent responding “always,” “frequently/many times/several times,” or “occasionally/once or twice”) * The 1999 (Zambia) and 2001 (Tanzania) figures refer to electricity; other figures refer to “enough fuel to cook your food” (Bratton et al. 2004, 11) ** GNI per capita, current US$, Atlas method used (www.worldbank.org); figures from 2000 and 2005. Sources: Bratton et al. (2004), Chaligha et al. (2002), REPOA (2006)

In both countries, lack of economic development is coupled with low levels of education and civic awareness. Less than 70 per cent of the population is literate.13 This is caused, in the first place, by poor access

26

Democratic Participation in Rural Tanzania and Zambia

to education and lack of resources such as buildings, textbooks, and teachers. According to Duncan et al., Zambia has a “chronic crisis in the education system” (2003, 23). An opinion piece in The Post, Zambia’s leading opposition newspaper, is descriptive of the crisis: Most children especially in rural areas don’t go to school either because the parents can’t afford, the school is very far maybe 40 kms or there is no school at all . . . . I was once showed a structure called a classroom for Grade 12 pupils, tears almost rolled over my [cheeks]. It can be equated to a piggery and I was told it was the best in that area.14

A further problem—compounding the lack of resources—is attitudinal: at least in some Tanzanian villages parents do not encourage their children to go to school because educated youth become detached from the adults and their communities (Swantz 1998). One can also question the adequacy and appropriateness of schools’ CE curricula. According to a government education official in Tanzania, “civics in schools happens in isolation”: what is taught at school is not retained or applied in practice because it differs from what pupils learn in their communities.15 For example, if pupils learn to wash their hands at school but do not see their family members wash theirs, real behavior change is unlikely. In Zambia the fact that civics has been taught on the eighth and ninth grades means that “most of the concepts may be a bit too heavy for the cognitive capability of those students.”16 And as in Tanzania, the CE curriculum has been too factual, theoretical, and “conformist.”17 “Rather than knowing the name of the president, it is more important to know the process of bringing in the president,” said one former Zambian civil society activist.18 The emphasis on facts has meant that civic awareness in Zambia is at the cognitive, but not yet at the pragmatic level, as the executive director of the Zambia Civic Education Association (ZCEA) put it. “People know what they’re supposed to do but will they do anything about it? No! They’re expecting somebody else to do it, not themselves.”19 Problems with the CE curriculum—not to mention civics being out of reach for most of the rural poor—has meant that awareness about rights, institutions, and the structure and functions of government is low. According to Ngware, “It is no secret that the average Tanzanian suffers from a total ignorance of basic legal rights” (1997, 246). Many Tanzanians have not even seen the country’s constitution.20 Also, Tanzanians have poor knowledge of their political leaders, especially on the national level—although they rank second highest in their own confession of political understanding among Afrobarometer countries. 21

Understanding the Context 27

Neither are Zambians any more knowledgeable of their political leaders, according to Afrobarometer data, except of one category of leaders: parliamentarians (Chaligha et al. 2002).22 Zambians lack understanding about the division of labor between the central and local government: “half of all . . . Zambians . . . think that central and local government are ‘the same thing’” (Bratton et al. 2005, 244), though this figure has since the early 2000s dropped to 27 percent.23 Confusion about the roles of different levels of government suggests that this is one issue that civic educators need to clarify. Another such issue is revealed by the findings of a previous study on Zambia: “the majority of citizens interviewed seemed to think that political participation begins and ends with voting. While most respondents vigorously asserted their right to vote, few understood that they could also hold their representatives to account between elections” (Bratton and Liatto-Katundu 1994, 561). And even when exercising their right to vote, people are susceptible to manipulation. As one Zambian government employee lamented, “The election year is the saddest year because that is when you see how the population is vulnerable to all sorts of political lies. There are no issues involved at all, just a bag of mealie meal or a tin of beer,” for which people sell their votes.24 4. NGOs: the Most Important Civic Educators

Low civic awareness has been tackled in both countries largely by the same group of actors: NGOs. Academics are also actively involved in awareness raising. According to one NGO employee in Tanzania, civic education is provided by NGOs because they are the only ones with the necessary resources (although not always adequate) and impartiality. 25 Also, although the Tanzanian law stipulates the National Election Committee and the Zanzibar Election Committee to ensure adequate civic education, these bodies have not fulfilled this task.26 In contrast to NGOs, political parties do not possess the impartiality required for effective civic education. For example, in the 2005 election campaign parties used a lot of abusive language to discredit each other, and “there [was] no party which [was] selling the party manifesto to the people.”27 Tanzanian non-state actors at large have lamented the “minimal efforts by political parties to publicize their manifestos to the larger public so that their election plans are known. Instead, candidates resort to statements that are aimed at slandering and causing social disharmony and political anarchy” (Agenda Participation 2000 2005, 24). According to an election observer, even President Mkapa (1995-

28

Democratic Participation in Rural Tanzania and Zambia

2005) said that other parties are “fools,” and that “if I were to act like other presidents, they [opposition members] would all be in prison.”28 A Tanzanian development worker said similarly, Last week . . . our president . . . kept repeating over and over in the radio . . . ‘Be careful about some donors who are engaging in financing civic education and voter education . . . . They want a certain person [in the office] and therefore they are financing to get [things] changed . . .29

Beyond local NGOs, NGOs’ umbrella organizations, and academics, civic education in both countries is provided and funded by a host of other actors. These include (1) international NGOs (INGOs), such as the Irish based Concern Worldwide and Transparency International; (2) political party foundations such as the Friedrich Ebert Stiftung (FES), involved in voter education in Tanzania; (3) foreign governments and multilateral organizations, such as USAID, UNDP, UNICEF, and the World Bank; (4) community based organizations (CBOs); and (5) churches and church related organizations such as the Catholics’ CARITAS and Zambia’s Justice and Peace Commissions. Although this is a varied group, the bulk of CE on the ground is given by NGOs and churches, entities that often are the only ones, in addition to some media outlets, that are able to reach unschooled citizens in rural areas. 5. Cultural Impediments to Participation: Passivity, Avoidance of Conflict, Corruption, and Low Status of Women

Finally, Tanzanian and Zambian cultures share some similarities that impede participation. The first of these is a sense of apathy or passivity, prevalent especially in the rural areas. This is a relic of the one-party era, during which citizens came to see the government as the source of all development. Thus even today, Zambians have a “laid back” attitude towards claiming their rights,30 while “Tanzanians are living in the past. They still believe that all they need will fall from the sky” (ACP-EU Courier 1999, 11). According to a government employee, some Tanzanians even expect that someone will tell them how to vote.31 Consequently, people do not easily criticize the government or their superiors. They have a desire to avoid conflict, and a great reverence toward power holders. They also lack understanding of the system and how to express criticism.32 As one development worker said, “Tanzanians avoid criticizing others, sometimes to the detriment of

Understanding the Context 29

themselves.” They would rather seek consensus. In Zambia, there is similarly a reluctance to criticize the government (Bratton and LiattoKatundu 1994). In the words of an NGO employee, “It is not the culture of Zambia to speak against authority as individuals. Instead, [people] do it through agents like the church and NGOs. Therefore it is difficult to find significant individual voices.”33 A third shared impediment to participation is the pervasiveness of corruption, which severely diminishes both the opportunities and incentives to participate. Donors identify it as one of the primary obstacles of development in Tanzania. Also, “Zambia’s track record regarding financial rectitude is unambiguously poor” (Bauer and Taylor 2005, 56). Typical of the African continent, corruption and patronage politics are so entrenched that they can be spoken of as institutions. For this reason, Hyden has said that “the creation of social capital through civic education is an especially tough challenge in societies like Tanzania” (2005, 8). And according to a Tanzanian government employee, the situation has worsened lately, with “affective kinds of relations” having become more entrenched in the dominant party.34 Finally, the low status of women significantly limits women’s participation. Some of the most notorious violations of women’s rights have to do with inheritance and land rights, including the infamous “property grabbing” affecting widows (Izumi 2007; Swantz 1998). The most violated are younger and older women, who are targets of female circumcision, and can be suspected and even killed for witchcraft. 35 Also, marrying off young girls takes place in both countries, which means that girls drop out of school earlier than boys. But often, rights violations remain hidden. As a longtime missionary to Tanzania said, “Tanzania’s is a ‘culture of concealing’: On the outside, you cannot detect that wives are being battered. [Yet they are.] And personally, I don’t know any man that does not have extramarital children.” Civic education has tackled these kinds of problems, but not without resistance: “in the beginning it was a horrific task because the men . . . were so harsh,” commented a Tanzanian NGO employee.36 In Zambia too men have been very resistant to incorporate women into decisionmaking.37 Differences between Tanzania and Zambia

While these kinds of historical-political, economic, and cultural similarities would suggest comparable opportunities and challenges for (the promotion of) participation in the two countries, there are also a number of factors that differentiate the country contexts and that would

30

Democratic Participation in Rural Tanzania and Zambia

lead one to expect different results. Some of them are in these same issue areas. Discussed below, the differences have to do with economic trends, attitudes toward the economy and government, disposition toward politics, and sense of efficacy. Differences also refer to government role in promoting participation and civic education, patterns of participation, and some aspects of culture, such as religion, traditional authority, and inclination to trust others. 1. Different Economic Trends and Attitudes toward Economy and Government

Beyond overall poverty, people’s attitudes and outlook are affected by the economic trends which they have seen in the past, and which they expect to characterize the future. Do the trends encourage or discourage participation? While poverty hinders participation in both countries, economic trends have been worse in Zambia, with a negative impact on Zambians’ attitudes. In fact, due to a long and continuous decline in the economy, Zambians “remain the most gloomy” of 12 nations studied in the Afrobarometer, with only 28 percent thinking “that economic life is better in their inflation-sapped economy, [and] 57 percent seeing it as worse than under the old order – to the point where some people have given up hope that any political or policy regime can bring about a recovery in living standards” (Bratton et al. 2005, 237). One Zambian NGO employee stressed how difficult it is to make people interested in participating in elections: “the past 15 years have been more disastrous to the people [than the period preceding it]. They have lost the little faith they had.”38 Because of this, CE is now an “uphill battle.” Industrialization has led to a great loss of jobs, a factor that a student of Zambia calls the “single most important factor in Zambian politics.”39 Probably related to different macroeconomic experiences, Tanzanians and Zambians also feel differently about their predemocracy governments. While “Tanzanians apparently still feel nostalgic about the previous political regime,”40 during which they accepted policies handed down by leadership (Chaligha et al. 2002, 36), Zambians espouse “little nostalgia for the former one-party regime” (Simutanyi 2002, vi). And unlike Tanzanians, they have experienced being “in the lead in making demands for change,” most importantly at the time of transition. Therefore, though both Tanzanians and Zambians are generally uncritical and want to avoid conflict, Zambians might, despite the impediment of dismal macroeconomic performance, be more likely to voice opinions in opposition to the government, especially in the region in which this study was conducted.

Understanding the Context 31

2. Different Citizen Disposition toward Politics and Sense of Efficacy

On the other hand, Tanzanians may be more predisposed to participate based on their interest in politics and their sense of efficacy. Political interest is higher in Tanzania—probably because the country’s singleparty rule was of a more mobilizing and politicized type (Chaligha et al. 2002). While 36 percent of Tanzanians say they are very interested in politics, only 22 percent of Zambians say so (ibid).41 In fact, compared with all Afrobarometer countries, Tanzanians have “some of the highest levels of interest in politics” (ibid, vi). With interest in politics being an important determinant of participation, this is noteworthy. Tanzanians also seem more efficacious than Zambians, with as many as six in ten (61 percent) believing that “in discussion with friends and neighbors, I can influence the opinions of others” (ibid, 27). “Turning to another indicator of efficacy, over three-quarters (78 percent) believe that ‘as a community, we are generally able to make our elected representatives listen to our problems’” (ibid, 27). In contrast, though the Afrobarometer question may not have been exactly the same in Zambia, Zambians appear less efficacious: 73 percent “cannot understand what goes on in politics and government. . . . Just over half also feel that they are not able to speak their minds freely about politics (52 percent)” (Simutanyi 2002, 5). Yet there are also alternative accounts according to which Zambians’ level of efficacy is actually quite high. For example, according to a study based on the focus group method, “[Zambians’] attitudes of competence are a major resource for consolidating democratic institutions” (Bratton and Liatto-Katundu 1994, 561). But clearly, in both countries there are those with low efficacy levels. Can civic education help promote their self-confidence? 3. Different Role for Government in Promoting Participation and Civic Education

The differences in individuals’ disposition need to be understood in light of a stronger government role in Tanzania in promoting both participation and civic education. This has shaped people’s attitudes toward, and relationship with, their government. Concerning participation, Tripp writes that “[u]nlike most African states . . . Tanzania had adopted from the outset an approach that saw political participation of people as vital to the success of the nation’s development plans” (1992, 228). Nyerere’s ideology was that “individuals [need to] participate in the process of governing

32

Democratic Participation in Rural Tanzania and Zambia

themselves; they think about and discuss politics with other individuals in the community” (Kweka 1995, 63). Party “[l]eaders and experts were required to ensure that the villagers participated in considering, planning and implementing their development plans” (ibid, 72).42 Similarly, Tanzanian village councils43 were established in the mid-1970s “from above,” with TANU organizing their elections (Killian 1998; Kweka 1995). In contrast, Zambian village committees only came into existence in the mid-1990s, thereby according Zambian citizens a shorter experience with participating in village governance. Another indication of Tanzania’s stronger legacy of participation is that Tanzania is further along in implementing decentralization. Though both countries inherited from the colonial masters a “dual system of administration” (Republic of Zambia 2003, 10) where the locals were governed through “Native Authorities” (Mwaipopo 2004, 2), Tanzania has been quicker to turn this system into one of real local government authority. By 2004, “responsibility for development ha[d] formally been devolved from the federal government . . . to the local government structures,” (Michael 2004, 8). In contrast, Zambia’s decentralization policy dates later (2003), with implementation scheduled to have started only in 2006. Therefore, decentralization had not yet had the opportunity to increase citizen-government interaction by the time data were collected. Also, unlike in Tanzania, in Zambia the central government plays an active role on the district level. For example, in Mansa District, the site of data collection, the local council’s jurisdiction only extends over the semi urban area of Mansa town, with central government being responsible for development in the villages. Furthermore, Zambia’s decentralization policy document interestingly emphasizes, as does the 1996 Constitution of Zambia, Zambia’s character as a unitary state, 44 suggesting a more challenged process of devolving of authority to the local level. This casts light on why some maintain that Zambian politicians only pay lip service to decentralization.45 As a result of its stronger legacy of participation and greater progress made in decentralization, Tanzania has more institutionalized structures for village-level citizen participation and interaction with the local government. For example, the Tanzanian development officers who are in closest contact with communities are usually civil servants based in the district council. Tanzanian villagers are in more frequent interaction with their local government also because, according to Tanzanian policy, all villages must forward their development plans to the district council.46 The council’s task then is to compile all village plans into a district development plan (Mwaipopo 2004). No such procedure exists in Zambia.

Understanding the Context 33

Finally, the Tanzanian government has also played a stronger role in the provision of civic education. In 2001-2004 it sponsored two programs as part of the Local Government Reform Programme (LGRP) to train village level leaders on the decentralization process and good governance, using local civil servants as trainers. The first of them reached 70,000 elected village leaders in all (10,000+) village councils.47 The intention was for trainees to share the lessons learnt with their communities. The second program was geared toward ward councilors, reaching over 3,400 of them in every ward.48 In Zambia, the later implementation of the decentralization policy means that also civic education—accorded an important role in implementation49—had not yet begun by the time of data gathering. Descriptive of Zambia’s situation, a government employee lamented, “there is not even one council in Zambia that could boast on doing civic education.”50 He added, Maybe civic education is not yet a priority here . . . . I have a feeling that politicians tend to enjoy ignorant crowds. . . . The more people know their rights, they’ll demand equal distribution of wealth, downward pulling of resources, retention of certain resources in their areas, development of certain industries in their areas. So at the end of the day, people will be empowered and not easily bribed by others.

4. Different Patterns of Participation

Different levels of institutionalization mean that participation has followed different patterns in the two countries. First, community meetings occur more routinely in Tanzania; in Zambia they are rather ad hoc. According to rules governing village meetings in Tanzania, both the village and the sub-village council are supposed to meet once a month, while the village assembly, consisting of all adult villagers, is stipulated to meet four times a year. At village assembly meetings, attendees and minutes are recorded and then sent to the ward level. In Zambia, there are no stipulations about the frequency of meetings, and “there are no meetings which villagers are obliged to attend, except those summoned by the chief.”51 Consequently, attending a community meeting is more common in Tanzania than many other African countries,52 although Afrobarometer data also suggest that attendance rates have doubled in Zambia in the past few years (see Table 2.2.). Table 2.2. compares data on attendance at community meetings and other forms of participation. The data are then discussed below.

34

Democratic Participation in Rural Tanzania and Zambia

Table 2.2. Democratic Participation in Tanzania and Zambia Tanzania

Zambia

2001

2003

2005

1999

2002

2005

1. Attendance at Community Meetings

75

83*

81*

31

62*

65*

2. Joining with Others to Raise an Issue

60

70*

70*

39

48*

42*

3. Turnout in Presidential Elections1

77 (1995), 84 (2000), 72 (Dec. 2005)

45 (1991), 58 (1996), 68 (2001), 71 (Sept. 2006) 36 39 52

4. Affiliation with Political 2 Party

79

69

76

5. Contacting Government Officials

8*3

43*4

29*4

22*3

n/a

n/a

82**

77**

61**

73*5

6. Affiliation with Local Associations Church/Mosque

n/a

66**

6

n/a

10**

Trade Union/ Farmers Assoc.

16**

23**

26**

7*

Women’s Group

11

n/a

n/a

n/a

n/a

n/a

Sports/Recreation

7

n/a

n/a

n/a

n/a

n/a

Educational Group

6

n/a

n/a

n/a

n/a

n/a

7

n/a

5**

Professional/ Business Group

6**

9**

9**

16*

Community Development Group

6**

22**8

20**8

31*9

n/a

11**8

Environmental Group

6

n/a

n/a

n/a

n/a

n/a

Human Rights/ProDemocracy

3

n/a

n/a

n/a

n/a

n/a

Notes: Percent of respondents saying they have participated in the activity at least once in the past 5 years (except when defined differently, below). * Percent of respondents saying they have participated in the activity at least once in the past year.

Understanding the Context 35

** Percent of respondents saying they are an “official leader” or “active member.” 1 Percent of actual voters of registered voters in all general elections during multipartyism. Figures are rounded to the full percent. 2 Percent saying they feel close to some political party. 3 Based on implication in Simutanyi (2002), this refers to state officials in general. 4 Contacting the ward councilor. 5 Percent saying they have attended a meeting of a church group (for other than religious services). 6 Percent saying they have “attended a trade union meeting at least a few times” (Simutanyi 2002, 5). 7 Percent saying they have “gone to a meeting of a local commercial organization” (Simutanyi 2002, 5). However, one should note that this question subsumes two groups of associations which in the Tanzanian survey are asked about separately: business groups and farmers association. The question in the Zambian survey is as follows, “Over the past year, how often have you attended meetings of a local commercial organization such as a business group or a farmers’ association?” (Cho 2002, Afrobarometer Codebook). 8 Refers to community development group or self-help association. 9 Percent saying they have “attended meetings of a group concerned with local matters such as schools, housing or rates” (Simutanyi 2002, 5). However, this statistic is problematic because the Zambian 1999 survey also asked two other similar questions, that is, about respondents’ attendance at meetings of (a) a local self-help association “such as stokvel, burial association, or neighbourhood watch” and (b) “[g]roup that does things for the community” (Cho 2002, Afrobarometer Codebook). Simutanyi reports that the combined percentage of people reporting attendance at these (that is, “a community self-help group or a group concerned with community issues”) “at least a few times” is 23 (2002, 5). “n/a” denotes either that data are not available, or that inferring the information from the data that do exist is difficult. Sources: Afrobarometer (2006), Bratton (1999), Bratton et al. (2005), Chaligha et al. (2002), Lolojih and Chikwanha (2006), Mulenga et al. (2004), Mutesa and Nchito (2005), and Simutanyi (2002)

The table shows that beyond attending community meetings, Tanzanians are more active in joining other people in raising an issue that they consider important. While in 2005 70 percent of them said that they have done so in the past year, the equivalent percentage in Zambia was only 42. It is interesting, and worth noting, that Mulenga et al. attribute the rise in Zambians’ attendance at community meetings and in

36

Democratic Participation in Rural Tanzania and Zambia

their proclivity to join others in raising issues to civic education— though they do not elaborate as to why this may be so: These improvements can largely be attributed to the civic education programs that have been introduced by such non-governmental organizations as the Civic Education Association and the Anti-Voter Apathy project, to mention just a few. The civic education programs have focused on encouraging people to participate in politics. Many have also learned the value of cooperating through participation in self-help schemes for housing provision (see Bratton et al., 1999) (2004, 14).

Third, Tanzanians have also tended to vote more actively than Zambians. Only in the most recent national elections have turnout levels equaled. That Zambians have voted less may be partly due to the fact that many Zambians distrust elections as a fair way to select leaders (Mulenga et al. 2004; Simutanyi 2002). In fact, “nearly one-quarter (22 percent) . . . [are] willing to consider alternative means of selecting political leaders because they feel that elections are fraught with many problems” (Mulenga et al. 2004, v). Low electoral participation was a problem already at the beginning of the multiparty era. Although “[a]n overwhelming percentage of respondents (94 per cent) acknowledged a civic duty to vote” (Bratton and Liatto-Katundu 1994, 546), turnout in 1991, at approximately 45 percent, was at an internationally low level (Bratton 1999; Mutesa and Nchito 2005). However, as Table 2.2. shows, turnout has since risen—first to 58 percent (1996), and most recently (2006) to 71 percent. Fourth, connected to Tanzania’s history of higher turnout rates as well as the stronger position of the dominant party, CCM, party affiliation is much higher in Tanzania. Seventy-six percent of citizens feel close to some political party, compared to 52 percent of Zambians. 53 Zambians’ current shunning away from party politics is significant because, as observed in an earlier study, Zambians who carried the membership cards of political party were consistently more likely to participate in national politics, in all its multiple dimensions . . . . [In fact] political party membership had a much more consistent effect (across all modes of participation) than any other institutional variable. And the effect was broadly civic rather than narrowly partisan, holding regardless of which political party— MMD or UNIP—was considered (Bratton 1999, 569).

Understanding the Context 37

This thus gives credence to the institutional explanation of political participation. Party membership is important also because it “more extensively” contributes to an individual’s interest in politics than registering to vote or joining an association (ibid, 580). In both countries, most of those feeling close to some party identify with the dominant party—in Tanzania the overwhelming majority of mainlanders with CCM, while in Zambia 64 percent with MMD. Yet, fifth, even with party affiliation being higher, Tanzanians do not appear to contact their government officials as often as Zambians. This is one of the two forms of participation in which Zambians appear more active. Table 2.2. shows that a much larger share of Zambians contacted their government officials at the turn of the century (1999/2001). This, Chaligha et al. (2002) suggest, is indicative of the fact that Tanzanians are not used to this kind of individualized participation. Finally, consistent with the bottom-up character of Zambia’s transition, the other way in which Zambians participate more actively is in local associations. In fact, they are “the most active in civic associational life” in all of Southern Africa (Simutanyi 2002, 2). In Bratton’s dataset, as many as “four out of five Zambians claimed some sort of affiliation with a community-based organization” (1999, 569), with “the most common form of organization to which the majority of Zambians belong [being] the church” (Duncan et al. 2003, 28). Restrictions were lifted and associational life expanded in the 1980s, when, “[i]n the context of growing political dissatisfaction, the churches and unions provided protected space within civil society for citizens to associate freely” (Bratton 1999, 561). In the post-1991 social transformation, the church has been a leading agent of change. 54 Although in Tanzania independent associational life was similarly suppressed during the single-party rule, Tanzania has been slower in reviving it since then. Consequently, “[a]ssociational life in Tanzania is quite weak, even by African standards” (Hyden 1999, 149). The only exception is church or mosque membership; “membership in other types of civil society groups is the lowest in any country where Afrobarometer surveys have asked similar questions” (ibid, 31). This suggests that mobilizing citizens to and by civic associations is easier in Zambia. 5. Cultural Differences: Religion, Traditional Authority, Trust

Lastly, beyond the cultural similarities that hamper participation, Tanzania and Zambia have at least three differences in their cultural contexts which predict diverse responses to civic education and the promotion of democratic participation. First, Tanzania’s population is

38

Democratic Participation in Rural Tanzania and Zambia

quite equally divided into adherents of two religions, Christianity and Islam, while Zambia’s is almost entirely Christian.55 This is relevant at least to women’s participation, facing more constraints in Islamic communities. There are also differences in advice given in religious institutions: while elders in the mosques in coastal Tanzania “critically oppose the youth’s support of multipartyism” (Swantz 1998, 153), many church leaders from various Christian denominations in northern Zambia encourage members’ going to the polls, albeit only after having cautiously evaluated the candidates.56 The two nations also differ in the extent to which citizens define themselves in religious terms, with “religious identity [being] most widespread in Zambia” of all countries in the Afrobarometer (Bratton et al. 2005, 189). While 36 percent of Zambians define themselves this way, “more even than feel themselves members of an occupation or class (25 and 23 percent respectively)” (ibid, 189), only five percent of Tanzanians do so (Chaligha et al. 2002). Instead, 76 percent of Tanzanians define themselves occupationally, more than any other Afrobarometer nation (ibid). Such a difference probably means that Zambians are generally more likely than Tanzanians to shape their behavior according to their religious beliefs. Second, traditional authority plays a bigger role in Zambia. Therefore opportunities for democratic participation there depend to a larger degree on the chief’s preferences and actions. According to one development worker, Zambian chiefs are very influential, especially in peripheral areas. Their role is to be “the custodian of land and the upholder of tradition” (Mickels-Kokwe and Kokwe 2004, 23). This is in contrast to Tanzania where traditional rule has clearly had a lesser influence; Nyerere abolished it to foster national identity (Chaligha et al. 2002). As a consequence, more Tanzanians (89 percent) today reject traditional rule as a form of governance than people in other Afrobarometer countries (Bratton et al. 2005). According to Bratton et al., the “Tanzanian result is understandable in terms of the central government’s systematic campaign to discredit the indigenous authority structure and replace it with a network of ten-house cells and party cadres” (2005, 80). Finally, important to one of this study’s hypotheses, extant studies seem to agree that institutional trust is very high in Tanzania, while in Zambia it is low. Bratton et al. refer to Tanzanians’ high trust in the state, while Chaligha et al. speak of “extraordinarily high levels of trust in government” (2002, 6). One piece of evidence could be that Tanzanians are “willing to wait for [economic] reform to produce results” (ibid, 24). These authors also find that “Tanzanians express extremely high levels of trust in each other” (ibid, 3). However, others

Understanding the Context 39

(Bratton et al. 2005; Hyden 2005) observe Tanzanians’ interpersonal trust to be much less,57 explaining it with the fact that villagization forced relative strangers to live in “close residential proximity” (Bratton et al. 2005, 195). In contrast, the literature is clear that Zambians are not very trusting either in their political institutions or politicians (Bratton and Liatto-Katundu 1994; Simutanyi 2002). Thus if the hypothesized negative relationship between civic education and trust is corroborated, in Zambia CE would reduce already meager amounts of trust, while in Tanzania it would bring excessive amounts of trust down to a “healthier” level. The Local Level and the Provision of Civic Education

These national-level data serve as a backdrop to the participatory patterns and conditions found on the local level. The rest of this chapter will therefore describe the regions, districts, and villages where data were gathered, and the civic education provided. Selection of regions began in Tanzania: the selected area was to be such that it had had enough exposure to civic education. After this, specific districts and villages were chosen on both theoretical and logistical grounds. This process was then repeated for Zambia. Below, the most important characteristics of the regions are described, followed by an outline of the most important CE programs in the districts selected, and finally a summary of demographic and other information from particular villages in which interviews were conducted. It is only with a detailed understanding of these conditions—the framework in which respondents live—that findings can be interpreted correctly. Selecting Regions within Tanzania and Zambia

The area within rural Tanzania that stood out in both its exposure to civic education and the need to promote democratic participation was Mtwara Region in the southeast. Along with neighboring Lindi Region, it has been the site of Tanzania’s first major area based participatory development cooperation program, the RIPS (Rural Integrated Project Support) program (1988-2005), as part of which civic education was carried out with the aim to empower rural citizens, promote their inclusion in decision-making, and enhance their relationship with the local government.58 To be sure, while RIPS has been the largest and the most important donor program in the area, other donors have also been active there.

40

Democratic Participation in Rural Tanzania and Zambia

Mtwara Region Mtwara Region is one of Tanzania’s least developed areas and very remotely located (see Table 2.3.). Poverty and remoteness make the promotion of participation particularly challenging. Therefore, if civic education increases participation in Mtwara, it is likely able to do so also elsewhere in Tanzania. Table 2.3. outlines some relevant demographic and other characteristics of Mtwara Region and its Zambian counterpart, Luapula Province. The characteristics that are most important for democratic participation and its cognitive prerequisites are briefly discussed below. Table 2.3. Demographic and Other Characteristics of Mtwara Region and Luapula Province Mtwara Region (2002)

Luapula Province (2000/2002-2003)

Population

1.2 million (i.e., 3.4% of national population)

853,000 (i.e., 7.8% of national population)

- of whom rural

80% (nation: 77%)

85% (nation: 65%)

- of whom younger than 15 years

37% (nation: 44%)

46% (nation: 46%)

- of whom female

53% (nation: 51%)

51.6% (nation: 51%)

Population growth

1.7% (nation: 2.9%), 1988-2002

3.2% (nation: 2.5%), 19902000

Population density (people/km2)

67; in 2005: 73 (nation: 39)

15.3 (nation: 13.1)

Distance from capital (by car)

10 hours (borders Mozambique)

8 hours (borders the Democratic Republic of Congo)

Household size

3.8 (nation: 4.9)

5.0 (nation: 5.4)

Incidence of poverty

“Most people” live on less than 0.15 USD/day (Voipio 1998)

“Extreme” poverty: 47% (nation: 46%)

Understanding the Context 41

Main livelihoods

Small-scale agriculture, fishing

Small-scale agriculture, fishing

Adult literacy (those over 15 years)

56%

61.5%

Kinship

Traditionally matrilineal, but increasingly patrilineal

Traditionally matrilineal, but increasingly patrilineal

Main religion(s)

Islam and Christianity

Christianity

Share of votes for dominant party’s presidential candidate in most recent elections

79% (nation: 80%), Dec. 2005

34% (nation: 43%), Sept. 2006

Sources: - Mtwara Region: 2002 Population and Housing Census (i.e., United Republic of Tanzania 2002); Issae (2005); National Electoral Commission of Tanzania (www.nec.go.tz) - Luapula Province: Living Conditions Monitoring Survey Report 20022003 (i.e., Republic of Zambia 2004a); 2000 Census (i.e., Republic of Zambia 2004b); The Electoral Commission of Zambia (www.elections.org.zm)

Mtwara’s remote location makes contact with (national) representatives difficult and affects awareness and motivation to participate. The remoteness is perpetuated by poor road network, the negative image of the region as a “punishment post” for civil servants, and the fact that all of Mtwara’s neighbors are poor as well (Issae 2005; Wembah-Rashid 1998). So remote is its location, and backward its reputation, that some people even talk about Mtwara as an island (it is very difficult to get in and out of the area).59 Its reputation as a punishment post means that for a long time, “[c]ivil servants who misbehaved elsewhere in the country were sent to this region as a disciplinary measure” (Wembah-Rashid 1998, 55). Unfortunately, “[t]his had an effect on the behaviour of these officers in their interaction with the inhabitants: they were either harsh to them or completely uninterested in their work” (ibid, 55). Also, according to Swantz, the Mozambican war (1975-1994) perpetuated Mtwara’s isolation: “the area was declared an emergency area where all travel was severely restricted. Interaction with the outer world was thus legally

42

Democratic Participation in Rural Tanzania and Zambia

curtailed” (Seppälä 1998, 15). Consequently, Mtwarans have a reputation of being more apathetic than Tanzanians in other parts of the country. As Wembah-Rashid describes sentiments in the area, [W]hen the majority of the people in a region are marginalized, isolated and inward oriented, they develop a behaviour best described as apathy. This includes feelings that they cannot succeed or do better even if they embarked on some project; they have not seen anybody succeed in their vicinity. This last statement may not be true today, but these are feelings which have persisted among members of the older generation (1998, 45-46).

Culturally, a distinctive feature of Mtwara—and potentially influential for participation—is that lineage is traditionally matrilineal. This is due to the prevalence in the area of the Makonde, the third largest of Tanzania’s approximately 120 ethnic groups,60 who constitute 86 percent of the coastal population in Mtwara District (Mihanjo and Luanda 1998). However, due to villagization, migration, the Mozambican war, and intermarriage, the region is increasingly mixed ethnically (RIPS 2005; Seppälä 1998), and no longer as matrilineal. Whereas lineage may still follow the matrilineal line, property rights do not (Seppälä 1998). As a missionary interviewed for the study said, if her husband dies, a woman is worth “nothing”; the husband’s relatives come and take away all her possessions. Observance of the matrilineal tradition is also reduced by Mtwara’s dominant religion, Islam. Islam has also contributed to the relatively low levels of education in the area, with Muslims in Tanzania being generally less likely than Christians to educate their children.61 This is partly because they see education as a western and Christian institution.62 According to the Mtwara regional education officer (REO), the region has about 55-60 percent literacy rate, lower than in the nation. The number of secondary schools, enrollment, and academic performance are low (Issae 2005). Finally, politically, Mtwara is known as a dominant party stronghold. This is due to “the historical strength of TANU/CCM as a culturally accepted form of patronage organization (even if people view it ironically in terms of efficiency and transparency)” (Seppälä 1998, 29). Therefore, residents view CCM as the upholder of peace, and consequently, opposition party activism as a “call for chaos” (ibid, 29). The popular sentiment is that “[t]he need for peace overrides the need for political change” (ibid, 29). CCM is popular also because many of the nation’s leaders—CMM party members—have come from southern Tanzania, including President Mkapa. Mtwarans are therefore easily

Understanding the Context 43

mobilized by the dominant party, with civic education, especially if seen as opposition party activity, potentially having a more difficult time changing mindsets and behavior in this region. Luapula Province In Zambia the area determined to most closely resemble Mtwara Region is Luapula Province. As shown in Table 2.3., Mansa, the provincial capital, is similarly remotely located: eight hours by car from Lusaka. Like Mtwara, Luapula suffers from “poor access to markets,” inadequate infrastructure (Ministry for Foreign Affairs of Finland 2001, 76), negligence, and “political disinterest” on the part of national leadership, even colonial administration, as a result of which it has remained relatively backward and dependent (Gould 1989). Similarities with Mtwara can also be found in culture, such as with kinship lineage being traditionally matrilineal. Besides marriage arrangements and the “mobility of males” (Gould 1989, 25)—with husbands traditionally moving to live in the bride’s locality—this means that descent and inheritance are primarily determined in the maternal line. However, as in Mtwara, observance of matrilineal traditions is diminishing, as demonstrated especially clearly by the frequent occurrence of property grabbing. In turn, while Luapula’s literacy rate is somewhat higher than that of Mtwara (see Table 2.3.), it too struggles with low enrollment rates, especially for girls. One hindrance to girls’ schooling is early marriages encouraged by many parents, who are motivated by the custom that the son-in-law begins working for them. Interestingly, according to the district planning officer at the Mansa District Education Board, particularly the mothers are difficult to convince of the utility of letting their daughters go to school. In turn, some of the most relevant differences between Mtwara and Luapula lie in religion and political attitudes. Whereas coastal Mtwara is predominantly Muslim, virtually no-one identifies themselves as nonChristians in Luapula. And whereas apathy may be said to characterize Mtwarans, Luapulans are more politically “enlightened.” 63 In the words of a Luapula ward councilor, “politically this place is quite alert and people are quite aware of their rights and ready to participate in political affairs of the country.” 64 Another interviewee said that Luapulans can communicate better than other Zambians in English and are not influenced by such a strong culture.65 And importantly, support for the dominant party is much lower in Luapula than it is in Mtwara (see Table 2.3.), reflecting also the lower level of support for the dominant party nationally. In fact, researchers of two Luapulan districts, Mansa and Samfya, say that “a hostile climate against government” in these districts

44

Democratic Participation in Rural Tanzania and Zambia

is a result of a lack of development activities and unfulfilled promises (Mutesa and Nchito 2005, 22). For example, Luapula seems to have received less foreign aid than Mtwara, including in civic education.66 Civic Education Provided

In both Mtwara and Luapula, tracking of CE activities was limited to one district: Mtwara Urban Disctict and Mansa District, respectively. Even so, it was extremely difficult to map CE exhaustively—due to the nebulousness and ubiquity of “rights education,” as well as a dearth of written records. Civic education is given not only at schools and through community programs but also through the media; it is also “conducted” informally by family and friends. And as mentioned, almost every organization is trying to participate in it. Nevertheless, based on information gathered from local government officials, NGOs, and the literature, there emerged a group of actors which can be considered the most significant CE providers in these areas. Mtwara Urban District Much of CE in Mtwara Urban District has been funded and/or provided by bi- and multilateral donors, (I)NGOs, and religious organizations, with local civil servants intricately involved in implementation. Of the CE providers in the area, five appear the most significant, either because of the length and/or geographical coverage of their programs, or that they have reached the villages chosen for this study. These are: civic education given as part of RIPS, child rights promotion by UNICEF, good governance projects by Hanns Seidel Foundation, and voter education by CHAWATA, a Tanzanian NGO for the disabled.67 In the first three, training was conducted by council civil servants, or “government staff,” with the last being given by an NGO. 1. RIPS Though the aim of the RIPS program throughout its presence in Mtwara (1988-2005) was to empower citizens and promote their democratic participation, civic education was not an explicit part of RIPS until from the year 2000 onward. CE was given both during community outreaches and in workshops held mainly in 2002-2003 in townships. Altogether over 360 villages in Mtwara and Lindi were reached, with CE messages focusing on the rights and responsibilities of villagers and leaders (RIPS Programme 2005). One former RIPS employee said,

Understanding the Context 45

In all the processes, civic education was seen as the entry point. We used to say that civic education is the entry point of any process from planning to implementation. You have to start with the rights and obligations. Are they aware of their rights? If you want them to be involved in the planning process, are they aware that this is their right? Participation—it is their right. So they have to be aware of that.68

RIPS sponsored a total of five workshops. The idea was to first train select civil servants in the council who would in turn train community leaders. The community leaders would then share the lessons learnt with others in their communities.69 However, as one former RIPS employee said, civic education was also taken up by NGOs that sought funding from RIPS: “Of course civic education then went beyond the rights and responsibilities. It . . . also involved the awareness of other issues, crosscutting issues like HIV/AIDS and environmental conservation.” RIPS also educated people on statutory community meetings, how often community members are supposed to attend them, and how to remove elected officials. It sponsored ethics workshops to NGOs, CBOs, the police, courts, and the Prevention of Corruption Bureau, using an “ethics manual” developed by two University of Dar es Salaam professors70 on citizens’ roles, equality, and ethical leadership. The workshops also utilized the UN Declaration of Human Rights (1948), the Tanzanian Constitution of 1977 (especially Chapter 3 on rights and responsibilities), and the PORALG guidelines stipulating the duties of village leadership and councilors.71 There has been both praise and criticism of RIPS as a development program and sponsor of CE. According to one development worker, CE given to the grassroots has really improved village governance: In most villages, leadership was a problem; people never knew how to change the leadership. But when they were given civic education, they knew their rights, and the way how to get rid of unwanted leadership. And they did it. And now they’re having very . . . dynamic leadership in those villages.72

Two other development workers similarly thought that civic education given as part of RIPS has bolstered people’s confidence to claim their rights—to “demand things.” People know what the local government is supposed to do. Critics refer to RIPS’ lack of transparency—“secretive administrative procedures”—and excessive use of expatriate staff, questioning the democratic nature of the program itself.73 According to an official in Mtwara’s regional government, RIPS projects were often

46

Democratic Participation in Rural Tanzania and Zambia

perceived as supporting the opposition. This may have affected their effectiveness in promoting participation. 2. UNICEF Another large area based development program, UNICEF’s Child Survival and Development Programme (CSDP), has also been operating in Mtwara Region since 1988. It aims to enforce, and trains people on, children’s rights to well-being, education, and participation in their community. The program is largely a response to the AIDS problem and high number of orphans, with as much as 40 percent of children in some communities being orphaned (Office of the Regional Commissioner 2004). The Most Vulnerable Children (MVC) program—a part of CSDP— is now in “full swing” in two of Mtwara’s five districts.74 UNICEF does not have any personnel in the regions; rather, as a cooperating partner of the Tanzanian government, its programs are implemented by local government officials, who train select community members in sessions of Community Justice Facilitation (CJF). Attendees learn how to identify the most vulnerable children living in their midst and ensure that they are cared for; they then share and implement these with others. Usually, participants are volunteers, as were the 25 attending the November 2005 training held in Mtwara town, the most recent CE program implemented in the Tanzanian site prior to data collection. So far, training on children’s rights has been provided to all the village chairpersons, village executive officers (VEOs), and vitongoji (that is, sub-village) leaders in the district. Like RIPS, CJF has been both criticized and praised. In the late 1990s Voipio wrote, The UNICEF-funded Child Survival and Development Programme (CSDP) in Mtwara region has had serious accountability problems. But it has also had remarkable success in promoting a spirit of shared responsibility and integration among local government officers from various sectors who have made up joint task forces to promote basic UNICEF-messages about mother and child welfare in the villages (1998, 98).

3. Hanns Seidel Foundation (HSF) The third significant CE provider in Mtwara Urban District is Hanns Seidel Foundation. This German based organization has sponsored training for the district’s community leaders on good governance, accountability, and the rule of law since 2003. It has also provided women’s leadership training, and supported decentralization in Tanzania

Understanding the Context 47

by strengthening the local government. As is evident on the organization’s website, civic or political education occupies an important place in its mission: Former German Federal President Roman Herzog once said that "education towards democracy" was the "permanent and real responsibility of political foundations". He stated that this education helped "citizens of an open society to participate in the developmental process of a democracy with as much knowledge as possible". The understanding of a democracy must be newly acquired in each generation. (www.hss.de).

In contrast to the multiyear programs of RIPS and UNICEF, the awareness raising conducted by HSF has been project work. “All [of the foundation’s] projects are designed in such a way that the countries or partner organisations can take them over themselves in the course of time” (ibid). But like the training funded by UNICEF, HSF sponsored projects have been extended to all district’s village chairpersons, VEOs, and vitongoji leaders. They were conducted in 2003-2004 by council staff, coordinated/supervised by a person from HSF, and held at two community centers in Mtwara town. One of their achievements has been the establishment of “suggestion boxes” at the district council, where people can drop off their comments, questions, and concerns. 4. CHAWATA The final significant CE program in Mtwara Urban District in recent history has been voter education conducted a few months prior to the 2005 general election by CHAWATA, a Tanzanian organization for the disabled. Like training provided by HSF, this civic education has been more seasonal and geographically limited than the efforts of RIPS and UNICEF. CHAWATA was assigned the task of voter education in part because the UNDP basket fund, from which finances came, sought to promote the participation of the disabled community in Tanzania (Hendra 2005).75 The process was initiated when CHAWATA, as one of over 350 NGOs responded to UNDP’s call to NGOs to conduct voter education. UNDP sponsored training to “core trainers who in turn taught the civic education curriculum to nearly 800 locally engaged community based educators from the funded NGOs” (UNDP, page 2). The theme in the voter education workshops in Mtwara was ”true democracy,” in which people were explained their right to vote, the importance of voting, meaning of citizenship, and contents of the

48

Democratic Participation in Rural Tanzania and Zambia

constitution.76 ”When you take the time to vote, you prepare the government which will be accountable to the people; in that sense you are the government’s employer,” the voter educator described his teaching. He used two different formats—informal (to those without education), which communicated the information via posters—and formal (to those who have gone to school), which employed a special training manual.77 The four villages that hosted the workshops were selected by the municipal director on the basis that they had many opposition supporters. In turn, participants were selected by VEOs and ward executive officers (WEOs) from among the various political parties and special groups such as women, youth, the disabled, and the elderly. Mansa District Compared to Tanzania, the most notable difference in civic education provided in Zambia is the absence of local government involvement. In Mtwara the council civil servants were involved in implementing all the CE programs described except voter education. In Mansa District, in contrast, virtually all civic education has been conducted by nongovernmental actors. The main providers, according to local government officials and NGOs, have been Anti-Voter Apathy Project (AVAP), Foundation for Democratic Process (FODEP), various women’s organizations, the Catholic Church, Southern African Centre for the Constructive Resolution of Disputes (SACCORD), and the Legal Resources Foundation. But because the activities of FODEP and SACCORD have been very limited in Mansa District, discussion here focuses on the four other (sets of) organizations. 1. AVAP A 10-year-old national NGO whose primary goal is to increase citizen participation in the electoral process, AVAP has provided civic education in Mansa since 2003. However, its CE agenda also includes the constitution and “other issues bordering on civic awareness.”78 Thus, although its focus is on elections, the organization works year-round, with the idea to “teach people how they can lobby for their own development.”79 One of the activities for which AVAP is well known is maintaining Democracy Information Centers where citizens can stop by to ask questions, or read newspapers and other materials. Such a center is also located in Mansa town. AVAP also organizes village and school visits, radio programs, and monthly discussion fora, where invited politicians, council staff, NGO personnel, political party members, and church leaders debate topical issues with local residents. During a

Understanding the Context 49

typical village visit, AVAP staff hold a public meeting in which they discuss the importance of responsible leadership and of carefully considering whom to vote. The organization also coordinates with church leaders about bringing CE to churches. At the time of data collection, AVAP’s voter education for the 2006 election was about to begin, including weekly radio programs. According to AVAP, its radio programs have been popular: Apart from phoning in a large section of residents have been writing letters to the programme as well as the democracy information centres bringing out their concerns especially in areas of governance and development. In their airing of grievances they have been mostly tak[ing] their leaders to task demanding for attention in areas they have been citing. This has strengthened the zeal for advocacy in that concerned offices have been getting demands from the general public directly through the media which has made it difficult for them to sit on and ignore.80

However, AVAP’s effectiveness is limited by lack of resources. For example, outreaches are limited to villages and schools within walking distance from town. To reach a village further away, staff would have to skip lunch so that they can still get back before sunset. 2. Women’s Organizations Women’s groups in Luapula seem to be considerably more active and influential than those in southeastern Tanzania. Luapula has a multitude of such groups that periodically hold workshops on women’s issues and also educate people in the communities. These include: the NonGovernmental Organisations Co-ordinating Committee (NGOCC)—an umbrella body for women’s organizations—one of whose members is Mansa District Women Development Association (MDWDA), Women for Change, Women in Law and Development in Africa (WiLDAF), Women’s Lobby, Society for Women and AIDS in Zambia (SWAAZ), Forum for Advancement of Women in Education in Zambia (FAWEZA), and Planned Parenthood Association of Zambia.81 Participants for the workshops are selected by village headpersons and/or the chief; they may also include men and the youth.82 Training, for example that given by MDWDA, focuses on women’s self-reliance economically, politically, culturally, and socially, so as to reduce women’s dependence on men.83 This involves leadership skills, business management, functional literacy, cross-cutting issues like HIV/AIDS, legal issues, and civic awareness. In October 2005, MDWDA also

50

Democratic Participation in Rural Tanzania and Zambia

organized a seven-day seminar on the 2006 elections for 25 women— one from each of the district’s area organizations. 3. The Catholic Church Often known for its active involvement in social issues, the Catholic Church has been another significant CE provider in Mansa District, where it has taken a proactive role in defending human rights. Two persons interviewed at the church’s Department of Development, Justice, and Peace (DDJP)—a branch of the national CCJDP—said, As a church, we are very much involved in civic education to make our people aware . . . . Civic education empowers citizens: once they know their legal rights, they know when [the rights] are taken away and are ready to defend them and prevent the abuse of political power. . . . Some churches don’t even touch upon human rights; they leave it to politicians. But we base our teaching on the Bible and justice. That’s why some churches always think that Catholics have taken up politics.84

The DDJP puts out information, in print and on the radio, organizes workshops, and gives personal legal advice. It conducts voter education, educates people on corruption—what it is, its forms and consequences, and how to fight it—and tells them about those “pieces of legislation that people should know about,” such as those dealing with sexual crimes, land distribution, and property grabbing.85 The organization also conducts research. According to the DDJP representatives, sexual crimes take place because the majority of the people do not know the law against them, amended in just 2005. The church cooperates with many NGOs including AVAP, NGOCC, and the Legal Resources Foundation, particularly through the church’s Yangeni radio station. 4. Legal Resources Foundation (LRF) Finally, a partner with the Catholic Church, the Legal Resources Foundation is similarly involved in providing legal aid and promoting human rights. It also litigates public interest, mostly on behalf of the poor. Since 2002 this NGO has had an office in Mansa as one of nine LRF offices nationwide. Staff normally receive citizens in the their office, though they occasionally also conduct community outreaches. However, some of the outreaches—such as the prison visits program and mobile clinics—were curtailed in 2004 due to a lack of funds. The most common problems for which citizens seek help are police brutality, wrongful arrests and detentions, delayed trial, and labor disputes. The

Understanding the Context 51

majority of the clients are men, though “we do promote women’s rights in our outreach programs.”86 Tanzanian Villages

Lastly, with an understanding of the main CE programs conducted in the research sites, the study set out to select the villages where interviews would be conducted. While logistics set the boundaries for how far the villages could be located (data would be gathered during day trips from the town), the initial logic was to select one village in which participatory approach to development has been rather successful (as determined by development workers), and another where it has not worked as well. Of course this presents a dilemma of prejudgment87 but in the course of data gathering awareness and participation indeed turned out to be much lower in one Tanzanian village (i.e., Mtawanya) than in the other (Mbae). This is despite the fact that both villages have been recipients of participatory aid. Another criterion was to select one village closer to town (Mbae)—with the goal to analyze the influence of proximity to a semi-urban area—and the other a bit further from town. While the two villages were the main Tanzanian sites, a third site, Shangani Ward, located in Mtwara town, was added so as to ensure an adequate number of certain types of respondents. This slightly increased the number of Christian respondents and those with secondary education. Some descriptive data about the Tanzanian sites and civic education provided in them are provided in Table 2.4.

52

Democratic Participation in Rural Tanzania and Zambia

Table 2.4. Characteristics of, and Civic Education Provided in, the Tanzanian Research Sites Mbae Village

Mtawanya Village

Shangani Ward

N

63

58

19

Location

App. 3 km from Mtwara town

App. 7 km from Mtwara town

In Mtwara town

Population (2004)

1,321 (of which those up to 17 years of age: 389, or 29%)

1,989 (of which those under 18 years of age: 758, or 38%)

15,600

Sub-village/ sub-ward areas

4

4

17

Facilities

Dispensary, mosque, no schools (though there is both a primary and a secondary school 1.5-2 km away in another village)

Dispensary, 2 mosques, 2 churches (both are Roman Catholic), 2 primary schools (though one is several km away) and a vocational school (carpentry, masonry). Both a primary and a secondary school are currently being constructed.

Hospital, dispensaries, mosques, churches, primary schools, at least one secondary school (and a Shangani secondary school under construction), vocational school; services of a town, including Internet cafes

Education

Only 1 female has secondary educ. (she is currently at school); in addition some boys are attending secondary school; only one male resident has finished (Form 4 of) secondary school

Data n/a

Data n/a

Understanding the Context 53

Table 2.4. Characteristics of, and Civic Education Provided in, the Tanzanian Research Sites (Continued) Mbae Village

Mtawanya Village

Shangani Ward

1. RIPS

Yes (2003-2004: participatory development— council staff urged villagers to join community groups; app. 16 people)

Yes (2003-2004, app. 25 people)*

Data n/a

2. UNICEF

Yes (leadership)

Yes (leadership (5), sharing message with app. 100 residents; pilot village)

Yes (11 days in Oct.-Nov. 2005)

3. HSF

Yes (2004, leadership: app. 6 people)

Yes (2003, leadership)

Data n/a

4. CHAWATA

Yes (June 5-7, 2005, in Mbae: 20 participants; Sept.23, 2005, 4 hours, elsewhere: 12 participants from Mbae)

No

No

Interview dates (2005)

Oct. 4-7, 15, 19

Oct. 11-14

Nov. 5-6

Civic education:

* The Mtawanya village executive officer pointed out that this training was for the previous village leadership, which is no longer there.

The first site, Mbae village, is the smallest of the six villages in Mtwara Urban District and only about 3 km from Mtwara town, the administrative and commercial center of the region. Because of that, a number of development projects/programs have been carried out in Mbae, including the CE activities listed in Table 2.4. Also, district council staff have visited Mbae more often than, for example, Mtawanya. Therefore, the general view expressed by civil servants and

54

Democratic Participation in Rural Tanzania and Zambia

NGO employees was that Mbae residents are more aware of their rights and participate more readily in development projects than residents in many other villages. Also, according to one community development officer, women in Mbae are more active than those in Mtawanya in contributing to village meetings and business activities. In Mbae residents’ “plans move faster,” as demonstrated by a faster completion of the village government building (ibid). In Mtawanya, the second biggest village in the district, the same CE programs have been conducted as in Mbae, with the exception that it did not receive voter education by CHAWATA. Zambian Villages

The Zambian villages were selected following the same logic as in Tanzania: one should be closer to the provincial capital, and another a bit further. Yet the primary concern in Zambia was to ensure that at least some respondents had been exposed to civic education, as the provision of CE has been much less systematic there. For this reason two villages close to town—Chamalawa and Makasa, located right next to each other, and often site to CE by AVAP—became sites for data collection. The third village, Mabumba, is further, 20 km from Mansa, but still next to the Mansa-Samfya road and therefore not really isolated. Village data are not as systematically recorded in Zambia as they are in Tanzania— therefore Table 2.5. with Zambian data (below) is not as detailed as Table 2.4. is on Tanzania. In both countries villages are divided into sub-sections, each of which has a set of elected leaders.

Understanding the Context 55

Table 2.5. Characteristics of, and Civic Education Provided in, the Zambian Research Sites Mabumba Village

N

Chamalawa & Makasa Villages 32 37

Location

3 km from Mansa town

20 km from Mansa town

Chiefdom

Mabumba

Mabumba

Population

800-900

700-800

App. 3,000

Sub-village areas

4

5

2 (though until 2000: 4)

Facilities

One basic school (grades 19) close to Makasa, closest secondary school in Mansa, no court (closest in Mansa), no clinic (closest in Mansa; another: 4.5 km away), several churches

One basic school, one secondary school, court, clinic, several churches, no police post

1. AVAP

Yes (several times since 2003; e.g., Nov. 2005: 18 participants from both villages—i.e., total: 36)

No

2. Other

None identified

- NGOCC initiated 2-day training as part of “Women in Public Life Project” (2003-2004) - Workshops on HIV/AIDS and rights of women, children (2004; app. 30 participants) - awareness raising at churches (e.g., on HIV/AIDS by staff of Mabumba clinic*)

Interview dates (2006)

Feb. 20-25

Feb. 6-11

71

Civic Education:

* They also regularly discuss health related issues with the village headmen and do community outreach once weekly.

56

Democratic Participation in Rural Tanzania and Zambia

According to the Chamalawa village secretary, AVAP is the only organization that has conducted civic education in Chamalawa and Makasa.88 Table 2.5. suggests that AVAP has held a number of public meetings in the two villages since 2003—according to the village secretary, every two to three months, and announced in churches. In addition to elections and voting, topics have included the rights of women and children, and the roles of members of parliament, councilors, and ministers. More women than men attend; of attendees, most are middle-aged.89 In October 2005 AVAP also organized a meeting between a ward councilor and the area’s residents; because Chamalawa and Makasa are located in this councilor’s ward, it is possible that some residents from those villages attended. Compared to Chamalawa and Makasa, Mabumba is large and much further from town. It is also home to Chief Mabumba, chief to all the villages in the study. This may affect residents’ motivation and/or opportunity to participate. Another significant difference is that Mabumba village has had more development programs. This is partly because it is also home to the mayor of Mansa and many retired civil servants.90 Nevertheless, in the absence of AVAP and other NGO programs, and with the long distance from town, civic education seems to have been more haphazard in Mabumba. According to village leaders, there are very few of those who are “sent” from Mabumba to attend the (irregularly held) workshops in town. Timing of the Study and Conclusion

Finally, when analyzing research results it is important to bear in mind not only the geographic, cultural, and historical-political circumstances of participation, but also the timing of this study. In both countries surveys were conducted in election year, although in Tanzania this was done closer to the general election. Whereas data collection in Tanzania ended in early November 2005—about 1.5 months before elections— Zambian data were collected about 7 months before the Zambian general elections. Nevertheless, this means that in both countries, certainly in Tanzania, respondents had had recent exposure to voter education; also, because of the proximity of elections, civic awareness and interest in participation should have been at its highest. Coupled with the fact that respondents were generally the most active, literate members of their communities (see Appendix C), this means that this study’s results should reflect the high end of civic awareness and participation in the study sites.

Understanding the Context 57

The picture that emerges from comparing the national, regional, and village level contexts is that although there are a lot of similarities between Tanzania and Zambia—perhaps more than between any other two Sub-Saharan African countries—there are also enough differences to lead us to expect different kinds of results. Most importantly, citizen disposition toward politics, and opportunities for and patterns of participation differ; in these respects one should perhaps expect Tanzanians to be more active politically. Among the factors affecting citizens’ opportunities for participation, the extent to which the decentralization process has been implemented is greater in Tanzania. Also rural Tanzanians appear to have been more systematically exposed to civic education—with the local government being more involved in it. Nevertheless, the Tanzanian sample would be more likely to possess certain cultural characteristics unfavorable to democratic participation, such as a sense of dependence on government, passivity, and even apathy. It does seem that Zambians would more readily criticize the government for wrongdoings and punish it electorally. Certainly, the Zambian civil society as a whole is more active and aggressive. These observations, coupled with the expectation that civic education influences individual forms of participation, more prevalent in Zambia, more than it does mobilized acts, more common in Tanzania, suggests that civic education might have an easier time promoting participation in Zambia. Yet this chapter will have made it clear that because of the multiplicity of factors involved in explaining participation—including philosophies propagated by former political leaders, and policies on decentralization—CE’s effects need to be examined through a contextual interpretation of multivariate statistical analysis. Due to poverty and the isolated location of the selected regions within Tanzania and Zambia, making the “case for civic education” will be challenging. Proximity of elections means that results should reveal the “most” that civic education can achieve in these types of contexts—as respondents should be as interested in public affairs and participation as they ever will be. Therefore, if civic education is to have an effect on rural citizens’ propensity to participate in southeastern Tanzania and north(western) Zambia, it should be during the time that data for this study were collected. The extent to which data demonstrate connections between CE exposure and the various cognitive factors is uncovered and explained in Chapter 3, with linkages between civic education and participation examined in Chapter 4.

58

Democratic Participation in Rural Tanzania and Zambia

Notes 1 Diamond (2002) speaks of “hybrid regimes,” a term subsequently adopted by others. 2 Freedom House rates them “partly free,” with Tanzania scoring 4 and 3 for political rights and civil liberties, respectively, and Zambia scoring 4 and 4 (1 denotes “free”). 3 Zambia had its first multiparty elections in 1991 and Tanzania had its in 1995. 4 In fact Kaunda “named one of his sons after Nyerere” (Ishumi and Maliyamkono 1995, 57). 5 The one-party periods ended about the same time—in Zambia in 1991 and Tanzania in 1992, with Kaunda staying in power until that date but Nyerere stepping down from the presidency in 1985. Tanzania’s one-party system ended in a constitutional change, while Zambia’s concluded at multi-party elections. 6 See Msabaha (1995) and Legum (1995) for good descriptions of ujamaa. 7 Nyerere was in fact elected the first president of the International council of Adult Education (ICAE; Mulenga 2001). 8 Personal interview of a student of Tanzania (July 19, 2005). 9 Ibid. Hyden elaborates, “Of course, people often elect representatives from their own communities, but appeals to tribal or ethnic values do not work in Tanzanian politics” (1999, 151). 10 Tanzania ranks 159th and Zambia 165th (2007/2008, www.undp.org). 11 Tanzania’s Gross National Income (GNI) per capita is $350 and Zambia’s is $630 (2006; www.worldbank.org). 12 But, as Bratton et al. (2004) stress, be cautious when comparing the results of Afrobarometer surveys over time—i.e., the sources referenced in Table 2.1. There are some differences in the wording of some questions in Afrobarometer Round 1 (1999/2001) and Round 2 (2002/2003). 13 The adult literacy rate in 2005 was 69.4 for Tanzania and 68 percent for Zambia (http://hdrstats.undp.org/indicators/3.html). 14 November 20, 2005, p.6/Comment. 15 Personal interview (November 2, 2005). 16 Personal interview of a donor representative, formerly a Zambian civil society activist (March 7, 2006). 17 Ibid. 18 Ibid. 19 Ibid. As a result of its problems, the Zambian CE curriculum is presently being reformed: civic education is being introduced at the high school level, with also new textbooks being developed (personal interview of ZCEA executive director, January 16, 2006). 20 This was pointed out by two government officials (personal interviews, October 15 and November 2, 2005) and one NGO employee (personal interview, October 31, 2005). 21 Chaligha et al. (2002). That is, as much as 46 percent of Tanzanians agree more with the statement, “I can usually understand the way that government works” than with the statement, “The way the government operates

Understanding the Context 59

sometimes seems so complicated that I cannot really understand what is going on.” In Zambia this self-reported confidence is much less, 18 percent, which means that among 12 Afrobarometer countries, Zambians are positioned ahead of only Lesotho (15 percent) and South Africa (12 percent; ibid). 22 Whereas only 8 percent of Tanzanians could name their member of parliament in 2001, 44 percent of Zambians were able to name theirs in 1999 (ibid). And, by 2005, Zambians’ knowledge of the name of their MP had risen to 63 percent (Lolojih and Chikwanha 2006). 23 Lolojih and Chikwanha (2006). The perception that central and local government are the same thing is ”especially widespread in former one-party states” (Bratton et al. 2005, 244). Also, people in both Zambia and Tanzania often confuse government in general and the dominant political party (Bratton and Liatto-Katundu 1994; Green 1995). 24 Personal interview (February 14, 2006). To be sure, lack of awareness and capacity are problems at the leadership level as well. For example, the local ward councilors in Tanzania “and other elected leaders” often have “very low” education and professional qualifications (Mwaipopo 2004, 14), while Zambians too complain about the “ignorance among councilors about their proper duties” (Bratton and Liatto-Katundu 1994, 553). Another problem is that some new councilors come to office with “aspirations [that] might not be in the best interest of the community” (FODEP training manual for ward councilors, no date, page 9). 25 Personal interview (November 10, 2005). 26 Personal interview of an NGO employee (November 10, 2005). 27 Personal interview of a Tanzanian election observer (October 20, 2005). 28 See the previous note. 29 Personal interview (September 22, 2005). 30 Personal interview of ZCEA executive director (January 16, 2006). 31 Personal interview (November 2, 2005). 32 This was pointed out by research assistant Anthony Nyange in Tanzania. 33 Personal interview (January 11, 2006). 34 Personal interview (November 10, 2005). 35 Personal interview of a Tanzanian development worker (November 10, 2005). 36 Personal interview (September 27, 2005). 37 Personal interview of a government employee (January 16, 2006). 38 Personal interview (March 14, 2006). 39 Personal interview (April 5, 2005). 40 This is evident from the fact that in 1999 only 62 percent would have rejected a one-party regime (Chaligha et al. 2002). 41 And importantly, 33 per cent of Zambians are “not at all” interested in politics (Bratton 1999). 42 As a consequence, the word “participation” has a special ring in Tanzania: it is associated with something that the government wants people to do (personal interview of a development worker, September 3, 2005). 43 Also called village governments. 44 “[C]ouncils will be given more authority and a higher degree of autonomy while maintaining sufficient linkages with the center as demanded in a unitary state” (Republic of Zambia 2003, 42).

60

Democratic Participation in Rural Tanzania and Zambia

45 Argued by a local government employee (personal interview on February 14, 2006). 46 Personal interview of a student of Tanzania (August 11, 2005). Villageinitiated planning is part of the government’s policy, Opportunities and Obstacles for Development (O&OD), which “has been developed to institutionalize the local government reforms in line with the government aspirations to devolve decision-making powers to the communities” (www.pmoralg.go.tz/programmes/view_programmes.php?intItemID=5). 47 Personal interview of a government official (November 10, 2005). 48 Ibid. 49 The policy states that one of the elements in the “principal focus of Government” will be “carrying out intensive sensitization of the population of the implications and obligations of decentralized governance so as to entrench the broad based support already demonstrated for the policy, and ensure the active participation of the population in the building of a democratic culture at all levels of our society” (Republic of Zambia 2005, 5). 50 Personal interview (February 14, 2006). 51 Some chiefs summon a meeting only about once a year (personal interview of a development worker, January 20, 2006). 52 According to Afrobarometer data from 2001, three-quarters of Tanzanians report having attended a community meeting at least once in the last five years—a figure that in that round of Afrobarometer surveys was surpassed only by Ugandans (81 percent; Chaligha et al. 2002). 53 In fact in the earlier rounds of the Afrobarometer, of the 12 countries observed, only one (Malawi) surpassed Tanzania in the extent of party affiliation, and only one (Uganda, a “no-party” state) was behind Zambia (Bratton et al. 2005). 54 Personal interview of an NGO employee (January 11, 2006). 55 According to Duncan et al. (2003), about 90 percent of Zambians are Christians. Carmody’s (2003) figure is lower, 70 percent, with 1 percent of the population being Hindu, Muslim, and Jews, and the rest (i.e., 29 percent) traditionalists. 56 Personal interviews of a Catholic Father (February 13, 2006), a Training Chaplain in the Anglican Church (February 16, 2006), and a Pastor at United Church of Zambia (February 28, 2006). The Zambian Catholic Church, for example, though advocating the separation of church and state, is clear about its expectations that Christians vote. 57 With ten percent of Tanzanians saying that most people can be trusted, placing Tanzania only ahead of the citizens of Lesotho among 11 Afrobarometer countries, Bratton et al. (2005) refer to “low level of interpersonal trust in Tanzania” (2005, 194). In Hyden’s words, “in Tanzania . . . social as well as political trust is low” (2005, 18). 58 RIPS was a development cooperation program between the governments of Finland and Tanzania, and involved Mtwara and Lindi Regions. It was one of the biggest (if not the biggest) development cooperation program that Finland has had with any country. Also it was significant for Mtwara and Lindi Regions. Voipio says, ”Though a minor donor in Tanzania as a whole, Finland has long been the most important donor in Mtwra and Lindi” (1998, 78). Especially after 1993 RIPS focused on supporting participatory development, with democratic

Understanding the Context 61

development and strengthening civil society organizations being a focus starting in 1999. 59 Personal interview of a missionary (September 16, 2005). 60 Other ethnic groups in the region are Makua, Yao, Mwera, Ngindo and Matumbi (RIPS 2005). 61 Personal interview of a development worker and a missionary (November 4, and September 16, 2005, respectively). 62 See previous note. As Swantz puts it, “Early islamization [of coastal areas in Southern Tanzania] prevented participation of the population at large in education because education was based on western models and thus had a Christian colour to it. The anti-education spirit hit women especially hard, because of Islam’s tendency to keep women out of the public eye” (1998, 172). 63 Personal interview of an NGO staff member in Lusaka (January 16, 2006). 64 Personal interview (February 23, 2006). 65 Personal interview of an NGO staff member (January 16, 2006). 66 As an example, in the one Zambian district to which research reported in this book was limited (see below), all donors have been either INGOs—such as Plan International or Family Health International—or multilateral agencies, such as FAO or the European Union, which had a multiyear micro-projects program in the province. (The EU’s program ended in 2002. ) The province’s few bilateral donors have largely focused on agriculture. 67 In addition, Concern Worldwide has conducted rights-based development projects in Mtwara. But because it has not done so in the district selected, its development projects are not discussed here. Also, while the local Ufukoni Theatre Group has performed drama about citizens’ rights and duties and topical issues to a significant number of people in the district, it is not included in the CE providers analyzed because the content of thegroup’s performances is difficult to know and villagers’ exposure challenging to define. 68 Personal interview (September 19, 2005). 69 Personal interview of a former RIPS employee (October 15, 2005). 70 Sirkku Hellsten and Azaveli Lwaitama. 71 Note, however, that the 1977 constitution did not have a bill of rights until 1984, which did not go into effect until 1988. Also, the Tanzanian bill of rights is problematic, containing “certain clauses and controversial principles which in themselves can allow human rights violation or nullify their own authority in the protection of human rights” (Hellsten and Lwaitama 2003, 60). Further, according to Hellsten and Lwaitama, ”the Tanzanian attempt to combine both individualistic and communitarian approaches to human rights within a single legal document has . . . created some self-contradictory principles within the Bill itself, between the Bill of Rights and the rest of the Constitution as well as between the Constitution and the national laws” (ibid, 59). 72 Personal interview (September 22, 2005). 73 Personal interview of an official in Mtwara’s regional government (November 2, 2005). 74 Personal interview of a UNICEF representative in Mtwara Region (November 3, 2005).

62

Democratic Participation in Rural Tanzania and Zambia

75

(ibid).

76

Contributions to the fund came from ten donor countries plus UNDP

Personal interview of the voter educator (October 3, 2005). Teaching materials were all in Swahili and they were funded by USAID (Hendra 2005). 78 Personal interview of AVAP programmes manager, Lusaka (January 20, 2006). 79 AVAP programmes manager, Lusaka (March 14, 2006). 80 AVAP proposal to Development Cooperation Ireland about planned activities in 2006-07, p.4. 81 These were mentioned by a NGOCC representative, Mansa (personal interview, February 14, 2006). 82 Ibid. 83 Personal interview of four MDWDA representatives (February 13, 2006). 84 Personal interview (February 14, 2006). 85 Ibid. 86 Personal interview of two paralegals at LRF (February 22, 2006). 87 One development worker said it is difficult to determine in advance where the participatory approach has worked, and where it has not worked so well. 88 Personal interview (February 14, 2006). Though this may be his opinion only, we could not identify any other CE projects/programs either. 89 See complete data on the sex and age of respondents in Appendix C. 90 Personal interview of the director of development planning, Mansa municipality (March 1, 2006). 77

3 The Effects of Civic Education on Cognition To understand how civic education influences participation, one first needs to uncover CE’s effects on cognition—the likely mediating factors. Therefore, this chapter presents findings on civic knowledge and democratic attitudes, testing Hypotheses 1-4 about CE’s influence on “first generation” civic knowledge, efficacy, trust, and interest in politics. It begins by describing the prevalence of CE exposure among respondents. Then, in connection with each cognitive element, it gives a detailed overview of respondents’ level of awareness and democratic attitudes. This demonstrates the kind of group among whom effects were discovered, the poor’s cognitive (un)preparedness for participation, and the kinds of challenges that civic educators and other development agents face. Notice that in this and the next chapter the explanatory sections always begin with a bivariate correlation of each dependent variable with CE exposure. All dependent variables are also dissected by sex (and, where relevant, age), facilitating an evaluation of Hypothesis 5: do the disadvantaged segments of the population (here, women) receive a relatively larger benefit from civic education?1 Following this, regression analyses reveal whether any impact of civic education survives a multivariate context—the presence of competing explanations and control variables.2 These variables include various forms of participation, as cognition could well be shaped by interaction with others. Notice also that in this chapter regression results for the various cognitive elements are presented graphically only when the dependent variable has an adequate relationship (i.e., about or above a .25 correlation) with any CE variable and a CE variable is statistically significant.3 This means the reader can conclude that civic education does not have a discernible impact on those cognitive dimensions that are not presented in tables—that is, when using this study’s definitions, methods, and samples.

63

64

Democratic Participation in Rural Tanzania and Zambia

Civic Education Exposure

As is to be expected considering this study’s contextual approach (that is, one not based on analyzing certain CE programs), in both countries there are more respondents that have not been exposed to formal civic education than those that have. Recall that civic education exposure was determined primarily by respondents’ self-assessment (“Has anybody ever told you about your rights?”),4 although village records were also consulted. However, in Tanzania these records only existed in Mbae village (in oral form), and in Zambia in Chamalawa and Makasa villages, but there only with regard to CE conducted by AVAP. Information obtained from the leadership of Mtawanya in Tanzania was too patchy to include in statistical analyses, while also that from Mbae and Chamalawa/Makasa proved incompatible with the information given by respondents. For example, while Mbae leadership indicated that 21 respondents in their village had received training by Hanns Seidel Foundation (HSF), only two respondents self-reported this training. Similarly, of the 49 respondents in Chamalawa and Makasa, who according to village records had received training from AVAP, only 12 self-reported this training. These discrepancies may simply be because of forgetfulness on the part of respondents, rather than village leadership giving false information, but it serves to reinforce the importance that CE studies should place on respondents’ self-assessment of exposure. It can be argued that important training events should be remembered. Due to the discrepancies, only self-reported exposure is analyzed here. And because the share of respondents that reported exposure to each of the main CE programs (that is, RIPS, UNICEF’s CJF, HSF, CHAWATA, and AVAP) was low, exposure will be outlined also with regard to the type of CE source. Figure 3.1. displays overall exposure, the share of those receiving civic education from a formal source, and those who mentioned any of the following formal sources: school, government staff, or NGO. In the figure and the ensuing discussion, RIPS, UNICEF’s CJF, and HSF are included in CE given by government staff, while CHAWATA and AVAP are NGOs. The categorization of all CE programs can be found in the first table of Appendix B.

The Effects of Civic Education on Cognition 65

%

55 50 45 40 35 30 25 20 15 10 5 0

Tanzania Zambia

Overall Exposure

Formal

School

Government Staff

NGOs

Figure 3.1. Self-Reported Exposure to Civic Education (I): Overall Exposure, Formal Sources, and Three Types of Formal Sources

The displayed percentages refer to those receiving civic education at least once.5 The figure shows that many more Zambians (49 percent) than Tanzanians (33 percent) reported exposure. It also shows that Zambians have been more exposed to all types of CE except that by government staff. These findings are not surprising; on the contrary, that the local government has been more involved in CE in Tanzania was to be expected based on the country’s higher level of decentralization and local government involvement in villagers’ lives. Beyond formal CE, respondents reported informal sources, displayed in Figure 3.2. % 50

45 40 35 30 25 20 15 10 5 0

Tanzania Zambia

Informal Village Religious Other Leadership Org. People

Media

Radio

Pol. Reps/Party

Figure 3.2. Self-Reported Exposure to Civic Education (II): Informal Sources, Five Types of Informal Sources, and Political Representatives/ Party as a Source

66

Democratic Participation in Rural Tanzania and Zambia

Figure 3.2. shows that Zambian respondents also reported much higher levels of exposure to informal CE.6 This could be because they really have learnt about their rights more often from such sources, but it is also possible that the difference is indicative of a different connotation the survey question may have produced in Bemba and Kiswahili— despite the fact that care was taken to translate the question as similarly as possible, in an all-inclusive way. Reported informal and semiinformal sources include village leadership (2 percent of respondents in Tanzania, 4 percent in Zambia) and political representatives/party—the most important informal sources to Tanzanians. In contrast, religious organizations/church were mentioned by no-one in Tanzania but by 6 percent in Zambia. This is in accordance with the important role that church plays among Zambians, and also that perhaps a limited amount of rights information is given at mosques in coastal Tanzania. Among media, while the radio is an important source of information to Tanzanians, too, they did not identify it as such in response to this question. Sources of rights information other than those mentioned here were relegated to the “other” category, capturing responses of five persons (4 percent) in Tanzania and two (1.4 percent) in Zambia. It will also be important to know who the CE recipients are. Are rights taught equally to men and women? If exposure to formal civic education is disaggregated by sex, data demonstrate that in Tanzania women are disproportionately non-recipients. In contrast, Zambian men and women participate roughly equally. This is displayed in Figure 3.3.7

% 80 70 60 50 Female

40

Male

30 20 10 0 Tanzania

Zambia

Figure 3.3. Self-Reported Exposure to Formal Civic Education by Sex

The Effects of Civic Education on Cognition 67

The difference in exposure between the sexes in Tanzania is significant8 regardless of whether referring to overall or formal CE. In contrast, in Zambia the small difference between the sexes is not significant. This suggests that in southeastern Tanzania CE is more selective—not surprising given the restrictions on women’s participation in that area. This suggests a first conclusion: for CE to reduce inequalities between the sexes, opportunities to receive it should be provided equally to women and men, or disproportionately to women. Civic Knowledge

The round of oral one-on-one interviews in the Tanzanian and Zambian villages revealed that civic educators have much to do in the area of civic knowledge; awareness levels are generally low. This can be established despite the fact that the study did not utilize such a measure of knowledge which would enable one to judge answers either correct or incorrect, except in clear instances of error. Indeed, it is very difficult to objectively determine the correctness of rights, responsibilities, and even government policies. Almost anything can be expressed as some form of a right, responsibility, or government policy—even if it was not expressly stipulated as such in some written document. Yet there were a few answers that clearly belonged to the “erroneous civic knowledge” category. Examples include voting as a children’s right, a male respondent saying that women have the “right” to cook for men, and that organizing a peaceful demonstration is a citizens’ responsibility. The almost all-inclusiveness of correct answers means that it is especially alarming if a person cannot identify a single item. The share of those not able to do so, or those being only able to identify one or two, is indicative of a real deficit in citizen understanding. Aggregate Civic Knowledge

Civic ignorance is clearly more prevalent in the Tanzanian site. For example, mean “aggregate civic knowledge,” which sums up all (1) government policies, (2) citizen’s rights, (3) women’s rights, (4) children’s rights, (5) citizens’ responsibilities, and (6) government’s responsibilities mentioned by the respondent, is considerably lower among Tanzanians (9.91) than Zambians (15.24). See Figures 3.4. and 3.5. This means Zambians know almost double the number of items. 9 Another difference is that, while men are more knowledgeable in both countries, there is a much larger difference between the sexes in Tanzania (means: 11.58 and 8.05, respectively) than in Zambia (16.01

68

Democratic Participation in Rural Tanzania and Zambia

and 14.51). Also, in Tanzania this difference is significant,10 while in Zambia it is not. This suggests that as with the opportunity to receive civic education, in Tanzania women are disadvantaged. This could be due to a number of things, with men having greater access not only to CE but education in general, and possibly mass media and other sources of information. %

12

Mean = 9.91

10

Std. Dev. = 5.083 8

N = 140

6 4 2 0 0

2

4

6

8

10

12

14

16

18

20

22

24

26

Figure 3.4. Aggregate Civic Knowledge: Tanzania %

12

Mean = 15.24

10

Std. Dev. = 4.964 8

N = 140

6 4 2 0 0

2

4

6

8

10

12

14

16

18

20

22

24

26

Figure 3.5. Aggregate Civic Knowledge: Zambia

28

30

The Effects of Civic Education on Cognition 69

How does age affect knowledge? Do the youth, generally more educated, know more, or are those more advanced in age better informed? In Tanzania the latter holds true: the 65-77 year-olds are the most knowledgeable, though this does not necessarily indicate any broader pattern as there are only three respondents in this group.11 Because of this, the difference between them and others is not significant. Interestingly, distribution of civic knowledge among the rest of the age groups12 in Tanzania is quite equal. But in Zambia, where the most knowledgeable respondents are the 25-34 year-olds, the difference between the mean score for them (17.41) and that for others (14.73) is significant.13 This suggests that knowledge in Zambia is concentrated among younger respondents—although multivariate analysis is needed to understand whether other factors like education account for this. Though the study did not hypothesize on aggregate civic knowledge, how does this variable “correlate” with CE exposure? In past studies, civic knowledge has been the area where CE has had the greatest impact. Results displayed in Figure 3.6. are consistent with such findings, giving initial support to CE’s positive impact. Importantly, the differences in means between those exposed and those not exposed to formal CE are significant.14 This is also true for overall CE. %

18 16 14

No Exposure to Formal Civic Education Exposed to Formal Civic Education

12 10 8 6 4 2 0 Tanzania

Zambia

Figure 3.6. Aggregate Civic Knowledge by Self-Reported Exposure to Formal Civic Education

Regression Results on Aggregate Civic Knowledge Consistent with previous studies, OLS regression analysis indicates that civic education promotes aggregate knowledge. Table 3.1. shows that it

70

Democratic Participation in Rural Tanzania and Zambia

influences Tanzanians’ knowledge more: only in Tanzania is CE significant.15 And importantly, much more of knowledge is explained by the variables on Tanzania (44 versus 24 percent), making the findings about Tanzania all the more important. Notably, CE’s contribution is almost as large as that of education in general or respondent’s position in the community. Also, CE is more influential than sex: although being a woman negatively affects civic knowledge, its effect is not as large as that of CE. For those seeking to empower women, this is encouraging. Table 3.1. Explaining Aggregate Civic Knowledge: Civic Education and Other Factors Compared Tanzania Zambia _____________________________________ Beta R2 Beta R2 (block) (block) Social Structure Position in community .241* Farmer .006 Male .196* Rural Cognitive Awareness Education .273* Media exposure .115 Institutional Influences Communing .099 Active partic. at mtgs Dem. Attitud. & Discussion Lack of efficacy Like discussing politics Civic Education Overall CE .214* Government staff -.060 Constant 4.261 Full Model

.280

.099 .054 -.210*

.121

.065 .219* .020

.012

.005 .091

-

.069 -.160 .182*

.025

14.833

.439

.238

Note: * p.3). Organization of the variables is adapted from Bratton et al. (2005). N = 140 per country.

What is the magnitude of impact? According to the “R2 block” value, CE exposure explains 2.5 percent of variation in aggregate civic knowledge in Tanzania, above and beyond the effects of structural

The Effects of Civic Education on Cognition 71

factors, cognitive awareness, institutional influences, and democratic attitudes.16 This is not a small percentage since the impact of such influential variables as education and media exposure has already been taken into account. In the Zambian case, while a major part of variance is explained by factors other than those in the model, civic knowledge appears to be much less structurally determined. For example, in Zambia one’s position in the community does not even correlate enough with civic knowledge to be included in the model, suggesting that there, knowledge is more equally available among the rural poor. Civic Knowledge by Type

But aggregate data do not explain much when one wants to know whether civic education promotes some types of knowledge more than others. Recall that Hypothesis 1 proposed: “Civic education promotes the knowledge of civil, human, and political rights and responsibilities more than it promotes the knowledge of socioeconomic rights.” This hypothesis will be tested below; but first, awareness will be described based on survey categories: rights (and responsibilities) of citizens, women, children, and the government, and government policies. Figure 3.7. shows that Zambians score noticeably higher in all areas of civic knowledge except one: children’s rights. In children’s right, the best known area of civic knowledge in both countries, Tanzanians score slightly higher. This will be discussed in more detail below. Otherwise, there are more Tanzanians than Zambians in each area of knowledge who cannot identify a single item.17 For example, more than half (52 percent) could not name a single government policy, despite the fact that the question used such a word for policy (sera) which according to local observers is very common in, for example, popular radio programs in Tanzania. The study accepted statements like “building a new clinic” as a government policy; thus it was not required that respondents reference the names of the policies. Mentioning a subject area was enough. In light of this, not being able to identify any is truly significant. Respondents who belonged to this category made statements like, “I don’t know even one,” and, “I don’t understand.” These serve to demonstrate the perceived difficulty of these issues to some citizens, and possibly a sense of distance they feel from the government.

72

Democratic Participation in Rural Tanzania and Zambia

Mean Number of Items Mentioned 3.5 Tanzania

3

Zambia

2.5 2 1.5 1 0.5 0 Gov't Policies

Citizen's Rights

Women's Rights

Children's Citizen's Gov't Rights Responsib. Responsib.

Figure 3.7. Civic Knowledge by Type: Government Policies, Rights, and Responsibilities

Beyond government policies, Tanzanians were largely unaware of citizens’ rights: one-third could not identify any. Recorded comments include, “I don’t know because I am illiterate,” and “I don’t know because our leader hasn’t told us.” Yet another set of rights which was poorly known—in both countries—is women’s rights: in Zambia respondents were the least aware of these rights, when considering the share of respondents who could name none, or only one.18 Interestingly, analysis by sex reveals that in both countries, men identify more women’s rights; perhaps this is due to women expressing these rights only in relation to how they judge they are being realized in practice.19 The fact that in Tanzania, children’s rights were so well known when compared to the sample’s overall pattern merits further analysis. Interestingly, this finding seems commensurate with the fact that one of the most extensive, recent—and ongoing—CE programs in the Tanzanian research sites has focused on children. This is UNICEF’s Community Justice Facilitation (CJF), which trains and equips people to promote the survival and livelihoods of the most vulnerable children. In fact, Tanzanians’ awareness of children’s rights is almost twice the magnitude of their awareness of any other right. In Zambia, awareness of the three types of rights is more equal.

The Effects of Civic Education on Cognition 73

There is no reason to expect that any of the formal CE programs, with the exception of that on children’s rights, would have promoted one type of knowledge, depicted in Figure 3.7., more than others. But importantly, when awareness of children’s rights is juxtaposed with exposure to formal CE, fascinating differences emerge: in both countries those exposed to CE are significantly more aware of these rights.20 Such differences, depicted visually in Figure 3.8., suggest that perhaps the CJF program has played a role in promoting Tanzanians’ awareness of these rights. However, it is too early to draw this conclusion as many other factors could also be at play. Mean Number of Children's Rights Identified 3.5 3 2.5

No Exposure to Formal Civic Education Exposed to Formal Civic Education

2 1.5 1 0.5 0 Tanzania

Zambia

Figure 3.8. Identification of Children’s Rights by Self-Reported Exposure to Formal Civic Education

Beyond knowing their rights, members of a functioning democracy ought to know their responsibilities. These are less known in Tanzania: almost half (45 percent) could identify no or only one citizen’s responsibility. In Zambia awareness was higher, although still a quarter was similarly unaware of their obligations. This clearly has implications for democracy: citizens cannot play their role in society fully unless they know what role they are supposed to play. Of the specific responsibilities mentioned, the rural poor were most aware of the duty to work: in Tanzania 54 percent and in Zambia 40 percent mentioned this obligation in one form or another. It is logical that working would be the

74

Democratic Participation in Rural Tanzania and Zambia

most frequently cited obligation: people have to work to survive. In many villages residents have the experience that usually, development— such as building a school or a clinic—requires their input. In Tanzania no other obligation was mentioned by nearly as many but in Zambia there was one as frequently cited obligation: 41 percent said helping others was their duty. This is significant: if so many people consider it their duty to help others, then raising awareness of rights, obligations, and other areas of civic knowledge should not be difficult, as citizens are willing to teach, and be taught by, each other. This should make reaching the poor with CE messages easier. Willingness to assist others would seem supportive of democracy, too. In contrast, in both countries few people mentioned standard democratic obligations for citizens. These include voting (7 percent of Tanzanians, 6 percent of Zambians), participation at community meetings or in other ways (4 percent in both countries), expressing one’s opinion or advising leaders/government (4 and 10 percent, respectively), reporting problems to authorities (no Tanzanian, 6 percent of Zambians), and paying taxes (only 2 respondents, both in Tanzania). Because many of these have to do with self-expression—a building block of democracy—the relatively low awareness of these suggest that citizen understanding of their responsibilities in these countries with roughly 15 years of experience with formal democracy is not yet consolidated. The other types of responsibilities of which citizens in democracies should be aware are those of their government. What do the rural poor expect from their government? Though one cannot necessarily judge whether answers are in conformity with democracy—because it may be difficult to determine whether a respondent rightly reflects knowledge of government’s responsibility or his or her dependence on the government—it is instructive to outline some of these. Providing for people materially—whether food aid, agricultural inputs, infrastructure, or something else—is understandably the most often cited duty.21 But a few also mentioned some democratic obligations, such as involving people in development (planning; 5 percent of Tanzanians, 3 percent of Zambians) and ensuring people’s rights (4 percent in both countries); they also articulated the standard expectation that government ensures law, order, peace, and/or harmony (28 and 33 percent, respectively). Testing Hypothesis 1 In relation to the rights and responsibilities of citizens, women, and children, the study tested Hypothesis 1. For this, answers were categorized into civil, human, and political rights and responsibilities on the one hand, and socioeconomic ones on the other.22 To get a more

The Effects of Civic Education on Cognition 75

thorough understanding of how Tanzanians and Zambians understand their rights, answers were further categorized into: (1) civil, human, and political rights as a whole, (2) rights and responsibilities related to “expression and initiating,” (3) rights and responsibilities related to participation and voting, and (4) socio-economic rights and its two subcategories: (5) specific material rights, and (6) ownership rights. 23 Appendix B details the kinds of items that fell under each category, with Figures 3.9. and 3.10. presenting awareness of these items visually. Mean Number of Rights and/or Responsibilities Identified 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0

Tanzania

All Civil, Human, and Political Rights

Rights and Resp. on Expression and Initiating

Zambia

Rights and Resp. on Participation and Voting

Figure 3.9. Civic Knowledge by Type: Civil, Human, and Political Rights, and Two Subcategories of Rights and Responsibilities Mean Number of Rights Identified 4 Tanzania

3.5

Zambia

3 2.5 2 1.5 1 0.5 0 All Socioeconomic Rights

Specific Material Rights

Ownership Rights

Figure 3.10. Civic Knowledge by Type: Socioeconomic Rights and Two Subcategories

76

Democratic Participation in Rural Tanzania and Zambia

The figures reveal a clear contrast between civic knowledge possessed in the two countries: while Zambians are more knowledgeable of non-material, “first generation” rights and responsibilities, Tanzanians know more socioeconomic rights. The differences between the group means are statistically significant in all “non-material” categories plus the category subsuming all socioeconomic rights. 24 Many of the items in whose knowledge Zambians excel have to do with self-expression (for example, freedom of expression). However, in contrast to the Afrobarometer (e.g., Bratton et al. 2005), findings reveal that rights on self-expression are not the only kind identified by Africans: Africans do also know socioeconomic rights. This is demonstrated by Figure 3.10. and further detailed in Appendix B. Although Zambians certainly mentioned many more “first generation” than “second generation” rights, they also did mention the latter. Moreover, the stark difference in the type of knowledge held by Tanzanians and Zambians suggests that one cannot speak of any “African” conception of human rights. Also, levels of knowledge are not uniform across the sexes. In accordance with findings on aggregate knowledge, Tanzanian women clearly lag behind their male compatriots in awareness of civil, human, and political rights.25 This is noteworthy as knowledge of these is likely more linked to participation than is knowledge of socioeconomic rights. While Tanzanian men identified on average 2.59 “first generation” items, women mentioned only 1.48.26 And in rights and responsibilities on participation and voting, men named on average 0.73 rights while women were able to identify 50 percent less, 0.36.27 Interestingly, Tanzanian women know their socioeconomic rights almost as well as men—perhaps because of their responsibilities at home and roles as caretakers. In contrast, in Zambia there is no large or significant difference in men’s and women’s awareness in any of the six areas of knowledge, suggesting that cause-and-effect relationships should be analyzed separately in each country. Indeed, in Zambia women’s level of knowledge is almost at par with that of men in every area. When data are correlated with exposure to formal civic education, very interesting but expected results obtain: there is a considerable and statistically significant difference in knowledge in both countries between those exposed and those not exposed to CE. This applies both to the aggregate “first generation” civic knowledge and each subcategory,28 while statistically significant differences are absent in all categories of knowledge of socioeconomic rights. These findings are presented visually in Figures 3.11 and 3.12.

The Effects of Civic Education on Cognition 77

Mean Number of Rights and/or Responsibilities Identified

7 6

Tanzania, exposed to CE

5

Tanzania, no exposure to CE

4

Zambia, exposed to CE

3

Zambia, no exposure to CE

2 1 0 All Civil, Human, and Rights and Resp. on Rights and Resp. on Pol. Rights Expression and Participation and Initiating Voting Figure 3.11. “First Generation” Civic Knowledge by Self-Reported Exposure to Formal Civic Education

Mean Number of Rights Identified 4 Tanzania, exposed to CE

3.5

Tanzania, no exposure to CE

3 2.5

Zambia, exposed to CE

2

Zambia, no exposure to CE

1.5 1 0.5 0 All Socioeconomic Rights

Specific Material Rights

Ownership Rights

Figure 3.12. Knowledge of Socioeconomic Rights by Self-Reported Exposure to Formal Civic Education

78

Democratic Participation in Rural Tanzania and Zambia

The figures and statistical testing give preliminary support to Hypothesis 1. Civic education indeed appears to promote the knowledge of civil, human, and political rights and responsibilities, such as freedom of expression and right to vote while it does not promote awareness of socioeconomic rights, such as right to education, work, or social services. In particular, level of awareness of specific material rights and ownership rights is almost equal among those exposed to CE and those not exposed to it, while ownership rights are actually referenced in both countries more by the latter. But do these preliminary findings survive regression analyses? How does CE compare to other explanations for civic knowledge? Regression Results on Types of Civic Knowledge While it confirms a lesser or non-relationship between CE and the knowledge of socioeconomic rights, as hypothesized, regression analysis does not confirm a link between CE and the knowledge of children’s rights. However, this may be because in Tanzania there were only 19 respondents who reported participation in CJF, the UNICEF sponsored program on the well-being of children. Thus, because of a small number of participants, CE by CJF turns out insignificant in the sample. It could still be, however, that CJF has increased awareness, not only among program recipients but also among other community members with whom participants are encouraged to share their knowledge. If so, it is only natural that exposure to CJF turns out insignificant in the full sample. Indeed, something must be causing Tanzanians’ relatively high knowledge of children’s rights. It is likely not coincidental that an extensive child rights program has been conducted in the area. Of course, a conclusion about the impact of this program could only be made after comparing the knowledge held by Mtwarans to that of Tanzanians living in areas with no child rights program. Returning to the question of whether CE promotes “first generation” civic knowledge, tests are provided in Table 3.2. below. A quick look at the three dependent variables—one on civil, human, and political rights, and two on the subtypes of “first generation” knowledge—suggests that CE does exert significant influence. With regressions also run for each dependent variable by sex, the table indicates that women in both countries are more susceptible to learning about this kind of civic knowledge through CE.29 This is consistent with interviews, with several observers, especially in Tanzania, indicating that women are more receptive to CE. This lends support to Hypothesis 5 which expected benefits to accrue disproportionately to women. Much of the rest of the data in Table 3.2. also provide support for both H1 and H5.

The Effects of Civic Education on Cognition 79

Table 3.2. Explaining “First Generation” Civic Knowledge: Civic Education and Other Factors Compared Rights and Resp. Civil, Human, and on Expression Political Rights & Initiating _________________________ ____________ Tanzania Zambia (w) Tanzania (w) _________________________ ____________ Beta R2 Beta R2 Beta R2 (block) (block) (block) _____________________________________________________________ Social Structure .320 .084 .203 Age -.093 Position in community .174* .462** Farmer .012 -.244* Rural -.111 Cognitive Awareness .141 .100 .110 Education .218* .295* Vocational school .332** Media exposure .228* .051 Institutional Influences .008 .093 Communing .054 Active partic. at mtgs -.120 Voted in last cmty el. -.204* Dem. Attit. & Discussion .068 Lack of efficacy -.196* Civic Education .083 .153 .410 Overall CE .133 .253* .202 School .129 Government staff .366* -.312* RIPS .145 CJF -.315* AVAP .244* Informal sources .619** Constant .460 3.485 .360 Full Model .552 .405 .817 Note: * p