INFORMATION TO USERS

This manuscript has been reproduced frommicrofilm the master. U M I films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

A Bell & Howell Information Company 300 North Zeeb Road. Ann Arbor. Ml 48106-1346 USA 313/ 761-4700 800/521-0600

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Order Number 9507836

War as a social trap: The case of Tanzania

Francis, Joyce L., Ph.D.

The American University, 1994

Copyright ©1994 by Erancis, Joyce L. All rights reserved.

UMI 300 N. Zeeb Rd. Ann Arbor, MI 48106

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. WAR AS A SOCIAL TRAP: THE CASE OF TANZANIA by Joyce L. Francis submitted to the Faculty of the School of International Service of The American University in Partial Fulfillment of the Requirements for the Degree o f Doctor of Philosophy i n International Relations

Signatures of Ccrmmittee:

Chair:

I/UXb- o CO (S tt)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. © COPYRIGHT

by

Joyce L. Francis

1994

ALL RIGHTS RESERVED

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. WAR AS A SOCIAL TRAP: THE CASE OF TANZANIA by Joyce L. Francis

Professor John M. Richardson, Jr. Principal Adviser

ABSTRACT War can entrap the victor as well as the vanquished. Tanzania's victory over Uganda in 1979 was quick and decisive, and it was won with surprisingly few battle deaths. Yet, for Tanzanians it proved a Pyrrhic victory. Despite the defeat of Amin's forces and a two-year occupation, chaos continued in Uganda. The war and occupation caused unintended consequences that were far more costly for Tanzania—politically, socially and economically—than anyone anticipated. An analysis of this war and its consequences permits a theoretical and policy-grounded examination of the efficacy of war as an instrument of public policy and conflict resolution. The disparity between the anticipated and actual outcomes of this war is explained through the depiction of war as a social trap—a set of circumstances in the decision environment that distorts the decision maker's utility calculations, creating a false picture of the probable outcomes of alternate courses of action, thereby leading the decision maker to chose a course of action contrary to her/his self interest. To the extent that leaders who initiate war are acting within a social trap, their ability to anticipate outcomes is distorted. This calls into question war's efficacy both as an instrument of public policy and as a means of resolving conflict.

i i i

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS

The road to this dissertation has been long, and there have been many that I honor for their contribution. Erik Mailer and Steve Lord provided me with the opportunity to live and work in Tanzania for three years, starting me along the road. My parents, Hazel and the late Gene Francis, gave me the support to stay the course, and Nils Pedersen shared the journey. The American University proved a stimulating environment for this work. Professor Ruth Lane asked the questions which stirred my curiosity, and Dean Louis Goodman challenged me to transform my curiosity into a research project. Colleagues Kim Alexander, Betsy Cohn, and Mike Nojeim followed every step, cheering small victories and transforming defeats into opportunities. Committee members, Professors David Hirschmann, Guy Martin, and Fantu Cheru, shared their extensive knowledge of Africa and challenged me to keep my research relevant to an African context. Doing fieldwork in Tanzania was a challenge in which I was greatly assisted. Katy Cooper, Martha Honey, and Tony Avirgan offered their knowledge and valuable contacts. Nassir Mawji and his family went out of their way to make my stay comfortable. The staff and faculty of the University o f Dar es Salaam, the Institute of Science and Technology, the President's Planning Commission, and the Daily News were most generous with their resources. Scores of

iv

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. interviewees, some named in the text and others who remain anonymous, took time to help me understand their experience. Above all, I am deeply indebted to Professor John Richardson, director of doctoral studies and chairman of my dissertation committee. He was enthusiastic about the project when it was just a glimmer of an idea and never wavered. His vision was grander than mine, so he expanded my thinking. His criticisms were always accompanied by helpful suggestions, and his praise was based on high expectations and a thorough reading, so it was greatly appreciated. There could be no better mentor.

v

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS

ABSTRACT i i i

ACKNOWLEDGMENTS...... iv

TABLE OF CONTENTS...... v i

LIST OF TABLES...... v iii

LIST OF ILLUSTRATIONS...... ix

INTRODUCTION...... 1

PART I. BACKGROUND TO THE STUDY

Chapter

1. TANZANIA GOES TO WAR...... 14

Counter-Coup Attempt and Mogadishu Agreement .... 17 Invasion and Counter invasion ...... 22 Drive to Kampala ...... 31 Post-War Occupation ...... 39 Pyrrhic Victory...... 44

2. ASSESSING WAR'S LEGACY: THEORY AND METHOD...... 47

Pyrrhic Victories, War Traps, and Paradoxes ...... 51 War as a Social Trap...... 55 A Typology of Social Traps...... 66 Methodology...... 72

3. STATE OF KNOWLEDGE ON WAR'S CONSEQUENCES...... 80

War's Consequences of the National Economy ...... 84 War's Consequences for the State ...... 96 War's Consequences for the Individual and Society. . . 101

v i

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PART II. TANZANIA'S WAR LEGACY

4. ASSESSING THE CONSEQUENCES OF TANZANIA'S WAR 114

Decision to Counter invade ...... 116 Assessing Actual Consequences ...... 121 Primary Consequences ...... 123 5. SECONDARY ECONOMIC CONSEQUENCES...... 131 Trade D e ficit ...... 134 Foreign Reserves ...... 139 Debt...... 140 Impact on Production ...... 143 Opportunity Costs of War ...... 146 Analysis ...... 152 6. SECONDARY POLITICAL CONSEQUENCES...... 156 Concentration ...... 157 Militarization ...... 160 Disengagement ...... 162 Challenges to the Nyerere ...... 171 International Relations ...... 176 Summary and Analysis ...... 183 7. SECONDARY SOCIAL CONSEQUENCES...... 189 Disability...... 189 Crime...... 191 Social Services...... 202 Acquired Immunodeficiency Syndrome (AIDS) ...... 210 Summary and Analysis ...... 213

PART ill. CONCLUSION 8. WAR AS A SOCIAL TRAP...... 220 Effectiveness and Effect of Tanzania's W ar ...... 220 Unfolding of Miscalculations ...... 223

BIBLIOGRAPHY...... 236

v ii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES

T ab le Page 1. Timeline of the C onflict ...... 43 2. Estimated Direct Costs of Counter Invasion and Occupation ...... 128 3. Actual and Peace Scenario GDP and OpportunityCosts. . . 151 4. Amount of Theft of Public Funds by PublicServants 200 5. A Typology of Social Traps of War...... 226

v iii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF ILLUSTRATIONS

Figure Page 1. Map of East Africa ...... 19 2. The Initiator's Decision Problem ...... 63 3. Overall Impact of War ...... 66 4. Imports of Trucks...... 125 5. Volume Index of Selected Major Exports ...... 135 6. Constant Value of Imports and Exports ...... 137 7. Trade Deficit in Constant Values ...... 138 8. Foreign Reserves ...... 140 9. Total Long-Term Debt...... 141 10. Debt Service on Long-Term Debt ...... 142 11. Percentage of Export Earnings Required to Service Long-Term D eb t...... 143 12. Annual Growth Rate of Manufacturing GDP at 1976 Prices ...... 144 13. Annual Growth of GDP at Constant P rices ...... 145 14. Average Annual Growth Rate of GDP at Constant Prices. 149 15. Central Government Expenditures as a Percentage of GNP...... 158 16. Total Formal Employment by Sector ...... 159 17. Total Armed Forces ...... 161 18. Official Producer Price of Cassava ...... 164 19. Purchases by the National Milling Corporation ...... 165 20. Maize Production and Percentage Officially Marketed. . . 166 ix

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21. Index of Total Per Capital Agricultural Production 169 22. Government Tax Revenue in Constant 1980 Values 170 23. Official Development Assistance to Tanzania ...... 178 24. Food Aid to Tanzania ...... 179 25. Homicide Rate ...... 195 26. Incidents of Theft of Public Funds by Public Servants. . 198 27. Education and Health Care Spending as a Portion of all Central Government Expenditure ...... 203 28. First Year Enrollment in Primary Education ...... 205 29. First Year Enrollment in Teacher Training Colleges. . . . 206 30. Number of Division Health Centers ...... 208 31. Number of Ward Dispensaries ...... 209

x

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INTRODUCTION

I first saw Tanzania in November of 1974, as I flew into Dar es Salaam to begin a three-year stay. This was during the height of the Ujamaa villagization program and five years before the war with Uganda. After I left in 1977, several friends and colleagues who continued to travel there kept me informed of changes, particularly those brought by the war. My doctoral studies provided me the opportunity to give full rein to my continuing interest and to explore the country's experience in a systematic way. The questions addressed in this research were crystallized for me when I came across a passage in a book that purported to capture the highlights of Tanzania's development experience. The author was quite critical of the government's socialist experiment and villagization program. On the other hand, the writer felt that the world owed Tanzania a debt for having rid it of the menace of Idi Amin. This perspective struck me as poorly informed. While Tanzania's Ujamaa program has been the topic of a great deal of research, its war with Uganda has almost entirely escaped the attention of scholars, except as a case study of the just war doctrine. The war's unintended effects have remained unexplored. My own sense is that this war had at least as much impact on the country's development as villagization. Therefore, I am surprised

1

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2

that so little has been written about its consequences. Perhaps this oversight is because this war is seen as just another among the all too many that have occurred in post-colonial Africa, while the villagization program, by contrast, was unique. Or perhaps the fact that the war was so decisively won in a relatively short time led scholars to underestimate its significance for Tanzania. Most distressing to consider is the thought that few scholars consider war and other forms of political violence to be a factor in development. This research seeks to redress what I consider an imbalance in the research on Tanzania's development. It also seeks to provide lessons from the Tanzanian experience of war that might inform policy makers who are considering the use of violent force to resolve conflict. War is a public policy that has enormous consequences for the victor as well as the vanquished. Those consequences must be assessed in order to evaluate the efficacy of war as a policy tool. Tanzania's war with Uganda provides a useful case in which to evaluate war's effectiveness at achieving desired outcomes and its effect, or unintended consequences. Tanzania won the war decisively and relatively quickly with surprisingly few losses on the battlefield. It was a victory that surprised even the victors, but it did not bring all the changes that were desired, and the final cost was enormous, far more than anyone anticipated. Therefore, an assessment of Tanzania's war offers particularly valuable insight

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3

into the nature of victory through war, since it is based on what is perhaps a best-case scenario of war. Any such analysis must take place against the background of a wide-ranging debate on Tanzania's "African experiment." (Yeager 1989) Tanzania's development experience has been unique and has caught the attention of countless scholars. Their assessments of the country's development run the full gamut of opinions, as the following two passages demonstrate:

Tanzania's economic crisis was caused more by internal policy failures than by outside factors. Even compared to other African countries that have confronted roughly similar outside factors, its economic performance has been so abysmal that the net effect of these factors cannot really be judged until the country performs up to its current agricultural potential. (Lofchie 1985, 60)

Even though such statistics are always subject to a degree of error there is no doubt that the available comparative empirical evidence cannot be used to suggest that Tanzania's experience, as regards social and economic change, has been significantly worse than that of other sub-Saharan economies. There is, thus, simply no defensible basis for the many studies that set themselves the task of explaining Tanzania's 'failure'. In these relative terms, there has been no 'failure', while in absolute terms, there has been 'failure' in most of Africa (Author unknown, quoted in Young 1988). Amongst those critics who see the Tanzanian experiment as a failure, there continues a debate concerning the source of the problem. The radical critique holds that the problem rests with the international capitalist system and contends that, despite the socialist rhetoric, Tanzania's experiment was a textbook case of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4

dependent development and peripheral capitalism. Tanzania's political process allowed such a paradox because Nyerere's brand of socialism called for political mobilization of various classes and ideologies into a united front rather than a revolutionary vanguard in the Marxist sense (Othman 1988). This form of political organization deepened rather than lessened class divisions because, once the unifying goal of independence was reached, bourgeois class interests could exploit inherent contradictions between segments within the party and capture party leadership.

The movement towards one-party system of government in Tanganyika was part and parcel of the response of the petty bourgeois class in power to consolidate itself and stifle political opposition in face of sharpening contradictions in the country resulting from its failure to deliver the goods of independence to the people. (Mwakyembe 1985, 35-56) This critique argues that the development plans that emerged from the party (TANU) leadership were in fact intended to further integrate Tanzania into the international capitalist system by creating conditions that would attract foreign capital and move local capital into the industrial sector. These party interests were in direct contradiction with those of the trade union movement, a "center of power" (Shivji 1986) and the only serious competition for TANU in the political arena. "Thus both at the political and economic fronts, the trade unions posed a major challenge to the new regime. From this perspective, it is easy to understand the government's nervousness and its eagerness to closely control trade union activity." (Nyang'oro 1989, 71-72)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5

TANU's development policy was captured in the Arusha Declaration, a populist document that elaborated Nyerere's vision of an African socialist society dedicated to self-reliance. Shivji charges that, rather than a socialist manifesto, "The Arusha Declaration was the most decisive turning point in the struggle between the petty bourgeoisie and the commercial bourgeoisie leading to the latter's disintegration," and that the nationalizations which followed "constituted the first open attempt on the part of the bureaucratic sector of the petty bourgeoisie to carve out an economic base for itself." (Shivji 1976, 79) In her critique of post-Arusha policy, Bolton supports this view, noting that nationalizations do not necessarily occur as an integral part of socialist strategy.

Nationalization strategy can, in fact, consolidate certain relationships and developmental directions which, theoretically and politically, might be conceived of as obstacles to a transition to socialism. . . . Nationalization within this context usually represents a strategy aimed at repairing or consolidating a capitalist economy. (Bolton 1985, 2) Bolton's case study of the nationalized sisal industry in Tanzania demonstrates the "unsocialist" nature of the nationalizations that took place in Tanzania after the Arusha Declaration. In a less ideological analysis, Coulson begins with the same assumption of a "bureaucratic bourgeoisie" capturing the political process. However, he comes to rather different conclusions about the motivations and allegiances of the Tanzanian ruling class.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6

We have seen that it was not a class of accumulators: it had little experience of industrial production and marketing; it had no experience of large-scale agriculture, and little faith in small-scale agriculture (indeed, most of its leaders and their parents had sacrificed to get their children into schools precisely in order to remove them from the necessity of hard work for little reward on the land). . . . In short, the Tanzanian ruling class was almost uniquely unsuited to bringing about economic transformation. It had learnt (mainly from the British) how to take over and operate the colonial institutions, and it was able to extend this and develop excellent relationships with a number of overseas governments willing to provide capital on concessionaire terms. But its allocation of resources internally was always likely to be unplanned, and in the extremely difficult world competitive conditions of the 1960s and 1970s this would not allow it to make much progress towards developing a strong integrated economy. (Coulson 1982, 326-7) Thus, in this analysis, the ruling elite were not the willing tools of international capitalism but rather the incompetent legacy of the colonial system, willing dependents of international aid. In contrast to the radical analysis, the liberal critique hold that the roots of the crisis lay in the government's economic interventions and preoccupation with transforming social and economic relations. This statist economic policy has proven itself inefficient and suppressed any incentive for expanding production. This view—advocated by the IMF, the World Bank, many business analysts and conservative economists—holds that further integration of Tanzania into the international economy, along with the introduction of market-oriented domestic policies, would create an "enabling environment" for growth.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7

Michael Lofchie, for example, argues that statist policies can be found in many countries of Africa, but in Tanzania they have been pursued more thoroughly and relentlessly, with disastrous results.

Four policies merit special attention because of their ruinous impact on the country's once-thriving agricultural system: (1) overvaluation of currency; (2) suppression of agricultural producer prices; (3) implementation of agricultural policy through a system of monopolistic parastatal corporations; and (4) pursuit of an industrial strategy based on the principle of import substitution. (Lofchie 1989, 119) Thus, Lofchie places the blame for what he see as Tanzania's failure on the state and its adherence to socialist development policies. In a pathbreaking analysis, Bevond Uiamaa in Tanzania: Underdevelopment and an Uncaptured Peasantry (1980), Goran Hyden charges that both the radical and liberal paradigms are inadequate to understand the challenge of development in Tanzania. "Marxist and marginalist economists alike have overemphasized the market as the determinant of African development." (Hyden 1980, 20) It is not, he says, exchange relations that are at the root of Tanzania's underdevelopment, but the "peasant mode of production" which gives rise to an "economy of affection." He points out that African peasants are unique in that they own the means of production—land. Thus, they are self-sufficient, unlike peasants in other parts of the world or other social classes in Africa. Their access to land and to the labor of family members offers them security against external (and exploitive) forces such as the market or the state.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8

[The] economy of affection is primarily concerned with the problems of reproduction rather than production. . . . In the economy of affection, economic action is not motivated by individual profit alone, but is embedded in a range of social considerations that allow for redistribution of opportunities and benefits in a manner which is impossible where modern capitalism or socialism prevails and formalized state action dominates he process of redistribution. Based on the social logic of the peasant mode, the economy of affection negates many of the power relationships that characterize the modern economies. (Hyden 1980, 18-19) Hyden argues that, while the economy of affection assures survival (reproduction) of Tanzanian peasants, it inhibits their development (production). The path to rural development lies in a thorough capitalist transformation of the means of production, rather than the socialist transformation that the country unsuccessfully attempted.

Building socialism requires people who have already been modernized: whose mental scheme is not religious but scientific; whose social position makes them accessible to alternative forms of social and political action. . . . It must be pointed out, however, that the task that socialist have set themselves is far more difficult than that of the capitalists. In the capitalist system, the market forces provide a free ride through history, so to speak. (Hyden 1980, 247) Thus, Hyden argues, Tanzania's development has been hindered by its attempts to build socialism from a pre-capitalist base without first passing through a capitalist revolution. Analysts who take a broad view of development (including economic, social and political factors) find a mixture of successes

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9

and failures in the Tanzanian experiment. Yeager, for example, concludes:

In terms of combining social equity and basic-needs satisfaction with governmental stability and mass political participation, the Tanzanian experiment has already accomplished more than can be claimed for most other African countries. Entrepreneurial freedom and economic growth have been sacrificed in the process. . . .

Since the Arusha Declaration, the state has indeed been used in attempts to mold Tanzanian society in the often flawed images of its leaders' perceptions and goals for the future. . . .

[Yet] Tanzania's domestic predicament is clearly related to its peripheral role in the international political and economic systems. Life at the periphery became more precarious during the late 1970s and early 1980s, when the industrial states adjusted priorities to compensate for their own international insecurities and domestic economic reversals. (Yeager 1989, 148-50) Yeager's sympathetic analysis is that Tanzania has made remarkable progress, yet its people have suffered both from its own ill- considered policies and from a hostile international economic environment Irrespective of their ideological perspective, all recent analysts have struggled to explain the dramatic decline in Tanzania's fortunes in the late 1970s, despite a consistent development policy. One observer notes:

For its first fifteen years of independence, up until 1977, Tanzania's development story was one of almost unbroken success. It had a trend growth of over 5 per cent; its external reserves were equal to five months of imports; it had an

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 10

overall food surplus and a record Recurrent Budget All the signs were that it had overcome the triple crisis of drought, food and oil prices of 1973-75. The country's socialist policies were clearly no impediment to its economic growth and development, and probably assisted it. Nor was its record visibly affected by the Arusha Declaration which was made a full ten years before the decline set in. (Legum 1988, 5) There is no longer any serious debate on whether or not Tanzania's development path led to failure. By 1983, the country was on the brink of disaster. The debate now revolves around whether or not the primary cause is internal policy or external shocks, outside the state's control. Those who primarily blame the crisis on internal policy continue the ideological debates of the first decade and a half—that Tanzania suffered from either too much or too little socialism. Those who primarily blame external shocks fo r the crisis generally agree on the list of causes. A representative list by Cheru (1989, 51) identifies the external factors that brought about the crisis as:

(1) deteriorating terms of trade resulting both from oil price hikes and decline prices for primary commodities; (2) breakup of the East African Community (EEC) in 1977, necessitating the replacement of previously shared services; (3) high interest rates, exacerbating the burden of debt service; (4) drought; and (5) war with Uganda.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11

While there is general agreement within this group of analysts as to the external sources of the crisis, there is some debate as to the relative importance of the factors. In this work, I do not attempt to enter the ideological debate on Tanzania's development policies and whether or not they have led to success or failure. I do offer an alternative argument as to the source of the crisis of the late 1970s. I argue that the war was a far more significant factor in Tanzania's crisis than has been acknowledged. Further, I argue in Chapters 1 and 2 that the war should not be considered external; rather it was the result of national policy, just as much as the villagization program of the early 1970s. Thus, I argue that it is appropriate to evaluate the efficacy of Tanzania's war policy, just as its villagization policy was evaluated. Thus, in this research, I assess the war's effectiveness at achieving the nation's goals and its effect on the nation's development. This assessment of Tanzania's experience of war explores the disparity between anticipated and actual consequences. It addresses three questions: (1) What were the actual consequences of the war throughout the decade following the war's onset, and how were these consequences experienced by different segments of the population? (2) Why were many of these consequences totally unforeseen and others so discounted by Tanzanian decision makers? (3) What generalized lessons can be drawn from this case about war's effectiveness and effect? I attempt to understand how rational

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12

leaders, despite worthy intentions, could set a nation on a war path that was ultimately so destructive. This dissertation is organized in three parts, following the sequence of the research. Part I sets the stage for the research. It reviews the history of the conflict, the theoretical and methodological approach taken in the research, and the literature on war's impact which informed the search for data. Part II presents the data on the war's consequences, distinguishing the immediate outcomes from the long-term consequences for the national economy, the state, and society. Part III assesses the war's effectiveness and effect and draws generalized conclusions about war as an instrument of policy.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PART I

BACKGROUND TO THE STUDY

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1

TANZANIA GOES TO WAR

We have only one task. It is. . . . to drive the snake out of our house.

Tanzanian President Julius Nyerere in his Declaration of War

The Tanzanian-Ugandan war made headlines perhaps only in East Africa. Few people are well informed about the war, even among those who follow events in Africa. Most remember Idi Amin and some of the horror stories of his rule, as well as the fact that he was ultimately overthrown. Few know that the Tanzanian army was primarily responsible for his overthrow. When apprised of that fact, people with whom I have discussed the war are invariably surprised. They think of President Julius Nyerere as a man of peace and Tanzania as a very poor country with too small an army to defeat what was considered to be one of Africa's most powerful militaries. While the purpose of this research is to assess the outcome of this war for Tanzania, the war's obscurity requires that this first chapter be devoted to a short history of the war and the events surrounding it. This background is essential to understand the provocation which led Julius Nyerere to declare war and order the invasion of Uganda. It is also important to know how the conflict 14

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 15

escalated to the point that the Tanzanian army drove Idi Amin out of the country and remained in occupation of Uganda for another two years, embroiling Tanzania in Uganda's continuing chaos that it was powerless to stop. Tanzania was "long regarded as black Africa's most peace- loving country." (African nation 1979, 41) Mainland Tanganyika experienced a peaceful transition from British colonial rule to independence. Though Zanzibar endured considerably more internal turmoil at independence, a union was peacefully forged in 1964, creating the United Republic of Tanzania. From independence until 1979, the Tanzanian government experienced no significant domestic or international threat. Julius Nyerere became the firs t president of both Tanganyika and the United Republic of Tanzania, serving until 1985, when he retired to the chair of the then-sole political party, Chama Cha Mapinduzi (CCM) or Revolutionary Party. His most remarkable legacy was the virtual elimination of ethnic conflict. Tanzania is the world's most culturally heterogeneous population, incorporating more than 130 ethnic groups (Kurian 1987, 1904). Such diversity, along with other factors, meant that no cultural group dominated the political arena, and a strong national identity emerged. Political conflict with a neighbor state ultimately threatened this relative tranquillity. On 25 January 1971, General Idi Amin, commander of Uganda's army and air force, overthrew President Milton Obote who was attending the Commonwealth Conference in Singapore. Obote immediately flew to Nairobi with the intention of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16

staging a counter-coup, but his efforts were thwarted by the Kenyan authorities who wished to have no part in the conflict.1 Obote's supporters inside Uganda were ineffective and were quickly crushed by Amin's forces. Within a few weeks, Amin had consolidated his hold on power and the presidency. Obote moved to Tanzania "where he was warmly received as Uganda's legitimate president." (Avirgan and Honey 1982, 34) Obote and Tanzania's President Julius Nyerere had a strong personal loyalty to one another. They—along with Zambia's Kenneth Kaunda and Kenya's Jomo Kenyatta—were founding fathers of their newly independent states. Obote, Nyerere, and Kenyatta had formed the East African Community. Obote and Nyerere had joined with Kaunda to form the "Mulungushi Club," a coalition of their three ruling parties (Avirgan and Honey 1982, 38). Though Obote's relationship with Kenyatta had grown strained due to ideological differences, Kaunda and Nyerere remained Obote's staunch supporters. Other African leaders shared Nyerere's reservations about Amin's coup. Heads of State of Kenya, Zambia, Ethiopia, Sudan, and Somalia joined Nyerere in withholding recognition of Amin's government. When, as part of the normal rotation, Uganda was scheduled to host the OAU Summit in June, 1971, many African leaders balked. A month before the summit was to convene, the OAU "decided to hold the meeting elsewhere and, not surprisingly, Amin

1 Obote's relations with the Kenyan government had been strained since 1969, when Obote expelled thousands of unskilled and semi-skilled Kenyans who were working in Uganda.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 17

had certain difficulties presenting his credentials." (Smith 1980, 164) The British government, however, had no such reservations and was very quick to recognize the Amin regime, much to the consternation of Obote’s supporters.2 The British government also brought its influence to bear on Nigerian leaders to offer recognition (Smith 1980, 83).

Counter-Coup Attempt and Mogadishu Agreement Tanzanian and Somali leaders immediately began planning support for an invasion of Uganda from northern Tanzania by Ugandan exiles. This plan was aborted because Nyerere feared that the British and Israeli military might intervene on Amin's behalf.3 Instead, training facilities for the Ugandan rebels were set up in both Tanzania and Sudan, from which a coordinated assault on

2 Britain's leaders had been distressed by Obote's ideological shift to the left, which began in 1968. Amin, it was hoped, would correct that shift. Moreover, at the Commonwealth Conference that had just taken place, Obote strongly challenged Britain's proposal to end the embargo against South Africa. Obote's eloquent counter to the proposal had visibly distressed British Prime Minister Heath (Mkapa 1993).

3 Israeli leaders had opposed Obote because he had refused the Israelis landing rights to supply the Anyanya, a rebellious tribe in southern Sudan. Amin, on the other hand, supported the Anyanya movement and welcomed the Israelis' support (Smith 1980, 70). Amin counted on Israeli support to the extent of flying to Israel in July, 1971, seeking its assistance to seize a part of northern Tanzania, thereby assuring access to the seaport of Tanga. The Israeli government refused, and Amin began courting Libya's President Qaddafi instead (Avirgan and Honey 1982, 11; Smith 1980, 84).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18

Uganda was planned for August, 1971. This second plan was also aborted when, in early 1972, Sudan's President Nimeiri made peace with the Anyanya rebels in the South. The settlement included an accord between Nimeiri and Amin in which each agreed to cease support for subversion of the other. Obote was then forced to move his supporters to Tanzania (Avirgan and Honey 1982, 34). By Summer 1972, pressure was mounting for Nyerere to approve an invasion of Uganda, to be carried out by Ugandan exiles from Tanzanian territory. Moreover, Amin was providing his own provocation. There had been numerous border clashes, and Ugandan planes had bombed a sawmill. In August, Amin publicly announced his plan to annex the Kagera salient, a triangle of approximately 700 square miles in northwestern Tanzania between the Ugandan border and the Kagera River (See Figure 1). With political pressure mounting from Ugandan exiles and Amin threatening invasion, Nyerere approved an invasion for September. The plan, however, was poorly coordinated and executed, and the invasion was a disaster. Most of the exiles that were not killed immediately were captured and later executed. The few that escaped retreated to Tanzania. In retaliation for Tanzania's complicity, Amin ordered the bombing of Bukoba and Mwanza, Tanzanian towns on Lake Victoria (Avirgan and Honey 1982, 35-36, and Smith 1980, 165). The invasion of Uganda cost Nyerere much of the diplomatic support—or at least neutrality—that he had enjoyed within the OAU for his opposition to Amin's rule. The governments of Nigeria,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19

Figure 1: Map of East Africa

SUDAN ETHIOPIA

ZAIRE

Lake SOMALIA Kyoea

KENYA

Kampala' / UGANDA

k a g e ra - s 4 Lake M t . Kenya Salient ^ Victoria

Nairobi RWANDA

Iwanza \BURUNDI

Mt. Kilimanjaro

Pemba Island

Zanzibar Dodoma i Island

Dar es Sjhlaam

Mafia Island

ZAMBIA

Miles

162 324 Kilometers

MOZAMBIQUE

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20

Guinea, Libya, and Egypt all issued condemnations of Tanzania's complicity in the invasion. Moreover, Libya's Qaddafi provided Amin with four hundred troops and material support against the possibility of another attack (Avirgan and Honey 1982, 36). Smith describes how the OAU ultimately was able to resolve the crisis.

The situation was becoming dangerous from the OAU point of view and it encouraged an initiative which was taken by the Somalian Government to call a meeting at Mogadishu in September 1972 to get Nyerere and Amin to agree upon a peace plan. With difficulty an agreement was reached. In essence it stated that operations of any kind against each other's territory should be stopped, also hostile propaganda directed against each other through the radio, TV or press, and that nationals or property of either state held by the other side should be returned. (Smith 1980, 166) The agreement also called for the Kagera salient to be made a demilitarized zone, a stipulation that would cost Tanzania dearly in 1978, when Amin's troops invaded with impunity. Nyerere kept to the letter of the Mogadishu Agreement. He closed Obote's training camps and, with the help of international aid agencies, resettled the approximately three thousand Ugandan exiles that had already fled Amin's regime, as well as those that continued to stream across the border. In all, nearly twenty thousand Ugandans would come to Tanzania during Amin's rule. Obote and his top aides were provided modest stipends but were kept on a short leash. Obote held no press conferences, and the few journalists that spoke to him were not permitted to quote him directly (Avirgan and Honey 1982, 37-38).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21

Both Obote and Nyerere did, however, keep up their campaign against Amin within the OAU. In May 1973, Obote sent a communique to all of Africa's heads of state charging that Amin had murdered tens of thousands of Ugandans in the two years since the coup and calling on African leaders to refuse to support Amin, either materially or morally (Avirgan and Honey 1982, 38). When Amin was offered another chance to host the OAU Summit in 1975—and with it the chairmanship of the OAU for the following year—Nyerere strongly protested.

By meeting in Kampala, the Heads of State of the OAU are giving respectability to one of the most murderous administrations in Africa. For this meeting will be assumed to have thrown the mantle of OAU approval over what has been done and what is still being done by General Amin and his henchmen against the people of Uganda. .. . Tanzania cannot accept the responsibility of participating in the mockery of condemning colonialism, apartheid and fascism in the headquarters of a murderer and oppressor, a black fascist and a self-confessed admirer of Hitler, (quoted in Smith 1980, 166-7) Despite the obvious truth of Nyerere's words4 and the political embarrassment Amin caused many African leaders, Nyerere's protest was not well received within the OAU, where perhaps half the heads of state reached their positions as a result of

4 It is widely believed that the eight-year Amin regime was responsible for the deaths of a half million Ugandans. For an elaboration of Amin's atrocities, see Kamau and Cameron (1979). Amin's reign was also deadly for the country's wildlife. During the Amin regime, the elephant population declined from about 30,000 to fewer than 2,000 (Africa News 38, 6 [April/May] 1993).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22 a military coup. Only Kaunda and Botswana's President Masire joined Nyerere's boycott of the OAU Summit in Kampala (Smith 1980, 167). For his part, Amin continued to harass Tanzania. On seven separate occasions over the next few years, he announced (falsely) that Tanzanians had planned or actually executed invasions of Uganda. "Further, Tanzanians living in or visiting Uganda were, on several occasions, abducted from their hotels and residences and killed by Amin's security forces." (Avirgan and Honey 1982, 37) With relations between the two leaders so hostile, the official communications and interactions that were required while the East African Community survived were quite often awkward.5

Invasion and Counter Invasion On 27 October 1978, two to three thousand Ugandan soldiers invaded the Kagera salient, demilitarized since the Mogadishu Agreement. Most observers believed the invasion was a decoy by Amin to distract the military from internal strife. One observer assessed Amin's motivation this way:

The invasion had its roots in a coup attempted on Oct. 8, when some 40 army officers attacked Amin's Presidential Lodge in Kampala. The "conqueror of the British Empire” and his family were ignominiously plucked from the scene by

5 Nyerere wished to have no interaction whatsoever with Amin. One former State House employee remembered that, when all three leaders East African needed to sign official EAC documents, the papers were to be brought for Nyerere's signature before they went to Amin so that Nyerere need not acknowledge that there was a relationship of any sort.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23

helicopter, but the mutiny later spread to the southern garrison cities of Mbarara, Mutukula, and Masaka, and 170 troops loyal to Amin were reported killed at the Bombo barrack just outside Kampala. By the third day of fighting, Amin was accusing Tanzania, a habitual whipping boy, of invading Uganda—his often-tried way of attempting to disguise civil disorder. (Clugston 1978, 41) Two hundred rebels had fled to Tanzania, pursued by Amin's soldiers. When the action to suppress the rebellion spilled over into Tanzania, Amin apparently took the opportunity to annex the long-coveted Kagera salient.6 The Ugandan army captured or killed the rebellious troops, then began a reign of terror on local civilians. The salient's population was fifty thousand, some of whom were Ugandans who had previously fled Amin's rule. Forty thousand of the area's inhabitants escaped south of the Kagera River. Fifteen hundred residents were killed and several hundred, mostly women and children, were abducted to Uganda. Rape of the prisoners was common. Ugandan soldiers destroyed the Kagera sugar factory (which produced five percent of Tanzania's total output) as well as a sawmill. They looted what sugar, coffee, tin roofs, and other items of value they could cart back to Uganda, also driving back twelve to fifteen thousand head of cattle. Fearing a counterattack, they blew

6 The ease with which this was accomplished was facilitated by the fact that the salient had been demilitarized since the Mogadishu Agreement. Tanzanian border troops were stationed at Kyaka, south of the Kagera River (Nsa-Kaisi 1993).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24

up the only bridge across the river large enough for tanks and artillery to cross. When word of the invasion reached Nyerere, who was in Songea in southern Tanzania, he called together his top military and security advisers.7 On October 30, after consultation, Nyerere ordered a general mobilization of the military, calling up what would ultimately be thirty-five thousand militia to add to Tanzania's standing army of twenty thousand. Amin then escalated the conflict by formally announcing on Ugandan radio that he had annexed the area south of its former border as far as the Kagera River and that "all Tanzanians in the area must know that they are under direct rule by the Conqueror of the British Empire." (Idi-otic Invasion 1978, 51) The following day, Nyerere issued a declaration of war against Amin in a fifteen-minute national radio broadcast that many Tanzanians still remember as his most moving speech. He stated, in part:

We have only one task. It is to hit him. We have the ability to hit him. We have the reason to hit him. And we have the determination to hit him. . . . We shall hit him. . . . We are all at war. All of us. (Daily News [Dar es Salaam], 3 November 1978) The United States, the United Nations, the OAU, and the Arab League pressured Amin to withdraw, and several African countries condemned Ugandan aggression, including Algeria, Angola, Ghana,

7 These included the late Prime Minister Edward Sokoine, Defense Minister Rashidi Kawawa, Home Affairs Minister Hassan Moyo, and Foreign Affairs Minister Benjamin Mkapa, as well as chiefs of intelligence, defense, and police (Kiwanuka 1985, 79).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25

Guinea-Bissau, Ethiopia, Kenya, Madagascar, Mozambique and Zambia (Komba 1979a, 12). While Tanzanians were very much behind their president, they were also confused and frightened. They had no experience of war. With no television, few radios, and only one newspaper whose delivery was sometimes unreliable, timely and accurate information was scarce and misinformation rampant. Not surprisingly, misguided action became all too common as paranoia spread. In Dodoma, Tanzania's nominal and centrally located capital, party leaders ordered secondary teachers to supervise the digging of trenches around the school. While Amin had boasted of bombing Dodoma, at no point was the capital at risk of invasion. In Tanga, on the northern coast, local authorities feared Amin would make good on his threat to occupy the port city and believed a tiny island just off the coast would be the ideal launching site for an invasion. Hence, they ordered all vegetation stripped from the island. What had once been a wildlife refuge drawing tourists literally disappeared as waves engulfed the island and swept it away. The armed forces were taking more effective action. Eight to ten thousand Tanzanian troops slowly moved into the Kagera region, no easy task as the front was 850 miles from Dar es Salaam on the coast—three days of hard driving over poor roads. The rainy season had just begun, greatly exacerbating the problems of troop movement. These monsoons were the heaviest in a decade, washing out roads and even bridges. Meanwhile, an air war ensued. Tanzania's artillery shot down several of Amin's planes, though they

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 26

also accidentally destroyed three of their own. Verbal threats between the two leaders also continued. Nyerere declared, "There will never be peace in East Africa until Amin goes," though he didn't specify whether this meant that Amin's troops must leave Tanzanian territory or that Amin must leave power (Behind the Lies 1978, 66). As international pressure mounted, Amin capitulated, and on November 8 he announced his intention to withdraw behind the traditional colonial border, though he did not renounce his claim to the Kagera salient nor were all troops in fact withdrawn. Tanzania's military was not, however, dissuaded from its war posture. By November 19, Tanzanian soldiers completed a pontoon bridge across the river and crossed without incident, as the bulk of the invaders had indeed retreated. They began clearing the area of land mines and small bands of Ugandan soldiers. Within five days, the damaged bridge was repaired, and tanks and heavy artillery poured into the salient as more troops arrived, including about twelve hundred Ugandan exiles mobilized from refugee camps. During a visit with an emissary from Nigeria's Head of State General Obasanjo, Nyerere showed signs of escalating the conflict. In early November, he had stated that the goal was "to drive the snake out of our house." (Daily News [Dar es Salaam], 3 November 1978). By mid-November, Nyerere's statements suggested that he now wanted retribution.

[Expelling Amin's forces was] no longer the problem. The problem is what next?. . . This is not simply a situation where the aggressor gets away with it. . . . Africa would be setting a dangerous precedent if it allowed matters to end there. The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27

OAU Charter has been violated. What is Africa going to do about it? (Daily News [Dar es Salaam], 18 November 1978) Nyerere visited the front and was convinced by his military officers that the Kagera salient could only be secured from the high ground a few miles across the Ugandan border at Mutukula. The Tanzanian military quietly began preparing phase one of its counter invasion. Tanzanians seemed ready for a fight after years of provocation from Idi Amin. Thousands gathered to cheer the troops as they departed from Dar es Salaam. They also seemed prepared to make the necessary sacrifices. The daily newspaper published lists of the citizens called up from the militia for active military duty. Sales taxes were raised on beer, alcohol, soft drinks, cigarettes, and textiles to pay for the war (PER 1979:1, 4). Government workers extended their workday by an extra hour to support the war effort. People patriotically conveyed stories of national pride and sacrifice.

The story is told of a rural farmer who appeared at a district office wanting to donate two cows to the cause, and of a wealthy Asian businessman who called a newspaper office and asked where he could send a contribution of 10,000 shillings (about $1300). (Ungar 1979,17) A government appeal for donations of cattle to feed the troops brought more than three times the cattle that the government had estimated it needed (Daily News [Dar es Salaam], 13 June 1979). Amin was also amassing his troops along the border, though the disarray and declining morale in the Ugandan army was taking its toll. Amin switched his front-line commanders at least twice and even personally assumed command for a time (Komba 1979a, 14).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28

During the next two months, numerous counter charges were made and border incidents reported as the two countries edged toward full-scale war. Nyerere was pressed from all sides to exercise restraint. The OAU Chairman, Sudanese President Jaafar Nimeiri, shuttled between Kampala and Dar es Salaam in an attempt to mediate. He was quickly rebuffed by Nyerere because, as spokesman for the OAU, he had failed to condemn Amin's aggression. Nyerere felt that only those who had condemned Amin could rightfully call for restraint. He blasted the OAU Charter as impotent against African fascists.

It does not matter what a Head of State does—he can kill as many people in his country as he likes. He will be protected by that Charter. That Charter is not there to protect the peoples of independent African countries. It merely protects their Heads of State. (Daily News [Dar es Salaam], 11 December 1978) The exiled Ugandan community was quick to take advantage of the shift in Tanzanian government policy from defense to retaliation. Milton Obote began holding press conferences for the first time since the Mogadishu Agreement. Ugandan exiles stepped up their coordination efforts, seeing the escalating conflict between Tanzania and Uganda as an opportunity for them to openly renew their struggle against Amin (Nabudere 1993). Ignoring all calls for restraint, Nyerere ordered Tanzania's troops to invade. On 21 January 1979, Tanzanian troops crossed the Ugandan border to take the high ground at Mutukula. Official statements, however, only allowed that, when Amin's forces again

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 29

attempted to escape across the border, they had been pursued into their own territory. The government made a point to state that it had no intention of occupying an inch of Ugandan territory (Daily News [Dar es Salaam], 23 January 1979), though it also claimed that Amin was planning a second attack on Tanzania which would be effectively resisted and "ensuing fighting will take place inside Uganda — not Tanzania." (Daily News [Dar es Salaam], 26 January 1979) In fact, Tanzanian forces met little resistance as they moved into Uganda; Amin's troops were so demoralized by this time that they simply fled ahead of the invading forces. Amin's internal troubles had reached a crisis point. The Tanzanian troops then proceeded to take their revenge for the massacres in the Kagera salient. The town of Mutukula was leveled, and all who didn't flee—military and civilian alike—were shot. This was not what Nyerere had in mind by retribution.

When news of the Mutukulu victory reached Nyerere, the Tanzanian president was predictably pleased, but he was appalled by the army's boastful report of the death and destruction they Isicl had wrought. Nyerere issued an immediate order that from that moment on a sharp distinction must be drawn between civilian and military lives and property and that the greatest care must be taken to protect anything nonmilitary. (Avirgan and Honey 1982, 70) Apparently, the troops had spent their rage in Mutukulu, and according to several observers, they largely heeded Nyerere's edict for the duration of the war. By February, Nyerere, encouraged by the ease of the first phase, decided to implement a second phase of the counter invasion—

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30

taking Mbarara and Masaka, major towns well within Uganda. Here lay the military barracks from which Amin's soldiers had staged the massacre in the Kagera Salient. Taking these targets was considered key to assuring that Amin could not again violate Tanzanian territory. Yet Nyerere was aware of the diplomatic storm he was creating and took steps to minimize the damage. Official government statements avoided any mention of Tanzanian troops in Uganda, saying rather that Mbarara and Masaka were besieged by "anti-Amin forces" and that Masaka had fallen to "advancing Ugandan resistance forces." (Daily News [Dar es Salaam], 23 February 1979) While this misinformation campaign fooled few international correspondents, it succeeded with private citizens. One businessman remembered that he first learned of Tanzanian soldiers fighting in Uganda after the war was over, when soldiers began returning across the border with looted items not available in Tanzania. With Ugandan troops in retreat, Amin was ready to negotiate and appealed to the UN, the OAU, and several countries to intervene diplomatically. The UN offered no response, as most members of the international community were appalled by the record of Amin's rule and looked forward to his demise. Even the Soviet Union, previously Amin's principal arms supplier, abandoned Amin when his defeat seemed imminent (Audifferen 1987, 272-3). The OAU, however, did respond, its members alarmed at the prospect of the war's escalation and the precedent that Amin's

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31

ouster would set. It appointed an ad hoc Committee of Nine, chaired by Nigeria, to meet in Nairobi with the foreign ministers of Tanzania and Uganda in a further attempt to bring about a cease-fire. When a delegation of committee members came to Dar es Salaam, Nyerere issued his conditions for ending the war. The OAU must officially condemn Amin's aggression—no OAU neutrality would be accepted. Moreover, Amin must renounce his claim to Tanzanian territory, give guarantees that no further aggression would occur, and pay compensation for the damage. Though Nyerere was expected to keep his word if these conditions were met, many African leaders believed he saw his chance to rid himself of the menace of Amin and set conditions that precluded any agreement. The OAU Charter contains no mechanism for the sanctioning of member states, merely stating that the organization "shall establish measures to be taken" in such an event (Audifferen 1987, 289). In its entire history, the OAU had never condemned a member state. Yet, Nyerere expected the OAU to take action against Amin's violation of Tanzania's borders. '"Such a move would split the organization wide open,' said an African diplomat. 'But unless Nyerere backs down, mediation efforts won't work.'" (Webb and Profitt 1979, 39)

Drive to Kampala If Nyerere was indeed only seeking to take the barracks at Mbabara and Masaka, as he maintained, his success presented him with a dilemma. Tanzanians had struck a serious blow to Amin's

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 32

army and its capacity to field another invasion. While this second phase had provided Tanzania with a buffer zone, would it remain if Tanzanians retreated? And what would become of the Ugandans who had welcomed Tanzanians as liberators? Apparently, no thought was given to retaining the position. The only choices considered were continuing or retreating. Despite the diplomatic storm that he knew would follow, Nyerere decided to continue, setting into motion phase three of the counter invasion, the taking of Kampala. This decision was influenced both by the OAU's failure to take action against Amin and the vulnerability of the Ugandan civilians who had welcomed Tanzania's "liberation" of southern Uganda.

Around Mbarara and Masaka the Tanzanian soldiers found dozens of notes addressed to Nyerere tacked to trees or left near watering holes. The notes, written in English and Swahili, begged the Tanzanians not to withdraw and said that if they did, the Ugandan civilians were coming with them. (Avirgan and Honey 1982, 86) Amin had taken large-scale retribution before when he felt he was betrayed, and the Tanzanians did not doubt that he would do so in this case. To lessen the political damage, Nyerere determined to continue downplaying Tanzania's role. At one point, he dispatched Dan Nabudere, secretary of the Political and Diplomatic Commission of the Ugandan government-in-exile, to the United Nations to convince delegates that Ugandans were playing an active role in the invasion (Nabudere 1993). Indeed, Nyerere did seek to involve

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 33

Ugandan exiles and anti-Amin forces recruited within Uganda, particularly around Mbarara and Masaka (now under joint command and known as the Uganda National Liberation Army or UNLA). His plan was to carry the battle to the outskirts of the capital, then leave the actual liberation of Kampala to Ugandans. Phase three would be a UNLA operation. These troops were proving less than effective, however. A planned attack on Amin's barracks along the Kenyan border had failed. Several UNLA fighters were killed while others were driven back into Kenya where they were arrested by Kenyan security (Avirgan and Honey 1982, 86-88). Nyerere's remaining self-restraint soon disappeared. In early March, Tanzanians received conclusive evidence that Libyan soldiers were assisting Amin. In a hard-fought battle for the Ugandan town of Lukaya, two hundred Libyans were killed. In all, the Libyan government had sent fifteen hundred to three thousand troops, as well as several cargo planes full of arms. Tanzanian soldiers feared the Libyans far more than Amin's forces, though they were later to learn that most of the Libyans were ill-trained militia members rather than seasoned troops. Nyerere's military advisers were expressing doubts that the UNLA could take Kampala unaided, but it was Libya's involvement that removed Nyerere's inhibitions. He reasoned that Libyan intervention justified Tanzania's assistance of the UNLA (Avirgan and Honey 1982, 93). He therefore ordered Tanzanian troops to participate in the Kampala campaign, much to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34

the delight of the Tanzanian field commanders.8 Phase three could be a joint Tanzanian-UNLA operation. Official government statements would reflect this shift, acknowledging the primary role that Tanzanian troops were playing in the invasion. As the invading armies approached the Ugandan capital, Qaddafi sent Nyerere a message via the Libyan ambassador to Tanzania demanding that Tanzanian troops leave Uganda within twenty-four hours. If Tanzania's troops were not withdrawn by the next day, Qaddafi vowed that Libyan forces would enter the war and help drive them out. Nyerere went public with the threat, noting that Libyan forces were already in Uganda yet were not stopping the Tanzanian army. Nyerere would be no more deterred by Qaddafi's threat than he had been by diplomatic pressure from the OAU. The following day, the Kenyan government also asked Tanzania to withdraw (Twaddle 1979, 219). Despite Kenyans' revulsion at Amin's policies, Kenyan President Moi feared that if Obote returned to power, there would be a Nyerere-Obote political alliance against him in East Africa. Therefore, Moi rebuffed Nyerere's repeated appeals since the start of the war to stop the flow of oil from Kenya's harbors to Uganda. Moreover, Moi allowed

8 At one point, Idi Amin was spotted from a reconnaissance post outside Kampala. No attempt was made to take action. The front-line commanders had previously agreed amongst themselves that they would not attempt to kill Amin, should the opportunity arise. They feared that he would be replaced by someone who would end the war and deprive them of the opportunity to take Kampala (Avirgan and Honey 1982, 95).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 35

refueling rights to Libyan aircraft bringing troops and military hardware to Amin. Nyerere saw Moi as openly siding with Amin, so he readily dismissed Moi's appeal for restraint. By mid-March, the Tanzanian forces reached Mpigi, high ground overlooking both Entebbe and Kampala. Nyerere, as well as many others, began to realize that Amin could indeed be ousted. Here Nyerere ordered the troops to halt in order to allow the political process to catch up with the military process. Nyerere and the exiles' leaders saw the need for a Ugandan government-in-exile to be defined prior to Amin's ouster. Nyerere had become increasingly aware of the animosity many Ugandans felt toward Milton Obote. Anxious to dispel charges from many quarters that his campaign against Amin was an attempt to return Obote to power, Nyerere organized a conference of one hundred representatives of various Ugandan exile groups. It was to convene at Moshi, on the slopes of Mt. Kilimanjaro in northern Tanzania.9 Seeing the conference as a threat to his authority, Obote opposed the conference, and Nyerere feared Obote's presence would sabotage the proceedings. "With considerable arm-twisting Obote was convinced not to attend, and he put his signature to a short letter of support for the conference's success." (Avirgan and Honey 1982, 105)

9 The Kenyan government, believing the conference was a ploy to legitimate an Obote presidency, prevented Ugandans in Kenya from leaving the country to attend the conference (Avirgan and Honey 1982, 112-13).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 36

The conference produced a Ugandan National Liberation Front (UNLF) under the leadership of a National Consultative Council (NCC), with Professor Yusufu Lule (former secretary general of the Association of African Universities and wartime administrator of the liberated town of Masaka) as NCC chairman. Obote had been effectively isolated, but that exclusion would come back to haunt the post-Amin period. Obote never fully forgave Nyerere for this isolation, and the two leaders' earlier rapport was forever changed. Entebbe fell on April 7. Most of Amin's forces had retreated to the capital, leaving Libyans to defend Entebbe, and three hundred Libyans were killed. Kampala was taken on April 10, but this time Nyerere offered the Libyans an escape route. Not wishing to further erode his relationship with Libya and other Arab nations, Nyerere sent Qaddafi a message informing him that the road east of Kampala would be left open for the evacuation of Libyan forces. This gesture, however, also provided an effective escape for Amin1** and his top aides. On 13 April 1979, at the capital's Parliament Plaza, Lule was sworn in as Uganda's new president. While the ceremony took place in Swahili, Lule spoke the last sentence of his acceptance speech in his Bagandan language. This dismayed the Tanzanians present, as the use of tribal languages is outlawed for all public speeches throughout Tanzania in order to lesson tribal conflict. If they had

10 Amin first fled to Libya, but he later settled in Saudi Arabia, where he still resides at the time of writing.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 37

understood, they would have been more than dismayed, as Lule's words foreshadowed the conflict to come when he said, "Now it is our turn." (Avirgan and Honey 1982, 153) Securing the rest of Uganda took only about six weeks. The timing of this and other phases was determined, not by the amount of resistance met (there was little), but by walking time. Tanzania had no transport for its troops. These soldiers literally walked from border to border across Uganda (Mkapa 1993). The last area to be secured was the West Nile, thought to be Amin's stronghold. Tanzanians later learned, however, that Amin had fled to this area with the fall of Kampala, but he had quickly left the country. Little resistance was actually met. The entire country was occupied by June 3, and the war was over. While Nyerere had outwardly appeared confident and self- assured throughout the war, he apparently harbored his own self­ doubts. He is a devout Catholic and a highly religious man, and at times his faith undoubtedly caused him to question his decisions. Two days after the war ended, he wrote a letter to a Baptist minister in the United States who had offered Nyerere his prayers in an earlier letter. Nyerere wrote:

I believe that in the march of history each one of us makes his or her own contribution towards whatever goal the Almighty has willed for humanity. I think nevertheless that it would be dangerous for any politician to regard himself even remotely as an instrument of divine will. We may feel right; we may even be morally and intellectually convinced that we are right: but we can never be sure. Personally I am not sure that this lack of certainty is necessarily wrong or even regrettable.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 38

After all, war always means the decreeing by humans that other human beings should kill and be killed. Should we feel certain? (Nyerere 1979) Indeed, the war had taken its toll. In all, forty to forty-five thousand Tanzanian troops had participated in the invasion, along with several hundred Ugandan exiles. Many more UN LA soldiers were recruited from within Uganda, particularly around Mbarara and Masaka. Remarkably, only 96 Tanzanian soldiers lost their lives in battle,11 though another 277 died in accidents, and approximately 1,500 civilians were slaughtered when Amin's troops overran the Kagera salient (Avirgan and Honey 1982, 196). In addition, 150 UNLA soldiers were killed. Again, the majority died when their boat sank in Lake Victoria on the way to the front. Though it is difficult to know with any precision, approximately six hundred Libyan and one thousand Ugandan soldiers were also killed (Avirgan and Honey 1982, 196). The Central Bank of Tanzania estimated the direct costs of the five-month operation at $70 million. If the economic dislocation caused by the war during those months was added, a modest estimate of the cost reached $150 million (QER 1979:3, 4). This was to be only the beginning; Tanzanians' involvement in Uganda was far from over. The two-year occupation which followed the war would prove far more costly—socially and politically as well as

11 This surprisingly low figure is a measure of the lack of resistance Amin's forces offered and suggests the possibility that means short of a full-scale invasion might have secured Tanzania from future invasion.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 39

economically—and embroil Tanzania ever more deeply in Uganda's chaos.

Post-War Occupation The new Ugandan government relied almost exclusively upon the Tanzanian troops and Tanzanian-led UNLA forces to secure the country and maintain order. Ominously, as the war ended, Uganda was already facing yet another political crisis. Defying the UNLF constitution which gave real political power to the National Consultative Council (NCC), Lule moved to strengthen the presidency, circumventing the NCC. The Council fought back and passed a no- confidence vote against him on June 20, a mere two months after he was sworn in. Geoffrey Binaisa12 was elected to replace him. Lule called upon his Baganda tribe to resist, and Tanzanian troops were required to control the conflict. Lule was put under house arrest, then moved to Tanzania where he was held for two weeks. He was ultimately released and flown to Britain after assurances that he would not engage in political activities (Avirgan and Honey 1982, 197-206).

12 Binaisa had fled Amin's Uganda and practiced law in New York. He was active in encouraging U. S. sanctions against Amin. Ironically, he was denied admittance to the Moshi Unity conference by the credentials committee (Avirgan and Honey 1982, 201). Perhaps he was chosen to lead the party in its time of crisis precisely because he had not participated in the previous political wrangling.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 40

In late July, twenty-five thousand Tanzanian troops, largely members of the militia, were demobilized and returned to Tanzania. Order, particularly around Kampala, was deteriorating as different political factions vied for power. The chaos spread to the Tanzanian and UNLA soldiers who had, since the massacre at Mutukula, impressed most observers with their discipline. Ugandan civilians began to associate the Tanzanians soldiers, not with liberation, but with the looting and raping which all-too-often accompanies occupation. Goods not available in Tanzania, particularly luxury items such as electronics, began to appear in towns in northern Tanzania. Returning soldiers often brought all they could carry or load into stolen cars. Some even returned with looted televisions, despite the fact that Tanzania had no television broadcasts.13 Indeed, more soldiers were killed during the occupation than during the war, usually at the hands of irate victims of the soldiers' abuse or their families. At one point, Nyerere complained to the Ugandan government about the killing of his troops, in no way acknowledging the declining discipline (Nabudere 1993). This lawlessness foreshadowed the chaos which was to characterize post-Amin Uganda and embroil Tanzania in a futile attempt to bring order to a nation it believed it was liberating from Amin's tyranny.

13 Tanzania still had no television broadcast in 1993, though there were plans to develop it.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 41

The root of the problem was that from Amin's time onward, the instruments of government, particularly the courts and police, were almost nonfunctional. The post-Amin governments all failed to get the bureaucracy working again. Therefore almost all constraints that a state normally exerts on human behavior were absent. These killings flourished in an environment void of social restraints and norms. (Avirgan and Honey 1982, 208) When the unrest did not improve toward the end of 1979, Nyerere threatened to remove his remaining twenty thousand troops if Binaisa did not restore order. Yet Binaisa's hold on power, as well as his relationship with Nyerere, was deteriorating. In early 1980, Nyerere withdrew another ten thousand troops and pulled back all Tanzanian troops from Kampala, turning responsibility for the capital over to the UNLA. By May 1980, the UNLF's Military Commission, led by Paulo Mwanga, staged a coup against Binaisa.14 Nyerere agreed not to challenge the coup if the Military Commission and the NCC, in accordance with the Moshi Unity Conference, would agree to hold parliamentary elections under Commonwealth observers. Further, the Military Commission must turn over Binaisa to Tanzanian troops who would hold him under house arrest until his fate could be determined by the newly elected government (Avirgan and Honey 1982, 210-220).

14 Some believed that this coup had the tacit approval of Julius Nyerere. He was distressed by Binaisa's independent action, particularly his move to improve relations with Kenya, whom Tanzanian regarded as having been pro-Amin during the war (Nabudere 1993).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 42

Milton Obote returned from Tanzania for the campaign, which was marked by continued violence. Nyerere pressed his demand that Commonwealth observers be present in order to dispel continued suspicions that Tanzania was imposing an Obote government. When the elections were finally held in December 1980, Obote's Uganda People's Congress (UPC) won a majority, and, though the election was chaotic and much contested, Obote was installed as president (Avirgan and Honey 1982, 220-230). With an elected government now installed in Uganda, Tanzania prepared to extract itself, though the Obote regime would prove no more of a salvation than the three that preceded it.15 In June 1981, the remaining ten thousand Tanzanian troops returned home, more than two years after the war ended. Table 1 recaps the major events of Tanzania's involvement. Despite its victory, these 2-1/2 years proved very costly for Tanzania. The counter-invasion and occupation had depleted the national treasury and directly cost the country over $500 million,16 or over ten percent of the country’s 1979 GNP. Tanzania could now turn its full attention to the problems the war had created at home.

15 Obote, seen by many to be as repressive as Amin, would himself be overthrown in a coup in July 1985. Chaos would reign in Uganda for a decade and a half, from the Amin coup in January 1971 until Yoweri Museveni took office in January 1986, six months after Obote was overthrown for the second time.

16 This figure was an estimate by the Tanzania government in an appeal for additional Western aid after the war. It included the costs of the first year of war and occupation (PER 1979, 4: 7).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 43

Table 1: Timeline of the Conflict

—Jan 1971 Idi Amin overthrows Milton Obote. Obote moves first to Kenya, then to Tanzania where he establishes a center of resistance to Amin's rule. Tanzanians assist by establishing training camps for Ugandan rebels.

—Sep 197 2 Ugandan exiles, with Tanzanian assistance, invade southern Uganda from Tanzanian bases but are quickly expelled. Amin and Nyerere sign Mogadishu Agreement whereby they agree to end hostility and aggression against one another.

- 1 9 7 2 - 7 8 Relations between Uganda and Tanzania are relatively quiet, but Nyerere continues political opposition within the OAU and Amin continues threats.

—Oct 1978 Amin's troops occupy Kagera salient, killing 1500 civilians, kidnapping hundreds of others, and looting..

— Nov 19 7 8 Nyerere declares war against Uganda, and the country mobilizes. Ugandan troops withdraw across the border prior to arrival of Tanzanian army.

—Jan 19 7 9 Decision Point #1 - Nyerere decides to invade Uganda to take high ground at Mutukula, overlooking the Kagera salient, in order to secure the border area.

— F e b ru a ry Nyerere issues conditions for war's end, including OAU condemnation of Amin's aggression. OAU balks. Decision Point # 2 - Failure of Amin to renounce claim on Tanzanian territory and of OAU to condemn Amin's aggression convinces Nyerere to continue. Tanzanian troops take Mbarara and Masaka barracks.

— M arch Decision Point # 3 - Nyerere decides to push on to Kampala, though intends for Ugandan exile forces to take the city. Libyan forces arrive to assist Amin, and Nyerere determines that Libyan involvement justifies Tanzanian involvement. Tanzanian government acknowledges its involvement, and Nyerere gives go ahead for Tanzanians to join in the assault on Kampala. He orders troops to halt for Moshi Unity Conference to establish Ugandan govemment-in- waiting.

—April Entebbe and Kampala fall.

—June Uganda is secured by Tanzanian troops and UNLA

—July 25,000 Tanzanian troops, largely People's Militia, return to Tanzania and civilian life.

—Early 1980 Another 10,000 troops are withdrawn.

— June 1981 The remaining 10,000 Tanzanian troops return to Tanzania.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 44

The war had devastated the economy, seriously depleted the road transport system, and brought a new level of lawlessness to Tanzania with the returning troops.

The flood of weapons from Uganda had spilled over into Tanzania, and violent crime was now a common occurrence in Dar es Salaam and other Tanzanian towns. Tanzanian police found that more than 60 percent of the people they arrested were militia members demobilized after the war. (Avirgan and Honey 1982, 236) A culture of violence spread through Tanzania which would exact social costs extending well beyond the limited period of war.

Pyrrhic Victory In June 1979, the Tanzanian government appeared to have done what it hoped to do—rid East Africa of Idi Amin. This victory had been achieved in only five months with surprisingly few losses. Yet, the true cost was to prove painfully high. While Tanzanian soldiers and their Ugandan allies had set out to liberate Uganda from the horror of Amin, Tanzanians found themselves deeply embroiled for another two years in Uganda's continued chaos and quite impotent to provide order. Ultimately, Tanzanians merely withdrew while the horror continued behind a new face.17 Tanzanians may have won the war, but they lost the peace.

17 It has been estimated that as many as 300,000 Ugandans died from atrocities committed during Obote's second reign. See Newell (1985) and Wilkinson (1985a & 1985b). It is instructive to note that this is more deaths per year of rule than occurred under Amin. It is interesting to ask why, then, Obote wasn't vilified in the same way as was Amin? When asked, some interviewees felt that

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 45

One might reasonably question this war's effectiveness at achieving Tanzania's goals. If the goal was to overthrow Amin in order to open the way for a better leader, Tanzanians' hopes must have been dashed by the brutality of Obote's second rule. If the goal was to protect Tanzania from further invasion by Amin's troops, some action short of a full-scale invasion may well have achieved that security without the high costs—perhaps heavily fortifying the border the highlands at Mutukula. In evaluating war as policy, it is important to assess the unintended as well as the intended consequences. In Chapter 4, I will return to the issue of anticipated consequences of this war before analyzing the actual outcomes. While the war was relatively short, those five months cost over one million dollars a day. The immediate economic costs equaled all 1979 development assistance combined, an enormous sum considering the fact that Tanzania was "the most aided country in Africa." (Ungar 1979, 16-17) Economic costs were certainly anticipated, though the extent of the dislocation was not. Other outcomes such as crime, AIDS,

Obote couldn't be judged by the same measure as Amin because his second rule occurred in the midst of what amounted to a civil war. Others suggested that this information was simply not known by the outside world. There was no voice to stand up and be counted as there had been with Amin, given that Nyerere was unlikely to criticize a leader he had been instrumental in bringing back to power. Moreover, Amin's murders were often so bizarre that they captured the attention of the international press. Still others, particularly Tanzanian officials, simply did not believe these figures.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 46

destabilization of the government, and near bankruptcy were certainly unintended and probably unanticipated. Pyrrhus, King of Epirus, said of his victory over the Romans at Ausculum in 279 BC, "Another such victory, and we are lost." (Maoz 1990, 250) Tanzanians' victory in Uganda can be considered such a Pyrrhic victory. Less was achieved than was hoped for, and victory was gained at far greater cost than was ever imagined. Why was it so? The following chapters attempt to answer that question and to find in the Tanzanian experience insights into the nature of victory through war.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2

ASSESSING WAR'S LEGACY: THEORY AND METHOD

War is public policy with enormous effects. Yet it is the least evaluated of public policies, when it should perhaps be the most carefully assessed. Scholars and politicians alike have failed to assess war's effectiveness and war's effect. Politicians may perhaps have reasons for wishing to avoid such evaluations, especially public ones. Scholars have no excuse.

Stein and Russett, "Evaluating War"

A thorough assessment of the consequences of Tanzania's war could offer important insight into the full costs of military victory and war’s efficacy as public policy, particularly in Africa. The toll of Africa's increasingly frequent wars has mounted since independence swept the continent in the 1960's. In the decade of the 1980's, for example, there were nine wars,1 costing the lives of two to three million Africans. Between nine and thirteen million were

1 War is generally distinguished from other forms of organized violence by the magnitude of participants or battle deaths. The most widely used categorization system is that by Singer and Small (1972) where an interstate war is one in which there are a total of 1,000 battle deaths and, for a civil war, the active involvement of 1,000 military troops or a minimum of 100 battle deaths.

47

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 48

displaced as they sought shelter from the fighting. The overwhelming majority of these wars' victims were civilians, particularly children (Copson 1991, 23-24). Africans who were not directly affected by war often suffered indirect consequences. National economies were devastated as human and material resources desperately needed for development were diverted to destructive purposes, and military expenditures claimed ever larger portions of governments' budgets. Natural resources were destroyed. Perhaps most costly, a culture of violence spread, leaving in its wake a legacy of ever higher levels of violence and despair. While Africa's conflicts have been increasingly deadly, they have been decreasingly responsive to peaceful resolution. Of all Africa's post-colonial wars, only fifteen percent were resolved through negotiation. The other eighty-five percent either ended in military defeat, devastating both sides, or continue to ravage the continent today (Rothchild 1992). Indeed, around the globe, while international and regional organizations such as the United Nations and the Organization of African Unity are increasingly becoming involved in conflict management, the success rate of these interventions is declining (Starr and Most 1985). This dismal record of conflict management once the conflict has turned violent suggests that efforts are better directed at discouraging the use of force when conflict does arise. Realistically anticipating the full consequences of the use of deadly force can reasonably be considered one approach to staying the hand of force. At the point of decision when war is chosen to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 49

resolve conflict, do decision makers realistically anticipate the costs and benefits? I believe they often do not. Dialogue on war as a strategy is seldom extended nor open; the debates that took place in the U. S. Congress and the United Nations concerning Kuwait and Somalia are rarities. Moreover, there is far too little research on the long-term impact of violence to inform such dialogue. Indeed, a former editor of the Journal of Conflict Resolution was so struck by the lack of serious scholarship on war's consequences th at she was led to lament "for most JCR contributors, once a war happens it ceases to be interesting." (Converse 1968, 476-477) There are good reasons to expand our understanding of war's consequences. Certainly, if Clausewitz is correct that war is politics by other means, war should be evaluated as rigorously as any other political means. Such evaluations of war that do occur too often lim it themselves to assessing its effectiveness at achieving certain goals, quite avoiding the issue of unintended consequences. Yet any thorough assessment of war's efficacy must consider the costs and benefits of unintended as well as intended consequences. Moreover, war's effects can themselves become determinants of subsequent violent conflicts. Thus, a rigorous assessment of prior wars' consequences should be a part of any thorough analysis of a subsequent war's causal factors (Stein and Russett 1980, 400). Tanzania provides a unique opportunity to study war's true legacy. Its war with Uganda was Tanzania's only significant conflict and was of limited duration. Political leadership and policies remained relatively constant, though there certainly were changes

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50

that occurred during the period under study. Precisely because of the decisive beginning and ending of hostilities, as well as the comparative stability of other factors, those economic, political, and social changes that did occur within Tanzania can be related more readily to the introduction of violence than in most cases of war. Therefore, an analysis of this war's legacy will expand understanding of the full consequences of war, providing important insight into the relationship between violence and development of African nations. Moreover, Tanzania's war can be considered a best-case scenario, if such a term can be applied to an event as destructive as war. Tanzania was victorious, and its victory was quick and decisive with relatively few casualties. Studying the vanquished adds little to our understanding; no one is surprised to realize the devastation of losing a war. What is more interesting and may give pause is to understand the full consequences for the victor, particularly one whose victory came so readily. The full consequences of this war were significant for Tanzania's development, and in assessing those consequences, I seek to provide a cautionary tale for decision makers who are considering the use of force to resolve conflict but who cannot reasonably expect a more beneficial outcome. This chapter outlines the theoretical and methodological approaches used in this case study of the war's consequences for Tanzania's development. In the first section, the concept of a "Pyrrhic victory" is elaborated and compared with similar concepts

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51

The second section argues the case that Pyrrhic victories occur in war because war is a social trap. The third section offers a typology of social traps, with examples from Tanzania. The fourth and final section discusses the applicability and generalizability of Tanzania's war to the exploration of war as a social trap. It also elaborates the sources used in the research.

Pyrrhic Victories, War Traps, Paradoxes, and Social Traps In Chapter 1, I characterized Tanzania's war in Uganda as a Pyrrhic victory and asked how Tanzanians could, despite their relatively quick and easy victory, gain so much less than they had hoped at a cost far greater than they had imagined. In this section, I explore the concept of a Pyrrhic victory and compare it to similar notions in the work of other scholars before proceeding to answer that question in the section which follows. A Pyrrhic victory is one that is achieved at staggering costs (Morris 1981, 1065). Such a victory occurs when the costs of success far exceed what was anticipated, and perhaps even exceed what the victor would have been willing to pay if an accurate assessment of the war's consequences had been possible beforehand. Certainly, the costs devastate the victor. It may also be that the benefits of a Pyrrhic victory prove to be less than expected or that the costs of the struggle are so great as to outweigh all possible benefits. Pyrrhic victories in war leave both the victor and the vanquished devastated by the struggle.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 52

One may envision a Pyrrhic victory in war as a trap, evoking the image of a nation or group being ensnared by its own victory and destined to suffer disastrous consequences. In his book entitled The War Trap (1981), Bruce Bueno de Mesquita presents a general theory of war's initiation and escalation, based upon decision makers' perceptions of maximum utility.2 Yet, Bueno de Mesquita's notion of a "war trap" is incompatible with the concept of a Pyrrhic victory, because he envisions that it is only the "victim, and not the initiator, that is trapped by conflict." (Bueno de Mesquita 1981, 182) He believes that initiators3 entrap their victims in war, while I argue that initiators, even if they are victorious, may be similarly entrapped by the unintended consequences of their choices. While not as catchy a title, Zeev Maoz's Paradoxes of War: On the Art of National Self-Entrapment (1990) comes closer to the notion of war's entrapment for both the victor and the vanquished. Maoz defines a paradox as "a causally induced contradiction between

2 A decision maker's utility is the value she/he places on a particular outcome.

3 Bueno de Mesquita distinguishes between "initiators" and "aggressors". Initiators may or may not have been the initial aggressors but are those that have "the last reasonable chance" to avert a war (1981, 99). By this definition, Tanzania must be considered the initiator of the war even though Uganda was the initial aggressor, as Tanzania had an opportunity to avert a war when Amin's troops withdrew and Amin sought diplomatic intervention. The fact that Tanzania found the conditions of cessation unacceptable (and thus chose to counter invade) does not negate the opportunity it had to avert the war.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 53 expectations and the consequences of behavior resulting from them." (1990, 13) Yet again, this concept of war's paradoxes misses the mark. Maoz argues that wars can occasionally be beset with contradictions, whereas I seek to convey that the process o f war is so wrought with contradictions that the outcome is likely to be paradoxical. The paradox is that the victor may lose by winning. Maoz further argues that war is generally treated with "linear logic", though some aspects of the process defy this rational, simple, and straightforward logic and are, therefore, paradoxical. I argue that it is the process of war, not only some aspect of it, that is paradoxical and non-linear,4 thus linear logic is flawed as a means o f choosing war over non-violent means of resolving conflict. Linear logic, or rational thinking, can lead one to chose war over other means, even though victory may be self-defeating. Thus, war can entrap both the initiator and the target, no matter which side wins. The concept that is most compatible with the notion of a Pyrrhic victory is the social trap,5 which comes from the fields of sociology and mental health. Social trap theory grew out of a

4 Here, I use the term "non-linear process" to describe a situation in which reasonable expectations produce behavior that leads to outcomes that could not logically be anticipated.

5 Social trap theory was first developed by four researchers with the Mental Health Research Institute of the University of Michigan—John Cross, Mel Guyer, Gardner Quarton, and John Platt. It was then elaborated in Platt (1973) and expanded in Cross and Guyer (1980).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 54

search for a new perspective on the origins of a wide variety of social problems, an alternative to the views that problems such as drug abuse, air pollution, and international conflict are either due to ignorance or to immorality (Cross and Guyer 1980, 3). Social trap theorists suggest that such problems result from social traps.

A social trap is a situation characterized by multiple but conflicting rewards. Just as an ordinary trap entices its prey with the offer of an attractive gain and then punishes it by capture, so the social situations which we include under the rubric of 'social traps' draw their victims into certain patterns of behavior with promises of immediate rewards and then confront them with consequences that the victims would rather avoid (Cross and Guyer 1980, 4). Another writer offers this definition.

[Social traps] explain the all too common social situations in which people behave contrary to their own self interest while making what appear to them to be perfectly rational decisions. . . . Social traps work on the differences between the perceived or apparent costs and benefits to the individual participants and the real costs and benefits. (Costanza 1984, 79-80) The converse situation, the countertrap, is one in which "an aversive bait causes the victim to avoid a course of action which, if followed, would have brought about a preferred consequence." (Cross and Guyer 1980, 17-18) More generally, social traps are situations in which the normal instrumental conditioning which causes learning and establishes useful behavior patterns becomes distorted. In these situations, the rewards which normally encourage behavior that is beneficial to our well-being or the punishments which normally discourage destructive behavior—and thereby encourage learning of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 55

behavior that will sustain us—are absent. Thus, the experiences which we count on to educate us for survival mislead us, teaching us behavior patterns which ultimately result in outcomes that we do not desire. Cigarette smoking or alcohol and drug use provide ready examples of social traps. Potential users, presented with immediate benefits (a high, feelings of well-being) which are balanced against immediate cost (price), might rationally find the benefits outweighing the costs. Over time, however, the benefits decline (less response per "hit") and the costs increase (addiction encourages higher usage, so price increases). Yet the users are now entrapped in addictive behavior, the cost of which may well be life itself (lung or liver disease, accidents). Users' original choices, which appeared rational to them, have led them to learn behavior that is contrary to their own self-interest.

War as a Social Trap In answer to the question posed in the previous section (How did Tanzania's war end in a Pyrrhic victory?), I assert that it is because war is such a social trap. That this can be so is best illustrated by the "dollar auction game."6 In this game, a dollar is auctioned to players, the catch being that both the highest and

6 This game was conceived by Martin Shubik and published in an article aptly subtitled "A Paradox in Noncooperative Behavior and Escalation" (1971). It is then used by Platt (1973) in his original elaboration of social trap theory.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 56

second highest bidders must pay the auctioneer their bids, but only the highest bidder receives the payoff of the dollar, in most instances, the bidding goes well above one dollar, a clearly irrational result despite the fact that, at each decision point, players consider themselves to be making rational choices. The bidding generally gets reduced to two players as others drop out. These two bidders soon reach a point where they are no longer seeking to maximize their gains but merely hoping to minimize their losses. That point is most pronounced when the bidding reaches one dollar. If the bidding is in $.05 increments and your competitor has bid $1.00, you have bid $.95. If you raise your bid to $1.05 hoping to win, you will lose $.05. If you don't raise your bid, you will lose $.95. So you raise, hoping to minimize your loss (Platt 1973). The ultimate winner achieves a Pyrrhic victory (by paying over a dollar for the dollar prize). This is because the game's structure is a trap, just as cigarette smoking or alcohol and drug use are traps. The trap or positive bait is the opportunity for gain. The countertrap or negative bait is the opportunity for loss. The only way to avoid the trap is to refuse to enter the game (Costanza 1984, 80-82), just as the only way to avoid addiction is to avoid using addictive substances. War can entrap like the dollar auction game because, while only the winning side receives the payoff (imposition of its will on the opponent), both the winner and loser must pay the price of participation (the cost of war). In war (as in the game), the cost of participation can be well above the value of the payoff, leaving both

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 57

the winner and loser worse off than had they not participated. Like the bidders in the game, warring sides reach a point at which the goal shifts from maximizing gains to minimizing losses. The perceived opportunity for gain is the trap that brings them into the war, but the investment they have already made (sunk costs) and the perceived negative consequences of losing are the countertraps which allow the war's escalation to a point that can devastate both sides. Given this insight, let me offer another definition of a social trap. A social trap is a set of circumstances in the decision environment that distorts the decision maker's utility calculations, creating a false picture of the probable outcomes of alternate courses of action, thereby leading the decision maker to choose a course of action contrary to her/his self interest. In this distorting environment, the real benefits of the chosen course of action are overvalued and the costs discounted. The decision maker is thus entrapped by the choice, destined to suffer consequences that were not part of the original decision calculus. The notion of war as a social trap is based on three premises. The first is that war is a choice to utilize violent force against an adversary, one option amongst several to resolve conflict. The possibility that alternate options may not be perceived is part of the trap and does not negate the presence of choice. A sub-premise follows from this; actors' decisions matter. The consequences that result from a choice to initiate war are substantively different than those that result from other choices. The second premise is that the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 58

choice to initiate war occurs through what the actors perceive to be a rational decision-making process in which decision makers conclude that violent force offers the maximum utility. The third and final premise is that decision makers' utility calculations are distorted by a misperception of war's real consequences because they are operating within a social trap, causing them to overestimate benefits and underestimate costs. I will elaborate on each of these separately. As to the premise of choice, there are those who assume that international political action, particularly war, is brought about by ''broad social forces and movements that are beyond the control of any particular rulers." (Luterbacher 1984, 166) While not denying the influence of such forces, I hold the belief that wars result from leaders' choices. Converse, a former editor of the Journal of Conflict Resolution, reviewed the journal's publications from 1957- 1968 and lamented the "vanishing choice-point" or failure by scholars contributing to the journal to identify choice-points in the war decision-making process. She makes the very common sensical argument that war is a choice because it requires one to take action.

A t some point war is a deliberate choice. . . . the making of international war, aggressive or defensive, requires the conscious setting in motion of innumerable preplanned, coordinated acts. International war is therefore caused, in the most proximate sense, by national decision-making. (Converse 1968, 477)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 59

Thus, mobilizing resources for battle is a choice, despite the presence of "broad social forces" that may influence or constrain that choice. As to the second premise of rationality, there are those who challenge rationality as unrealistic. Against this criticism, I question the alternative. "If we assume decision makers are irrational, then we are compelled either to study each war as a unique event or to study war from some perspective other than decision making." (Bueno de Mesquita 1981, 32) The same reasoning has been attributed to Hans Morgenthau in Politics Among Nations.

Morgenthau explicitly acknowledged that the assumption of rationality was not descriptively accurate . . . but he believed that it could be used as a baseline, which could be 'tested against the actual facts', making a theory of international politics possible. That is, even though such an assumption is not always descriptively correct, it serves a valuable theoretical function. (Keohane 1986, 11-12) Others would challenge the degree of rationality in choice. Simon (1976, 1982), speaking for many behaviorists, associated rational choice with perfect rationality. He saw this perfect rationality as an impossibility in the real world, so he offered the concept of "bounded rationality," or rationality constrained by such things as imperfect information and uncertainty. This bounded rationality forces decision makers to "satisfice" by choosing a satisfactory option rather than an optimal one.7

7 Van Witteloostuijn sees this debate as reflecting a contradiction only "if textbook versions of behaviorism and choice theory are compared." In actual research, however, he found that both approaches yield comparable results. He concludes, therefore,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 60

I see war as a choice, yet I accept the notion of constraints on rational choice, or bounded rationality. Rational actors need not have complete information, perfect foresight, or "polyocular vision."8 They are assumed only to make choices as if to maximize their welfare and, by assumption, the welfare of those to whom they are accountable, based on the information available to them. Decisions to initiate or escalate war arise from decision maker's calculations of "expected utility," or the value a decision maker places on a particular outcome multiplied by the perceived probability of its occurrence (Bueno de Mesquita 1981, 34). It should be said at this point that these two assumptions, which could be combined under the term rational choice, do not mean that wars result from a single choice. On the contrary, they result from a series of choices, just as there were a series of bids in the dollar auction game that escalated the competition. In conflict, as in the game, each actor's choice is dependent upon the previous choice of the adversary. A potential war initiator in a conflict faces two choices: (1) initiate war, and (2) do not initiate war. If war is initiated, a defender also faces two choices: (1) resist the

that "maximising and satisficing decision rules are equivalent rather than opposite principles." (1988, 289-290) Nicholson (1992, 55) also dismisses the debate by suggesting that, in a complex decision-making environment, rationality requires satisficing behavior.

8 This term, coined by Magoroh Maruyama (1976), refers to the added perception one might have with more than two eyes. It is a metaphor for a multi-dimensional perspective.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 61

aggression, and (2) yield to the initiator's demands. If war is initiated, the warring sides face similar choices concerning escalation, as did Nyerere when he moved his troops from Mutukula to Mbarara and Masaka, then on to Kampala. Depending on the choices of both sides, the conflict has four possible outcomes: (1) war in which the initiator wins, (2) war in which the defender wins, (3) defender yields to initiator's demands without war, or (4) the status quo—neither side initiates war or yields (Maoz 1985).9 The choices facing each side can be modeled as a decision tree, as shown in Figure 2. In addition to choices, the decision tree incorporates a means of modeling the expected costs and benefits of potential outcomes. Outcomes are modeled in terms of expected utilities—the value (positive value of victory but negative value of defeat) of the expected outcome times the anticipated probability of its occurrence—and the expected costs. Expected costs differentiate between the cost of prosecuting a war and the political cost of being an initiator even though war does not result, such as international condemnation or economic boycotts.

9 It is important to note that, since the initiator is not necessarily the aggressor, it is possible that there was an act of aggression by one side that led to the war's initiation by the other side. This was the case in the Ugandan/Tanzanian war, as Amin occupied the Kagera salient, then retreated under international pressure. Tanzania then initiated war. Thus, status quo in this case means the status prior to war's initiation but after an act of aggression. In Maoz's model, he adds an extended game which includes the possibility of simultaneous initiation, though this scenario is not included here.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 62

The third premise concerns the trap-like nature of war. Figure 2 models the conflict situation before the outcome of the series of decisions is determined. Yet, to assess the efficacy of war as a means of achieving policy goals (and to support the notion of war as a social trap), it is necessary to evaluate the actual utility of war and compare the expected outcomes that informed a decision for war with the actual consequences. Significant disparity between the anticipated and actual outcomes would give rise to the search for an explanation for this disparity. We wish to know, based on empirical evidence, how accurate decision makers predictions are about war's outcomes. Both Bueno de Mesquita (1981) and Maoz (1982) did extensive quantitative studies of wars in the 19th and 20th centuries and found that initiators were victorious far more often than not.10 Since it can reasonably be assumed that war is initiated only when there is an expectation of victory, their work suggests that decision makers who initiate war indeed do predict the probability of victory fairly accurately. However, the projections of gains and losses associated with victory and defeat, as well as the costs of prosecuting a war, are quite another matter. The costs of prosecuting a war have proven far greater than anticipated and the

10 Of the fifty-eight interstate wars fought during the period from the Congress of Vienna to 1974, the initiators won a statistically significant forty-two times (Bueno de Mesquita 1981, 22). In another study of interstate wars, Maoz found that being the initiator was a far more accurate predictor of victory than was relative power (1982, 4).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. gains often less, thus the Pyrrhic victory (Maoz 1990). So, a general conclusion to be drawn from these studies is that the nature of war's outcome (victory or defeat) is fairly accurately predicted, but the consequences of those outcomes (costs and benefits) are not. How can this disparity between anticipated and actual consequences be explained?

Figure 2: The Initiator's Decision Problem

Initiator's Defender's Potential Expected Utilities Choices Choices Outcomes ______and Costs

In itia to r Wins EU . * "EC Defender r ^ victory war Resists j \ Defender i- P i\ wins In itia l ► E U «defeat war

Defender Yields Victory’ political Status Do Not in itia te ^ status quo

Legend: Pr = Anticipated probability of defender's resistance; Pi = Anticipated probability of initiator winning war; EU = Expected utility; EC = Expected costs. Source: Adapted from Maoz (1985, 79).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 64

Luard (1987) studied wars from 1400 to the present and was led to question why leaders resorted to wars when they were so frequently ineffective at achieving the victor's intended consequences. He concluded that there were two main factors that distort decision makers' rational calculations of war's consequences—the costs and benefits of victory and defeat. The first factor is the uncertainty that surrounds the decision-making process. It is not possible to predict with any degree of certainty the choices the adversary will select nor the consequences of those choices. Moreover, information that does exist may simply not be available to the decision maker. It does not necessarily follow that unintended consequences are undesirable, however the study by Maoz (1990) suggests that they usually are. The second factor Luard found is that costs and benefits of war are not borne by the same segments of society.11 A decision maker is likely to be closer to the segment of society that will receive a disproportionate share of the benefits while bearing a minimal share of the costs. This disparity between social segments can reasonably explain the behavior of leaders whose actions appear irrational when acting out revenge, for example. They will experience the benefits, but they either will not perceive or they will tend to discount the costs of their war decisions because they will not bear them.

11 Richardson and Samarasinghe (1992) also make this point.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 65

In addition to these limitations of uncertainty and perspective, there are constraints on the decision process itself. A diplomatic and military crisis which might lead to war can be extremely stressful for decision makers, impairing their decision performance. The crisis can come without warning, and its severity requires quick decisions. Gathering and absorbing information to guide the decisions can be very demanding of decision makers and bring about fatigue. The stress induced by such a crisis brings about a "shortened and narrowed perspective" (George 1992, 466-7). In other words, decision makers cannot accurately anticipate the full range and intensity of consequences nor the length of time over which the consequences will be felt. The suboptimal decision process that results from these constraints and distortions causes leaders to pay less attention to the longer-range consequences as well as the unintended (and usually negative) consequences. The net result is that decision makers overestimate benefits and underestimate costs of war. This shortened and narrowed perspective can be represented in Figure 3.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 66

Figure 3: Overall Impact of War

Anticipated by Decision Makers> Unanticipated by Decision Makers> Net - B enefits

Net Costs

0 Time/Intensity/Breadth of impact—>

A Typology of Social Traps Having elaborated the premises on which the notion of war as a social trap are based, let us now explore how social trap theory illuminates the disparity researchers have found between anticipated and actual consequences of war. As the theory has developed, several types of traps have been identified. Here I present a taxonomy from Cross and Guyer (1980), with supporting material from Platt (1973) and Costanza (1984). I will give examples here from the Tanzanian-Ugandan war, but others present themselves in the analysis of the war's consequences in Chapters 4 through 7. Also, new traps emerge out of this case study. Thus, this taxonomy provides a vocabulary with which to proceed but will be amended in the concluding chapter. In this taxonomy, traps are differentiated primarily based upon who is entrapped by the decision.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 67

A. Self Traps12 entrap the individual decision maker. A decision maker's consequences are dependent upon her/his choices alone and do not depend on the presence of other decision makers or other potential victims. An individual can be entrapped by alcohol abuse without her/his actions necessarily affecting others. If others are affected, they will not be more affected than the decision maker. While Self Traps entrap individuals, they become of concern to society when: (1) several decision makers are entrapped and their interactions become mutually reinforcing (the phenomenon of group think), (2) non-decision makers are significantly affected (the health hazard of passive smoke), or (3) society can have an influence on a decision maker's choices that affect others (a populace that is war weary or has war fever). This category is further differentiated by the mechanism by which the individual entraps her/himself. 1. Time-Delav Traps occur because there is a time lag between an action and the experience of consequences from that action. This lack of timely feedback impairs learning from experience. Moreover, it impairs a decision maker's ability to perceive long-range consequences. In the case of Tanzania, Julius Nyerere decided to move his troops into Kampala and install a new government before he could know

12 Social trap theorists have referred to this category as "Individual Traps", but to me the relevant point is not that an individual is entrapped but that the individual entraps her/himself.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 68

anything about the stability of the leaders chosen at the Moshi Unity Conference. 2. Ignorance Traps occur through ignorance of the true consequences of one's actions, either through lack of information or unwillingness to utilize the information available. Such willing ignorance can be caused by an unrealistically high perceived value of the possible gain (the Midas effect13). For Nyerere, Obote's appeal as an alternative to the hated Amin may have distorted Nyerere's assessment of the consequences of another Obote government in Uganda. 3. Diminishina-Returns Traps14 occur when benefits diminish without the change being perceived by the actor. Use of heroin is a classic example, as the original dosage becomes less and less able to induce a high. Over time the goal of a high is replaced by the goal of simply removing pain, just as in war when the goal shifts over time from maximizing gains to minimizing losses. For Tanzanian leaders, the security they anticipated would come by removing Amin

13 Midas was the king to whom the god Dionysus gave the power of turning all he touched into gold (Morris 1981, 829). The Midas effect, then, is the consequence of projecting such superhuman powers upon another.

14 Social traps theorists have referred to this trap as the "sliding reinforcer trap". I find this term to be specific to psychology and prefer the term "diminishing returns traps" as conveying meaning to a wider audience.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 69

dissipated as Tanzanian troops brought their newly acquired weapons and undisciplined behavior back to Tanzania. The anticipated benefit did not live up to its promise, and an international threat became a domestic threat. B. Multiple-Person Traps entrap two or more people who are interdependent, such that the consequences each will experience (the payoff matrix) are, in part, dependent upon the choices of another (or others). Unlike self traps, multiple- person traps would not occur in the absence of an interdependent interaction; it is the interaction that is entrapping. "Prisoner's dilemma" (Rapoport and Chammah 1965) is the classic example. Arms races are a form of prisoner's dilemma and a multiple-person trap. Libya's support of Amin was too, as it changed the payoff matrix for Tanzania.15 The presence of Libyan soldiers in the war increased the potential costs for Tanzania by increasing Tanzania's perception of the enemy's power. It also reduced the potential costs by giving Tanzanian troops a justification for intervening on the side of Ugandan exiles. C. Collective Traps entrap a large number of people who constitute a collectivity. Like multiple-person traps, collective traps involve an interaction, but collective traps do not occur because of the interaction but because a decision maker's payoff matrix from choices differs from the payoff

15 It is also true that Tanzania influenced Libya's payoff m atrix.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 70

matrix of the larger population that will experience the consequences of those choices. It is this disproportionate experience of consequences between the decision maker and the collective that causes the trap. This category is further differentiated by whether the decision maker disproportionately enjoys the benefits or would be uniquely required to bear the costs of her/his decision, relative to the collectivity. 1. Commons Traps occur because individuals are motivated by personal gain to take actions that are harmful to society as a whole. The decision maker gains at the cost of the collective. The classic example is the "tragedy of the commons" (Hardin 1968), which creates individual goods but collective bads because an individual can benefit from using a public asset (graze cattle on public lands), but the collective asset is damaged and the collectivity harmed. In the previous section, I discussed the fact that, in war, the costs and benefits are not borne by the same segments of society. War makers are likely to benefit more and sacrifice less than the general population. 2. Missinq-Hero Traps occur because, in order for society to avoid negative consequences, a decision maker must be willing to suffer disproportionate costs. This is Schelling's mattress problem (1971) in which a horrendous traffic jam cannot be eliminated until someone is willing to stop and remove a mattress that is blocking one of two lanes on a

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 71

road, thereby sacrificing for the collective good. In this case study, Tanzanians, Ugandans, Libyans and others suffered through a war because no one of influence (leaders in Tanzania, the OAU, the Ugandan exile community, or others) was willing to suffer the political cost of pressing for a non-violent resolution to the conflict. Clearly, as the several examples from the Tanzanian war exemplify, in a situation as complex as war, any number of these types of traps may be present. Cross and Guyer (1980) offer of category of "hybrid traps" that include various combinations of the previous types of traps. Thus, war can be a hybrid social trap in which all of the various types of traps may be present. As the consequences of Tanzania's war are assessed, I will seek to elaborate on these social traps already identified and to expand this taxonomy, specific to the instance of war and based on the experience of Tanzania. One question that scholars often hesitate to ask about their research is, so what? If I can indeed make a convincing case that Tanzania's war was a Pyrrhic victory because its leaders were operating within a social trap when they chose war and that this case lends support for the assertion that war, in general, can be a social trap, what will be gained? There is laboratory research which suggests that actors who have information about the nature of decisions that they face, including warnings about social traps, avoid those traps more frequently than actors who do not (Rapoport 1988a). Moreover, when faced with a multiple person or collective

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 72

trap, actors who were given full information about the nature of the situation they faced proved more cooperative—and therefore more effective at avoiding negative consequences—than those who did not (Rapoport 1988b). Thus, awareness of the trap-like nature of war may improve the efficacy of conflict decisions and, hopefully, stay the hand of force in some conflict situations.

Methodology Clearly, any successful attempt to assess the efficacy of war and to explore its trap-like nature requires a very thorough analysis of the war's anticipated and actual consequences. I have chosen to do this through a case study of Tanzania and its war with Uganda. Within the field of International relations, case study methodology has played a role in knowledge development in various ways, ranging from exploratory pilot studies that laid the basis for further research and theory development to the corroboration or negation of well-developed theory (Maoz 1990, 21). In the first instance, it has contributed to the development of "grounded theory" or "theory that is initially derived from observations, not spawned wholly out of logic and imagination." (Eckstein 1975, 107) In the latter instance, the approach has served to test the utility of theory. The use of the case study in this research attempts to do the former. The Tanzanian case serves as a uniquely appropriate context in which to elaborate a theory of war as a social trap because it is a best-case scenario of war, yet it ended in a Pyrrhic victory.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 73

While case studies are valued for their internal validity, they are also criticized for their lack of generalizability (Maoz 1990, 22). Why, one might ask, should establishing that Tanzanian leaders were caught in a social trap which led them into a war with staggering costs offer support for a general theory of war as a social trap? The answer to that question is that generalizability from one case can be greatly increased through the selection of a "crucial" case, and, in this study, a "least-likely" case. A "least-likely" case is one that ought not to conform to the theory, if any case can be expected to do so. (Eckstein 1975). Maoz's Paradoxes of War (1990) and Tuchman's The March of Follv (1984) offer many cases of wars that do conform to the theory, cases in which apparently rational actors underestimated the costs and overestimated the benefits of war. These authors do not, however, suggest that these cases point to a generalizable outcome of war. This case study of Tanzania's war with Uganda serves as a crucial case to elaborate a theory that war is a social trap precisely because Tanzania was perhaps one of the least-likely nations to find itself entrapped by war. First, it was least-likely to declare a war, given its peaceful history and leadership. Moreover, because of its quick and decisive victory with relatively few losses, Tanzania as a warring nation was least-likely to suffer deleterious effects. Evidence that this least-likely nation indeed found itself entrapped by war, despite its victory, would provide a compelling argument for the premise that war, in general, can entrap the victor as well as the vanquished.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 74

A major part of this study is the detailed assessment of the war's consequences, i have chosen to do this by expanding upon an approach first proposed by Richardson and Samarasinghe (1992) In their analysis of the costs of conflict in Sri Lanka, they differentiate between direct (primary) and indirect (secondary) economic consequences. I also make this distinction, but I choose to further differentiate by assessing social and political as well as economic outcomes. I do this because I firmly believe that many of the myriad motives that drive actors to choose violent force are not economic and cannot necessarily be assessed in monetary terms. Some political and social consequences can, however, be given a price tag. I have done so in this study, for example, in the discussion in Chapter 7 of AIDS transmission thought to have occurred because of the war. I elaborate further in Chapter 4 on the approach to assessing consequences used in this study. In addition to generalizability, the value of a case study also depends upon the validity and availability of data. Tanzania is a useful case for an analysis of the consequences of war for these very reasons. Many of the crucial actors that were involved are still living. Moreover, the political leadership is open to this sort of analysis, a situation that is all too rare among politicians. The Tanzanian government issued formal permission for this research and provided institutional support through the University of Dar es Salaam and the Institute of Science and Technology. Much of the research was carried out in Washington, D.C. through such institutions as the Consortium of University Libraries,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 75

the Library of Congress, the World Bank, and the International Monetary Fund. Moreover, I was able to interview Tanzanians living in the United States as students, diplomats, and immigrants. Other valuable data, however, was gathered during a three-month stay in Dar es Salaam. This field research included formal interviews with over twenty informants, about half of whom were in government service at the time of the war. It also involved a review of the library files of the President's Planning Commission and the archives of the party organ, The Daily News. In addition, many informal interviews took place over transactions in the market or on the street or upon receiving an invitation to dine out or in someone's home. A foreign woman working alone in Tanzania is quite uncommon, so my presence sparked the curiosity of many people who offered assistance and opinions. Some of the formal appointments were arranged through the party headquarters, but I arranged most of them myself by building up a network of contacts over time through Tanzanians in the United States or Americans who have worked there. My initial list of names of desired interviewees primarily consisted of Julius Nyerere and those close advisors who consulted with him on the war, as well as a few civic leaders. In the process of attempting to set up these interviews, however, I was often directed to other useful informants o f whom I was previously unaware. Thus, as some individuals were unwilling or unable to be interviewed, others were added to my list. In two cases, I encountered Tanzanian researchers

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 76

in the library and archives who took an interest in my project and assisted me in locating war veterans and AIDS workers. Generally, my questions sought to discover what the informants' experiences of the war had been and how the war had impacted them and their families. In the case of government leaders, I sought to learn what consequences were anticipated at the time of the war. I also shared some of the data that I had gathered and asked both how the anticipated consequences compared with actual outcomes and how the interviewee would explain any disparity. In sum, this analysis is based upon primary data collected through formal and informal interviews and secondary data such as macroeconomic data from the IMF, archival material from Tanzania, press accounts, and previous researcher's analyses. No attempt was made at surveying or other large-scale data gathering. In addition to the usual constraints, this project was constrained by the political sensitivity and complexity of the subject and the necessity to gain Tanzanian government approval of all steps in the field research. The evidence presented here is indicative of the war's consequences, as opposed to establishing a definitive connection. That level of certainty in a subject as complex as war is outside the scope of any methodology. Indications of a connection between observed phenomena and the war take several forms: (1) proximity in time to the war and occupation, (2) comparisons with non-warring neighbors, (3) comparisons with other time periods within Tanzania, (4) assertions by interviewees of a connection, based upon their

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 77

expertise and experience, (5) common sense, and (6) consensus in the literature on consequences of war. In some cases, several of these factors support the connection. In others, evidence is less clear. In each case, I assess not only the consequence but also the strength of the evidence. The Tanzanian leadership was unusually willing to reflect on its experience of war, and the timing of the application for the field research proved extremely fortuitous. In Summer 1992, Tanzanians voted to institute a multi-party political system, ending the one- party rule that had continued since independence. Though the process is not envisioned to be fully implemented until the 1995 presidential elections, several changes were implemented immediately, including granting permission for opposition parties to form and to publish newspapers. These changes have brought about a remarkable opening of the political dialogue which allows for critical analysis of the past, including the rule of Julius Nyerere. Many Tanzanians and others with experience in Tanzania doubted my ability to gain research clearance, but timing proved decisive. My application for research permission received approval in late 1992 when I don't believe it would have just months earlier. Several government officials that I interviewed told me quite candidly that they would not have met with me the previous year. I cannot claim any foreknowledge or unusual insight in this matter of timing. It was pure serendipity. Once approval was in hand, I was not denied access to any source of information, as far as I know. Some documents were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 78

sought which could not be found, but this is not surprising given the limited resources that Tanzania has to maintain archives, indeed, I was pleasantly surprised that so much was available under such constrained conditions. Only one interview was denied—former president Julius Nyerere. Though that was admittedly a major loss, I am convinced that it was not because of the topic but due to a general policy the he only grants interviews on his current work, the South Center. One other interview scheduled with the wartime Defense Minister had to be canceled, along with all his interviews and official duties, as he experienced a health crisis. While these foregone interviews were disappointments, I met Nyerere's personal assistant,16 Ms. Joan Wicken, as well as other members of his staff and government. I also interviewed at length a man many consider Nyerere's philosophical and political heir, Colonel (retired) Kabenga Nsa-Kaisi. Moreover, the wartime Minister of Foreign Affairs, Benjamin Mkapa—now Minister of Science, Technology and Higher Education—offered me rare insight into the high-level meetings at the time of the invasion. During a political science conference in Dar es Salaam, I was fortunate to interview Dan Nabudere, a leader of Ugandan exiles in

16 Many believe that "chief of staff" might be a more appropriate title to describe this very astute lady's role in Nyerere's presidency, though Ms. Wicken eschews any such notion of influence. She has worked with Nyerere for over thirty years, during both his presidency and, since his retirement, in his capacity as President of the South Center. This organization seeks to strengthen ties, particularly economic trade, among countries of the South, thereby increasing their self-reliance and international influence.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 79

Tanzania prior to the war and a member of the Ugandan government that was established at the Moshi Unity Conference. His input, along with discussions with others who had lived in Uganda, provided me with a Ugandan perspective of the war and its outcome. Many other Tanzanians within and without government, as well as several in the opposition parties, spent time with me, relaying both their personal experiences of the war and their sense of the war's consequences over time. These interviewees will remain anonymous here, mostly at their request, because identification with a critical perspective might put them in jeopardy should the current open political climate change. Moreover, many of their names would add nothing to their testimony. This chapter has outlined the theoretical and methodological approaches used in this study as well as the sources of data for analysis. In this endeavor, as in any such study, prior research has informed the search for data as well as the method of analyzing that data. It is to this prior research that we now move. The next chapter will assess the state of knowledge on war's consequences and develop from it a kind of road map to guide the search for data on the consequences for Tanzania's war.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3

THE STATE OF KNOWLEDGE ON WAR’S CONSEQUENCES

We are as ignorant about war as the physicist is of the true nature of matter.

Braudel, The Mediterranean and the Mediterranean World in the Age of Philip II

Reviewing the literature on war's consequences serves two purposes in this research. First, an analysis of the outcomes of Tanzania's war necessitates gathering an enormous amount of information. To set out on such a search without first pausing to seek guidance from previous research on war's consequences would be folly. Much can be learned from previous studies about which information might be most useful and how it might appropriately be analyzed. Second, an assessment of the state of knowledge on war's consequences offers an idea of what outcomes might legitimately have been foreseen by Tanzanian decision makers. Any disparity between consequences anticipated in the literature and either the anticipated or actual outcome of war in Tanzania would offer interesting avenues of inquiry. Three characteristics of the literature on war's consequences are quickly apparent. The first is its paucity. Since the end of World War II, scholarly attention has focused upon causes rather

80

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 81 than consequences of war. The possibility that the devastation of one war could become the cause of another is often ignored (Stein and Russett 1980, 399-400). Most studies consider the conditions preceding war, "with little attention to the outcomes of war beyond how long it lasted, how many people were killed, or which state was victorious." (Van Raemdonck and Diehl 1989, 249) The second notable characteristic of this literature is its Western bias. While it may be understandable that the two World Wars are, by far, the most studied wars, it is certainly less understandable that the studies of those wars' impacts are almost entirely limited to Western European nations and the United States, hardly the only nations impacted. In fact, while there has been a noticeable increase in studies of war's consequences and a decline in Western nations' involvement in war since World War II, studies continue to focus on the West's experience of global war. Despite the finding that "the great preponderance of violence in the contemporary international system has erupted in Third World areas" (Starr and Most 1985, 33), studies of war's impact on Third World nations are almost nonexistent. Africa's wars have been as neglected as those of other Third World areas, a regrettable situation considering the fact that "Africa accounts for 35 percent of the nation-years of war over the 1967-1976 period." (Starr and Most 1985, 44) The third characteristic of this literature is the nearly complete lack of comprehensive—or even consistent—conclusions. Studies tend to focus on a narrow band of indicators such as the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 82 effect of war on imports or the rate of technological innovation. Moreover, many studies reach contradictory conclusions, particularly in the economic realm, as they generalize from different sets of indicators, indeed, the conclusions drawn are so limited, diverse, and even conflicting, that the researchers of war's consequences seem "blind" to other perspectives and could be likened to the blind men seeking to learn about an elephant in Saxe's poem (1936, 521-522).

THE BLIND MEN AND THE ELEPHANT It was six men of Indostan To learning much inclined, Who went to see the elephant (Though all of them were blind), That each by observation Might satisfy his mind. The First approached the elephant, And, happening to fall Against his broad and sturdy side, A t once began to bawl: "God bless me! but the elephant Is nothing but a wall!"

The Sixth no sooner had begun About the beast to grope, Than, seizing on the swinging tail That fell within his scope, "I see," quoth he, "the elephant Is very like a rope!"

And so these men of Indostan Disputed loud and long, Each in his own opinion

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 83 Exceeding s tiff and strong, Though each was partly in the right, And all were in the wrong!

So, o ft in theologic wars The disputants, I ween, Rail on in utter ignorance Of what each other mean, And prate about an elephant Not one o f them has seen!

Consequently, and perhaps understandably, none of the studies can offer a comprehensive assessment of the nature of the "beast." An understanding of the gestalt of war's legacy, it seems, eludes us. These limitations in the state of knowledge clearly provide room for ignorance traps to distort a leader's expected utility from the use of force in conflict and demonstrate the need for further studies on war's consequences, particularly in developing nations. Mindful that one purpose of this survey is to draw guidance for the data search on the Tanzanian-Ugandan war, I regularly question the validity of these studies as a guide to the consequences one might reasonably anticipate for Tanzania. In the chapter's three sections, I assess the state of knowledge on war's consequences for: (1) the national economy, (2) the state, and (3) the individual and society. Admittedly, this is an artificial separation of consequences. What impacts the individual and the society clearly also impacts the state and the national economy, and vice versa. However, this typology reflects the disparate, separated nature of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 84 the literature and provides a manageable way of organizing the many studies considered here.

War's Consequences for the National Economy The largest segment of the literature is concerned with the relationship between war and the national economy, and here the lack of consensus is most pronounced. As with the blind men and the elephant, these scholars study different wars and different aspects of war—using different indicators—and come to all possible conclusions: 1. War stimulates economic growth. 2. War retards economic growth. 3. There is no significant relationship between war and economic growth (or no relationship can be found). 4. War both stimulates and retards economic growth, the net result depending on various factors. It is not possible to compare these arguments directly because each scholar seems to have hold of a different part of the elephant. Therefore, in order to survey the full spectrum of arguments, each of these four conclusions will be discussed separately.

Positive Impact A t one end of the spectrum, those who see war as a stimulus to economic growth view war as a catalyst for increasing social discipline which extends to the forces of production, improving efficiency, and thereby stimulating economic growth. Nincic (1980)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 85

surveyed the American experience of recent wars and found that "the nation's economy experienced particularly pronounced surges of growth during wartime." He concluded that:

The evidence does imply that additional claims were placed on the nation's productive capacity, that available economic resources were used more intensely and that eventually, expanded productive capacity was established in response to increased needs. (Nincic 1980, 106) For example, Biersteker (1979, 98) notes that war often isolates a nation through a decline or cessation of trade and foreign investment. This isolation stimulates innovation and entrepreneurial effort by forcing the populace to rely on its own resources. Moreover, ''the reduction in foreign investment during these periods enable[s] local entrepreneurs to initiate and expand their activities without being displaced by large multinationals." Deprived of previous alternatives and free from foreign competition, individuals develop low-cost or intermediate technologies. Indications of war's economic consequences for developing nations can be only narrowly inferred from studies of the consequences of defense spending,1 based on the common sense assumption that war results in marked increases in defense expenditures. If it is granted that one outcome of war is increases

1 Ball (1988) offers an excellent review of the major empirical studies of this growth-military expenditure relationship. Note particularly Chapter 10.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 86 in defense expenditures, then the impact of these increases can be considered to be a consequence of war. Benoit (1973; 1978) surveyed 44 developing countries' defense expenditures and growth rates (as measured by non-defense output) between 1950 and 1965. To his surprise—-and to the consternation of many scholars—he found a strong positive correlation. Countries with high defense expenditures generally experienced the highest economic growth rates, while those with the lowest defense expenditures showed the lowest rates of growth. He explains this result by pointing to the modernizing influence of the military establishment, as well as the military's contributions to civilian sectors such as public works and transferable training. In a later study to reproduce Benoit's results, this positive relationship could only be confirmed for countries that were not constrained by a high debt-service ratio and limited access to external credit. For "financially resource constrained" countries, no significant correlation between defense spending and economic growth could be found (Looney and Frederiksen, 1986). What might this discussion indicate about the outcome for Tanzania's war? How might war affect Tanzania's economic institutions? Under Tanzania's socialist policies, many of these came under government control through a myriad of parastatals. Mobilization and war might place increased claims on the nation's productive capacity and impose a discipline on the government in all areas, extending beyond the war, thereby increasing productive capacity. However, Tanzania would not experience isolation which

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 87

might stimulate the development of new technologies and a more efficient use of local resources. As to the military's influence, certainly Tanzania's military has had a significant role in development programs, particularly in rural areas. If it is assumed that war increases the size of the military and if the military were to remain involved in development activities in rural areas after a war, then this could potentially have a positive impact. However, a poor nation such as Tanzania must be considered "financially resource constrained" (Looney and Frederiksen, 1986), so there is no reason to make the general assumption of a positive economic impact from expanded defense spending.

Negative Impact The other end of the literature spectrum holds that war's most salient characteristics are waste and destruction. Any benefit th at accrues from technological advances, efficiency, or added exploitation of resources is erased by the losses war entails. Wheeler argues:

Negative effects of war which cannot be counterbalanced by putative benefits include: indiscriminate destruction of capital; increases in industrial output which cannot be sustained after, leading to increased unemployment, recession and depression; permanent maldistribution of the work force; manpower losses which often involve many of the healthiest, most able and most creative citizens of the state; destruction of trade patterns; and disruption of communications and interference with transportation development needed for peaceful uses. (Wheeler 1975, 48)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 88 Wheeler studied the impact of war between 1870-1956 on industrial growth—indicated by energy consumption—and concluded that war negatively impacts industrial growth.2 This proved a consistent pattern for both large and small powers, developed and developing nations, winners and losers. He further noted th at the number of lives lost is the best indicator of the cost to industrial growth. In a review of the American experience of serious war from the Revolution through World War II, Wright dismisses the efficiency and innovation arguments.

War doubtless stimulates research and invention along a limited number of lines and these have been greatly broadened in scope with the mechanization of warfare, but it is not clear that it promotes genera! technological progress for peace­ time purposes. . . . In times of war the pressure of competition is greatly reduced. (Wright 1943, 14) In addition to these consequences for productivity, war incurs long-term financial obligations—war debt and veterans' benefits. In the United States, war "contributes significantly to the nation's debt, indeed "practically the entire federal debt has been a war debt." (Musgrave and Culbertson 1953, 101) Another long-term financial drain from American wars has been that "veterans benefits for our major wars during the past century have averaged 1.8 times the original cost of those wars." (Clayton 1969, 662) Interest

2 In a later publication, Wheeler (1980, 284) notes that World War II was an exception to this pattern.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 89 payments on the war debt averaged another fifty percent of the original cost. The Vietnam war provides an example of the extent of these costs. The initial, direct cost of the war was estimated at $110 billion. Adding the costs of veterans benefits and interest payments, the figure grows to $330 billion! Clayton puts this figure into perspective for us:

Compared with other federal expenditures during the same period (fiscal years 1960-70), the war in Vietnam has cost ten times more than Medicare and medical assistance, sixteen times more than support for education, and thirty-three times more than was spent for housing and community development. . . . Put another way, the war has cost us one-fifth of the value of current personal financial assets of all living Americans, a third again as much as all outstanding home mortgages, and six times the total U.S. money now in circulation. (Clayton 1969, 663) As to development and military expenditures in the Third World, several scholars have challenged Benoit's findings (Deger and Smith 1983; Grobar and Porter 1989; Gyimah-Brempong 1989). They acknowledge the positive impacts of military expenditures pointed to by Benoit—increased demand, modernization of attitudes and civilian spin-offs. They found, though, that these positive impacts are "swamped" by the negative effect of decreased savings and investment (Gyimah-Brempong 1989, 79). Moreover, Ball (1991) points out that increased emphasis on a defense-industrial sector intensifies the urban bias at the expense of agriculture and favors capital-intensive over labor-intensive investment. War, then, exacerbates an existing impediment to economic development.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 90 Tanzania's war was fought on Ugandan territory, so it was spared much of the destruction of productive capital that others experience in war, although its transportation system was badly damaged, if the number of lives lost provides a valid indicator of the cost to industrial growth, as Wheeler (1975) suggests, then Tanzania's loss should not be substantial from this war. Debt, however, might well be incurred and would be a long-term drain on the nation's meager resources. The cost of veterans benefits is an unknown factor. Having had no wars prior to that with Uganda, the nation established no system of compensation for veterans. It would only become a cost if the government should decide to establish such a system.

No Impact or Inconclusive There are scholars who believe that war has no significant impact on economic growth. These authors suggest that the prewar economic growth pattern returns after the war, irrespective of the course of the war. While losers are negatively affected in the short-run, over a period of 15-20 years "the effects of the loss dissipate; losers accelerate their recovery and soon resume antebellum status." (Organski and Kugler 1977, 1347) The authors term this phenomenon the "phoenix factor." Interestingly, in a later work, Organski and Kugler found that postwar assistance "is not a significant factor in the loser's recovery rate." (1980, 145) Raster and Thompson (1985, 513) support this conclusion. They further distinguish between interstate and global wars. The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 91

impact of interstate wars on economic growth was found to be "statistically insignificant" while the impact of global wars was significant but "primarily temporary." In another supporting study which utilized the number of telephones as an indicator of economic progress, Barbera (1973, 121) concluded that the relationship between war and economic development, as well as war and global inequality, was "irrelevant." The one study to research war's impact on the national economies of developing countries proved inconclusive. Ammons (1989) compared the economic growth rate of African nations at war with those not at war. While she found no significant differences between the two groups, she did not conclude that war had no impact on economic growth. Rather, she pointed to the high level of conflict on the continent, both interstate and domestic, and the resultant, unprecedented numbers of refugees. She concluded that these factors slowed development in non-warring countries just as war retarded the development in warring countries. Therefore, war was only one factor impeding economic growth on the continent. These studies provide no real guidance for this research except to suggest that the time frame chosen for analysis of consequences, namely the decade following the war, may be too short. In other words, any short-term economic consequences of the war that might be found would be eliminated if the analysis were to wait until the performance of two decades could be evaluated. At the time of writing, the war is only a decade and a half old, so I can only hope that this study stimulates another scholar to compare my

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 92 findings with two full decades of postwar economic performance in Tanzania in order to test this proposition.

Mixed Impact Finally, there are those scholars who find both positive and negative consequences of war, the particular mix of factors depending upon the nature of the war and the postwar policies. Van Raemdonck and Diehl, again studying the West's experience o f global war, concluded that, while the "immediate economic effects are negative, the long-term impact of war and the utility of external aid are uncertain. The impact of war seemed to be less intrinsic than it was conditioned by the policies pursued after the war." (Van Raemdonck and Diehl 1989, 249) Deane surveyed the British experience of two centuries of war and its impact on Britain's industry. She found both growth- stimulating and growth-retarding effects. She concluded that growth is stimulated if the demands of war:

1. draw into productive use under-employed factors; 2. stimulate output in industries whose expansion reduces costs or creates opportunities for other branches of industry; and 3. precipitate fiscal or financial or organisational developments which redistribute incomes or opportunities in favour of innovating enterprise. (Deanne 1975, 91) War retards growth if it involves:

1. physical destruction of real capital; 2. diversion of scarce labour, capital and raw material resources from productive to unproductive uses; and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 93 3. an increase in the risks and uncertainties of mercantile or manufacturing enterprise. (Deane 1975, 91) Within this "mixed" group, there is also a group of scholars of the economic development consequences of military expenditures. They attempt to balance the costs and benefits of defense spending and find the results do not provide an unambiguous answer. In other words, if asked whether defense spending stimulates or retards economic growth, they would be forced to answer, "It depends." Kennedy (1974; 1983) points out that real costs of defense spending are primarily the opportunity costs of alternative uses of these funds. He notes that it is spurious to assume that the alternative uses would have a more growth-producing impact than does defense spending. He concludes:

In what way defense spending in a particular country is contributing to, or competing with, development will depend upon circumstances and will not follow some general law applicable to all times and places." (Kennedy 1974, 189) Ball supports this view.

What the armed forces are buying with their money can be divided into salaries, purchases of operating material, weapons procurement, construction, and research and development (R&D) costs. Each of these categories of expenditure can affect a country's economy in different ways. (Ball 1991, 278) Clearly, the diversity of opinions on war's economic consequences, based so predominantly on the West's experience of global warfare, offers limited guidance and forces us to extrapolate in order to anticipate how the Tanzanian economy might respond to war. In such diversity of conclusions, it seems most reasonable to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 94

take the "mixed impact/it depends" perspective and then search for characteristics in the Tanzanian economy that might provide insight into war's economic consequences, based upon what guidance the literature provides. Tanzania's economy is overwhelmingly dependent upon its agricultural production. In the early 1980s, approximately 83 percent of its economically active population was employed in agriculture, contributing 40 percent of the nation's GNP and 80 percent of its export earnings. Three-quarters of these farmers produced on their own land. Potentially productive land was plentiful as only 5-10 percent of the country was cultivated. Tanzania led the world in arable land per capita. Average farm size was, however, small at around two hectares. An insignificant portion of total land holdings were above ten hectares. Productivity was low (25-30 percent of the land's potential) due to a lack of inputs and soil depletion (Kahama et al 1986, 49; Boesen et al 1986, 109). In addition to being the nation's primary enterprise, Agriculture is also the key to Tanzania's industrial sector. At independence, the country's industrial development was minimal. President Nyerere's guiding development philosophy, articulated in the Arusha Declaration of 1967, foresaw agriculture providing the surplus needed to fuel industrial development under state guidance. Unfortunately, this vision did not manifest itself. Any decline in agricultural output would hamper industrial development and severely restrict foreign exchange for imported industrial inputs to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 95 those industries that were developed, largely through aid projects (Skarstein 1986). Declining agricultural productivity would then bring about a concurrent decline in industrial productivity. Simply stated, agricultural labor is key to Tanzania's economy, and there is no surplus o f agricultural labor. The wealth of land means all can be employed, if not productively so. War's most important consequence for the country, then, would seem to be that it would drain away desperately needed labor from agricultural production to unproductive military campaigns. If we apply this understanding of Tanzania's economy to Deane's criteria (1975, 91) for determining the economic consequence of war (see previous discussion on pages 91-2), we can readily conclude that war might retard growth in Tanzania. Of the three criteria that stimulate growth, the first two are not applicable to Tanzania. Its only underemployed factor of production is land, and it has few industries to be stimulated. As to the third criteria, one might imagine that the increased demand for food in wartime (because soldiers have quit farming but are still eating) might, in many countries, raise prices and stimulate innovation in agriculture. However, the agricultural marketing board in Tanzania was nationalized, and prices were controlled, so this third criteria for war's growth stimulus is also missing. Of the three criteria for economic decline, the second (diversion of scarce resources, both human and material) is clearly the most relevant to Tanzania. Half of the invading force was called up from the country's militia. Tanzania's militia were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 96 overwhelmingly farmers. There was no one to replace them, so production might well have declined. Thus, war in Tanzania could be anticipated to result in the following long-term consequences for the national economy: 1. A diversion of labor to the military might bring about a decline in agricultural production. 2. Declining agricultural production might restrict the availability of foreign exchange. 3. War could incur international debt, the repayment of which might further restrict the availability of foreign exchange. 4. Lack of foreign exchange might restrict the import of necessary inputs for industrial production, thereby causing a decline in industrial output. 5. The above factors might combine to retard the nation's economic production.

War's Consequences for the State Throughout the literature on the development of the modern state, war is generally considered to have played a major role in the "origins, development, consolidation, and viability of the state." (Stein and Russett 1980, 408) Indeed, Tilly (1975, 74) noted that "preparation for war has been the great state building activity." By way of explanation, Ames and Rapp (1977, 177) pointed out that the state is the monopoly provider of two public goods—defense and justice. That is, the state provides protection for its citizens from threats by both foreigners (defense) and fellow citizens (justice).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 97

Historically, the need for this protection provided the impetus for the political centralization and coordination, which resulted in the formation of the state. While war was not a sufficient condition for the rise of the state, it was a necessary one (Carneiro 1970, 734). 'Thus the threat of war makes the state." (Stein and Russett 1980, 408) Returning to the analogy of the blind men and the elephant, it would seem that these researchers of war's impact on the state are standing at one end of the elephant—the western end. This view of war reflects the West’s bias mentioned at the beginning of the chapter. In fact, a nation's experience of war depends upon whether it wins, loses, or is caught in the path of others' wars. Western states did indeed expand through war, but whose territory and resources were usurped to feed this expansion? For the Third World in the age of imperialism, war impeded development of—or even destroyed—the state. Fortunately, not all researchers are "blind" to the Third World's perspective, nor are they all "men". Nicole Ball has written extensively on the consequences of conflict for Third World nations. She points out that Third World wars have politicized the military and enhanced its political power relative to other groups. All too often, the military has used this enhanced power to seize control of the state, thereby weakening its legitimacy.

A frequent outcome of wars and long-term conflicts is the increased involvement of the security forces in the political systems of the states party to the conflict. . . .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 98 Regrettably, the involvement of the security forces in politics only makes the process of creating political systems capable of mediating among the interests of different societal groups more difficult. . . . They have increasingly assumed the positions of mediator among competing elite groups and of guarantor of elite-dominated political and economic systems. Security forces . . . have posed a serious obstacle to political development in the Third World . . . (Ball 1991, 289- 291) In Africa, loss of the state's legitimacy has led to a "silent revolution" (Cheru 1989) or widespread "disengagement" (Azarya 1988) whereby citizens distance themselves from state control through noncompliance with laws and a retreat into a subsistence or parallel economy. Ultimately, the state becomes impotent, a "Lame Leviathan" (Callaghy 1987) or a "state without citizens". (Ayoade 1988) If Ames and Rapp (1977, 177) are correct that the state's role is to provide defense and justice, then it would seem that defense has come at the expense of justice for the state's citizens, and ultimately at the expense of the African state. Traditionally, war has been seen as a catalyst not only for the territorial spread of the state but also for the increase in the state's scope or level of penetration into civil society. "During wars, the effort to create military force, to sustain the war effort, and to discipline the population in both pursuits multiplies state powers, bureaucracies, finances, and interventions in private life.” (Tilly 1992) A t war's end, the state generally retains some, if not all, of this newly acquired power relative to civil society. Lane (1958, 414-415) offers an explanation grounded in economic theory for this power shift. The state holds a monopoly

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99 over the provision of security (and, thereby, stability), a basic ingredient in other products produced by the society. During wartime, the need for protection increases, as does the leverage power of the state's monopoly. Therefore, the state can leverage its protection monopoly to gain control over other areas of production, particularly those closely related to military power such as transportation and nuclear energy, retaining this control in the postwar period. One manifestation of this expanded control of the state is the growth of taxes. "By and large, wars have always provided the principal occasions on which states have increased their levies of resources from their subject populations." (Tilly 1978, 211) Resources are expended more rapidly in war than in peacetime, "and wars thus act as a persistent stimulus to increase the fiscal burden." (Stein and Russett 1980, 409) It has been widely noted that taxes seldom return to their prewar level once the war is over. Postwar public revenue and expenditure levels remain higher than can be fully accounted for by such war-related costs as debts and veterans' pensions. This has been dubbed the "displacement effect" (Peacock and Wiseman 1961, Rasler and Thompson 1985). Growth of the military is one reason for this expanded spending. A t war's end, the size of the armed forces and the military budget have generally not returned to prewar levels. In fact, a study of American military personnel numbers from the Civil War through the Korean War found a virtual doubling of armed forces

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100

after every war. This mechanism is known as the "ratchet effect" (Russett 1970, 2-5).3 Another manifestation of expanded state control is a process known as "concentration." (Peacock and Wiseman 1961, Stein 1980) This process involves a centralization and growth of authority (or concentration of power) of the central government relative to local authorities.

Wartime mobilization engenders political concentration because wars are not run out of local field offices. Unlike education policy, it is not possible for each locality to establish its own war policies, even when it might like to ." (Stein 1980, 17) As with taxation, this shift of power to the central authority often remains when peace returns. While the concept of war as "state builder" was not appropriate for African states, war as "state power grabber" may well be. Wars and the accompanying military spending have long demanded a burdensome share of Africa's ali-too-limited resources. From 1960-1987, military spending in Sub-Saharan countries as a percent of GNP increased by over 500%, from 0.8% to 4.2% (Sivard 1991, 53). Thus, war in Tanzania might well result in an increase in the state's reach, concentration of resources, and taxation, as well as growth of the military. Yet a higher level of extraction from a populace often living at the edge of subsistence could bring war weariness and

3 Stein (1980) found, however, that the Vietnam war was an exception to this pattern.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 101 detract from credibility and legitimacy of the state, ultimately leading to the disengagement discussed earlier. The initial buildup of state power, ironically, ultimately could weaken the state. Based on this discussion, anticipated consequences of war for the Tanzanian state include the following: 1. Mobilization and war's prosecution might bring about an increase of resources in the central government relative to civil society. 2. Initial mobilization might evoke willing sacrifices, but over time the populace might weary of continued sacrifice and government intrusion, spurring massive "disengagement" from government control. 3. Postwar military size and budgets might remain well above prewar levels (the "ratchet effect"). 4. Higher taxation levels reached during the war might remain after the war (the "displacement effect"), yet with a shrinking tax base due to a ravaged economy and disengagement, revenues might decline.

War's Consequences for the Individual and Society War's impact on individuals has been compared to an infectious disease, suggesting that both have comparable patterns of incubation, outbreak, and contagion. In outbreak, those who are susceptible will undergo "something in war analogous to the rise of temperature in fever." They rush to the cause of war, though clearly in varying ways depending upon their age, gender, and circumstances.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 102

Next comes contagion, as "fighting is infectious," and those first infected with "war fever" infect others with their fever. The infection is borne by individual conversations and the mass media. In time, the cycle winds down. Long exposure confers immunity, so that one no longer wishes to fight or to support the fight (war weariness), though this acquired immunity is not permanent and fades in a decade or two (Richardson 1960, 232). This analysis finds that there is a cyclical nature to a citizenry's support for war. At the outset of war, support might be high, but it declines over time, ultimately turning to an aversion lasting beyond the period of actual fighting which might contribute to staying the sword when new conflicts arise. War takes its toll on those in its path. The devastation of war cannot help but affect all who reside in a nation at war, though to greatly varying degrees (Jervis 1976, 239). Those most affected—trauma survivors such as soldiers, prisoners, forced-labor camp internees, and noncombatants victimized by invaders—often exhibit diminished psychological functioning or even anti-social behavior, in addition to limitations from physical injuries they may have suffered (Helmer 1974, 256-299). This impact is greatest on the most vulnerable—the youngest, poorest, and least educated (Williams and Tarr 1974, 214). Thus, beyond the death toll, war brings with it a long-term loss of human capacity both from physical and psychological injuries. There seems no reason to believe that war would affect Tanzanians differently than other nations' citizens. Its peaceful

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 103 history precluded an acquired immunity to war fever, so early support for the war might have been high, but as the sacrifices accumulated and endured, Tanzanians could grow weary of the war and occupation. Though the war was comparatively short, the violence continued through occupation, so we might expect physical and psychological crippling over the full period of two and a half years. Wars also bring about social change, for better or worse. These might include an increase in the crime rate or a higher demand for—and willingness to support—social services. The extent of the change depends upon the length of the war, the level of commitment of resources, and the war's outcomes—whether the nation won or lost. Vanquished nations engaged in long wars that involve a major portion of the nation's resources and population experience the greatest change (Dunn 1974, 244-245). Tanzania's war was short and victorious, but the occupation dragged on for two years, demanding a very large portion of the country's resources. Thus, the sacrifices demanded of the general populace might well have been high relative to the nation's resources, suggesting significant social changes. Social historians of war—those scholars concerned with the impact of social changes wrought by war—have not come to a consensus on the nature of the social changes brought by war. These studies can generally be divided into three groups, each focusing on different areas of concern and concluding that war's impact on the society is positive, negative, or mixed. The first group of scholars,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 104

the "positives", concerns itself with the impact of war on social institutions and concludes that war strengthens them. The second group, the "negatives", studies the affect of war on the balance of power between the state and civil society and concludes that war empowers the state at the expense of civil society. These first two perspectives are generally large, broadly-based studies of global wars. The last group, the "mixed", studies the shifts in competition within society itself that war brings and concludes that they are both positive and negative. This mixed perspective is the growing area of the literature and incorporates empirical methodologies more often than the other two. As was true of the literature on economic consequences, these different perspectives cannot be compared directly as they each have hold of a different part of the elephant, so I will discuss them separately.

Positive Impact - War and Social Institutions This perspective is based on the realization that prosecuting war necessitates a reorganization of society. The term society here is utilized in the larger sense of the nation-state, encompassing both the state and civil society. These scholars see war as a supreme test of society's institutions, causing those that do not serve society to decline or disappear and those that do serve to reorganize toward greater efficiency. The demands of waging warfare impose a social discipline which improves social institutions by making them more responsive and efficient.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 105 Thus, war is seen as "the great auditor of institutions." (Barnett 1963, 11) One scholar (Marwick 1974, 7) believes that this perspective is well expressed in lines Orson Welles wrote for his character, Harry Lime, in the film The Third Man:

. . . you know what the fellow said. . . . In Italy for thirty years under the Borgias they had warfare, terror, murder, bloodshed — they produced Michelangelo, Leonardo da Vinci and the Renaissance, in Switzerland they had brotherly love, five hundred years of democracy and peace, and what did that produce ...? The cuckoo clock. (Greene and Reed 1968, 114) This perspective may not provide relevant guidance for Tanzania where, as in other African nations, modern social institutions (that is, those beyond the extended family or ethnic/religious group structures) are not well-developed, and those that do exist are quite fragile. Thus, if war tests these institutions, it would seem more likely that they would collapse rather than become more responsive and efficient, as was found in the West. Moreover, in Tanzania, it was by design that one of the strongest modern social institutions was the army. Under Nyerere's guidance, it was intended to be a major force in development, assisting communities in translating the party's political ideology into social action (Nsa-Kaisi 1993). During the war and occupation, the military was clearly otherwise occupied and could not fulfill this social mandate. Thus, it would seem reasonable to expect that modern social institutions in Tanzania might have suffered serious setbacks from the war. It may be, however, that if modern social structures were weakened by the stress of war, traditional social structures would reemerge.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 106 Negative Impact - War and the Nexus of State and Civil Society This perspective focuses on the nexus of state and civil society—the distribution of power between the government and society—and concludes that war shifts that nexus in the direction of the state. In war, the state lays claim to a larger share of the nation's human and material resources, so it wields increased power. In wartime, defense spending increases dramatically, and plowshares are turned into guns. These plowshares are the social opportunity costs of war. For example, in one study of U.S. military expenditures in the years 1939-1968, Russett (1970, 141 & 150) found that, for every dollar added to national expenditures for defense spending, forty- two cents was extracted from personal consumption, while seventeen cents was extracted from investment in housing. He found further that state and local government expenditures declined by thirteen cents, including a decline of eight cents for education and two cents each for both health and welfare. Rising defense spending during wars proved to be an enduring outcome, "the ratchet effect", so this shift of resources lasted far beyond the war. U. S. citizens paid dearly to prosecute the country's wars. There are no comparable studies of war's social opportunity costs in the Third World. Most studies on developing nations have concerned themselves with economic rather than social impacts. However, it seems reasonable to believe that the power shift from civil society to the state is a universal outcome of war. Indeed, it is the state that prosecutes war and thus must garner the necessary

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 107

resources. Common sense would dictate that the decline in personal consumption and social spending accompanies all wars, as military spending precludes other expenditures. Thus, Tanzania might well experience similar shifts in resource use. While studies of spending shifts in developing nations during wars are lacking, there is useful information on war's impact in the Third World on human resource allocation. In looking at the effect of military spending on human resources in Africa, Ball (1991, 282- 288) notes that the military often competes with the civilian sector for skilled manpower. An expanded military draws the best and brightest away from social and economic needs during their most productive years. Moreover, the training that these recruits receive is seldom transferable to civilian life. Thus, expansion of the military drains precious human resources away from productive employment. In addition, war's concentration of financial resources increases opportunities for state corruption, further depleting the resources available for social spending. Thus, in war the state may pull human and material resources from civil society, expanding opportunities for state workers to divert these expanded resources for personal gain, and reducing the resources available for social needs. Moreover, these changes may endure long after the war ends. Civil society's power to direct the nation’s resources toward its goals are thus diminished, while the government's goals are increasingly defined by the military. All of these outcomes might occur in Tanzania as a result of war, just as it has in other African nations.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 108

Mixed Impact - War and Social Participation. Inequality. & Cohesion This more recent literature is concerned with shifts of social status within segments of society—such issues as economic participation, inequality, and cohesion—and concludes that the net result is both positive and negative. Wartime mobilization increases the level of participation in society because it demands expanded production to support the war effort. Previously marginalized societal groups—such as women and minorities—are drawn into the expanded production system and, thus, into participation in the wartime economy. These newly integrated groups gain an enhanced social and economic position relative to the previously dominant sectors, thereby decreasing inequality. Moreover, wartime taxation, typically more direct and progressive than peacetime taxation, also leads to a decrease in inequality (Stein 1980). Yet this shift towards economic equality brings new social tensions, as competition and conflict between society's segments is intensified. When an external threat exists, as it does in war, there is an initial increase in cohesion. However, the process of mobilization to meet the threat involves an increase in the depth and breadth of extraction of resources from society. This engenders resentment and competition for remaining resources and tends to create disunity (Stein 1976; 1978). Newly integrated groups demand a reallocation of political power and rewards to reflect the new economic realities. These demands for change generate conflict and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 109

potential violence with those who held power before the war and who resist change that would reduce their power (Stohl 1976; 1980). Thus war increases the level of participation which decreases inequality, but this can threaten social cohesion. Adding to a breakdown in social unity (or perhaps because of it), war seems to bring about an increase in social violence (Stein 1980, 49).4 A path breaking study of 110 (Archer and Gartner 1984) consistently found postwar increases in rates of homicide. These increases occurred in both great and small wars, amongst victorious as well as defeated nations, and despite gender or age. The researchers found that the number of battle deaths relative to the population was the best predictor of the size of the increase in violent behavior in society. Seeking to explain this connection between war and violent crime, Archer and Gartner determined that exposure to authorized or sanctioned killing that war entails offered the best explanation. Soldiers, having once learned to kill, sometimes continue killing even after the war has stopped. Like the shift of the nexus between state and civil society, there seems no reason to believe that war's impact on levels of participation, inequality, cohesion, and violence in Tanzanian society would be different than in any society. As workers are called up and go off to war, heretofore unemployed or underemployed workers

4 Not all scholars have come to this conclusion, however. In his study, "The Impact of the First World War on British Society," Marwick (1968, 63) concluded that war did not foster a growth of violence.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. no might join the formal workforce. These workers are then participating more actively in the society, yet they are competing with other social groups for diminished personal and social resources as the war drains away the country's wealth. Social conflict might increase. As soldiers return from war, they might also exhibit a tendency for both corruption and increased violence.

Anticipated Social Consequences of War for Tanzania Based on this review and discussion, we can develop a list of long-term consequences Tanzanians and their society that could be expected to occur as a result of war. 1. Some individuals, particularly those who directly suffer war's violence or lose loved ones, might experience diminished capacity from psychological or physical trauma. 2. Some participants might manifest violent behavior after returning from the war, increasing rates of violent crime. 3. Pre-existing social institutions, already fragile, might be destabilized. Community development efforts in villages might decline as military personnel are drawn from villages to prosecute the war. 4. As military spending increases, social spending to meet such needs as health and education might decline, causing services to be cut back. 5. Opportunities for formal employment might increase as workers are drawn away to war, increasing the participation

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 111

and status of previously marginalized groups and decreasing inequality. 6. However, the concentration of resources for war might also increase opportunities for corruption by those in positions of authority, thereby increasing inequality. 7. The decline of personal resources that results from mobilization might increase competition and social conflict, threatening social cohesion. In this chapter, I have surveyed the literature on war's consequences, in order to anticipate potential social, political, and economic impacts of the Tanzanian-Ugandan war on Tanzania's development. These anticipated consequences serve both to inform the search for (and analysis of) data on Tanzania's experience of war and to suggest those consequences which might conceivably have been foreseen by decision makers. I am interested in any disparity between actual and anticipated consequences, as well as any disparity between those consequences anticipated in this review and those anticipated by Tanzanian decision makers. It is the two factors—the anticipated and actual outcomes of Tanzania's war—that are to be compared as a means of evaluating the efficacy of war. The Western bias of this literature required that I extrapolate in many instances from the West's experience of global war, sometimes a poor guide to the experience of war in Africa. Moreover, the limited and often narrow interest in the subject of war's consequences has required that I attempt to build a broad framework of consequences from often spotty evidence. These

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 112 limitations do, however, highlight the need for case studies of individual wars and their consequences, particularly in the Third World, and add import to the type of research undertaken in this study. This is the end of Part I, which has provided the background to this research. Part II is concerned with the analysis of Tanzania's experience of war, and it includes four chapters. Chapter 4 investigates the decision for war that was made by Tanzania's leaders. The method of assessing actual consequences of the war is elaborated, distinguishing between primary and secondary consequences, and the direct consequences of Tanzania's war are assessed. Chapters 5-7 then assess the secondary consequences of this war on Tanzania's national economy, the state, and society.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PART II

TANZANIA'S WAR LEGACY

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 4

ASSESSING THE CONSEQUENCES OF TANZANIA'S WAR

The main key to the economics of the postwar world is a simple truism . . . the "Bathtub Theorem.” Production may be likened to the flow of water from the faucet, consumption to the flow down the drain. The difference between these two flows is the rate at which the water in the bathtub—the total stockpile of all goods—is accumulating. War drains the economic bathtub in a great waste of consumption. Kenneth E. Boulding, The Economics of Peace

From the time of the Mogadishu Agreement in 1972 until the invasion of the Kagera salient in 1978, Amin and Nyerere danced around one another in an uneasy truce, a truce which, at times, exploded into violent conflict along the border between the two countries. Amin's final provocation came as a total surprise to Tanzanians and was the most serious of all. Fifteen hundred civilians living in the Kagera salient were killed. Crops, cattle, and all other items of value were looted. Productive assets (including a major sugar factory) worth $108 million were destroyed (PER 1979:4, 7). Moreover, Amin announced his intention to permanently annex the territory. Benjamin Mkapa, wartime Foreign Minister,

114

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 115

noted that, because the conflict with Amin had dragged on for eight years, this sudden escalation caught Tanzanians unprepared.

Yes, but no one really believed that he would actually occupy. He caught us really unawares. You know, you used to dismiss him as a kind of buffoon with all these dreams about military grandeur and medals and everything, but he did do it. He really did do it. (Mkapa 1993) It is from that point in time, late 1978, that we begin an analysis of the choices made by Tanzania's decision makers and the resulting consequences. We are concerned in this study with the disparity between anticipated and actual outcomes of war, exploring Tanzania's initiation of war with Uganda and the consequences of that decision. Therefore, the consequences of Amin's invasion of the Kagera are not considered, other than as provocation for Tanzania's counter invasion. Only those costs and benefits attributable to Nyerere's decision to initiate war are assessed. This chapter explores those consequences of war that were anticipated by Tanzania's leaders and begins an analysis of actual outcomes. The first section analyzes the decision process which followed Amin's invasion of the Kagera salient and the anticipated consequences of a counter invasion and full-scale war. The second section distinguishes between primary and secondary consequences of war, defines the period under study, and elaborates the method used to assess actual consequences. The third and final section assesses the primary consequences of the decision for war.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 116

The Decision to Counter Invade When word of the invasion reached Dar es Salaam, Nyerere convened an emergency meeting of his advisers.1 There was no question within the group that Amin's troops must be removed from Tanzanian territory, but this was achieved without confrontation. Amin capitulated to international pressure and withdrew his troops before Tanzanian forces could be mobilized and transported to the Kagera region. However, once Tanzanians regained control of the Kagera salient, several options for resolution of the conflict were available to Tanzanian decision makers, including: (1) continuing to press for resolution through the OAU or United Nations, (2) stationing several thousand troops along the border to secure Tanzanians against another such invasion, (3) taking the highlands at Mutukula in Uganda, providing for a more secure border and adding to the pressure on Amin to resolve the conflict, or (4) initiating a full- scale war to oust Idi Amin. Nyerere, as Commander-in-Chief of the armed forces, ultimately chose to initiate full-scale war.2

1 These included Prime Minister Edward Sokoine, Defense Minister Rashidi Kawawa, and Foreign Affairs Minister Benjamin Mkapa, as well as chiefs of the defense forces, police, and intelligence (Kiwanuka 1985, 79).

2 In this work (and consistent with Bueno de Mesquita's [1981, 99] definition), Tanzania is considered to be the initiator of the war. For a fuller explanation of this distinction, see Chapter 2, footnote 3.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 117

Understanding the factors that influenced this decision is critical to assessing the anticipated consequences of this war. One factor was the intense pressure on the decision makers. A serious crisis was facing the country when its leaders gathered for consultations, and a sense of urgency surrounded the decision process. As one participant remembered it:

I have been in the government for many years. The work we did for these two days [October 31 and November 1, 1978] was unbelievable. We never slept, never rested at all. It was an emergency, and we acted under great pressure. War decisions are terrible. (Quoted in Kiwanuka 1985, 80) There was also pressure for action from the military establishment. When the Tanzanian government provided support for the invasion of Uganda by exiles in 1972, Amin retaliated by bombing the northern towns of Bukoba and Mwanza. The Tanzanian Peoples Defense Forces (TPFD) had wanted a showdown with Amin then, but Nyerere insisted that the fight was between Amin and the exiles. This time, Amin had picked a fight with Tanzanians, and the TPFD welcomed the confrontation. Having just successfully completed large-scale military maneuvers, the TPFD also felt they were ready. Most of the country's leaders apparently agreed, as very few doubts were voiced about the feasibility of defeating Amin's forces (Kiwanuka 1985, 80-84). Benjamin Mkapa remembered the debate.

[When] a lot of that assessment was done, we realized that his [Amin's] army was not as well-trained, it wasn't led as well. He had good officers, but every time they proved themselves, he would remove them and put in an incompetent, you see. And so, when it is not well led, quite clearly his fighting capacity is limited. He had the arms, but he lacked the skills to use the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 118

arms. And then look at our own positive things. At least some of our boys had some combat experience. They'd been, you know, working with the liberation movements in Mozambique and so on for quite some time. They were better officers, they were better trained. The Chinese had trained us. The Russians had trained part of our army. So, we felt this was do-able. (Mkapa 1993) Most observers conclude that Amin's blunder into the Kagera gave Nyerere the excuse he had long wanted for taking direct action to oust Amin. Dan Nabudere, a leader of the Ugandan exiles, opined:

I am quite convinced that Tanzanians were determined to pursue Amin because they had had difficulties with him and he had made raids into their territory before, and I think they intended to get rid of him. . . . Nyerere himself wanted to get rid of that problem once and for all. (Nabudere 1993) Nyerere hated Amin, whom he saw as a usurper of power from the legitimate leader, a murderous buffoon that was unfit to rule, and a threat to peace in the region. Only Nyerere's commitment to the principle of national sovereignty and self-determination had stayed his hand in the past. Amin's actions in the Kagera removed any constraints and provided Nyerere with what he considered a moral crusade and a just war. In this atmosphere, options other than war were scarcely considered.3 When I asked Benjamin Mkapa why the leaders didn't chose the option of fortifying the border with troops—thereby

3 Major-General Danjuma, Nigeria's Chief of Staff, visited Nyerere bearing war maps, hoping to dissuade Nyerere from a military confrontation with Amin. "President Nyerere's reaction was to ask the Major-General to fold his maps and go back to Lagos." (Kiwanuka 1985, 91)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 119

insuring against further threat from Amin—and then pursuing a diplomatic resolution, he dismissed the option:

You would have had to station a very large army up to the northwest, a very large portion of our armed forces there. . . . We said, "No, no, lets pulverize the fellow." . . . And that's what happened, he was pulverized . . . If you like, it strengthened your defensive position if you pulverized the fellow. . . . The [decision for] the total removal of Idi Amin was inexorable. (Mkapa 1993) At the time war was declared, Tanzania's leaders didn't envision that their soldiers would actually march into Kampala. This was both politically and militarily risky. Rather, they anticipated that, with their troops pushing back Amin's troops from the border and liberating southern Uganda, anti-Amin Ugandans would organize a general rebellion that would oust Amin; thus, the three stages of decision described in Chapter 1. The first stage, taking the highlands of Mutukula, was an immediate reaction of self- defense. When that proved easy enough, a second stage was initiated whereby Tanzanian soldiers moved on to Amin's strongholds in southern Uganda, the barracks at Mbarara and Masaka. At this point, the Tanzanian army invested time and effort recruiting and organizing a Ugandan force. Yet, the hoped-for rebellion didn't come, the Ugandans proved less organized and able than Tanzanian leaders had hoped, and Tanzania's military losses were proving less than once feared. As Benjamin Mkapa remembered:

I can assure you we kept getting surprises as we continued fighting Idi Amin, because his rhetoric got more and more confident, more outrageous the more we advanced. He would say, "I have put up here fortifications; this is impregnable."

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 120

And then the thing would just wilt away. So, I must say that at first we envisaged that it would take much longer. (Mkapa 1993) Moreover, Libyan forces intervened to support Amin. That gave Nyerere political justification for the involvement of Tanzanian troops in stage three, taking the capital and installing a new government. Tanzanian involvement in the removal of Amin was no longer so militarily or politically risky. Thus, the trap was sprung for both Amin and Nyerere by which a declaration of war to defend the Kagera salient became an all-out war to oust Idi Amin. From Nyerere's perspective, the anticipated benefit was the removal of a hated enemy and long-time threat to Tanzanian security. Removal of Amin would bring sweet revenge for violation of the Kagera salient, pride of victory over a fierce enemy, and peace of mind. Moreover, Tanzanian leaders anticipated that Amin would be replaced by a government chosen from among the Ugandan exile community, and Tanzania's constant support of this community assured that future relations with its neighbor would be cordial. And, finally, because of Amin's stupidity and Qaddafi's intervention, the Tanzanian army could remove Amin without political cost, as Tanzanians saw themselves fighting a just and moral war against evil. What of the costs? These were ill-considered. Tanzanians had no experience of war, so they could not properly envision what war might do to the country. Thus, with no real sense of what havoc war could bring to the country, the leaders decided that Amin had to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 121

be removed at whatever cost. As Benjamin Mkapa remembered, "It was a question of national honor. Whatever it takes, it must be done." (Mkapa 1993) In the vocabulary of social traps, the trap for Tanzania was set because its leaders' perceptions of gain from overthrowing Amin (honor, revenge, security) outweighed their perceptions of costs— costs that were unrealistically low or completely disregarded due to their lack of experience with war and its destruction. The countertrap was the perception of consequences that would accrue from the alternatives. Amin might invade again as the OAU could not be relied on to control him, and maintaining a sufficient defensive force along the border seemed too costly.

Assessing Actual Consequences In their pioneering study of economic costs of violent political conflict in Sri Lanka, Richardson and Samarasinghe (1992) distinguish between primary and secondary costs.4 In this study, I retain this distinction and expand it to incorporate all consequences, both costs and benefits. Primary consequences are directly and unambiguously attributable to the war. These impacts accrued only

4 Richardson and Samarasinghe (1992) include a third category, tertiary costs. These result from a "sense of instability an uncertainty" that accompanies violent conflict. Thus, for example, a foreign investor may decide to cancel a planned manufacturing facility, costing the country potential income and jobs. As the war between Tanzania and Uganda took place on Ugandan soil, Tanzanians did not suffer these costs, so only primary and secondary consequences are analyzed.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 122

during the actual fighting and occupation. Primary costs include loss of life, destruction of assets, and extraordinary military outlays for men and material to prosecute the war and support occupation. Primary benefits include the achievement of the goals of the war. Assessing these primary consequences is a rather straightforward task and will be done in the next section. In contrast, secondary consequences are spread over a period of time long after the shooting stops and troops return. The relationship between these long-term consequences and the war are less direct and more ambiguous. While Richardson and Samarasinghe confined their analysis to economic consequences, I chose to explore secondary political and social consequences as well. The increase of national debt, the shifting of international political relations, and the decline of social services are all examples of secondary consequences of war. Ambiguity about the cause of these secondary consequences occurs from the fact that many factors other than the war might have intervened to bring about some of the same outcomes experienced in war. This ambiguity is particularly pronounced in the economic arena. Around the time of the war, Tanzania experienced several other shocks which had important economic consequences. Separating war from these other causal factors cannot be done with exactitude. At times, patterns will be discerned that coincide in time with the initiation of war. At other times, it will be instructive to compare Tanzania's performance indicators with those of neighboring countries that experienced similar shocks,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 123

except war. In some instances, common sense may dictate a connection. The assessment of secondary consequences is considerably more complex than is the task of assessing primary consequences. The time period surveyed is, wherever possible, the decade preceding and the decade following the war. Often, however, the lack of data requires analysis of a shorter period. Moreover, in assessing secondary consequences it is necessary to make assumptions about outcomes that might have occurred under peaceful conditions by extrapolating from the period prior to the war. These secondary consequences for the Tanzanian economy, the state, and the society are analyzed in Chapters 5-7.

Primary Consequences Recall that the war's primary consequences were the direct result of the war and accrued during the time of the war and occupation. Below, I assess both primary costs and benefits.

Primary Costs Deaths - The Tanzanian government has confirmed the deaths of 711 Tanzanian soldiers (Daily News [Dar es Salaam], 7 July 1984). A surprisingly small number—only ninety-six—were actually killed in action during the counter invasion. Another 277 were killed in accidents during the war, such as drowning, when their boat capsized trying to cross Lake Victoria (Avirgan and Honey 1982, 196). The remaining 338, nearly half of all confirmed deaths, were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 124

killed during the two-year occupation. In addition, the government considered 67 soldiers to be missing in action (MIA) but assumed dead (Daily News [Dar es Salaam], 7 July 1984). Making the assumption that the MIA's are indeed dead, the total number of deaths during the war and occupation was 778. Destruction of Assets - This war was fought on Ugandan territory. Ugandan soldiers did indeed invade the Kagera salient, but they withdrew before the Tanzanians could mobilize. Thus, the war that we are assessing was not fought on Tanzanian soil but in Uganda. Although there were instances of bombings around the Bukoba area on Lake Victoria, these bombs were poorly aimed and often fell in the lake or the forests around the towns. Thus, there was minimal damage and no loss of civilian life. What was an important factor, however, was the loss of transport capacity. Many of the soldiers and much of the material to support them had to be transported from Dar es Salaam on the coast—a distance of 850 miles over poor roads—to the Ugandan border. The monsoon rains that hit just as the war was beginning added to the difficulties. Troops and war material were primarily transported via trucks, often simply expropriated when needed from drivers who happened to be passing by. These trips over rapidly deteriorating roads were very hard on the trucks, and, after a couple of trips to the front, the vehicles often broke down. Those that might have been repaired often had to be abandoned because import restrictions precluded importing spare parts.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 125

Trucks were the backbone of the goods transport system in Tanzania. Those that were incapacitated were both privately and publicly owned, and private owners whose trucks were expropriated for government use were sometimes compensated and sometimes not. Thus, there is no way to accurately assess the value of this lost capacity. However, we can get some indication of the extent of the loss. Figure 4 shows the dramatic increase in imports of trucks during 1978 and 1979, the years of mobilization and war. Imports during this period increased by 100 percent. If we assume that all this increase in truck imports was war-related, then approximately 5,500-6,000 transport vehicles were lost in the war. While we cannot accurately evaluate this cost, the loss of this transport capacity proved significant and will be reflected in longer-term assessments of the economy.

Figure 4: Imports of Trucks

7000 T £ 6000-• = 5000-• £ 4000-■ Z 3000 --

1000

io lO'Dr^coCT'O-r- <\iro r-r^r^r-r-r^r-

Source: Bureau of Statistics, Selected Statistical Series: 1951-1988 (Dar es Salaam: President's Office - Planning Commission, 1991), 92.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 126

Increased Military Outlays - Scholars1 estimates of the direct costs of this war have ranged from $500 million (Cheru 1989, 51) to $1 billion (Yeager 1989, 109), though they include the value of assets destroyed by Amin's troops in the Kagera salient. By disaggregating the various estimates, it is possible to subtract the costs borne by Amin's actions in the Kagera and attempt a more precise assessment of the direct costs of the actions taken as a result of Nyerere's decision to counter invade. Observers assessed the average daily cost during the five- and-a-half months of fighting (January to June) at over a million dollars (Ungar 1979), suggesting a total direct cost of military outlays at over $170 million. In his budget speech in June 1979, Prime Minister Edward Sokoine set the nation's outlays between late October (initiation of mobilization) and early April (installation of the new government in Kampala) at $209 million, a figure not out of line with the $170 million estimate if it included the added costs of mobilization and arms purchases. Moreover, Sokoine estimated that costs from April to October 1979 would be $182 million. By year's end, these costs were still considered accurate (Reuters. 22 June 1979; QER 1979:4, 7). By extrapolating from this last estimate, we can develop assumptions with which to estimate occupation costs. War costs during the six months from April to October 1979 included four factors: (1) two months of cleanup operations in western Uganda, (2) support of 45,000 troops in occupation for two months, (3)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 127

demobilization of 25,000 of these troops, and (4) support of 20,000 troops who remained in occupation the last two months. It must be remembered, however, that the smaller force that remained consisted of regular army whose upkeep in a long-term, stationary post would be higher than was true with the maintenance of the militia during the war. By allocating the government's estimated cost of the last six months ($182 million) over the four cost factors, we might reasonably assume the following costs: 1. cleanup operations for two months = $60 million (roughly $1 million per day), 2. occupation by the full force for two months = $50 million, 3. demobilization of 25,000 troops = $30 million, and 4. occupation by smaller force = $30 million (or $15 million per month) This force of 20,000 remained about another six months at an estimated cost, based on the above assumption of $15 million per month, of $90 million. At that point, another 10,000 were demobilized at a cost assumed to be approximately $15 million. The last occupation force of 10,000 remained another 15 months until June 1981. Again basing our estimate on extrapolated assumptions, this last phase of the occupation cost $105 million ($7 million per month), and the final demobilization cost another $7 million. These extraordinary military outlays, over and above the costs of maintaining the military during peace time, total $608 million, as summarized in Table 2. In order to gain some perspective

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 128

with a figure this large, it is informative to compare it with other activities within Tanzania at the time. The war's military outlay is larger than all official development assistance to Tanzania in 1979 ($589 million).5 It was also more than the value of all goods exported from Tanzania in 1979 ($511 million). Exports reached an all-time high of $564 million in 1981, declining thereafter (UN International Trade Statistics Yearbook!. Thus, the war's direct military outlays exceeded all 1979 development assistance, as well as even the most successful export year.

Table 2 Estimated Direct Costs of Counter Invasion and Occupation Date Cost in Millions Ite m 10/78-4/79 $209 war & mobilization 4/79-10/79 182 war & occupation 10/79-4/80 90 occupation 4/80 15 demobilization 4/80-6/81 105 occupation 6/81 7 demobilization

TOTAL DIRECT COSTS $608 million

5 This figure is not inflated by donations due to sympathy with the overthrow of Amin. While Tanzania did make an extraordinary appeal for foreign assistance in June 1979, none was forthcoming.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 129

Remember, this figure is just the cost of extraordinary military outlays and does not include other costs such as loss of human resources through death or injury or the destruction of material resources such as trucks and roads. To give one idea of what the nation had to forego in order to prosecute the war, it is instructive to look at the cost of Tanzania's largest development project. The Tazara or Great Uhuru railway line extends from Dar es Salaam (on Tanzania's coast) 1,100 miles to Kapiri Mposhi (in central Zambia). This railway line is significant for Tanzania as well as Zambia as it transports 80 percent of all Zambia's exports and 45 percent of its imports through Tanzania (Yeager 1989, 130- 131). The line was completed in 1975 at a total cost of $450 million, or the equivalent of $596 million in 1979 dollars. Thus, for this war and occupation, the military costs alone were more than the country's largest development project.

Primary Benefits At the end of the war, Tanzanians had good reason to believe they had achieved their goals, even if their entanglement was greater than they had originally envisioned. Tanzania's army had driven Idi Amin and his Libyan supporters out of Uganda and installed the government chosen at the Moshi Unity conference. The country had rallied in a show of political unity that was paralleled only at the time of the Arusha Declaration.6 Tanzanians were victorious and proud, and the world took notice.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 130

In the Spring and Summer of 1979, observers of this war might well have thought the benefits had been worth the costs. Tanzania's army had been victorious, their loss of life had been surprisingly low, and half their troops were coming home. There was no more to fear from Amin, and friends now headed the Ugandan government. Recall that, in dollar auction game discussed in Chapter 2, after the bidding reached $1, players realized that they could not gain, even if they won, and the strategy shifted to minimizing losses. Tanzania was reaching a similar point in this war, a point at which the gains were maximized and losses were yet come, yet this point was far less apparent in war than in the case of the auction. Over the next few years, the victory would lose some of its luster as Uganda fell back into chaos, embroiling the Tanzanian occupiers. Obote would prove to be at least as deadly as Amin, so Tanzanians could no longer take pride in having "liberated" Uganda. A sense of security would erode as some of Uganda's chaos spread to Tanzania, and the economic cost would climb as war debts came due. The remaining undeniable benefit was honor, and the costs were beginning to be apparent. These and other secondary consequences to the national economy, the state, and the society are assessed in the following three chapters.

6 The Arusha Declaration of 1967 proclaimed a series of major initiatives to achieve "socialism and self-reliance". Major enterprises such as banks, import houses, and sisal plantations were nationalized, and beginning steps were taken to collectivize agricultural production. These would ultimately lead to the Ujamaa villagization program in the early 1970s.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 5

SECONDARY ECONOMIC CONSEQUENCES

We won the war, and we came back poor.

Colonel (retired) Kabenga Nsa-Kaisi

The financial aspect of warfare is something that needs to be constantly borne in mind. It has done immense damage to the economy, there's no doubt about it.

Wartime Foreign Minister Benjamin Mkapa

In an appeal to donor nations for economic assistance as the war in Uganda was ending, the Tanzanian government acknowledged that the war's primary costs were only the beginning.

The economic effect of the war has been far more pervasive than its direct financial cost. Although difficult to quantify, it has involved the postponement of development projects, diversion of transportation facilities for war purposes and, in short, further weakening of the country's export capability and growth performance. (Quoted in Honey 1979) The journalist who reported this had been living in Tanzania for many years. She observed that the economic crisis at the end of the war was the worst in the country's eighteen years of independence. The war had drained the nation's foreign reserves to such an extent that the berths in Dar es Salaam harbor stood empty; there was no

131

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 132

foreign exchange with which to import goods. Thus, its ability to function was severely hampered by the lack of necessary imports. Industries and hospitals alike went without. Such essential commodities as sugar, rice, and toilet paper were scarce (Honey 1979). Unfortunately, this was not to be the country's low point. Bad as the economy might have seemed in 1979, it would worsen during the next four years. This chapter surveys the secondary economic consequences of the war. I begin with economic consequences because, in many instances, the political and social consequences elaborated in the next two chapters are actually outcomes of economic crises to which the war contributed. The decline in social services, for example, was a consequence both of shifts in spending priorities necessitated by war and the general economic decline whose beginning preceded the war. Thus, I attempt to focus first on the secondary economic consequences that contributed to social and political outcomes. Recall that secondary consequences are longer- term and less readily attributable to the war that primary consequences. This ambiguity of connection with the war is particularly true for secondary economic consequences because, at the very time the war was draining a large portion of the country's already-limited resources, other economic calamities were also befalling Tanzania. The East African Community was disbanded in 1977, costing Tanzanians dearly. It required that they replace some previously- shared services for such activities as civil aviation, post and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 133

telegraph, and rolling stock maintenance. The direct cost of this transition was estimated at $100 million and had to be met almost entirely in foreign exchange. Secondary costs were also high, as all communication links were severed between Kenya and Tanzania, and "almost everyone who had to choose to do business with one or the other preferred Kenya." (Ungar 1985, 403) In 1979, there was a second major increase in world petroleum prices. In addition, the world prices for the commodities Tanzania produced were declining. At the National Party Conference on 20 October 1982, Nyerere lamented:

To buy a seven-ton truck in 1981, we had to produce and sell abroad about four times as much cotton, or three times as much cashew or three times as much coffee or ten times as much tobacco as we had to produce and sell in 1976/77. (Quoted in Cheru 1989, 51) Lastly, there was a drought during the 1980-81 growing season, severely limiting production for export and further exacerbating the trade deficit. Thus, it is within the context of this generally dismal economic climate that one must consider the impact of the war on Tanzania's economy. The war did not cause Tanzania's economic malaise, but it certainly did exacerbate problems already apparent. The war could not have come at a worse time for the country. Tanzania was facing a series of economic challenges, like other non­ oil producing, developing nations, and these crises came on the heels of a decade of general economic decline. The war added another serious setback to an already weakened economy, and its impact was

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 134

to significantly diminish the nation's capacity to survive these hard times and to lengthen the time necessary for recovery. While there were many factors responsible for Tanzania's economic malaise, an assessment of the war's share of those troubles will be made at the end of the chapter. The focus of the chapter will be upon the war's impact on Tanzania's international trade and the resultant consequences for the country's comparative economic position. In the next chapter, I will discuss the disengagement of farmers from the formal economy by smuggling or retreating into subsistence production as a strategy to weather the country's economic crisis. While the peasants survived through these strategies, there is no information on this economic activity to be analyzed. The only economic information available is the formal economy which, in Tanzania, is directed toward the export of primary commodities. Therefore, this discussion includes sections on the trade deficit, foreign reserves, debt, and domestic production. A fifth section will compare the country's economic performance with non-warring neighbors in order to assess the extent to which the war with Uganda contributed to Tanzania's economic decline. The final section with analyze how accurately these costs were foreseen and those social traps which may have caused these consequences to be disregarded or discounted.

Trade Deficit Throughout the decade of the 1970s, the volume of exports from Tanzania generally declined. Figure 5 shows an index of export

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 135

volumes of the major export crops from 1970 to 1985. The crops which experienced the most serious decline were sisal and cashew nuts, both of which fell to a fraction of their original volumes. Cotton exports declined roughly by half. In contrast, tobacco exports

Figure 5: Volume Index of Selected Major Exports

250.00

O 200.00 - ...... ■------Sisal

Cotton £ 150.00 ------♦------Coffee

------0— Tea 1 100.00 Tsg 'V ------A ------Tobacco

------£----- Cashew Nuts 50.00 A

0.00 o CM co o

expanded through the middle of the 1970s but dropped back to the previous level by 1985. Tea and coffee exports had small fluctuations, but export volumes were generally flat. The reasons for this general decline in export production are analyzed

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 136

elsewhere1 and are not the subject of concern here. It should merely be noted as background for the discussion which follows. While volumes were declining, so were the country's terms of trade, the ratio of prices of its exports to its imports. This combination of declining volume and terms of trade resulted in a serious decline in the nation's export earnings and its ability to pay for imports of necessary goods. Mobilizing for war and maintaining troops in Uganda, however, required a massive infusion of goods from abroad. Figure 6 shows the values of both imports and exports in constant 1985 prices. In the two years prior to mobilization, real export earnings declined by nearly a third. Yet, at precisely the time export earnings were plummeting, the mobilization and prosecution of the war required a one-third increase in imports.

1 See, for example, Kahama et al (1986).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 137

Figure 6: Constant Value of Imports and Exports

in 35000 co 2o* 30000

| 25000

§ ~ 20000 Exports U 9 “JZ•> 15000 Imports cn 10000

5000

o-t\ifOTrinvDr^(D(Lo — CMhO\fin r^-r^r^r^r^r^r^r-r^^-QDcoaDcooooo Year Source: IMF, International Financial Statistics Yearbook 1990 (Washington DC: International Monetary Fund, 1991), 708-709. Conversion to constant values by author, based on GDP Deflator.

With export earnings dropping and imports increasing, the trade deficit jumped sharply. Figure 7 shows the trade deficit in constant 1985 values. In 1978, the year the nation mobilized for war, the nation's trade deficit more than quadrupled in real terms. Although it dropped somewhat in the next few years, the deficit remained far above the level of most pre-war years. The war came at a time of general economic decline. Yet, war required goods, many of which had to be imported, such as weapons, vehicles of all sorts, and oil. The nation's already declining foreign exchange earnings could not support this level of import consumption, and the country was plunged into a foreign exchange crisis.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 138

Figure 7: Trade Deficit in Constant Values

18000 V) CO ot 16000 14000 (9 12000 B•» ^ e «> 10000 CJ 9 8000 I I 1 £ > 1 - I I I I 6000 4000 2000 0 -I O'—cvJio'^’ insor-

Source: IMF, International Financial Statistics Yearbook 1990 (Washington DC: International Monetary Fund, 1991), 708-709. Conversion to constant values by author, based on GDP Deflator.

Recall in the introduction that I mentioned the surprising lack of interest in Tanzania's war relative to the interest in its Ujamaa villagization program. It is instructive to compare the pattern of trade during the two time periods. The economic environment in which both occurred is comparable. The height of villagization, 1973-74, occurred at the same time as the first oil price increase and a drought. Figures 6 and 7 reveal that the impact on trade during the Ujamaa period was comparable to that of the war period. That is to say, export earnings were declining, and the disruption of villagization, an oil price hike, and added food imports to compensate for drought resulted in a marked increase in imports.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 139

While the impact on trade in both cases is comparable, it should be noted that the resulting trade deficit at the time of the war was far more severe and longer-lasting than at the time of villagization.

Foreign Reserves This dramatic increase in Tanzania's trade deficit quickly drained the country's foreign reserves, as shown in Figure 8. Note that, by 1977, the national economy had recovered from the dislocation of villagization and had built up reserves of $282 million, "thanks to an unprecedented rise in coffee prices." (Kahama et al 1986, 114) Yet, the mobilization in 1978 quickly depleted these reserves. One study of Tanzania's economic crisis lends support to the assertion of the war's importance as a causal factor in the decline of foreign reserves. The authors compare Tanzania's reserves at the time of the war to other developing countries and find the Tanzanian situation unique.

In 1976, Tanzania had US$112 million in foreign reserves, enough to cover 2.4 months of imports. In the same year, the foreign reserve import coverage averaged 2.4 months for other low-income countries .... By 1980, Tanzania's foreign reserves had declined by over 80 per cent to US$20 million. Only one other country suffered such a sharp decline. In fact, most countries were able to increase the amount of their foreign reserves. Tanzania's import coverage had fallen to only 0.2 months, while the average for low-income countries had risen to 5.3 months. (Kahama et al 1986, 7) Other developing countries were facing the same declining terms of trade as Tanzania. But Tanzania was at war, and its financial donors

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 140

were not interested in paying for it. War-related expenditures led Tanzania into a foreign exchange crisis, setting off a chain reaction of events that brought Tanzania to the brink of economic disaster.

Figure 8: Foreign Reserves

300 •» - 250

£ 200 tn S *- 150 e «s ©e 100 •p**» •I £

oo-tMto^rinsor-cooto — (siro^m'or- r'-r'-r'-r'-r'-r'-r^r'-r'-r'-cocococDeocoeoco Year Source: IMF, International Financial Statistics Yearbook (Washington DC: International Monetary Fund, 1991), 706-707.

Debt When the purchase of war material quickly depleted the nation's foreign reserves, the Tanzanian government turned increasingly to borrowing in order to pay for imports, particularly increasing its loans from private sources. Figure 9 shows the country's total long-term debt. Note that there had been a general lull in the growth of debt in the late 1970s due to the jump in coffee prices which built up the country's reserves. However, beginning in 1980, when foreign reserves were approaching zero, debt jumped

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 141

markedly and the portion of debt from private sources particularly increased.

Figure 9: Total Long-Term Debt

3000

2500 4* CO 2000 S *> Private Creditors s © 1500 IP Official Creditors 1000 E 500

Year Source: World Bank, World Debt Tables (Washington DC: World Bank, various years). Note: When figures differed from publication to publication, the most recently published figure was used.

More important for the immediate crisis, Tanzania's debt service increased dramatically during the mobilization and war. Material for war, particularly weapons, had to be purchased on private credit, and the terms of these loans were decidedly less favorable than for official loans. The interest rate of all new loan commitments made to the Tanzanian government during the years of mobilization and war increased 150 percent (from 3.1 percent in 1977 to 4.7 percent in 1979), and the maturity shortened by over ten years, from 34.7 to 24 years (World Debt Tables 1981). Thus, debt service more than doubled during these two years. Figure 10 shows

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 142

the growth of debt service from 1975 to 1985. Note that, during the years of mobilization, war, and occupation (1978-81), debt service nearly doubled and continued through 1985 at far above pre-war levels.

Figure 10: Debt Service on Long-Term Loans

80

^ Principal

H Interest

in r- co o ■— co to tr in r'-r'-r'-r'-r'-eococococDeo Year

Source: World Bank, World Debt Tables (Washington DC: World Bank, 1986/87).

This increased debt service required ever greater sacrifices. In the earlier section on trade, it was noted that export earnings generally declined during the 1970s and 80s. With debt service dramatically increasing at the time of the war and occupation, the ratio of debt service to exports earnings nearly doubled between 1977 and 1980, from around six percent to eleven percent. The ratio of debt service on long-term debt to export earnings is shown in Figure 11.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 143

Figure 11: Percentage of Export Earnings Required to Service Long-Term Debt

14

12 10 n I t» Vkm I I a P I l l l l I I I I I I

1 I I I I I invDr-cooo— f\iroTT r^r'-r'-r'-r'-coeocoeooo Year

Source: World Bank, World Debt Tables (Washington DC: World Bank, various years).

Impact on Production The combination of the foreign exchange crisis and the increased indebtedness had a serious long-term impact on Tanzania's ability to function. The country simply could not afford to the imported goods necessary to maintain production. By 1983, the real value of imports plummeted to nearly half the prewar level (see figure 6 above). At the same time, the portion of total imports allocated to food increased, while that allocated to support production declined. In 1980, 80 percent of the nation's export earnings were spent on oil and food. Imports to supply industry and the increasingly import-dependent agricultural sector were severely curtailed, as were the items necessary to maintain the country's infrastructure (roads, buildings, and transport equipment).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 144

Industrial production fell to as low as 25 percent of capacity (Svendsen 1986, 63 & 68). This decline, combined with a massive retreat into subsistence agriculture (see following chapter), meant that productive capacity in Tanzania was slowly grinding to a halt. This decline was most pronounced in manufacturing and can be seen in the annual growth rate of the manufacturing sector's contribution to the gross domestic product (GDP) at constant prices, as shown in Figure 12. From this chart, it is possible to identify three periods of production. Note that the 1970s were years of some growth, though it was certainly inconsistent. By 1980, however, the

Figure 12: Annual Growth Rate of Manufacturing GDP at 1976 Prices

20

15 e Of 10 ll > e HI

3 - 5

-10 -15 r'-r'-r'-r'-r^r'-r'-r'-r'-r'-cocooocococococo Year Source: Bureau of Statistics, National Accounts of Tanzania 1976-1991 (Dar es Salaam: President's Office - Planning Commission, 1992), 11.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 145

import restrictions were limiting capacity, and manufacturing production declined markedly for four consecutive years, before beginning a very shaky recovery in 1984. This pattern of manufacturing decline after the war is consistent with the experience of the economy as a whole. Figure 13 shows the annual growth rate of the nation's real GDP. Note that, from 1970 until the war in 1979, there were widely fluctuating growth rates, but the rate never fell below 2.5 percent. However, beginning in 1979, there were five straight years of growth below two percent, and in two of these years the economy actually contracted. It wasn't until 1984 that the economy returned to a

Figure 13: Annual Growth of GDP at Constant Prices

7 6

E«> 1 a 0 -1 -2 J------o-wto^in'j3r‘ (DO\o-wto^in^r-(o r'-r'-r'-r-r'-r'-r'-r'-r'-r'-ODCOooeoeococococo Year

Source: IMF, International Financial Statistics Yearbook 1990 (Washington DC: International Monetary Fund, 1991), 162-3.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 146

growth rate matching even the lowest prewar figures, though this growth did not last. Recovery did not seem assured until 1987. The concurrence of this economic slump with the timing of the war certainly supports the argument that the diversion of resources to war instead of production had a serious, negative effect on the already declining national economy. In the next section, I attempt to assess what part of the decline can be attributed to the war.

Opportunity Costs of War In the introduction to this chapter, it was noted that the war was one of several economic calamities that befell Tanzania in the 1970s and 80s. Yet, it is important to try to distinguish the war's consequences from these other calamities. War's secondary economic consequences can be considered opportunity costs, or the costs of "foregone production" that occur due to resources being directed towards military activity rather than production and the disruptions that occur as a result of that policy (Richardson and Samarasinghe 1992). These consequences can be assessed by comparing Tanzania's production (measured as GDP) with that of its neighbors that did not experience war at the time under consideration.2 This approach makes two important assumptions: (1) that the neighbors

2 Tanzania has more bordering neighbors than any other country in Africa—eight. Its non-warring neighbors are Burundi, Kenya, Malawi, Rwanda, and Zambia. Its three other neighbors (Mozambique, Uganda, and Zaire) experienced war during the period under consideration, 1970-87 (SIPRI 1985).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 147

experienced the same external economic constraints that impacted Tanzania—namely oil price hikes, declining terms of trade, and drought, and (2) that, because other disruptions were experienced in common, differences that did occur in the pattern of economic production can be attributed to the disruption of the war. These assumptions are admittedly bold, particularly the latter one. Common sense would suggest that the first assumption cannot be wildly out of line. None of Tanzania's neighbors are oil exporters, and their economies are, for the most part, based on similar primary commodities and are, therefore, subject to the same constraints. The second assumption—that differences can be attributed to war—has more utility than demonstrable validity. Despite numerous similarities, Tanzania's economy undoubtedly differs from the economies of its non-warring neighbors in some ways—size, level of industrialization, dependency on oil, etc. However, in order to make any assessment of the war's economic consequences, it is necessary to make some assumption about the conditions that might have occurred without war, and comparatives are most helpful in developing this peace scenario. By using five countries, it is hoped that these differences cancel one another. Moreover, Tanzania's average rate of growth in the nine years prior to the war does compare with its neighbors' growth rates, which provides a justification for the use of these countries as a basis for Tanzania's peace scenario. In this analysis, economic costs will be considered equivalent to economic consequences, as no significant economic benefits

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 148

accrued to Tanzania as a result of the war that would offset costs. Immediately after the war, all intercontinental trade with landlocked Uganda was transported through Tanzania rather than Kenya, but this trade quickly reverted to Kenya because it was usually more cost effective. Transport through Kenya was also more reliable since much of Tanzania's transport capacity was lost in the war. Moreover, some trade with Uganda developed during the occupation, but this disappeared as soon as Tanzanian troops returned. Finally, as will be discussed in the next chapter, food aid to Tanzania increased during the two years of occupation, one means by which donors could provide quick and politically acceptable assistance in response to Tanzania's appeals for help after the war. These events can indeed be considered beneficial. However, their impact was so inconsequential by comparison to economic costs that they will be disregarded. As a country's pattern of growth is inconsistent from year to year, it is useful to look at average rates of growth over a period of time. Previously, it was noted that Tanzania experienced three distinct patterns of growth during the period of study, namely pre­ war (1970-78), war and immediate post-war (1979-1983), and longer-term post-war (1984-86). This assessment will utilize these time periods as the basis for comparison with non-warring neighbors. Figure 14 compares Tanzania's growth rate in each period with the average of its non-warring neighbors. As mentioned above, Tanzania's growth rate nearly reached that of its non-warring

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 149

neighbors during the nine pre-war years, despite the disruption of Ujamaa and the breakup of the East African Community. Over the five years of Tanzania's war, occupation, and immediate post-war period (1979-83), Tanzania and its non-warring neighbors experienced an economic decline, but Tanzania's was far more severe. The average growth rate of the non-warring neighbors declined by 2 percent, but Tanzania's growth rate fell 4 percent to nearly zero. During the three years of the post-war period (1984- 76), Tanzania recovered slightly to nearly 1 percent, while its non­ warring neighbors grew at a rate of 3 percent. By 1987, the nation

Figure 14: Average Annual Growth Rate of GDP at Constant Prices

5 , ------

H Tanzania

H Non-warring Neighbors

70-78 79-83 84-86 87-88 Years

Source: IMF, International Financial Statistics Yearbook (Washington DC: International Monetary Fund, 1990), 162-3. Note: Tanzania's non-warring neighbors are Burundi, Kenya, Malawi, Rwanda, and Zambia. Tanzania's three other neighbors (Mozambique, Uganda, and Zaire) experienced war during the period under consideration, 1970-87 (SIPRI 1985).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 150

seemed to have recovered from the impact of the war, as Tanzania experienced two years of solid growth and surpassed its non­ warring neighbors' average growth rate, lending support to the assumption that Tanzania's growth would have matched that of its neighbors had the war not occurred. Based on the preceding information and assumptions, a scenario can be developed concerning Tanzania's growth without war. In this case, it will be assumed that Tanzania would have experienced a 2.5 percent growth rate each year from 1979-83 and a 3 percent growth rate during the years 1984-86, as did its non­ warring neighbors. Thus, this peace scenario assumes that the growth rate would have been 2 percentage points higher than it actually was during the full period 1979-86. It also assumes that normal growth—or growth unaffected by the war—resumed in 1987. This scenario can then be compared with actual growth figures, the difference between the two being the opportunity costs (or secondary economic costs) of the war. Table 3 summarizes the results of this approach. It shows the real GDP (1985 value) that actually occurred as well as the real GDP that would be anticipated in this peace scenario and the differences between these two outcomes, the war's opportunity cost for Tanzania. This total opportunity cost of nearly TShs 90 billion or US$5.12 billion3 is more than three-quarters of the size of the total

3 The 1985 exchange rate was 17.472 TShs/US$ (IMF 1990, 685).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 151

GDP in 1979. The war and subsequent occupation cost the Tanzanian economy an average of TShs 11,248 million or $644 million each and every year over the eight years the war's economic consequences

Table 3: Actual & Peace Scenario GPP & Opportunity Costs Actual Peace Opportunity Year Growth Scenario Growth C ost Growth Real GDP* Growth Real GDP (M il Rate* (Mil 1985 Rate (Mil 1985 1985 (%) TShs) (%) TShs) TShs)

1978 - - 115,862 - - 115,862 0

1979 1.2 117,295 2.5 118,759 1,464

1980 0.8 118,278 2.5 121,728 3,450

1981 -1.1 116,976 2.5 124,771 7,795

1982 1.3 118,486 2.5 127,890 9,404

1983 -0 .4 117,968 2.5 131,087 13,119

1984 2.5 120,892 3.0 135,020 14,128

1985 -0 .2 120,621 3.0 139,070 18,449

1986 0.4 121,066 3.0 143,243 22,177

Total Cost = TShs 89,986 million ______or US$5.12 billion *Source: IMF, International Financial Statistics Yearbook (Washington DC: International Monetary Fund, 1990), 163 & 685.

retarded growth. Recall from the previous chapter that the total primary economic costs equaled $608 million, though this was 1979 dollars. Converted to 1985 dollars, the primary costs equaled $869 million. Thus, the economic disruption and dislocation of the war

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 152 caused the nation annually to forego production equal to, on average, three-quarters of the direct costs of the war each year 1979-86. For a small country like Tanzania, $5.12 billion is an enormous price to pay. To gain some perspective on just how great a sum this is, note that the total Tanzanian debt in 1985 stood at only $3.6 billion and the entire 1985 GDP was $6.9 billion. The war's opportunity costs were nearly one-and-a-half times the size of the national debt and three-quarters the size of the country's total annual output. This was indeed a very costly war.

A nalysis This overview of the war's secondary economic consequences for Tanzania generally supports the propositions developed from the literature reviewed in Chapter 3, though their impact was exacerbated by a decline in the country's terms of trade that was not anticipated. Agricultural production declined after the war, though that was a pattern established before the war. This declining production, combined with a decline in the nation's terms of trade, resulted in a fall in export earnings just as the country depleted its reserves in the mobilization and incurred added debt. To deal with the massive trade deficit which resulted, the government severely restricted imports, many of which were essential as inputs for further production, thereby creating a spiral of declining production that lasted until 1987. The net result was a opportunity cost of $5.12 billion in foregone economic growth spread out over a period of eight years.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 153

How accurately were these costs foreseen? In several addresses to honor Hero's Day of 1979, three months after the war had ended, President Nyerere exhorted the nation to pull together and raise production.

Let us apply the same discipline we demonstrated during the war and redress our economy. Failure to do so could make the country an object of derision before the outside world with claims that our economic problems stemmed from sheer pride to fight Amin. (Daily News [Dar es Salaam], 31 August 1979) He acknowledged that there were "hardships, shortages, and economic troubles" resulting from the war, but he predicted that these problems would last only eighteen months, then things would get back to normal (Daily News [Dar es Salaam], 2 September 1979). In fact, the economy seriously worsened over the next eighteen months. By then it was 1981, and the economy actually contracted one percent in real terms (See Figure 13), the worst economic year in Tanzania's history. No one seriously imagined the depth of the crisis nor the length of time the country would suffer. One former government employee reflected on the miscalculation, believing the government waited far too long to exact taxes to support the war, thereby misleading the country about the true costs.

It was both an underestimation of the real costs and an underestimation of the people of Tanzania, I believe. [First], not realizing both what the costs would be, and secondly, wanting everybody to be wholehearted in support of the war and frightened that if you put taxes up, they might not be. (Identity Protected)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 154

Thus, there was an Ignorance Trap, the result of willing ignorance, that prevented decision-makers from fully anticipating costs. They wished to retain support for the invasion, so they perpetuated the ignorance by delaying costs (through borrowing rather than taxing to finance immediate war expenditures), thereby creating a Time-Delay trap for the populace. Tanzanians were lulled into believing that, once the war was over, the greatest costs were behind them, when massive costs were actually still several years down the road. Moreover, the opposite of a Diminishing-Returns Trap, what I'll term a Spiraling-Costs Trap, was at work. Over time, the opportunity costs were actually growing. Table 3 shows this very clearly. As the decline of production proceeded, the annual opportunity costs grew, shown in the right-hand column of the table. Each year, the costs were greater than the previous year. At the same time, the benefits that were celebrated at the war's end were eroding, a Diminishing-Returns Trap. While the border was more secure, political stability did not come to Uganda during the occupation. Indeed, chaos continued until 1986 when Museveni replaced the deposed Obote, and some of this chaos spilled over into Tanzania. These are some of the effects that are discussed in the next two chapters. These traps distorted the ability of Tanzania's leaders to accurately anticipate the war's political and social consequences, as well as the economic consequences reviewed in this chapter. The next chapter assesses consequences for the Tanzanian state, the extent to which these political outcomes were unanticipated, and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 155

the traps which may have inhibited the leaders' ability to accurately foresee these consequences..

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 6

SECONDARY POLITICAL CONSEQUENCES

During wars, the effort to create military force, to sustain the war effort, and to discipline the population in both pursuits multiplies state powers, bureaucracies, finances, and interventions in private life. . . [However], the accumulation of debts and political commitments during a war effort, the stress of demobilization, and the dismantling of wartime controls all combine to make states more vulnerable to their domestic opponents.

Charles Tilly, "The Future of the State System"

This chapter reviews and evaluates the evidence of the war's consequences for the Tanzanian state. As this was an interstate war, there were international as well as domestic consequences. Domestic effects will be the primary focus of this analysis, though some attention will also be given to the war's impact on Tanzania's international relations. War's domestic political consequences fall under the general heading of changes in the state's power relative to civil society, including society's reaction to this shift. Thus, the discussion which follows includes four sections which deal with this shift: (1) the state's concentration of resources, (2) militarization, (3) civil

156

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 157

society's disengagement from the state, and (4) political challenges to the government. A fifth section discusses changes in Tanzania's political and economic relations with its neighbors and donors. In a sixth and final section, the war's political consequences will be summarized and analyzed in light of the view that this war was a social trap.

Concentration Concentration refers to the size of the state relative to civil society. The division of power between the government sector and the private, or civil, sector can be determined by the resources that are allocated to each. One means of measuring that allocation of resources is to compare the central government's budget to the gross national product (GNP), as shown in Figure 15. Recall in the previous chapter that a similarity was noted in the trade deficit both at the time of villagization and the war. Here again, we see a repeated pattern. The increase in the government's share of GNP that occurred during the war also occurred during the villagization program of the mid-1970s. Once again, the change was more extreme during the war than during villagization. Central government expenditures as a percent of GNP averaged around 25 percent during the early 1970s. In the mid- 1970s, the period of villagization, government expenditures grew to over 30 percent of GNP, then fell back again to the level of the early 1970s. When the war began in 1979, this percentage jumped again, this time to over 35 percent and remained comparatively high for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 158

four years before falling again to levels comparable to the early 1970s.

Figure 15: Central Government Expenditures as a Percentage of GNP

38 36 34 „ 32 o5 30 $ 28 °* 26 24 22 20 r'-r'-r'-r’-r'-r'-r'-r'-r'-coeocococoflocooo Year

Source: Central Government Expenditure - United Republic of Tanzania, Economic Survey (Dar es Salaam: Ministry of Planning, various years). GNP - IMF, International Financial Statistics Yearbook (Washington DC: International Monetary Fund, 1991), 708-709.

In addition to financial resources, both government and private sectors utilize human resources. Another measure of relative allocation of resources is the relative employment levels in the public and private sectors. Figure 16 shows the formal employment by sector. There was a general increase in the government’s formal employment throughout the 1970s. Once again, there were two periods of a larger-than usual increase, villagization and the war. In 1973, the number of public employees grew by 16

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 159

percent, while this number grew by 24 percent in 1978, the year of mobilization for war. Note that private sector employment remained fairly constant throughout, so employees were primarily drawn from non-formal rather than from private employment.

Figure 16: Total Formal Employment by Sector

500

^• 400 o 0 5 300 Public *> Private 200 |> 1 100

0 o — fMioTrin<£>r-floa*0'^-CMKjTr r^-r-r-r-r-r^r-r-r-r^-floflococooo Year

Source: Bureau of Statistics, Selected Statistical Series: 1951-1988 (Dar es Salaam: President's Office - Planning Commission, 1991), 16.

The expansion of public employment during villagization was largely a result of the need to staff education and health facilities that were developed in the new collectives. The expansion brought about by the war is more difficult to explain, as it can only partly be accounted for by military recruitment. Over sixty thousand employees were added to the government payroll in 1978, yet only an estimated twenty-five thousand of these new government employees

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 160

were drafted into the military. Note also that, while the rate of expansion in government slowed, the number of public employees remained high through 1984, representing nearly a 50 percent growth over pre-mobilization levels. Thus, the war was accompanied by a substantial extension of the government's control over the nation's resources, both human and financial. In mobilizing for war, the government claimed an additional ten percent of the national product and five percent of ail formal employment. While the expansion of public employment was an enduring consequence, the growth in government's share of GNP returned to prewar levels by 1983. In both instances, this shift of the state-society nexus in the direction of the state was similar to—though more pronounced than—the shift during the Ujamaa villagization period.

Militarization One factor contributing to the growth of government was the expansion of the armed forces, an increase that endured long after the war was won and the occupiers were back home. By 1987, eight years after the war, Tanzania's armed force was more that twice as large as it had been prior to the war, a "ratchet effect" comparable to that found in a study of U. S. wars and military size (Russett 1970). Figure 17 shows that, after Uganda's invasion of the Kagera salient in 1978, Tanzania increased its armed forces from under 20,000 to a peak of 51,850. Throughout most of the next decade, this number remained above 40,000.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 161

Recall from above that the growth In government's share in human resources was an enduring outcome of the war while its expanded share of GNP was not. In the long term, this means that a constant level of resources were increasingly shifted to wages rather than other government expenditures, and from figure 17, it is clear that these wages employed soldiers as opposed to other national needs such as teachers and nurses. This effect will be seen in the next section on social consequences. What is important to note is that the portion of the nation's resources allocated to national security significantly expanded during the war and that this effect remained throughout the next decade.

Figure 17: Total Armed Forces

W7_ 60000 8 50000 40000 It | 30000 II II < 20000 II II |o 10000 WTTTI I I I II " 04 IKK .1 111 11 11 I CM V <0 co o CM r r <0 r - r - r- r^ CD CD CO CD Year

Source: Legum, Colin, ed., Africa Contemporary Record (London: Rex Collins, 1972- 1988).

There was one other aspect of Tanzania's militarization. Under Nyerere, there had always been a policy of moving people from the military to government service and vice versa. This was in line

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 162

with a deliberate policy of politicizing the military (Nsa-Kaisi 1993). When the soldiers returned victorious, their political clout increased, and many of them were given government positions. Although there is no data demonstrating this shift, Minister Mkapa acknowledged the change when he said, "There is no doubt that there were more of them [military] in the political establishment than before." (Mkapa 1993)

Disengagement Disengagement can be defined as "the tendency to withdraw from the state and keep at a distance from its channels as a hedge against its instability and dwindling base." (Azarya 1988, 7) Two trends that have already been discussed converged at the time of the war to stimulate a massive disengagement: (1) the nation faced an economic crisis (discussed in the previous chapter), and (2) the government expanded its control of the nation's resources (discussed in the first section of this chapter). In response to these changes, individual agricultural producers disengaged from the government's control in one of two ways: (1) by retreating to a subsistence economy, halting production for the market, or (2) by shifting to alternative marketing channels such as smuggling and black markets. Both shifts can be seen in the pattern of agricultural production and marketing after the war. An example may be instructive here. In Tanzania, a major instrument of government extraction was marketing parastatals such as the National Milling Corporation (NMC). This parastatal was

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 163

granted a monopoly on the wholesale purchase of grain for the retail or export market. Only grains produced for household consumption could legally circumvent the National Milling Corporation. Similar monopolies were granted to various parastatais for other commodities. The government could then set both the producer price and the retail price for these commodities, thereby collecting revenue from the difference between these two prices. For export commodities, the export price was determined by the world market, but the government still controlled the producer price and thus collected the difference. For some commodities considered essential for the domestic market, the government subsidized the producer price in order to provide the commodity to urban consumers at affordable prices. In most cases, however, the government gained on the transaction, in all cases, government revenues from these commodity sales benefited by keeping producer prices low. Figure 18 shows the effect of price controls on the producer price of one sample crop—cassava, a staple of the Tanzanian diet. Note that, in the case of cassava, as government's concentration of resources increased during the years of mobilization, war, and occupation (1978-81), the real producer price dropped by more than one-third. This decline continued through 1982-3 as the country faced a deepening economic crisis. In an effort to extract more resources for government use, the producer price of cassava was depressed relative to other goods and services. Once again, this pattern is a repeat, though more pronounced, of the pattern seen in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 164

the mid-1970s, when government resources were stretched by the demands of villagization.

Figure 18: Official Producer Price of Cassava

650 T

-k .= 500

■^•ino — w ro f in o r'-r'-r'-r'-r'-r'-cooococococococo Crop Year Ending

Source: Bureau of Statistics, Selected Statistical Series: 1951-1988 (Dar es Salaam: President's Office - Planning commission, 1991), 54.

The case of cassava is typical of many other commodities, and producers reacted quickly to the depression in prices. During the period of the war and occupation, the government dramatically increased its margins on crops. Not surprising, this increased the producers' incentive to avoid official marketing channels, and many farmers sold on the black market, where prices could be as much as 70 percent higher (Kahama et al 1986, 54-55). The resulting large- scale abandonment of the National Milling Corporation can be seen in Figure 19, which shows the purchases of selected food crops on the official market. Note that NMC purchases grew steadily during the latter half of the 1970s. However, after the war, when producer

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 165

prices were dropping so dramatically, officially marketed quantities plummeted from 368,800 to 151,000 metric tons, a drop of nearly 60 percent. Compare Figures 18 and 19 and note the synchronicity in the 1975-82 period; marketed quantities rose and fell with producer prices, not an illogical relationship.

Figure 19: Purchases by the National Milling Corporation 400 j ys ■■ ¥7 350 o o 300 ■■ o» 250 -• Wheat

*»c 200 -■ e H* 150 ■■ L j Paddy z<** 100 ■■ V r 50 -I Maize 0 4

Production Year Ending

Source: United Republic of Tanzania, Economic Survey (Dar es Salaam: Ministry of Planning, 1982), p. 102.

Some producers were able to circumvent the National Milling Corporation and market their crops on the black market or smuggle them out of the country, mostly to Kenya. The ability to do this greatly depended on the crop produced, the proximity to the Kenyan border, and the producer's access to transport that could to bring the produce directly to consumers. These constraints clearly favored

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 166

the large producers in the highlands along the Kenyan border—already the wealthiest of Tanzania's agricultural producers—and shut out the small producers in the interior of the country. The extent of the black marketeering and smuggling can be seen in Figure 20, which shows the pattern of another crop, maize (corn), one of Tanzania's most important. Note that production increased through the early and mid-1970s, but it declined in the years 1979-81. However, in 1982, production took off, nearly tripling by 1986.

Figure 20: Maize Production and Percentage Officially Marketed

3000 Ac % Marketed 2500 ••

2000 ■■

1500 ■ - Production 1000 -■

to'tfiftvor'-CDCho — r'-r'-r'-r'-r^-r'-r-cocococoeococo Production Year Ending

Source: Production - FAO, Production Yearbook (Rome: United Nations Food and Agriculture Organization). Marketed Quantity - Bureau of Statistics, Selected Statistical Series: 1951-1988 (Dar es Salaam: President's Office - Planning Commission, 1991). 53.

Contrast this growing production with the percentage of maize produced that was marketed through official channels. During

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 167

the years of forced villagization in the early 1970s, the percentage reached a peak of nearly 25 percent. This was a period of unprecedented control of agricultural production by the government, so this might be considered a potential maximum, with the remaining 75 percent largely going to household consumption (maize is a staple of the Tanzanian diet). As the villagization program began to unravel and the government's control loosened, this percentage dropped back to around 15 percent. The marketed percentage increased again briefly the year of the war, but with the steep decline in producer prices after 1979, the officially marketed percentage plummeted to under 5 percent. Peasants were finding alternative markets for their maize. This expansion of a black market created conflict for the government, not only with producers but also with consumers. By 1981, as black marketeering grew, the government's attempt to stem the tide was often targeted at the black market's customers. As was discussed in the previous chapter, one manifestation of the economic crisis at the time of the war was a serious imbalance of payments and an elimination of the country's foreign exchange reserves. The government responded by establishing ever more severe restrictions on imports in order to conserve foreign exchange. Thus, imported consumer goods were nearly non-existent. However, farmers who were able to smuggle their crops into Kenya could exchange them for imported consumer goods which would bring a high price on the Tanzanian black market.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 168

In an attempt to stop this smuggling and black marketeering, the police often searched urban consumers, looking for foreign items obviously obtained on the black market. The actions of police in these efforts were often extreme and proved another means of fueling official corruption (see section on crime in the following chapter). One businessman remembered being taken to the police station, questioned, and threatened because he was found in possession of one tube of toothpaste not available in Tanzania for which he had no receipt. He was ultimately released, but his toothpaste was retained by the police. While some producers were able to access alternative markets, many were not. These producers chose the other alternative in response to the government's increased control, retreat into subsistence. One estimate found that "subsistence production constituted as much as 40% of total GDP in 1980 against 29% in 1970." (Svendsen 1986, 64) This was certainly the most common response, as per capita agricultural production in Tanzania, both food and cash crops, began a disastrous decline after 1979. Figure 21 shows that per capita agricultural production generally increased from 1974-1979. This is total production, not officially marketed produce. Production dropped precipitously in 1980 as producer prices dropped, and it continued declining through 1987. This data demonstrates quite clearly that the nation's agricultural producers, the backbone of the Tanzanian economy, responded to the economic crisis and government's suppression of producer prices at the time of the war by simply disengaging. Those

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 169

farmers that had alternatives simply abandoned the official marketing channels, smuggled their goods outside the country, and sold them for higher prices or exchanged them for consumer goods in great demand on Tanzania's black market. Others who could transport their goods to urban areas sold their produce on the black market at prices far above the official price. Most, however, did not have access to either option. These farmers, usually small-scale producers in the country’s interior, merely abandoned commercial production altogether and produced only for household consumption or for local barter.

Figure 21: Index of Total Per Capita Agricultural Production

105 e 0 J 100 _ ■ - I eo 1 ©* r* 95 J co 0) 1 , , > 90 J 1 •» (0*> CQ 85 lllll tT Ln sor-oooto — fMio^insor- r'-r^ r-r'-r-r^cocococococococo Year

Source: FAO, Production Yearbook (Rome: United Nations Food and Agriculture Organization, 1985 & 1989).

This disengagement had an impact on the state's revenues. Figure 22 shows real tax revenues collected by the state. Through

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 170

the decade of the 1970s, tax revenues generally increased. Though there was a serious drop in 1976, the increase was back on track two years later. During mobilization in late 1978, the government increased sales taxes on luxury items (PER 1979:1, 4). At the war's end in June 1979, as the costs kept mounting, taxes on these items as well as gasoline were again raised (Honey 1979). Despite these increases, revenues dropped somewhat in 1979. They recovered again the following year but remained flat thereafter, despite tax increases and an expanding population. This would suggest a shrinking tax base, just what one would expect at a time of large- scale disengagement from the formal economy, the source of tax revenue.

Figure 22: Government Tax Revenue (in constant 1980 values)

7000 -

* 6500 - ■ - ■ l . l a a e £ 6000 -

£ 5500 - m i m \ e T i 11 xt» 5000 - tft 4 5 0 0 - ■ I III 9 4000 - cvjfOTj-in^Dr-coCT'O — c\j k ) tT ld r'-r'-r'-r'-r'-r'-r'-r'-cocococococo Year Ending Source: IMF, Government Finance Statistics Yearbook. (Washington DC: International Monetary Fund, 1982 & 1990).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 171

This disengagement by the nation's agricultural producers also reflected a loss by the government of a certain credibility that it had previously enjoyed. The peasant's generosity that was evidenced at the war's outset turned to hardened resistance and even hostility as the cost proved more than anticipated and the government continued demanding sacrifice. When the war was so quickly and decisively won, the average citizen assumed the worst was over. After the war was won, President Nyerere predicted that the hard times would be over in eighteen months (Daily News [Dar es Salaam], 2 September 1979). By 1981, when the hardships continued and even worsened, citizens were tired of the government's use of the war to excuse all problems. Something other than the war had to be at fault, and the government received its share of blame. With both its resources and credibility diminished, the Tanzanian state was weakened. The government that had mobilized the nation to defeat one of Africa's most powerful armies was being abandoned. The state that had committed itself in the Arusha Declaration to increase peasant participation, both economic and political, found itself in circumstances that discouraged the peasant from participating. Like many other African nations, Tanzania was in danger of becoming a "state without citizens" (Ayoade 1988), only inhabitants.

Challenges to Nyerere The first sign of serious malcontent and organized political resistance to the government's policies came in March 1982. A

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 172

group of Tanzanians, led by an army officer, hijacked a plane from the national airline, forcing the pilot to fly to Kenya, Saudi Arabia, Greece, and England, "along the way radioing a description of the plight of the poor in Tanzania and a demand for Nyerere's resignation." (Ungar 1985, 404) Moreover, they required that all negotiations be communicated through Oscar Kambona, a former Foreign Minister in Tanzania and bitter opponent of Nyerere who had fled into exile in 1967 (Legum 1981-82, B279). The group's leader, Moussa Member, claimed to speak for a 3,000-member group called the Tanzanian Youth Democracy Movement that he founded in 1979. Though there was neither independent corroboration that such a group existed nor that Kambona had a part in the plot, the hijackers' attempt to seek his leadership for their rebellion was an indication of the beginnings of a united resistance. After a 26-hour siege, the hijackers surrendered and released their hostages. They were ultimately convicted in British courts (Legum 1982-83, B288), but the dramatic challenge to Nyerere's rule (the first since the Tanganyika Rifles mutinied over salaries and Africanization of the military in 1964) was noted both at home and in the world press. The government was aware that the rising level of discontent with the shattered economy, widespread corruption which had reached the senior ranks of the army, and dissatisfaction of younger officers with both military appointments and budget cutbacks made the situation in Tanzania ripe for a coup attempt.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 173

Indeed, the Tanzanian army became a center of discontent with and even rebellion against Nyerere's regime. In early 1983 a group of officers stationed near the Uganda border mutinied; before they could articulate their grievances, their rebellion was put down, but it sent a chill through the ranks of the government. (Ungar 1985, 403) In addition, the same year of the hijacking, a coup plot was uncovered, though it was not military officers but prominent civilians who were thought to be the leaders. The government charged that the group planned to assassinate Julius Nyerere in January 1983 and take over the government. Twenty-nine people were arraigned on charges of treason. The alleged leader, a senior civil servant and former Commissioner of the Kagera Basin Development Organization, was accused of plotting with Oscar Kambona. Others charged were a university lecturer, a pilot for the national airline, several businessmen, and a prominent industrialist, as well as three army lieutenant-colonels and seventeen junior officers (Legum 1982-83, B287-8). The case was given major coverage in the domestic press, which merely highlighted the vulnerability of a government once considered inviolate. Long-time Tanzania watcher Sanford Ungar observed, "the real change after the war in Uganda was the end of illusions about and within Tanzania," characterizing the country as one of Africa's "fallen stars." He noted that many former champions of Tanzania now argued that "Nyerere could not be permitted indefinitely to substitute prestige for performance." (1985, 403 & 405) The Economist called Tanzania "never-never land" (20 October 1984, 33), and others declared Nyerere and his policies a "failure."

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 174

(Roehikepartain 1985) One sympathetic observer remarked that only the "extraordinary passivity" with which Tanzanians faced severe economic conditions prevented the sort of coup that would result elsewhere in Africa (Landor 1985, 242). Domestic critics of Julius Nyerere were now "more vociferous and were often to be found within his own close circle." These challenges to Nyerere's leadership were taken so seriously that the President felt the need to make a four-hour speech in defense of his leadership to the ruling party (CCM) conference in 1982. He said that, while some were charging that Tanzania was "finished", he continued to believe that recovery would come, rather defensively claiming, "We have good policies. We have good plans. We have good leadership." (Legum 1982-83, B283) This refrain no longer struck a chord among Tanzanians. When Nyerere made similar claims at a celebration of the anniversary of the Arusha declaration, "one third of the vast crowd summoned to hear his speech walked out." (Harman 1987,15) Dissent against the government, including Nyerere and his policies, grew. In 1982-83, the country's economy was in crisis, the worst since independence, and "political demoralization" was rampant. One observer noted, "The stability of the Tanzanian state today derives principally from its capacity to identify and repress dissent rather than from the consent of the peasant majority." (Lofchie 1985, 160) Though the country was in desperate need of outside assistance, the IMF's conditions of reform were inimical to Nyerere.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 175

In 1983-84, the government instituted its own form of structural adjustment, paced more slowly than the IMF would have it but considered politically palatable. When this did not prove sufficient, pressure for more fundamental changes in economic policy built. In 1985, Julius Nyerere left the presidency to Ali Hassan Mwinyi, a move generally interpreted to mean that he had accepted the necessity of economic reform, but that he was unwilling to preside over the dismantling of his socialist vision. Mwinyi quickly reached an accommodation with the IMF which led to far-reaching economic reform (Lofchie 1988, 209). Although it cannot be claimed with any authority, it is arguable that the war and the economic disruption that it wrought was the straw that broke the back of Nyerere's socialist path in Tanzania. What is clear is that the economic crisis to which the war contributed resulted in serious disillusionment with Nyerere and his policies and an eventual shift away from those policies. Tanzania's socialist development path is a thing of the past, and like so many other African nations, its new path is more the vision of the IMF than the country's leaders. Whether this shift in economic policy is for better or worse is a debate that is outside the purview of this dissertation.1 What is important in this study is the extent to which the war contributed to the crisis that forced the shift. It is ironic is that, while Tanzanians believed Amin to be a threat to their

1 See Lofchie (1988) for a positive review and Cheru (1989) for a critical review of these changes.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 176

sovereignty, their war to oust him may have resulted in a far greater challenge to their political and economic independence (Francis 1993).

International Relations Despite the fact that Tanzania's ouster of Idi Amin set a dangerous precedent in Africa, Tanzania's international relations were not significantly affected by the war. It is noteworthy that the UN Security Council never debated the invasion of Uganda and overthrow of Idi Amin, despite the fact that the action is generally considered to have been in violation of international law.2 This neglect is not satisfactorily explained by indifference, because the Tanzanian action was a case, unique in independent Africa, of a state overthrowing the government of another state. This was certainly not a matter of indifference. Neither can the UN's neglect be explained by acceptance of Nyerere's claim that the Tanzanian troops were merely finishing what Amin had started; this was widely considered a pretext. The UN silence had more to do with the distaste for Amin and the perception of Nyerere's intentions, which were thought lacking in expansionist ambitions. No territory was annexed nor reparations sought (Installed . . . 1979, 17). Though UN members chose to remain silent, despite Tanzania's violation of international law, Tanzania's supporters did

2 See, for example, Burrows (1979), Chatterjee (1981), and Hassan (1981).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 177

not come to its rescue. In June 1979, after the war was won, Tanzania requested $366 million in extraordinary assistance from nine of its donors to help it survive the immediate economic crisis caused by the war. Only four of the nine countries responded, and their combined assistance was small-scale (PER 1979:4, 7). Many Tanzanians were surprised and disappointed by this reaction. As one former official put it:

We took people at face value when they said Amin's a terror and must be got rid of. And when we found ourselves, without wishing to, involved in the job of doing it, we expected there'd be a little bit of money where the mouth was. (Identity Protected) Others were less optimistic. Benjamin Mkapa didn't expect any outside assistance beyond largely-moral support from countries Tanzania had assisted, Mozambique and Angola. Rather, his best hope was that all would refrain from aiding Amin or obstructing Tanzania. He reasoned:

Our best aid donors are the Scandinavian countries, and by definition they don't like violence and war. . . . And so, generally, while they didn't like Idi Amin and they thought his removal would be a good thing, I don’t think they liked the way we did it. I don't think they were prepared to underwrite it. (Mkapa 1993) While Tanzania's donors did largely ignore its immediate request for assistance, perhaps wishing to avoid any appearance of support for the overthrow of Amin, they did seem to sympathize with the country's economic plight and provide what assistance they could within the limitations of the aid system. As the U. S. Ambassador to Tanzania, James Spain, explained:

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 178

If they want to talk to us about building a paper mill four years from now, then we can help them. But we simply have no mechanism for paying for a war that's already been fought." (Quoted in Honey 1979) However, it was possible for the donors to increase aid flows not already in the "pipeline" and to provide food aid more immediately. Figure 23 shows Tanzania's official development assistance (ODA). In 1981, two years after the war (about the time it takes to initiate changes in aid flows), Tanzania's portion of all ODA to low- income countries did increase nearly 1 percent, suggesting the possibility of some sympathy with Tanzania's ouster of Amin. However, that figure began declining again in 1982, falling below prewar figures, as Tanzania and the World Bank ran into conflict over the terms of structural adjustment.

Figure 23: Official Development Assistance to Tanzania (as a percentage of all ODA to low-income countries, except China and India)

6

o —

Source: World Bank, World Development Report (New York: Oxford University Press, 1986) Table 22.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 179

Foreign assistance in the form of food aid can serve two purposes. It can provide food as humanitarian relief when there is an extraordinary event such as crop failure or dislocation that disrupts normal production. Alternately, food aid can be used to replace normal food imports, thereby saving a country foreign exchange expenditures. Figure 24 shows food aid (cereals) to Tanzania. A comparison of this chart with Figure 21, agricultural production, shows that, while per capita agricultural production did decline somewhat in the two years following the war, food aid increased dramatically, far more than would be explained by declining food production.

Figure 24: Food Aid to Tanzania

^ 350 | 300 o 250 © 200 z 150 £ 100 S 50 o 0 •sriflvor'-cooo— cMKj'srin r'-r-r-r'-r^r'-cocococococo Year

Source: FAO, Food Aid in Figures (Rome: United Nations Food and Agriculture Organization, 1984 & 1991).

Like the rise in ODA, this is hardly definitive evidence of sympathy with the war. Yet, it may be that donors were providing

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 180

Tanzania with food aid to save the country foreign exchange, suggesting a desire to provide a more immediate and politically palatable form of assistance than financial aid. It is possible that donors were indeed glad to be rid of Amin and were not too particular how it was done. However, international law was violated, and a dangerous precedent was established. Although donors may well have been sympathetic to Tanzania's economic crisis, they undoubtedly wished to show that sympathy in a way that did not appear to be supporting the ouster of Amin. While the UN was silent about the war, the OAU was not, though it was clearly ineffective at stopping it. The OAU's attempts at mediation failed, in part, because Nyerere insisted on official OAU condemnation of Amin's actions while the organization had no such sanction clause (Audifferen 1987, 289). Had such a clause existed, it is unclear whether enough unity would have existed to sanction Amin, given Libya's support of Amin's regime. However impotent the organization was to take action, OAU members were clearly disturbed by the precedent being set by Tanzania's overthrow of Amin. Tanzania's immediate neighbors were particularly disturbed. They were not so quick as donor countries to ascribe benign motives to President Nyerere. Benjamin Mkapa, wartime Minister of Foreign Affairs, discussed the tension and Tanzania's actions to counter it.

[Our neighbors] said they were not concerned, but in fact they were. And we were also extremely anxious to set [them] at ease. I was sent by my President to some of our neighbors whom we thought would be a little more anxious. I went to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 181

Malawi because there is a considerable exile community here, and [I] had a very good reception with President Banda in which we again explained our extremely limited objective, that certainly we had no territorial ambitions at all, that our neighbors should rest [easy]. So, I went to Malawi, I went to Ruanda and Burundi, you know the smaller countries. I also went to Kenya because there was a time during that war when there were fears that, after we got through to the eastern border, we might just keep going. So, we were very conscious about that, and we took deliberate steps to restate our intentions to respect borders and everything, that this was an exceptional situation. (Mkapa 1993) The OAU held its annual summit in Monrovia, Liberia, in July 1979, three months after a new government was installed in Kampala. In his opening speech, the outgoing OAU Chairman, President Numeiry of Sudan, offered a scathing attack of both Nyerere and Tanzania's actions. He recapped the OAU's efforts at mediation and blamed Nyerere's intransigence for its failure. Nyerere countered that Numeiry was blaming the victim. The mediation efforts failed, he said, because of the OAU's impotence to act against African aggressors (Audifferen 1987, 304-306). Next, Nigeria's Head of State General Obasanjo joined the attack against Nyerere,3 and Uganda's President Binaisa jumped to Nyerere's defense. The debate was getting nowhere, as African

3 This was not surprising. The Nigerian and Tanzanian governments had been at odds on several occasions. Nyerere had recognized the breakaway Biafra in its civil war against the Nigerian government in the 1960s. The Nigerian government was one of the first to recognize the Amin regime, and when Obasanjo's emissary came to Dar es Salaam hoping to dissuade Nyerere from invading Uganda, he was summarily dismissed.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 182

leaders were clearly divided on the issue. Thus, a group of moderate Francophone countries convinced the Assembly to shelve the debate, and Nigeria's attempt to censure Nyerere failed. The incoming OAU Chairman, President Tolbert of Liberia, addressed the Assembly and spoke of the more important issue of the OAU's impotence to stop aggression. Once again, the proposal for a continental peacekeeping force was brought to the table, and once again no action was taken (Audifferen 1987, 306-308). Thus, Nyerere's part in overthrowing Amin was effectively forgotten by the rest of Africa, and Tanzania's relations with the world were not noticeably affected. One foreign relationship was dramatically changed, however, and decidedly in Tanzania's favor—that with Uganda. Changing the leadership, and thus this relationship, was the primary point of the war, though Tanzanians were also motivated to help Ugandans rid themselves of Amin's tyranny. The removal of constant tension with the leader of a neighboring country and the continuing threat he posed was the single, unambiguous benefit of this war for Tanzania. While admitting that the war did not achieve stability for Uganda as was hoped, Benjamin Mkapa remarked, "For us, I think basically, we have achieved our political goal, we have been able to live cooperatively with all the regimes that followed." (Mkapa 1993) Thus, Tanzania's invasion of Uganda did succeed in removing a painful thorn in its side, significantly improving relations with its neighbor without seriously damaging its relations with other countries. Even Libya backed out of the war rather quietly, having been given a face-saving means of retreat. Qaddafi did not

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 183

participate in the attack on Nyerere at the 1979 OAU meeting, and Numeiry and Obasanjo let the matter drop after their initial tirade failed to find support of other members. The rest of the world seemed to be pleased to find Amin ousted and rather admiring of Tanzania's courage and commitment to take him out at great cost, though many were also clearly troubled by the precedent that was set.

Summary and Analysis This review of political consequences generally supports the propositions developed in Chapter 3. Power did shift from the civil society to the government as the government extended its control over an additional ten percent of the financial resources and five percent of the human resources that had previously been in the private sector. There was a "ratchet effect", expansion of the military, which grew to two and a half times its prewar size during the war and remained at double its prewar strength long after the war and occupation were ended. Faced with a severe economic crisis and a government that was claiming a larger portion of the shrinking economy, a significant portion of the population, primarily small agricultural producers, retaliated by disengaging, removing the fruits of its labor from the government's reach. Some peasants found alternative marketing channels, either smuggling or black marketeering. Most, however, merely retreated into a subsistence economy as the government's heightened extraction no longer made their market

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 184

labors worthwhile. Disengagement deprived the government of desperately needed revenue, and despite several increases in taxes, revenues remained flat as the tax base shrank. As economic conditions in Tanzania deteriorated, so did political support for the government. Many merely retreated from the government's reach, but others openly criticized the government's policies and challenged its authority. Criticism of Nyerere's leadership was more vocal and closer to the center of government. Two acts of open rebellion, a plane hijacking and a coup plot, revealed a weakening of Tanzanians' support for Nyerere and his government. They were the first such challenges in well over a decade and would have been considered quite unthinkable prior to the war. Ultimately, Nyerere and his policies were discredited, and his 1985 resignation marked a dramatic shift in government policy. Against these costs must be balanced the clear benefit of removing the constant tension and insecurity of living next to Idi Amin. While the hoped-for stability in Uganda proved elusive, the chaos was no longer the serious threat to Tanzania that it had been previously. Tanzanians were grateful for this renewed security and proud of their soldiers' victory, even if they were unprepared to pay the costs involved. This certainly gave the government's support an initial boost that undoubtedly cushioned the later assaults. Tanzania's international standing was minimally affected by the war. Tanzania's actions were interpreted as disproportionate to the threat from Amin and, as such, both a violation of international law and a dangerous precedent. Yet the delight at the demise of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 185

Amin's rule and Tanzania's lack of territorial ambitions combined to mollify international criticism. Even vocal critics within the OAU let the matter drop once the war was over. While the country's donors did not offer direct support for the war, they did expand official development assistance, if briefly and marginally, and provide significant amounts of food aid.. While there is no certainty, the lack of other explanations suggests the possibility that this aid increase reflects a desire on the part of donors to help defray war costs without publicly acknowledging support for the war. When Tanzania's leaders determined to oust Idi Amin, they hoped their actions would not seriously affect their relations with other states. As they hoped, what international criticism did emerge was limited to the OAU and was short-lived. They may also have anticipated that the war would require an extension of the government's reach and an expansion of the military, though their positions would dictate that these changes would be perceived as beneficial. Four of the seven leaders that gathered to decide how to respond to the invasion of the Kagera4 were members of the security forces. All were senior party members who had established a policy of socialism and the key role of the military in the country's development. Thus, they would consider an expansion of the government and the military to be not only necessary for the country's defense but also its development.

4 See Kiwanuka (1985), page 79.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 186

While these leaders read the international political climate quite accurately, they badly misjudged the domestic political situation. The war did not actually benefit most Tanzanians and came at a cost they were unwilling to pay. The sacrifices required by the war and other economic shocks brought economic crisis, widespread political disillusionment with government policies, and open dissent against Julius Nyerere. What traps might have been operating to distort the leaders' perceptions of war's consequences for the Tanzanian State? Certainly, a Collective Trap was at work. Sanford Ungar opined that the war "did a great deal for the morale and self-respect of the Tanzanian elite." (1985, 403. Emphasis mine.) While leaders, particularly the military, benefited from the war and particularly from the government's (including the military's) increased power, the general populace paid the price. A sort of myth had been perpetuated by the Arusha Declaration that the government and the people were one and the same. But the government's increasing control of the nation's resources, as well as civil servants' corruption and the military thievery, had replaced the myth with a tug-of-war over resources between the state and ordinary Tanzanians. The people lost their trust that the state would act in their interest. What the leaders truly failed to anticipate was the disengagement that resulted and the constraints on government that would result. The leaders had enjoined the populace to relocate into villages during the early 1970s. Many peasants resisted and were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 187

eventually coerced, but villagization was completed and the country returned to normal. While the Ujamaa villagization was disruptive, the economy recovered fairly quickly, and the leaders had reason to believe that their detractors were won over again. They may well have anticipated the same would be true of the war, falsely believing that a war won so decisively and quickly would prove no more disruptive than villagization. This was a serious miscalculation that may have been the result of willing ignorance (an Ignorance Trap), since to believe otherwise went against their political philosophy. It is undoubtedly true that the degree to which the populace abandoned the government and rejected its leaders was the result of a combination of events in addition to the war. The serious dislocation of villagization may well have made peasants weary of further sacrifice, no matter what point of national honor was at stake. The oil price hikes, declining commodity prices, and the drought made the hardships that much greater. These factors joined with the war costs to seriously threaten the survival of many. But their proximity in time to the war obscured these other factors, and the government became a victim of its own propaganda that the sacrifices would only last eighteen months. The people saw a government continuing to blame a war that the peasants had been told was behind them. Thus the government's explanations for the hard times were seen as excuses, and the peasants withdrew, depriving the government of both credibility and resources.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 188

Another Ignorance Trap, this time a result of genuine lack of understanding, may have prevented leaders from anticipating the discontent that festered in the demobilized militia. These militia members had little desire to return to the hard life of a village peasant, made even harder by the war, after their experience of power and material plenty in Uganda. Tanzania's leaders, as members of the country's elite, had little understanding of the life of a village peasant; they couldn't foresee the resistance that would grow out of this change. There were, of course, social as well as economic and political consequences of this war. Some resulted from the economic crisis, while others resulted from the experience of the war's participants. These social consequences are the focus of the next chapter.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 7

SECONDARY SOCIAL CONSEQUENCES

A great war leaves the country with three armies—an army of cripples, an army of mourners, and an army of thieves. German Proverb1

The previous two chapters reviewed the war's secondary economic and political consequences. This chapter completes the assessment of the longer-term consequences by reviewing and evaluating the war's outcomes for the society. The chapter includes four areas of impact on individuals and society—disability, crime, social services, and AIDS. In addition, there is a concluding section which summarizes these findings, analyzes the extent to which the outcomes were anticipated by decision-makers, and discuss the traps which may have inhibited their foresight.

Disability The Tanzanian government has confirmed that 693 soldiers suffered serious and permanent physical injury during the war and

1 Browns, Ralph Emerson, ed. 1965. New Dictionary of Thoughts. 714. Standard Book Company.

189

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 190

subsequent occupation. Five years after the war ended, the Deputy Minister for Defense and National Service, Eslie Mwakyambiki, acknowledged in the national assembly that none of these trauma survivors had been compensated or aided in rehabilitation by the government, as they had been promised, though he assured the assembly that this would be done (Daily News [Dar es Salaam], 7 July 1984). In early 1993, I was told by wounded veterans and other, however, that no compensation was forthcoming. I encountered some of these disabled veterans in Dar es Salaam where they had joined the ranks of beggars patrolling the vicinity of the tourist hotels. One such man, having lost his legs, maneuvered on a wooden, wheeled platform through the use of great arm strength, begging outside the Kilimanjaro Hotel. With apparent bitterness, he told me through an interpreter that he had received nothing from the government, despite many public pronouncements to the contrary. A grass-roots political organizer on the island of Pemba reported that these disabled veterans, as well as their families, friends, and neighbors, were easy recruits for the opposition parties because they were very bitter towards a government they feel betrayed them. Because no psychological services were provided for veterans, there are no figures on the number of trauma survivors who suffered psychological disabilities. Yet there are individual stories. I learned of one man, a well-educated and talented civilian attached to the army for the course of the war, who returned quite changed, his ability to be productive much diminished. Although I

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 191

was unable to speak to him personally (his physical health is also now in jeopardy), I met with his daughter and two former colleagues. They told me that, upon his return from the war, this previously very capable man was unable to resume his former position because he was deeply depressed and had difficulty focusing on any task for long. After a long period of recuperation, he attempted to open a small service business but continued to be incapacitated. Family members quietly took over his duties. All three interviewees attributed his disability to his experience in war. This man and the physically disabled veterans I encountered are illustrative of the "army of cripples" this chapter's epilogue notes as a consequence of war.

Crime Violence There were also trauma survivors who manifested stress through anti-social behavior, including acts of violence. One veteran admitted what journalists had reported at the time,2 that Tanzania's troops were involved in crime and corruption both during the occupation of Uganda and upon their return to Tanzania.

Let me tell you one thing, the people who came back from Uganda were very different people. They were changed in the sense that the indiscriminate acquisition tendencies developed in such a manner that Tanzania had never experienced. It was not all peaceful at that time.... They [Tanzanian troops] did commit a bit of crimes. . . . But, the lesson we learned is that war is very

2 See Avirgan and Honey (1982) and Ungar (1979).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 192

costly, both in terms of resources and in terms of life. To a certain extent, vou end u p with a crippled mentality. . . . [Tanzanian troops] get exposed for about two years to this very confusing situation, very chaotic situation. How can you expect these people to get away untainted? (Identity Protected, Emphasis mine) Some of the theft that occurred was well coordinated and took place quite openly. One businessman remembered an army vehicle arriving at his neighbor's house in Dar es Salaam after the war. He watched as number of armed soldiers in uniform held the neighbor at gun point while the valuables in the house were loaded into the truck. The neighbor, he said, made no attempt to file charges, fearing that the police were collaborating with, or at least willfully ignoring, these rampaging soldiers. Benjamin Mkapa, wartime Minister of Foreign Affairs, admitted individual incidents of crime perpetrated by returned troops, but he denied that there was organized looting. Explaining how soldiers might become criminals, he said:

I remember my father saying to me, "These young people you have recruited urgently, trained them, they've fought a war, they've shed blood, they've come back here, they need to have some medicine to have a bath in, in order to excapate themselves of all these things. Otherwise they'll be crazy in their lives, they'll be crazy in their behavior, and stuff like that." (Mkapa 1993) While acknowledging the wisdom of his father's words, Minister Mkapa admitted that there had been no such attempt on the part of the government to deal with this aspect of the demobilization. Despite official disclaimers, there is evidence of organized extortion by returned veterans. The people's militia comprised the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 193

largest group of Tanzania's troops in Uganda. In addition to being a reserve army for duty such as the war and occupation in Uganda, the militia is mandated as an instrument of law and order in Tanzania. Militia members have the same powers of arrest as do the police. After the war, there was a scarcity of common consumer items, particularly 1981-82, due to import restrictions. During this period, members of the militia became notorious for harassing the public by indiscriminate searches, often appropriating what they found (Shaidi 1989, 256-7). Recall the example in the previous chapter of the businessman whose tube of foreign toothpaste was appropriated at the police station. This criminal behavior, combined with the flood of weapons into the country, resulted in a marked increase in violence (Avirgan and Honey 1982, 236). This can best be seen in the data on homicides committed in Tanzania through the 70s and 80s, as shown in Figure 25. There was a general decline in homicides in Tanzania through the 1970s, interrupted by a jump in 1975 & 76 that was presumably a consequence of the disruption of villagization. There was also a general, though less dramatic, downward trend .in the average homicide rates of three of Tanzania's non-warring neighbors during both the 70s and the 80s.3 While we are interested here in

3 While cross-national data on crime is problematic, homicide figures tend to be more reliable than other types of crime statistics. Moreover, in a comparison of four different data sets, Bennett found that "the data sets provide consistent results when they are used to assess the direction of change in crime over time and across nations." (Bennett and Lynch 1990, 176) The general decline in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 194

shifts in patterns, it is noteworthy that Tanzania's homicide rate is generally higher than its neighbors, surprising considering its peaceful history. One researcher suggests that this may be due to the inclusion of suicides in murder statistics in Tanzania (Owomero 1988). Despite these problems with the data, shifts are apparent and informative. Tanzania's downward pattern is broken in the mid- 1970s, but by 1978 the homicide rate returned to levels that might have been anticipated, based on the pattern of the early 1970s, . The war years marked the beginning an abrupt and sizable increase in homicides in Tanzania, while its non-warring neighbors continued to experience declines. In 1980, after the war and of the return of 25,000 demobilized militia and 10,000 soldiers of regular army, the homicide rate increased after four years of steady decline. It continued upward during 1981, when the last Tanzanian troops returned, and showed a very sharp increase in 1982. These last soldiers to return had remained for the full two years of occupation in an increasingly chaotic Uganda and seem to have brought the violence back to Tanzania. The homicide rate remained high for another five years, finally subsiding slightly in 1987. Significantly, 1987 is also the year that the economy seemed to return to a normal growth pattern (see Chapter 5).

homicide that these four nations exhibited is consistent with the pattern observed among developing nations with comparatively low growth rates, a category which includes the four nations noted here. Generally, however, in Bennett's findings, development showed a positive relationship with theft (Bennett 1991, 348).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 195

Figure 25: Homicide Rate

11

,*#- Villagization

Average (Kenya/Malawi /Zambia)

Tanzania

■ 4 4-1 I I-I I I I l-l I I I I I I ) I 70 72 74 76 78 80 82 84 86

Year

Source: Crime (1970-84) - Richard R. Bennett, Correlates of Crime Data Set; Crime (1985-87) - Interpol, International Crime Statistics (Lyons, France: ICPO- Interpol); Population - World Bank, International Economic Department, World Tables of Economic and Social Indicators. 1950-1987 [Computer file]. (Washington DC: World Bank [producer], 1988. Ann Arbor, Ml: Inter-university Consortium for Political and Social Research [distributor], 1990).

It should not be surprising that the increase in Tanzania's homicide rate was greater after the occupation than after the war. Archer and Gartner (1984) studied the connection between war and crime (discussed in Chapter 3) and found that soldiers' exposure to sanctioned killing brought about an increase in the homicide rate at home once the soldiers were demobilized. Chaos in Uganda worsened considerably during the occupation (as did the death rate of Tanzania's soldiers), and those Tanzanian soldiers who remained for the full two years had the greatest exposure to violence. Thus, their

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 196

return in 1981 could be expected to introduce the greatest increase in homicide rates. In addition, Figure 25 provides another informative comparison of the relative impact of villagization and war. It reveals a repetition of behavior in both events; homicide rates increased markedly after both villagization and the war and occupation. In both cases, there was lag time of approximately one year from the height of each event (relocation, demobilization of militia, and return of occupiers) and the occurrence of a noticeable increase in rates of homicide. This repetition of patterns between these two events, both highly disruptive to the nation, is something that has been seen throughout this study. It should be noted, however, that the rate of increase in homicides, though not the actual rate, was greater during the war and occupation than it was during the forced relocation of millions of peasants. Moreover, the effect on homicide rates lasted far longer after the war than after villagization. Although the relocation impact reversed itself almost immediately, the war's impact lingered for several years. The Tanzanian government had been unable to pay these soldiers or even to provide them with necessary supplies during much of the time of occupation. Consequently, they had grown accustomed to taking what they wanted at the point of a gun. This behavior, as well as other abuses of power, resulted in deaths of Tanzania's occupying soldiers as well as Ugandan citizens (PER. 1980:3, 4 and 1981:3, 4; Nabudere 1993).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 197

Not surprisingly, this behavior continued once the soldiers returned to Tanzania, often with deadly consequences.

Theft bv Public Officials Like homicide rates, reported incidents of theft by public officials significantly increased during the years of mobilization, war, and occupation (See Figure 26). The economic crisis the country faced after the war brought with it a drastic fall in the purchasing power of civil servants' salaries which, aggravated by a shortage of essential commodities, created a black market for these items at greatly distorted prices. As the living standards of public servants declined, public institutions came to be regarded as "legitimate areas for plunder." (Shaidi 1989, 263) As one high- level, retired civil servant put it:

So, you started getting shortages, and also you started getting corruption which we hadn't had until that time. 1978 was definitely the turning point on corruption in Tanzania, without any question. I don't know how it started, one never knows how these things start. And I'm not saying it never existed before, I'm saying it wasn't a problem, because people were watching. It could have been somebody saying, "Don't commandeer my lorry, I'll give you this." I don't know how it started, but I do know you can time the beginning of corruption being a problem from 1978.. .. And it could have only been the war, there's nothing else it could have been, because after that you stared getting shortages, you started getting constraint on money. (Identity Protected, Emphasis mine) As the government's credibility declined, the populace occasionally took action themselves:

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 198

[l]n May 1982, four villages in the Maswa district of the Sonyanga Region decided to boycott implementing any more agricultural projects until there had been an proper accounting of the returns on their products sold for the previous year. Their suspicion was that payments due to them had been squandered on high living by senior officials. (Legum 1982-3, B280)

Figure 26: Incidents of Theft of Public Funds by Public Servants

2000

1800 4» at I I «>«»1600 ft« o 1400 «> a E 1200 llll

1000 n i l 1111 so i"- co cx» o CM KJ r- r- r" r- co co CO CO os os os OS OS

Year

Source: Leonard P. Shaidi, Crime, Justice and politics in contemporary Tanzania: State power in an underdeveloped social formation, International Journal of the Sociology of Law 17 (1989): 263.

Incidents such as these helped spur the government to act, and Prime Minister Edward Sokoine announced a campaign against "economic sabotage" 18 March 1983.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 199

[It was] an effort to convince the IMF and other outside donors and lenders that the government could gain effective control over the economy. In its initial phases, more than a thousand people were detained as "economic saboteurs," and nearly $2 million in proceeds from smuggling, illegal currency dealings, and other rackets was seized by the police. (Ungar 1985, 405) This campaign may explain the sharp drop in incidents of theft in 1983 that can be seen in Figure 26 above. Although the incidents of public theft increased after both the war and occupation, the value (as opposed to the number) of these thefts did not grow significantly until 1981 (See Table 4). That year, the amount of public theft more than doubled in real terms (constant 1985 TShs), from TShs177.2 million to TShs 436.7 million. The amount stolen per incidence of theft also nearly doubled, from TShs 121,000 to TShs 232,000, and the amount recovered (already pitifully small) dropped by more than half, from TShs. 9.6 million to TShs. 3.8 million, despite a doubling of the amount stolen. This difference in amount of theft between the end of the war (1979) and the end of occupation (1981) is similar to that seen in the previous section on homicides. It suggests a difference between the experience of soldiers in war and that of soldiers in occupation. The occupation of Uganda was more chaotic and deadly for Tanzania's soldiers than the war and provided more opportunities for abuse of power. The soldiers returning in 1981 had remained a full two years in increasingly chaotic circumstances in which their discipline had seriously declined. Their experience may well have encouraged

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 200 corrupt behavior, or as one anonymous interviewee said above, a "crippled mentality."

Table 4 Theft of Public Funds by Public Servants (in constant 1985 TShs)

Amount Amount stolen Am ount Year sto len per incident recovered (millions) (thousands) (millions) 1976 113.0 96 1.3 1977 125.4 103 2.7 1978 155.0 104 2.9 1979 107.1 59 10.4 1980 177.2 121 9.6 1981 436.7 232 3.8 1982 239.7 143 6.3 1983 249.8 188 4.0 Source: Leonard P. Shaidi, Crime, Justice and politics in contemporary Tanzania: State power in an underdeveloped social formation, International Journal of the Sociology of Law 17 (1989): 263. Conversion to constant values by author, based on GDP Deflator from IMF, International Financial Statistics 1990 Yearbook (Washington DC: international Monetary Fund, 1991), 709.

Although one can only surmise, a plausible scenario can be developed connecting the experience of war and occupation with the increase in theft of public funds. Upon their return, these occupying soldiers were brought into the civil service in unprecedented numbers, as noted in the previous chapter. They had experienced increasing violence, lawlessness, and corruption during their two

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 201

years in Uganda, and they returned to a devastated economy and austere conditions. Those that were given government positions had new opportunities to act for their personal gain. One businessman even suggested that they often felt it was their due. The soldiers had sacrificed for their country, only to be denied the standard of living to which they had grown accustomed, due to their power as occupiers. Thus, some of these veterans may well have taken advantage of the opportunity for gain where they found it, in the public coffers to which their newfound positions within the government gave them access. Interestingly, while Tanzanians blame the chaos in Uganda for teaching their soldiers criminal behavior, Ugandans blame Tanzanians for teaching them looting. A Tanzanian businessman who had spent several of the post-war years in Ugandan explained it this way:

Quite a lot of Ugandans have a resentment, not to the Tanzanian in general, but to the Tanzanian army. Basically what they say, and it was a sentiment I heard over and over, 'You guys taught our guys how to loot.' Before the war, any trouble Uganda had, there was no such thing as looting. Even Amin's soldiers were not thought of as looters. When Obote was overthrown [the first time], there was no widespread looting. But when Obote was overthrown by Tito Okello, you can't believe what happened. I still can't believe it myself. . . Everything was looted. . . . When I asked how this could happen, "You all live together, how can you do this to one another?" they would say, "Well, blame it on the Tanzanians. When we had this war over Amin, the Tanzanian soldiers felt they had nothing to lose. It wasn't their country, so they'd take back anything and everything they could. They felt they were liberating this country, yet they

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 202

were not getting anything out of it, a small salary. They saw that the Ugandans were wealthier than Tanzanians, so they took what they wanted." Ugandans believed this was the logic of the Tanzanian soldiers, and they [Tanzanian soldiers] then taught the Ugandans how to loot. (Protected Identity) Perhaps a more appropriate analysis is that war—and the lawlessness and economic deprivation that invariably accompanies it—can corrupt those who experience it, victors and vanquished alike. Thus, war leaves in its wake, as the German proverb quoted at the beginning of the chapter suggests, "an army of thieves" as well as "an army of cripples." Fortunately, the effects of war on crime that were surveyed in this section do not seem to be long lasting, as increases in homicides and theft subsided during the period analyzed.

Social Services Social Service Spending Not surprisingly, the war-related expenditures displaced spending on needed social services. There was no outside assistance forthcoming to finance the war, so the war had to be prosecuted at the expense of other activities. Defense spending as a portion of all government spending jumped from 12 percent in 1976 and 1977 to 15 percent in 1978 and to 24 percent in 1979 (Economic Survey). This reallocation affected all areas of the budget, but the concern in this section is with the impact of that spending shift on the delivery of social services. Figure 27 shows the portion of the central government's expenditures that went to education and health care.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 203

In 1979, there was a decrease of nearly one-quarter of social service spending.

Figure 27: Education and Health Care Spending as a Percentage of all Central Government Expenditure

22 20

~ 18 16 S m 14 1,2 @ Health ■ Education ? 10 8 u Vu 6 o. 4 2 0 r-r-r-r-r-r^r-r^r-coooooco Budget Year

Source: United Republic of Tanzania, Economic Survey (Dar es Salaam: Ministry of Planning, various years). Note: Years refer to end of budget year.

Education spending as a portion of the central budget reached is peak in 1972. For the years 1972-75, the critical years of the Ujamaa villagization programs, this percentage dropped nearly three points but recovered again 1976-78. In 1979, the percentage dropped three points in that year alone. This time, the recovery of the education budget was much slower; it had not regained its

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 204

prewar peak by 1983. As was the case with homicide rates in the previous section, it is noteworthy that the war's impact on the education budget was even greater than, the impact from the villagization program. Health spending in the prewar years, by comparison, was more stable than spending for education, and the impact of the war's budget shift on health spending was less severe. Prior to the war, health spending as a percentage of the government's budget slowly increased to a high of over 7 percent, but this fell to 5 percent the year of the war. Health spending, like education spending, did not returned to its prewar peak. In fact, its priority continued to decline slightly. Although this funding reallocation during a war is not surprising, our concern here is its impact on the delivery of social services.

Impact on Education During the 1960s and 1970s, Tanzanians achieved success in education that was both spectacular and unique in Africa. In line with the Arusha Declaration of 1967, a program of Education for Self Reliance was implemented, which sought to make primary education relevant to the needs of the rural peasant rather than preparation for higher education to which very few could realistically aspire. Enrollment in primary schools expanded rapidly, particularly in the early 1970s, and in 1977, the government made a commitment to universal primary education with the goal of providing primary education to every Tanzanian.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 205

The economic dislocation of the war and the economic crisis which followed would prove to be, however, the beginning of the decline of this education miracle. The segment of the school population that was most affected was first-year primary students. First-year primary enrollment had increased steadily through the early and mid-1970s, reaching a peak in 1978. As seen in Figure 28, in 1979, this first year enrollment began a two-year decline that is as dramatic as the previous years' increases. Despite a 30 percent increase in population, it would take a decade, until 1989, for the first year primary enrollment to reach prewar levels.

Figure 28: First Year Enrollment in Primary Education

600

550 fltl I «>e 500 e P 450 ■ i n n n m u n 11 in SO I"- oo o —

This decline in new enrollment in primary schools was matched by an even more precipitous decline in new enrollment in teacher training colleges (See Figure 29). Enrollment in teacher

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 206

training took a leap in the early 1970s to meet the demand for new schools brought into being by resettlements under the Ujamaa program. However, in 1980, the number of first year teacher trainees dropped precipitously, and it again declined noticeably in 1981. By 1988, the last year figures are available, first-year enrollment in teacher training had not regained its prewar level.

Figure 29: First Year Enrollment in Teacher Training Colleges

8000

6000 •>e E £ 4000 m i ue W 2000 in i m

kO p- co O *- CM to p- p* p- P- CO CO CD CO Year

Source: Bureau of Statistics, Selected Statistical Series: 1951-1988 (Dar es Salaam: President's Office - Planning Commission, 1991), 104.

Overall school enrollment, including primary and secondary education, as well as the number of schools and teachers, expanded after the war. However, this expansion was far slower and, indeed, was not able to keep up with the expansion of the population. Thus, the shift of funds away from education that was necessitated by the war considerably slowed the spread of educational facilities,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 207

particularly primary schools, that had occurred in the 1970s. Budget priorities necessitated by war placed education out of reach for an increasing number of Tanzanians, because new school places were no longer created as rapidly as the population was expanding, despite the government's clear commitment universal primary education.

Impact on Health Care Delivery As with education, Tanzanians made impressive early gains in health care delivery, based on two policy initiatives of the 1960s: (1) the Titmuss Report4 of 1964 advocated an expansion of rural health care and an emphasis on preventative, rather than curative, services, and (2) the Arusha Declaration of 1967 re-emphasized the government's commitment to rural development. It was not until the early 1970s, however, that rural health care delivery claimed a major portion of health care spending and villages actually gained access to health facilities to any significant extent. By 1979, one study found that ninety-two percent of the population lived within ten kilometers of a health facility, and seventy percent were within five kilometers. Forty-five percent had a health facility in their own village (Heggenhougen 1986). Also important to the nation's health, by 1977, over half of the population had piped water (Yeager 1989, 77). These were impressive advances for a poor developing country like Tanzania.

4 Titmuss, R. M. et al, The Health Services of Tanganyika—A Report to the Government (London: Pitman, 1964).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 208

Unfortunately, changes in spending priorities necessitated by the war and the economic crisis which followed meant the end of this growth in health care. Take the case of health centers, the health delivery unit designed for the division level, a large administrative unit of which there are twenty in Tanzania. As Figure 30 makes clear, the spread of health centers abruptly slowed in 1979 and halted in 1980. The number of health centers remained nearly constant for seven years, while the population and, presumably, health care needs increased by 21 percent.

Figure 30: Number of Division Health Centers

300

r^r^r-r^r-®floooooOCO^’ ttOOCM^’ sO Year

Source: Bureau of Statistics, Selected Statistical Series: 1951-1988 (Dar es Salaam: President's Office - Planning Commission, 1991), 114.

This pattern was repeated at other administrative levels. At the ward level, a much smaller unit comprising an average of four villages and providing health care through dispensaries (Heggenhougen 1986, 311), growth of care was also halted (See

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 209

Figure 31). The number of dispensaries grew rapidly during the 1970s until 1979 , then halted for several years.

Figure 31: Number of Ward Dispensaries

3000

S „ 2 5 0 0 . 9 £ Z 2000 - «n e s 1500 k . 965407 0) & io o o o 500 IIIIII III II 0 r-r-r-i^r-cocDaDoo Year Source: Bureau of Statistics, Selected Statistical Series: 1951-1988 (Dar es Salaam: President's Office - Planning Commission, 1991), 114.

Troublesome as this patterns appears, the reality was actually much bleaker. Though health facilities indeed continued to be constructed after the war, many lacked equipment, medicine and other supplies, and adequately trained personnel (Heggenhougen 1986, 311). Construction funds had been in the development budget, or the "pipeline", for some time and would have been difficult to halt. Funds to provide the human and material resources required to make these facilities operational, however, came from recurrent expenditure and could be more readily diverted to other priorities. As with the public corruption that was discussed in the previous section, the war was the turning point in education and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 210

health care delivery in Tanzania. Yet, it must be acknowledged that these changes were exacerbated by the other economic shocks that hit Tanzania in the late 1970s and early 1980s. The war and its economic dislocation may have been the catalyst for the initial decline of social services, but other shocks to the nation’s economy intensified the impact and greatly extended the length of time it would take to recover.

Acquired Immunodeficiency Syndrome (AIDS) When asked about the consequences of the war with Uganda, most interviewees mentioned AIDS. A Tanzanian clerk at the American Embassy voiced an opinion that I would hear repeatedly, "The war cost us AIDS.” During the decade of the 1980s, AIDS spread throughout eastern and central Africa. Although this spread was mirrored in many parts of the world, there is compelling evidence that the war contributed to the rate of infection. The first reported case of AIDS in Tanzania was in 1984 in the Kagera Region, along the Uganda border. However, AIDS experts believe that "extensive spread of the human immunodeficiency virus (HIV) probably began in the late 1970s and early 1980s" (World Bank 1992, 10), precisely the time of the mass movement of Tanzanian troops across the Tanzanian-Ugandan border. Following identification of the first AIDS case, the number of reported cases in Tanzania escalated rapidly.

As of end 1990 [sic], a cumulative total of 21,175 AIDS cases had been reported throughout the country. The number of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 211

officially reported cases is thought to understate considerably the true number. . . . The NACP [National AIDS Control Program] estimates that the true number of AIDS cases from the start of the epidemic through 1990 is more than 100,000 (or about 5 times the number of reported cases). (World Bank 1992, 11) As in the rest of Africa, AIDS in Tanzania is transmitted primarily through heterosexual intercourse, which accounts for 75-80 percent of AIDS infection (World Bank 1992, 20). For the purposes of assessing the war's impact, it is instructive to note that the rate of HIV seroprevalence (rate of infection) differs widely from region to region within the country. The Kagera Region on the Ugandan border has the highest rate of infection for urban areas in Tanzania—over 10 percent.5 The region's capital of Bukoba has an estimated seroprevalence rate of 31 percent for adults in the age group 25-34 years. Three other areas have significantly higher rates of infection than the rest of the country: (1) Mwanza, on Lake Victoria and adjacent to Kagera Region, (2) Mbeya, on the main transportation route from Dar es Salaam to landlocked Zambia, and (3) the city of Dar es Salaam. Each of these areas has an extremely high rate of urban infection (7-10 percent) compared to the rest of the country (1.4-5.3 percent). (World Bank 1992, 16) What is common to these four areas of high

5 "An approximate rule of thumb for assessing the impact of a given, constant level of HIV infection on the mortality rate of sexually active adults in Africa . . .[says that] a seroprevalence rate of 10 percent will . . . double the adult mortality rate, and roughly half of all adult deaths will be due to AIDS. A seroprevalence rate of 20 percent will triple adult mortality, and two-thirds of adult deaths will be due to AIDS." (Ainsworth and Rwegarulira 1992, 2)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 212

seroprevalence is that they are ail centers of cross-border movement of people and goods. The AIDS crisis in Uganda is more critical yet. Already in 1988, Ugandan health ministers were estimating a national HIV seroprevalence rate of 10 percent! (Christianity Today. 8 April 1988, 37) There is no way to firmly establish that the war contributed to the spread of AIDS. A medical director of the World Bank's AIDS Program would only suggest that a scenario of the war facilitating transmission was "not inconsistent with the evidence." (Malangalila 1993) A former member of the USAID mission to Tanzania was far more sure when he advised, "You would be missing what is perhaps the single most important consequence of that war if you didn't look at AIDS." Certainly common sense would suggest that the movement of 45,000 soldiers into an area now known to be one of the world's highest concentration of HIV infection for a period of several months to two and a half years would put many of them at high risk of infection. Their return to Tanzania would then hasten the spread of the disease throughout the country, particularly in high transit areas. If this widely held perception that the war significantly contributed to AIDS were accurate, what might be the impact? The World Bank has done extensive estimates of Tanzania's AIDS epidemic and its potential consequences.

The number infected will reach 5.8 to 17.4 percent of the population by the year 2010 . . . . The work force will become younger (average age 29 instead of 31 in 2010) and less experienced, and will have less education and training. . . . The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 213

economy will be adversely affected .... GDP will grow more slowly; by 2010 it could be 14 to 24 percent lower than it would have been if there had been no AIDS epidemic. (World Bank 1992, i) Given a projected population of nearly 50 million in the year 2010, there could be 2.9-8.7 million Tanzanian AIDS victims . For each adult AIDS victim, the country would bear the burden of direct costs (cost of treatment) and indirect costs (value of healthy years of productive life lost to the society), estimated by the World Bank at $2,462 to $5,316 (1985 US dollars) (Over et al 1988). It can be reasonably assumed that the war made some contribution to the spread of AIDS. If the war were assumed to have increased AIDS transmission by 5 percent, this would translate to a social cost of between $7 and $46 billion!6

Summary and Analysis This overview of the evidence of social change supports some, but not all, of the propositions developed in Chapter 3. There were, as anticipated, trauma survivors with disabilities, both physical and psychological. Though the government acknowledged the problem and its debt to the wounded soldiers, the lack of compensation for these veterans left them, in many cases, destitute and bitter.

6 Five percent assumed war-related infection of an infected population estimated at (minimum) 5.8% of 50 million = 2.9 million to (maximum) 17.4% of 50 million = 8.7 million. 5% of these figures at a social cost of $2,462 to $5,316 per victim = $7 to $46 billion.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 214

As anticipated, crime increased, as indicated by rising homicide rates and theft by public officials. The evidence suggests three reasons for these increases. First, there was a large influx of weapons from Uganda after the war. Secondly, the Tanzanian soldiers in occupation in Uganda were often left without support and allowed to fend for themselves. With a gun in their hands and the authority of occupiers, they learned lawless behavior which continued once they returned to Tanzania. Thirdly, returning veterans were brought into government service in record numbers, providing them increased access to public resources. Increased defense expenditures during the war and occupation displaced development programs in health and education. In education, this resulted in limited new access to education, as first-year primary school enrollment dropped for the first time since independence. First-year teacher training enrollment also dropped. Unfortunately, these proved to be enduring trends. The nation that had once made great strides towards universal primary education found that, due to budget constraints imposed by war and the economic crisis that followed, education eluded an increasing number of its citizens. In health care, displacement was less severe, but the impact was also devastating. While the growth of new facilities was halted for several years, there was no decline. However, existing facilities often suffered from lack of supplies, thus limiting their effectiveness. One observer lamented:

Tanzania's medical system . . . has all but ceased to function, because of the unavailability of critically necessary medical

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 215

equipment and medicines. And in major regions, the nation's educational system, hampered by shortages of basic school supplies, functions in name only. (Lofchie 1985, 160) The nation also experienced an expansion of employment from the war, largely in the public sector. This was anticipated as a social consequence, though it was discussed in the previous chapter as a consequence of the government's concentration of resources and was therefore not elaborated in this chapter. Two other consequences predicted from a review of the literature were: (1) a decline or destabilization of social institutions, and (2) a decline in social cohesion. No evidence was found to either support or refute the first contention beyond the impact on education and health care institutions already mentioned. It is only possible to surmise that, as most modern village social institutions were the product of government development programs whose funds were displaced by defense expenditures, these programs and the institutions they supported would have suffered from a lack of resources. This is, however, merely reasoned conjecture. As to a decline in social cohesion, the only evidence that was found was an increase in some types of crime, one indication of a breakdown in cohesion. One potentially significant consequence clearly could not have been anticipated—AIDS. Evidence suggests that the movement of troops into Uganda increased the rate of spread of AIDS. This troop movement was tragically coincident with what experts identify as the outbreak of HIV infection in the region and took place

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 216

in precisely the location of highest infection. The circumstances could not have been worse. While it is impossible to know what impact the war might have had on the spread of AIDS, an estimate of a 5 percent increase could translate into a social cost of billions of dollars. AIDS could be the most devastating consequence of this war. Several of the consequences discussed here, with the exception of AIDS, were anticipated in the literature. Were these consequences also fully anticipated by Tanzanian decision-makers? The transmission of AIDS was clearly unknowable. The disease had not yet been identified at the time of the war, so decision-makers could not have anticipated this outcome. Other social consequences considered in this chapter, however, were foreseeable but were not a deterrent to the initiation of war. Why not? The Time-Delay Trap meant that the benefits of war (revenge, honor, and security from future invasion) were felt rather quickly. Certainly, these were celebrated throughout Tanzania as the first troops returned. Yet, the costs that were felt at the same time were only a small portion of the total cost. The long-term impact of the war and two-year occupation, terminated rather ignobly, was a debt that had not yet come due. By the time the last troops returned home in 1981, the costs were causing suffering, and there was far less to celebrate. To the extent that they considered costs, Tanzania's leaders undoubtedly foresaw such losses as disability and dislocation of human and economic resources. These are inevitable outcomes of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 217

war, and the decision-makers were surely prepared for the country to bear some burden. However, the Ignorance Trap, due to the lack of experience with war, may well have prevented them from accurately assessing the extent of the loss. Moreover, the "Midas effect" of their perceptions of Tanzanian social discipline and unity, reinforced by early support of the war and an orderly mobilization, may have prevented them from anticipating the impact of the war on crime. Their misperception led them into a Diminishing- Returns/Spiraling Costs Trap in which a breakdown in discipline led to an increase in crime, and heavy, long-term sacrifice undermined national unity. I would also argue that a Collective Trap prevented the leaders from fully anticipating social costs. As elites, they would not bear the brunt of the social costs. Their positions were secure, and perhaps, in the case of military leaders, even enhanced by the victory. They would not suffer from disabilities or decreased access to education and health care. This is not to say that Tanzanian leaders were less in tune with their people than are other leaders. If anything, they were perhaps more so, as party leadership under Nyerere carried with it an ethic of humility and involvement with, and accountability to, the citizenry. I believe, rather, that this insulation from consequences is inevitable, to greater and lesser degrees, in all elites, thus preventing decision-makers from fully considering the true consequences of war. In these last four chapters, we have reviewed the war's primary consequences, as well as the secondary economic, political,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 218

and social consequences which followed. Dividing costs into these categories has been useful in that it has allowed an in-depth look at many different factors of the Tanzanian experience of war. However, it has also obscured the interconnectedness of these factors. The political disengagement of the populace was related to economic consequences, and both were related to a decline in government revenues and social spending. The various social traps which distorted the leaders' perception impacted their assessment of all three categories of consequence. In the concluding chapter, the various categories will then be recombined to assess the gestalt of war and the social traps in which Tanzanian leaders found themselves.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PART III

CONCLUSION

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 8

WAR AS A SOCIAL TRAP

War is the unfolding of miscalculations.

Barbara W. Tuchman, The March of Follv

You can no more win a war than you can win an earthquake.

Jeannette Rankin, Noted Pacifist and Former U.S. Senator from Montana

It is time to draw some conclusions about war. In the first section, I assess the overall consequences of Tanzania's war, both intended and unintended. I then reconsider these consequences from the perspective that this war was a social trap, which leads even the most successful of warriors to Pyrrhic victories. In the second and final section, I elaborate on the previous typology of social traps, developing an expanded typology of the social traps of war.

Effectiveness and Effect of Tanzania's War There are two aspects to the question of war's efficacy: (1) effectiveness, the extent to which it achieves the nation's goals, and (2) effect, the cost that it incurs. In removing Idi Amin, Tanzania's leaders had two goals. They wanted to exact revenge, removing the

220

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 221

source of continuous threat to Tanzania's people and border, and they hoped to liberate Ugandans from tyranny, bringing political stability to their neighbor. The first goal was unquestionably and readily achieved. Tanzanians need no longer fear their northern neighbor. For the past decade and a half, they have been able to live peacefully with all of Amin's successors. Just as unquestionably, the second goal was not achieved, at least not within the time period of Tanzania's involvement. When Amin's tyranny ended, Obote's began. Amin's eight-year regime was thought to be responsible for a half million deaths (Kamau and Cameron 1979), while three hundred thousand Ugandans were killed during Obote's four-year, postwar rule (Newell 1985; Wilkinson 1985a & 1985b). Per year, Obote's rule was actually deadlier than Amin's. To be sure, Obote inherited mayhem that Amin created, so many consider him less to blame for the killings than Amin. However, for the average Ugandan, its matters little if death comes from tyranny or mayhem. The world of the average Ugandan did not improve when Amin fled. Indeed, the war merely added to the mayhem. Some Ugandans blamed Tanzanians for imposing a second Obote regime upon them during occupation. Dan Nabudere, one of the leaders of Ugandan exiles in prewar Tanzania and a member of the first postwar government, was one who initially blamed Tanzanians for increasing Uganda's suffering. Over time, he came to believe that Nyerere genuinely sought peace and stability in Uganda and that, when the first postwar Ugandan governments failed, Nyerere

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 222

believed Obote was the only one with a chance to succeed. If that was so, then Nyerere must be faulted with serious misjudgment of Obote, despite his honorable intentions. When I spoke to Nabudere, he also questioned whether or not Ugandans were ultimately aided by Tanzanian intervention. He remained convinced that Amin's regime was crumbling and would have soon fallen without Tanzanian intervention. Moreover, he believed that, without Tanzania's assistance, Obote would not have returned to power because he faced too much opposition.

Milton Obote would never have come back. Its impossible, because his army was divided.. . . the Ugandan group of Obote . . . was basically coordinated by the TPDF [Tanzanian military]. It was not really under the command of Obote, because under the command of Obote, the other groups were not prepared to cooperate. With the Tanzanian command, they were prepared. (Nabudere 1993) Thus, it can be argued that, had the Tanzanians stayed out of Uganda, Amin would have fallen soon and a better leader than Obote might have emerged. It is impossible to know. What is clear is that, for Ugandans, the war failed to liberate them; indeed, it worsened their plight. When Tanzania's leaders determined to oust Idi Amin, they believed it would be worth whatever it cost. Their army succeeded, but for Tanzanians, the war was a Pyrrhic victory. The costs were staggering, as the last four chapters made clear. The country's economy floundered for eight years, as production declined and debt grew. The opportunity costs ($5.12 billion) amounted to nearly a year's GDP. The war seriously exacerbated an economic crisis that

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 223

was already in the making. On the heels of the breakup of the East African Community, a sharp decline in terms of trade, oil price hikes, drought, and misguided government's policies, the war assured that recovery would be long and painful. As if this were not enough misery, AIDS emerged more forcefully than anywhere on the globe. The combination of factors was devastating for Tanzania. Incidents of crime and corruption increased dramatically. Social benefits which Tanzanians had grown to expect from their government disappeared. AIDS cases skyrocketed, revealing a time-bomb waiting to explode. In a desperate attempt at survival, peasants abandoned the formal economy. The crisis brought open challenge to the government and its policies, seriously destabilizing the twenty-plus-year rule of Julius Nyerere. The epigraph at the beginning of Chapter 5, "We went to war, and we came back poor" applies to more than the economic costs of this war. No one imagined the cost of victory would be so great. It is hard to believe that, had the leaders had the benefit of accurate foresight, they would have risked so much to gain far less than they desired. Yet, a decade and a half later, many said, "We had no choice." This seems the final trap that leads to war.

Unfolding of Miscalculations Recall from Chapter 2 my assertion that war can be a social trap, or a set of circumstances in the decision environment that distorts the decision maker's utility calculations, creating a false

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 224

picture of the probable outcomes of alternate courses of action, thereby leading the decision maker to choose a course of action contrary to her/his self interest. Kenneth Boulding once pointed out, "The analysis of these processes of perverse dynamics is the key to successful intervention in human betterment." (Quoted in Costanza 1984, 79) Social trap theory provides a means of analyzing the "perverse dynamics" that entice national leaders (or leaders of other political units) to initiate wars that turn victors and vanquished alike into losers. Understanding these processes may provide leaders with a means to escape war's traps. As I mentioned then, social trap theory developed from a search within the field of sociology for an explanation for such social problems as drug and alcohol abuse that could provide an alternative to the previous approaches relating to actors' ignorance or immorality. Such a search is also appropriate in seeking to understand Tanzanian leaders' decision to invade Uganda. I certainly do not believe, nor would any serious student of Tanzania suggest, that these leaders were either evil or fools. They were and are mostly gifted and dedicated leaders. I believe they, like many—if not all—leaders who initiate war, misperceived the likely outcomes of their decision because of the "perverse dynamics" of war. In Chapter 2, a typology of social traps was offered, based on the work of several social trap theorists. In this section, I expand on that to develop a typology of war's social traps, adding new traps that have emerged in the analysis of the Tanzanian case. In some cases, the added traps were subcategories in the previous typology

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 225

that I believe constitute a category of their own. In one case, a new trap is identified. Table 5 depicts this typology of war's social traps. As with the previous typology, war's social traps are differentiated first by who is entrapped by the decision. However, I shift the language to focus on political leaders, war's decision makers. Note also that several of the traps interact to exacerbate the effect on leaders' perceptions of likely outcomes of their choices. A. Self Traps entrap the individual decision maker.

1. Time-Delav Traps occur in war because many of the war's costs continue long after the shooting stops but the benefits of victory, or lack thereof, are apparent when the war ends. At war's end, Tanzanians celebrated their victory over the hated Amin with a sense of renewed pride and security. Over the next three years, however, they would see their economy devastated and their government greatly discredited. 2. Sunk-Cost Traps are the mirror image of Time-Delay Traps. They occur in war because some costs must be incurred before potential victory (and its anticipated benefits) can be assured. This is the trap in which bidders find themselves during the dollar auction game (See Chapter 2) as their bids approach one dollar. If the game continues, bidders begin to realize that, even if they win, the potential

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 226

______Table 5: A Typology of War's Social Traps Self Traps entrap individual leaders.

❖ Time-Delay Traps occur because there is a time lag between war and many of its consequences.

0 Sunk-Costs Traps occur because leaders increase the expected utility of war as the costs directs cost of assuring victory rise.

❖ Myopia Traps occur because leaders are only capable of limited foresight.

0 Ignorance Traps occur because leaders lack information to accurately anticipate consequences of war.

4 Wishful-Thinking occur because leaders' desires alter their Traps perceptions of likely outcomes (the Midas effect).

4 Diminishing-Returns occur because war's benefits decline over Traps time without the change being anticipated. 4 Spiraling-Costs Traps occur because war's costs increase over time without the change being anticipated.

4 Paradigm Traps occur because leaders' beliefs limit their perception of the options available to resolve conflict.

Multiple-Person Traps occur because the consequences of leaders' entrap all leaders involved actions are dependent upon the responses in conflict. of others (the Prisoner's dilemma).

Collective Traps entrap the populace involved in conflict. 0 Commons Traps occur because leaders stand to gain more and lose less from war than the general populace.

❖ Missing-Hero Traps occur when a collectivity can avoid negative consequences only if a decision maker is willing to take a politically- risky course of action Source: Adapted from: Platt (1973), Cross and Guyer (1980), Costanza (1984), and Maoz (1990).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 227

for gain is disappearing. The potential for loss, however, is increasing. Despite the fact that bidders will lose money even with a victory, they carry on because they stand to lose even more money by losing than by winning. Thus, the potential for loss rather than gain motivates the bidders to continue. Yet, there will be no winners of this game once bidding passes one dollar, only relative losers. In war, the zero-payoff point is far less apparent, and leaders have less incentive to acknowledge it, even to themselves. To do so would be to admit their misjudgment. Thus, they resolve the inherent irrationality of continuing the war by adjusting upward their expected utility of victory. The prize of victory seems more valuable as the costs rise, just as the desirability of an elusive lover increases, prolonging the pursuit. This is the trap that kept Tanzanians in occupation in Uganda far longer than they anticipated, a period that proved more devastating that the actual war. They fought a war partly to liberate Uganda from chaos, and they kept hoping that, in time, they could impose order, but their investment kept rising while order eluded them. 3. Myopia Traps occur in war because leaders are better able to anticipate short-term than long-term consequences. Tanzanian leaders could more accurately predict the war's military expenditures than they could the long-term economic displacement. This trap applies to benefits, as

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 228

well. Nyerere and his advisors could more accurately anticipate the end of Idi Amin's reign of terror than they could the deadly rule of Milton Obote. In war, Myopia Traps combine with Time-Delay Traps to inflate decision-makers' perception of benefits and deflate their perceptions of costs. Because of the Time- Delay Trap, long-term costs are delayed while potential benefits occur more quickly. Because of the Myopia Trap, costs that are more distant in time are discounted relative to benefits. This can lead to Pyrrhic victories like Tanzania's. 4. In this typology, Ignorance Traps incorporate only that which occurs from true lack of information or understanding. The stress and time pressure that surrounds the environment in which war decisions are made can exacerbate the problem of Ignorance Traps. Tanzania's leaders, for example, had no way to know that their soldiers would be at risk from the unknown but deadly AIDS. Their lack of experience with war prevented them from anticipating the enormity of the war's economic displacement. In many cases, their education and elite status precluded an understanding of the life of the average peasant and the resistance a soldier might have at returning to that life, made harder by the war's sacrifices.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 229

5. Wishful-Thinking Traps i occur because wars are fought over issues about which leaders feel very strongly. Their preferences for particular outcomes cause actors to discount the expected utility of other outcomes that might not necessitate the use of violent force. In the case of Tanzania, Nyerere's hatred of Amin and his friendship with Obote possibly caused Amin's removal to be valued more highly than other outcomes that might have met the nation’s security needs without war. Others leaders shared this view, leading them to anticipate that highly fortifying their border to discourage any further incursions by Ugandan forces would be more costly than a war to overthrow Amin. Moreover, leaders' hesitancy to impose war taxes, thereby dampening the war fever, led them to discount the full costs they were incurring by their decisions. 6. & 7. Diminishinq-Returns Traps and Spiralina-Costs Traps occur because the magnitude of war's consequences change over time; incremental benefits, or returns, decline while incremental costs increase. Yet, due to the Myopia Trap, this change is not anticipated. The security from Amin's menace (a benefit) that Tanzanian leaders thought they had assured through war eroded as a new threat—this time internal—emerged. The economic crisis and declining social

1 In the previous typology, these limitations were called "willing ignorance", a subcategory of Ignorance Traps.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 230 security fostered internal dissatisfaction and open challenge to—as well as widespread abandonment of—the government and its policies, seriously undermining its authority and credibility. When Amin threatened Tanzania's national sovereignty, he was stopped. Yet the economic crisis to which the war contributed brought a new and significant threat—although a threat of a different sort—to national sovereignty in the form of IMF conditionalities. That war's annual costs can increase over time is well demonstrated in Table 3 of Chapter 5, which shows that annual economic opportunity costs of war increase each year. This phenomenon is also true of social and political consequences. Tanzania's war expenditures displaced social spending which, among other outcomes, caused a decline in the number of new primary school entrants. Every year that the school population was limited to finance war costs, more students were denied an education because the population was expanding. This meant that the increase in population of uneducated and unskilled citizens (social opportunity costs) escalated, just as economic opportunity costs grew over time. 8. Paradigm Traps are an addition to the typology and occur because of psychological constraints which limit a leader's perception of the options available for resolving conflict. This is a new category of trap, as the social trap literature does not discuss belief systems and their impact on

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 231

decisions. Different disciplines have used different terms for this concept.

Systems engineers, operational researchers and some experimental psychologists have used the term "models." Some philosophers and logicians have used the term "logics." Anthropologists and psycholinguists have used the terms "cognitive structure," and more recently "epistemology" or "epistemologies." Sociologists have also used the term "epistemology," but more in the sense of "sociology of knowledge" that in the sense of experimental study of cognitive structure. More recently sociologists began using the term "paradigms" mainly because of their effort to gain their inspiration from physics, and because Thomas Kuhn's work on paradigms in physics has become respectable among sociologists. (Maruyama 1973, 2-3) Here I use the term paradigm to describe the beliefs that structure reasoning and thus guide the decision-making process. Leaders who have chosen to use force invariably say, "We had no choice." Yet, history reveals numerous examples of leaders who chose other options against powerful odds and prevailed. King Christian X of Denmark counseled calm and saw his country overrun by Hitler's forces in four hours, but the Danes proved unconquerable. Anwar Sadat reversed decades of violent conflict to make peace with Israel. Mohandas K. Gandhi led a non-violent campaign which ended Britain's domination of India. To the extent that leaders believe force is their only option, they are entrapped by the paradigm that conflict can be resolved by violence, while non-violent means are ineffective. Yet

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 232

the continued prevalence of war suggests that, although one conflict may be resolved for a time, others are created, often by the consequences of previous wars. When Uganda invaded the Kagera salient, Nyerere and his advisers believed that the only way to assure Tanzania's security was to remove Amin by force (Mkapa 1993). Yet, Tanzania's leaders were entrapped in the same paradigm as Amin—that one can gain through the use of force. Despite a very clear difference in motivation, both sides lost significantly. Tanzanian leaders' argument was with Amin, not with the Ugandan people. They went to war because they believed that war would resolve their conflict. In the end, Amin lost his rule, but Tanzanians' loss was perhaps more profound.

The real change after the war was the end of illusions about and within Tanzania. The country lost its image as a conciliator that could not easily be drawn into using the crass methods of the rest of the world; Tanzania's hungry and poorly disciplined soldiers behaved just as badly as any others elsewhere in Africa. (Ungar 1985, 403) B. Multiple-Person Traps entrap all leaders involved in war, because the consequences of initiating war depend upon what the adversary decides to do, as shown in the decision tree in Figure 2, Chapter 2. Clearly, the cost of initiating a war differs dramatically if the target succumbs without a fight than if it resists. Moreover, there may be other actors who influence the consequences of war, such as third parties who

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 233

intervene or new leaders that emerge to replace an ousted ruler. When Libya responded to Amin's appeals for assistance, this changed the calculus of risks for Tanzania. The ability of Uganda to resist was enhanced, but the political cost of Tanzania's invasion was diminished. When Kenya decided to allow war supplies to be transported from its port to Uganda during the war, despite Nyerere's appeals to stop it, this increased the risks for Tanzania. Multiple-Person Traps can combine with other traps to exacerbate the effect. For example, Nyerere's decision to send his troops into Kampala and to oust Amin was based upon his anticipation that the government that emerged out of the Moshi Unity Conference would provide a stable alternative to—and an improvement on—Amin's rule. That the "unity" that the conference was designed to promote was already disintegrating before the war was won. As political conflicts grew amongst factions vying for power in post-Amin Uganda, the costs of occupation grew and the benefits of the war diminished. This example can be seen as a Multiple-Person Trap, because the consequences for Tanzania were dependent upon actions of Ugandans. It can also be a Diminishing-Returns or Escalating Costs Trap because the benefits of the war (Tanzanians could no longer take pride in having ended the chaos in Uganda) and its costs (soldiers continued to die) shifted over time.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 234

C. Collective Traps entrap the populace involved in war. They occur because the consequences of war are different for leaders than they are for the general populace. There are two possibilities here; they can either gain more or lose more. 1. Commons Traps occur because leaders are in a position to gain more and lose less from war than the general populace They are unlikely to lose their life on the battlefield or suffer serious deprivations to support an armed struggle; these are the plight of the general populace. Yet, they stand to gain—at least politically—from victory whereas the general populace would gain far less. In Tanzania, the military leaders clearly enhanced their power after the war, as the size of the military stabilized at twice the prewar size. Yet, the general populace lost, if it can be agreed that Tanzania's borders were no more threatened in 1987 than they were in 1977. The cost of maintaining this expanded military comes at the expense of other uses for those resources. Some level of social services decline was inevitable. 2. Missina-Hero Traps occur when a collectivity can avoid negative consequences only if a decision maker is willing to take a politically-risky course of action. The leaders of the OAU failed to condemn Amin's invasion of the Kagera Salient, and the organization lost credibility with both its own members and the outside world. War fever was high in Tanzania after Amin's invasion of the Kagera salient, and the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 235

military, as well as the Ugandan exile community, were pushing for an invasion. It would have been politically risky for any Tanzanian leader to push for a diplomatic settlement. Yet, had such a leader emerged and prevailed, Tanzanians could have avoided the devastation of their Pyrrhic victory. All of these traps can be found in war. They may not impede the initiator’s ability to anticipate victory, but they seriously distort leaders' ability to anticipate the consequences of that victory. Although there may be victors in war, there are no winners—only relative losers. This is the cautionary tale that I have sought to tell.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. BIBLIOGRAPHY

Africa's most curious war. 1979. Time 113 (April 16): 58.

African nation that stands up to Idi Amin. 1979. U. S. News and World Report 86 (March 26): 41-42.

Ainsworth, Martha, and A. A. Rwegarulira. 1992. Coping with the AIDS epidemic in Tanzania: Survivor assistance (The World Bank Africa Technical Department. Population, Health and Nutrition Division. Technical Working Paper No. 6).

Ames, Edward, and Richard T. Rapp. 1977. The birth and death of taxes: A hypothesis. Journal of Economic History 37: 161- 178.

Amin fights another day. 1979. Economist 270 (March 24): 69.

Amin's blitzkrieg. 1978. Newsweek 92 (November 13): 87.

Aminery. 1978. Economist 269 (November 41: 62.

Ammons, Lila. 1989. Consequences of war on African countries' social and economic development. Ph.D. diss., University of Maryland.

Archer, Dane, and Rosemary Gartner. 1984. Violence and crime in cross-national perspective. New Haven: Yale University Press.

Audifferen, Ekundayo. 1987. The O.A.U. and crisis diplomacy: A study in regional conflict management. Ph.D. diss., University of Aberdeen, U. K.

Avirgan, Tony, and Martha Honey. 1982. War in Uganda: The leaacv of Idi Amin. Dar es Salaam: Tanzania Publishing House Ltd. 236

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 237

Ayoade, John A.A. 1988. States without citizens: An emerging African phenomenon. In The precarious balance: State and society in Africa, ed. Donald Rothchild and Naomi Chazan, 100- 118. Boulder: Westview Press

Azarya, Victor. 1988. Reordering state-society relations: Incorporation and disengagement. In The precarious balance: State and society in Africa, ed. Donald Rothchild and Naomi Chazan, 3-21. Boulder: Westview Press.

Ball, Nicole. 1988. Security and economy in the Third World. Princeton: Princeton University Press.

______1991. The effect of conflict on the economies of Third World countries. In Conflict resolution in Africa, ed. Francis M. Deng and I. William Zartman, 272-291. Washington DC: Brookings Institution.

Barbera, Henry. 1973. Rich nations and poor in peace and war: Continuity and change in the development hierarchy of seventy nations from 1913 through 1952. Lexington: Lexington Books.

Barnett, Corelli. 1963. The swordbearers. London: Eyre and Spottiswoode. Quoted in Arthur Marwick. War and social change in the twentieth century. A comparative study of Britain. France. Germany. Russia, and the United States. 7. New York: St. Martin's Press, 1974.

Behind the lies. 1978. Economist 269 (December 2): 66.

Bennett, Richard R. 1989. Correlates ofcrime (Data Set). Washington, D.C.: The American University.

______1991. Development and crime: A cross-national, time- series analysis of competing models. The Sociological Quarterly 32 (3): 343-363.

______and James P. Lynch. 1990. Does a difference make a difference: Comparing cross-national crime indicators. Criminology 28 (1): 153-182.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 238

Benoit, Emile. 1973. Defense and economic growth in developing countries. Lexington: Lexington Books.

______1978. Growth and defense in developing countries. Economic Development and Cultural Change 2 (January): 271- 280.

Biersteker, Thomas J. 1979. Distortion or development? Contending Perspectives on the multinational corporation. Cambridge: MIT Press.

Boesen, Jannik, Kjell J. Havnevik, Juhani Koponen, and Rie Odgaard, eds. 1986. Tanzania: Crisis and struggle for survival. Uppsala: Scandinavian Institute of African Studies.

Bolton, Dianne. 1985. Nationalization—A road to socialism? London: Zed Books.

Boulding, Kenneth E. 1945. The economics of peace. New York: Prentice-Hall.

Braudel, Fernand. 1973. The Mediterranean and the Mediterranean world in the aoe of Philip II. Volume II. Translated by Sian Reynolds. New York: Harper and Rowe.

Bueno de Mesquita, Bruce. 1981. The war trap. New Haven, CT: Yale University Press.

Bureau of Statistics. 1991. Selected Statistical Series: 1951- 1988. Dar es Salaam: President's Office - Planning Commission.

Bureau of Statistics. 1992. National accounts of Tanzania 1976- 1_991. Dar es Salaam: President's Office - Planning Commission.

Burrows, Noreen. Tanzania's intervention in Uganda: Some legal aspects. World Today 35 (Julyl: 306-310.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 239

Butler, David, with Nicholas Proffitt. 1979. Amin: Now you see him. Newsweek 93 (April 16): 60-62.

Callaghy, Tom. 1987. The state as Lame Leviathan: the patrimonial administrative state in Africa. Chapter 4 in The African state in transition, ed. Zaki Ergas. New York: St. Martin's Press.

Carneiro, Robert L. 1970. A theory of the origin of the state. Science 169 (August): 733-738.

Chatterjee, S. K. 1981. Some legal problems of support role in international law: Tanzania and Uganda. International and Comparative Law Quarterly 30 (October): 755-768.

Cheru, Fantu. 1989. The silent revolution in Africa: Debt, development and democracy. New Jersey: Zed Books.

Clayton, James L. 1969. Vietnam: The 200-year mortgage. Nation. 26 May, 661-663.

Closing in on Amin. 1979. Newsweek 93 (March 29): 59.

Clugston, Michael. 1978. Big Daddy finds a convenient war. Macleans 91 (November 13): 41.

Converse, Elizabeth. 1968. The war of all against all: A review of the Journal of Conflict Resolution. 1957-1968. Journal of Conflict Resolution 12: 471-550.

Copson, Raymond W. 1991. Peace in Africa? The influence of regional and international change. In Conflict resolution in Africa, ed. Francis M. Deng and I. William Zartman, 19-41. Washington, D.C.: Brookings Institution.

Costanza, Robert. 1984. The nuclear arms race and the theory of social traps. Journal of Peace Research. 21 (January): 79-86.

Coulson, Andrew. 1982. Tanzania: A political economy. Oxford: Clarendon Press.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 240

Cross, John G., and Melvin J. Guyer. 1980. Social traps. Ann Arbor: University of Michigan Press.

Deane, Phyllis. 1975. War and industrialisation. Chapter 5 in War and economic development: Essavs in memory of David Joslin. ed. J. M. Winter. Cambridge: Cambridge University Press.

Deger, Saadet, and Ron Smith. 1983. Military expenditure and growth in less developed countries. Journal of Conflict Resolution 27 (June): 335-353.

Dunn, David. 1974. War and social change. In The use of force in international relations, ed. F. S. Northedge. New York: Free Press.

Eckstein, Harry. 1975. Case study and theory in political science. In Strategies of inquiry, eds. Fred I. Greenstein and Nelson W. Polsby, 79-137. Reading, MA: Addison-Wesley.

End of a nightmare. 1979. America 140 (April 23): 344-345.

FAO. 1984 & 1991. Food aid in figures. Rome: United Nations Food and Agriculture Organization.

______Various years. Production yearbook. Rome: United Nations Food and Agriculture Organization.

Final hours seem the longest. 1979. Macleans 92 (April 16): 31-32.

Fitzgerald, Mary Ann. 1979. East Africa's waning chances for cooperation. Business Week (August 6): 48.

Francis, Joyce. 1993. Tragedy of the commons. Daily News. 9 March, 4.

George, Alexander. 1992. Adapting to constraints on rational decisionmaking. In International politics: Enduring concepts and contemporary issues. 3d ed., ed. Robert J. Art and Robert Jervis, 451-71. New York: HarperCollins.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 241

Greene, Graham, and Carol Reed. 1968. The third man (screenplay), 114. Hollywood: Script City. Quoted in Arthur Marwick. War and social change in the twentieth century. New York: St. Martin's Press, 1974.

Grobar, Lisa M., and Richard C. Porter. 1989. Benoit revisited: Defense spending and economic growth in LDC's. Journal of Conflict Resolution 33 (June): 318-345.

Gyimah-Brempong, Kwabena. 1989. Defense spending and economic growth in Subsaharan Africa: An econometric investigation. Journal of Peace Research 26: 79-90.

Hardin, Garrett. 1968. The tragedy of the commons. Science 162: 1243-1248.

Harmon, Nicholas. 1987. East Africa: Turning the corner. The Economist. 20 June, 3-18.

Hassan, Farooq. 1981. Realpolitik in international law: After Tanzanian-Ugandan conflict "humanitarian intervention" reexamined. Willameette Law Review 17 (Fall): 859-912.

Heggenhougen, Harald Kristian. 1986. Health services: Official and unofficial. In Tanzania: Crisis and struggle for survival, eds. Jannik Boesen, Kjell J. Havnevik, Juhani Koponen, and Rie Odgaard. Uppsala: Scandinavian Institute of African Studies.

Helmer, John. 1974. Brinaina the war home: The American soldier in Vietnam and after. New York: Free Press.

Honey, Martha. 1979. Tanzania appeals for aid to rescue war- battered economy. The Washington Post. 13 July, 29(A).

Hyden, Goran. 1980. Bevond Uiamaa in Tanzania: Underdevelopment and an uncaptured peasantry. Berkeley: University of California Press.

Idi-otic invasion. 1978. Time 112 (November 13): 51.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 242

IMF. 1991 & 1990. International financial statistics yearbook. Washington DC: International Monetary Fund.

______. Various years. Government finance statistics yearbook. Washington DC: International Monetary Fund.

Installed by invasion. 1979. Economist 271 (April 21): 16-17.

Interpol. Various years. International Crime Statistics. Lyons, France: ICPO-lnterpol.

Jervis, Robert. 1976. Perception and misperception in international politics. Princeton: Princeton University Press.

Kahama, C. George, T. L. Maliyamkono, and Stuart Wells. 1986. The challenge for Tanzania's economy. Portsmouth, New Hampshire: Heinemann.

Kamau, Joseph, and Andrew Cameron. 1979. Lust to kill: The rise and fall of Idi Amin. London: Transworld.

Kennedy, Gavin. 1974. Defence and growth. Chapter 10 in The military in the Third World. New York: Charles Scribner & Sons.

______1983. The impact of defense spending. Chapter 8 in Defense economics. New York: St. Martin's Press.

Keohane, Robert R. 1986. Realism, neorealism and the study of world politics. In Neorealism and its critics, ed. Robert R. Keohane, 1-26. New York: Columbia University Press.

Kiwanuka, Katoroogo Mayanja. 1985. Foreign policy decisionmaking in Africa: The case of Tanzania. Ph.D. Diss., Syracuse University.

Komba, Marcelino. 1979a. Amin's pillage in the Kagera. Africa 89 (January): 12-17.

______1979b. On the brink of fresh fighting. Africa 90 (February): 25-29.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 243

______1979c. Nyerere warns Amin. Africa 91 (March): 23-24.

Kurian, George Thomas. 1987. Encyclopedia of the Third World. Volume III. New York: Facts on File.

Landor, Alfred. 1985. Nyerere in the Tanzanian dilemma. Contemporary Review 246 (May): 242-246.

Lane, Frederic C. 1958. Economic consequences of organized violence. Journal of Economic History 18,4 (December): 401- 417,

Last legs? 1979. Economist 270 (March 311: 44-45.

Legum, Colin. 1988. The Nyerere years: A preliminary balance sheet. In Tanzania after Nverere. ed. Michael Hodd, 3-11. New York: Pinter Publishers.

______, ed. Various years. Africa contemporary record. London: Rex Collins.

Lofchie, Michael F. 1985. The roots of economic crisis in Tanzania. Current History 84 (April): 159-163f.

______1988. Tanzania's economic recovery. Current History 87 (May): 209-212f.

Looney, Robert E. and P. C. Frederiksen. 1986. Defense expenditures, external public debt and growth in developing countries. Journal of Peace Research 23: 329-338.

Luard, Evan. 1987. War in international society. New Haven, CT: Yale University Press.

Luterbacher, Urs. 1984. Last words about war?. Journal of Conflict Resolution 28 (Marchl: 165-182.

Malangalila, Dr. 1993. Interview with author, Februaryl 5.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 244

Maoz, Zeev. 1982. Paths to conflict: International dispute initiation. 1816-1976. Boulder, CO: Westview.

______1985. Decision-theoretic and game-theoretic models of international conflict. In Dynamic models of international conflict, ed. urs Luterbacher and Michael D. Ward, 76-111. Boulder, CO: Lynne Rienner.

______1990. Paradoxes of war: On the art of national self- entrapment. Boston: Unwin Hyman.

Maruyama, Magoroh. 1973. Paradigmatology and its applications to cross-disciplinary, cross-professional and cross-cultural communication. A unpublished paper presented at the EDRA Conference, 18 April.

______1976. Toward cultural symbiosis. In Evolution and consciousness: Human systems in transition, eds. Erich Jantsch and Conrad H. Waddington, 198-213. Addison-Wesley.

Marwick, Arthur. 1968. The impact of the first world war on British society. Journal of Contemporary History. 3: 51-63.

______1974. War and social change in the twentieth century: A comparative study of Britain. France. Germany. Russia and the United States. New York: St. Martin's Press.

Matatu, Godwin. 1979. Amin hangs on the precipice. Africa 92 (April): 35-38.

Mkapa, Benjamin. 1993. Interview with author, March 3.

Moreau, Ron. 1979. Uganda starts anew. Newsweek 93 (April 30): 49-50.

Morris, William, ed. 1983. American heritage dictionary of the English language. Boston, MA: Houghton Mifflin.

Musgrave, R. A., and J. M. Culbertson. 1953. the growth of public expenditures in the United States, 1890-1948. National Tax Journal 6 (June): 97-115.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 245

Mwakyembe, Harrison G. 1985. The parliament and the electoral process. In The state and the working people in Tanzania, ed. Issa G Shivji. Dakar, Senegal: Codesria. Quoted in Julius E. Nyang'oro. State corporatism in Tanzania. In Corporatism in Africa: Comparative analysis and practice, ed. Julius E. Nyang'oro and Timothy M. Shaw, 67-82. Boulder, CO: Westview Press, 1989.

Nabudere, Dan. 1993. Interview with author, January 21.

Newell, David. 1985. Butchery at Namulonge. Newsweek 106 (August 26): 29.

Nicholson, Michael. 1989. Formal theories in international relations. New York: Cambridge University Press.

______1992. Rationality and the analysis of international conflict. New York: Cambridge University Press.

Nincic, Miroslav. 1980. Capital, labor, and the spoils of war. Journal of Peace Research 17: 103-117.

North, David, and Dan Turner. 1979. Uganda: Butcher on the block. Macleans 92 (April 23): 33-36.

Now, revenge. 1979. Economist 271 (June 9): 77-80.

Nsa-Kaisi, Colonel (retired) Kabenga. 1993. Interview with author, February 28.

Nyang'oro, Julius E. 1989. State corporatism in Tanzania. In Corporatism in Africa: Comparative analysis and practice, ed. Julius E. Nyang'oro and Timothy M. Shaw, 67-82. Boulder, CO: Westview Press.

Nyerere, Julius, President of Tanzania. 1979. Letter to Pastor Neal T. Jones of Columbia Baptist Church, Fall Church, Virginia, 5 June.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 246

Organski, A. F. K., and Jacek Kugler. 1977. The costs of major wars: The phoenix factor. American Political Science Review 71: 1347-1366.

______1980. The costs of major wars. Chapter 3 in The war ledger. Chicago: University of Chicago Press.

Othman, Haroub. 1988. Nyerere's political legacy. In Tanzania after Nverere. ed. Michael Hodd, 158-64. New York: Pinter Publishers.

Over, M., S. Bertozzi, J. Chin, B. N'Galy, and K. Nyamuryekung'e. 1988. The direct and indirect cost of HIV infection in developing countries: The cases of Zaire and Tanzania. In The global impact of AIDS, eds. Alan F. Fleming, Manuel Carballo, David W FitzSimons, Michael R. Bailey, and Jonathan Mann. New York: Alan R. Liss, Inc.

Owomero, Basil 0. 1988. Crime in Tanzania: Contradicions of a socialist experiment. International Journal of Comparative and Applied Criminal Justice 12 (Winter): 177-189.

Peacock, Alan T., and Jack Wiseman. 1961. The growth of public expenditure in the United Kingdom. Princeton: Princeton University Press.

Picking up the pieces after Idi Amin. 1979. U. S. News and World Report 86 (April 23): 15.

Platt, John. 1973. Social traps. American Psychologist. 28: 1243- 1248.

QER (Quarterly Economic Review of Tanzania. Mozambiouel. 1979:1 - 1981:3. London: Economic Intelligence Unit Ltd.

Rankin, Jeannette. Quoted in Kevin S Giles. Flight of the dove: The story of Jeannette Rankin. 153. Beaverton, OR: Touchstone Press, 1980.

Rapoport, Anatol. 1988a. Experiments with n-person social traps I. Journal of Conflict Resolution 32 (September): 457-472.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 247

______1988b. Experiments with n-person social traps II. Journal of Conflict Resolution 32 (September): 473-488.

______and Albert M. Chammah. 1965. Prisoner's dilemma: A study in conflict and cooperation. Ann Arbor: University of Michigan Press.

Rasler, Karen A., and William R. Thompson. 1985. War making and state making: Governmental expenditures, tax revenues, and global wars. American Political Science Review 79 (June): 491-507.

Richardson, John M., Jr., and S. W. R. deA. Samarasinghe. 1992. Measuring the economic dimensions of Sri Lanka's Ethnic Conflicts. In Economic dimensions of ethnic conflict, eds S. W. R. deA Samarasinghe and Reid Coughlin. London: Francis Pinter.

Richardson, Lewis F. 1960. Arms and insecurity: A mathematical study of the causes and origins of war. Edited by Nicolas Rashevsky and Ernesto Trucco. Pittsburgh: Boxwood Press.

Roehikepartain, Eugene C. 1985. Dreams and realities in Tanzania. Christian Century 102 (11 December!: 1151-1152.

Rothchild, Donald. 1992. Conflict resolution in Africa. Lecture given at The American University, November 17.

Russett, Bruce M. 1970. What price vigilance? The burdens of national defense. New Haven: Yale University Press.

Saxe, John Godfrey. 1936. The blind men and the elephant. In The best loved poems of the American people, selected by Hazel Felleman. New York: Doubleday & Company.

Schelling, Thomas. 1971. The ecology of micromotives. Public Interest 25: 61-98.

Shaidi, Leonard P. 1989. Crime, justice and politics in contemporary Tanzania: State power in an underdeveloped

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 248

social formation. International Journal of the Sociology of the Law 1989 17: 247-271.

Shivji, Issa G. 1976. Class struggles in Tanzania. New York: Monthly Review Press.

______1986. Law, state and the working people in Tanzania. Portsmouth, NH: Heinemann.

Shubik, Martin. 1971. The dollar auction game: A paradox in noncooperative behavior and escalation, Journal of Conflict Resolution 15: 107-113.

Simon, Herbert A. 1976. Administrative behavior: A study of decision making processes in administrative organizations. 3rd ed. New York: Free Press.

______1982. Models of bounded rationality. Cambridge, MA: MIT Press.

Singer, J. David, and Melvin Small. 1972. The wages of war. 1916- 1965: A statistical handbook. New York: John Wiley.

SIPRI (Stockholm International Peace Research Institute). 1985. SIPRI yearbook: World armaments and disarmanent. London: Taylor and Francis.

Sivard, Ruth. 1991. World military and social expenditures. Washington DC: World Priorities.

Skarstein, Rune. 1986. Growth and crisis in the manufacturing sector. In Tanzania: Crisis and struggle for survival, eds. Jannik Boesen, Kjell J. Havnevik, Juhani Koponen, and Rie Odgaard, 79-104. Uppsala: Scandinavian Institute of African Studies.

Slow end to one Amin. 1979. Economist 271 (April 71: 64-65.

Small battle, how many dead? 1978. Economist 269 (November 18): 57-58.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 249

Smith, George Ivan. 1980. Ghosts of Kampala. New York: St. Martin's Press.

Someone's fighting. 1979. Economist 210 (March 31: 60.

Starr, Harvey, and Benjamin A. Most. 1985. Patterns of conflict: Quantitative analysis and the comparative lessons of Third World wars. In The lessons of recent wars in the Third World. Volume I. eds. Robert Harkavy and Stephanie G. Neuman, 33-52. Lexington: Lexington Books.

Stein, Arthur. 1976. Conflict and cohesion. Journal of Conflict Resolution 20 (March): 143-172.

______1980. The nation at war. Baltimore: Johns Hopkins University Press.

______, and Bruce M. Russett. 1980. Evaluating war: Outcome and consequences. In Handbook of political conflict: Theory and research, ed. Ted Robert Gurr, 399-422. New York: Free Press.

Stohl, Michael. 1976. War and domestic political violence: The American capacity for repression and reaction. With a Preface by Ted Robert Gurr. Beverly Hills: Sage Publications.

______1980. The nexus of civil and international conflict. In Handbook of political conflict: Theory and research, ed. Ted Robert Gurr, 297-330. New York: Free press.

Svendsen, Knud Erik. 1986. The creation of macroeconomic imbalances and a structural crisis. In Tanzania: Crisis and struggle for survival, eds. Jannik Boesen, Kjell J. Havnevik, Juhani Koponen, and Rie Odgaard, 59-78. Uppsala: Scandinavian Institute of African Studies.

Tilly, Charles, ed. 1975. The formation of national states in Western Europe. Princeton: Princeton University Press.

______1978. From mobilization to revolution. Reading: Addison-Wesley. Quoted in Ted Robert Gurr, ed. Handbook of political conflict. 315-316. New York: Free Press, 1980.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 250

Tilly, Charles. 1992. The future of the state system. An April 22 speech presented at The American University.

Tuchman, Barbara W. 1984. The march of follv: From Trov to Vietnam. New York: Knopf.

Twaddle, Michael. 1979. The ousting of Idi Amin: Regime's swift collapse took Tanzania by surprise. Round Table 275 (July): 216-221.

Uganda will need rebuilding from the ground up. 1979. Economist 271 (April 21): 75-76.

UN Department of International Economic and Social Affairs. Annual. international trade statistics yearbook. New York: United Nations.

Ungar, Sanford J. 1979. Tanzania goes to war. New Republic 180 (March 10): 16-18.

______1985. Africa: The people and politics of an emerging continent. New York: Simon and Schuster.

United Republic of Tanzania. Various years. Economic survey. Dar es Salaam: Ministry of Planning.

Van Raemdonck, Dirk C., and Paul F. Diehl. 1989. After the shooting stops: Insights on postwar economic growth. Journal of Peace Research 26,3: 249-264.

van Witteloostuijn, Arjen. 1988. Maximising and satisficing: Opposite or equivalent concepts? Journal of Economic Psychology. 9: 289-313.

War is hellishly expensive. 1979. Economist 271 (April 141: 81-82.

Webb, Peter with Nicholas Profitt. 1979. Idi Amin at the brink. Newsweek 93 (March 12): 39.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 251

Wheeler, Hugh. 1975. Effect of war on industrial growth. Society 12 (May/June): 48-52.

______1980. Postwar industrial growth. In The correlates of war II: Testing some realoolitik models, ed. J. David Singer. New York: Free Press.

Wilkinson, Ray. 1985a. The ouster of Milton Obote. Newsweek 106 (August 5): 32.

______1985b. Muwanga: Worse than Amin? Newsweek 106 (August 12): 38.

Williams, Alden, and David W. Tarr. 1974. Modules in Security Studies. Lawrence: University of Kansas Press.

World Bank, International Economics Department. 1988. World Tables of Economic and Social Indicators 1950-1987 [Computer file]. Washington, D.C. [producer]. Ann Arbor, Ml: Inter-university Consorium for Political and Social Research [distributor], 1990.

______Population and Human Resources Division, Southern Africa Department, African Regional Office. 1992. Tanzania AIDS assessment and planning study (Report No. 9825-TA).

______1986. World development report. New York: Oxford University Press.

______Various years. World debt tables. Washington DC: World Bank

Wright, Chester W. 1943. The more enduring economic consequences of America's wars. Journal of Economic History 3(supplement): 9-26.

Yeager, Rodger. 1989. Tanzania: An African experiment. 2d ed. Boulder, CO: Westview Press.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 252

Young, Tom. 1988. Introduction to part vii: Political issues. In Tanzania after Nverere. ed. Michael Hodd, 155-7. New York: Pinter Publishers.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.