Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The process and level of military intervention in the states of tropical Africa, 1960-1971 Latouche, Daniel 1973-12-31

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


UBC_1974_A1 L38.pdf [ 24.88MB ]
JSON: 1.0058299.json
JSON-LD: 1.0058299+ld.json
RDF/XML (Pretty): 1.0058299.xml
RDF/JSON: 1.0058299+rdf.json
Turtle: 1.0058299+rdf-turtle.txt
N-Triples: 1.0058299+rdf-ntriples.txt
Original Record: 1.0058299 +original-record.json
Full Text

Full Text

c. f THE PROCESS AND LEVEL OF MILITARY INTERVENTION IN THE STATES OF TROPICAL AFRICA, 1960-1971 by DANIEL LATOUCHE Bacc. , Sc. Soc, Universite de Montreal, 1964 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department, of POLITICAL SCIENCE We accept this thesis as conforming to the required standard THE UNIVERSITY OT BRITISH COLUMBIA May, 1973 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver 8, Canada (i) ABSTRACT THE PROCESS AND LEVEL OF MILITARY INTERVENTION IN THE STATES OF TROPICAL AFRICA, 1960-71 by Daniel Latouche Chairman: Michael D. Wallace Recent studies on political development and military institutions have tended to explain political intervention by the armed forces of developing countries by examining either the organizational charac teristics of the military establishments or the development context in which these military establishments operate. This study presents the view that while corporate factors may explain the process of intervention in the short run, changes taking place in the social, economic and political environment can best explain the level of military intervention existing in a specific country. After the presentation of a brief survey of the literature and the elaboration of an analytical grid, Part II of this study surveys the major organizational transformations which have affected African military forces and which can serve to explain their decision to intervene actively in the political process. This historical (ii) reconstruction of the process of military intervention led to an explanation which stresses the vulnerability rather than the strength of African military organizations as the major reason for their involvement in politics. Because of the nature of their output, their recent creation and the permeability of their boundaries, African military organizations are shown to be easily threatened by changes taking place in their environment. Military coups occur when a military organization decides to use its institutional weight to modify its environment so as to insure its corporate survival. Part III of this study investigates by a quantitative methodology some of the changes taking place in the economic, social and political environment of Tropical Africa to determine if they can be correlated with the present level of military intervention in these states. Six major environmental changes were identified (economic development, social mobilization, political participation, party institutionalization, government penetration and social conflict) and operationalized through the use of factor analysis. The level of military intervention was operationalized by examining the extent to which a military establishment breaks out of its organizational boundaries to occupy roles outside those normally associated with its defence function. Data on the level of military intervention was collected for each year of the 1960-71 period and then aggregated into a final index of military intervention for the 32 states of Tropical Africa. Three analytic techniques were then used on these data. (iii) By simple bivariate correlation it was established that the level of military intervention was positively associated with the level of internal conflict and of party institutionalization in a given society. Levels of government penetration and of political participation were also associated with the dependent variable, but in a negative way and at a less significant level. When step-wise regression analysis was employed, the same results emerged. Furthermore, the six independent variables taken together explained 40 per cent of the variance in the level of military intervention. Using the technique of dependent analysis developed by Boudon, it was established that only the level of social conflict made a direct and substantial contribution to the level of military intervention. The influence of the other variables was apparently dependent on this last variable. In conclusion, it is stressed that the process of military intervention is amenable to an organizational and historical analysis and that the characteristics of the socio-political environment in which a military establishment operates can make a significant contribution to an understanding of the level of military intervention existing in the states of Tropical Africa. (Thesis Supervisor) (iv) TABLE OF CONTENT INTRODUCTION PART I MILITARY INTERVENTION IN THE STATES OF TROPICAL AFRICA. A SURVEY OF THE LITERATURE AND A FRAME WORK OF ANALYSIS CHAPTER 1. MILITARY INTERVENTIONS IN THE STATES OF TROPICAL AFRICA. A STUDY OF THE LITERATURE 8 The Organizational Approach to Military Coups in the States of Tropical Africa .- 10 1. The orientation'to violence of African armies as a cause of coups d'etat 11 2. The national orientation of African armies as cause of coups d'etat 13 3. The social background of the officer corps as a cause of coups d'etat 14 4. The corporate format of African armies as a cause of coups d'etat 18 5. The ideological format of African armies as a cause of coups d'etat 21 The Developmental Approach to Military Coups in the States of Tropical Africa 28 Notes and References to Chapter I 51 CHAPTER II THE MILITARY AND ITS ENVIRONMENT: A FRAMEWORK FOR ANALYSIS 55 (v) The Military Organization as an Open System and an Institution 55 Military Intervention as an Institutional Response 68 The Organizational Parameters of the Process of Military Intervention 73 1. The nature of the organization's output 73 2. The stage of development of the organi zation 74 3. The permeability of the organization's boundaries 6 4. The integration of sub-units in the organization 78 The Environmental Parameters of the Level of Military Intervention 79 1. The complexity of the environment . . 79 2. The nature of environmental changes . 81 Notes and References to Chapter II 88 PART II THE PROCESS OF MILITARY INTERVENTION: THE ORIGINS AND TRANSFORMATIONS OF AFRICAN MILITARY ORGANIZA TIONS CHAPTER III THE AFRICAN ARMED FORCES AS ORGANIZATIONS AND INSTITUTIONS 90 The Pre-Colonial and Tribal Military Tradition in Tropical Africa 90 The African Colonial Forces and European Do mination (1880-1945) 91 The African Colonial Forces and the Transition to Independence (1945-58) 106 (vi) The Transformation into National Armies (1958-63) 139 Political Interventions by African Military Organizations (1963-70) 162 Notes and References to Chapter III 181 PART III THE LEVEL OF MILITARY INTERVENTION: THE ECONOMIC, SOCIAL AND POLITICAL ENVIRONMENT CHAPTER IV THE LEVEL OF POLITICAL INTERVENTION BY THE AFRI CAN MILITARY FORCES, 1960-71: DEFINITION AND OPERATIONALIZATION OF A CAUSAL MODEL . . . 195 The Measurement of Military Intervention . . . 200 The Identification of the Environmental factors 211 1. The level of economic development and mi litary intervention 213 2. The level of social mobilization and mi litary intervention 237 3. The level of political participation and military intervention 246 4. The level of political institutionaliza tion and military intervention .... 248 5. The level of government penetration and military intervention 251 6. The level of internal conflict and mili tary intervention 255 Validity and Reliability of the Indices . . . 259 Notes and References to Chapter IV 274 (vii) CHAPTER V EMPIRICAL RESULTS: BIVARIATE AND REGRESSION ANALYSIS 278 Bivariate Analysis 282 Step-wise Regression Analysis 293 Notes and References to Chapter V 328 CHAPTER VI EMPIRICAL RESULTS: DEPENDENCE ANALYSIS . . . 332 The Dependence Approach to Causal Modelling . 332 The Findings of the Causal Model 345 The Dynamics of the Causal Model 350 The Content of the Causal Model 36Notes and References to Chapter VI 387 CHAPTER VII INTERPRETATION OF THE EMPIRICAL RESULTS . . . 390 Political Parties and Conflicts during the Struggle for Independence 394 Parties and Conflicts after Independence . . 411 Notes and References to Chapter VII 423 CHAPTER VIII SUMMARY AND CONCLUSION 424 An Environmental Paradigm 42The Process of Military Intervention .... 429 The Level of Military Intervention 436 Critical Assessment and Future Research . . . 443 APPENDICES BIBLIOGRAPHY (ix) LIST OF TABLES TABLE I MILITARY EXPENDITURE BY GREAT BRITAIN ON ITS MILITARY FORCES, 1950-59 (IN POUNDS) . . . 108 TABLE II MILITARY EXPENDITURES BY LOCAL GOVERNMENTS IN SELECTED BRITISH TERRITORIES,' 1955-60 (IN POUNDS) 110 TABLE III DAILY PAY FOR EUROPEAN AND AFRICAN OFFICERS IN THE GHANAIAN REGIMENT IN 1954 (IN GHANAIAN SHILLINGS) 114 TABLE IV NUMBER OF INCIDENTS OF MILITARY INTERVENTION BY COUNTRY (1960-71) 197 TABLE V DISTRIBUTION OF INCIDENTS OF MILITARY INTERVEN TION BY PERIOD AND TYPE (NUMBERS AND PERCEN TAGES) 199 TABLE VI ANNUAL INDEX OF MILITARY INTERVENTIONS BY COUN TRY (1960-71) 205 TABLE VII LEVEL OF MILITARY INTERVENTION BY TWO SCALING METHODS 209 TABLE VIII THEORETICAL, EMPIRICAL AND OPERATIONAL ELEMENTS IN THE DEFINITION OF THE LEVEL OF ECONOMIC DEVELOPMENT 227 TABLE IX ROTATED FACTOR MATRIX OF THE INDICATORS OF THE LEVEL OF ECONOMIC DEVELOPMENT 232 TABLE X CORRELATIONS BETWEEN 12 INDICATORS OF THE LEVEL OF ECONOMIC DEVELOPMENT 234 TABLE XI ROTATED FACTOR MATRIX OF SELECTED INDICATORS OF THE LEVEL OF ECONOMIC DEVELOPMENT .... 235 TABLE XII ROTATED FACTOR MATRIX OF SELECTED INDICATORS OF THE LEVEL OF SOCIAL MOBILIZATION 245 TABLE XIII ROTATED FACTOR MATRIX OF SELECTED INDICATORS OF THE LEVEL OF POLITICAL PARTICIPATION . . . 249 TABLE XIV ROTATED FACTOR MATRIX OF SELECTED INDICATORS OF THE LEVEL OF PARTY INSTITUTIONALIZATION . 252 (x) TABLE XV ROTATED FACTOR MATRIX OF SELECTED INDICATORS OF THE LEVEL OF GOVERNMENT PENETRATION ... 256 TABLE XVI ROTATED FACTOR MATRIX OF SELECTED INDICATORS OF THE LEVEL OF INTERNAL CONFLICT ...... 260 TABLE XVII FACTOR SCORES OF EACH COUNTRY BY THE SIX COM POSITE INDEPENDENT VARIABLES 261 TABLE XVIII ROTATED FACTOR MATRIX OF ALL INDICATORS .... 266 TABLE XIX CORRELATION MATRIX OF THE INDICATORS OF THE LEVEL OF ECONOMIC DEVELOPMENT AND THE IN DICATORS OF THE LEVEL OF SOCIAL MOBILIZA TION 269 TABLE XX ROTATED FACTOR MATRIX OF SELECTED INDICATORS OF THE LEVEL OF SOCIAL MOBILIZATION AND LEVEL OF ECONOMIC DEVELOPMENT 270 TABLE XXI CORRELATION OF THE COMPOSITE INDEPENDENT VA RIABLES AND THE DEPENDENT VARIABLE 283 TABLE XXII CORRELATIONS OF BASIC INDICATORS AND THE DE PENDENT VARIABLE 285 TABLE XXIII STEP-WISE REGRESSION ANALYSIS USING THE SIX COMPOSITE VARIABLES AS PREDICTORS 301 TABLE XXIV COMPARISON OF THE CORRELATION AND REGRESSION COEFFICIENTS OF THE COMPOSITE INDEPENDENT VARIABLES AND THE DEPENDENT VARIABLE .... 302 TABLE XXV STEP-WISE REGRESSION ANALYSIS USING THE 9 BASIC INDICATORS OF THE LEVEL OF ECONOMIC DEVELOP MENT INDEX 305 TABLE XXVI STEP-WISE REGRESSION ANALYSIS USING THE 8 BASIC INDICATORS OF THE LEVEL OF SOCIAL MOBILIZA TION INDEX .306 TABLE XXVII STEP-WISE REGRESSION ANALYSIS USING THE 4 BASIC INDICATORS OF THE LEVEL OF POLITICAL PARTI CIPATION INDEX 310 (xi) TABLE XXVIII STEP-WISE REGRESSION ANALYSIS USING THE 4 BASIC INDICATORS OF THE LEVEL OF PARTY INSTITUTION ALIZATION INDEX 311 TABLE XXIX STEP-WISE REGRESSION ANALYSIS USING THE 5 BASIC INDICATORS OF THE LEVEL OF GOVERNMENT PENE TRATION INDEX 312 TABLE XXX STEP-WISE REGRESSION ANALYSIS USING THE 7 BASIC INDICATORS OF THE LEVEL OF INTERNAL CONFLICT INDEX 313 TABLE XXXI COMPARISON OF THE CORRELATION AND REGRESSION COEF FICIENTS OF THE BASIC INDICATORS 315 TABLE XXXII STEP-WISE REGRESSION ANALYSIS USING ALL INDICATORS OF EACH INDEPENDENT VARIABLE 320 TABLE XXXIII STEP-WISE REGRESSION ANALYSIS USING THE 6 BEST BASIC PREDICTORS 321 TABLE XXXIV SUMMARY OF THE MAJOR FINDINGS OF THE BIVARIATE AND REGRESSION ANALYSIS 324 TABLE XXXV THE SYSTEM OF DEPENDENCE EQUATIONS 337 TABLE XXXVI CORRELATION MATRIX OF THE INDEPENDENT COMPOSITE VARIABLES 344 TABLE XXXVII DEPENDENCE COEFFICIENTS FOR THE INDEPENDENT VARIA BLES IN THE CAUSAL MODELS 346 TABLE XXXVIII COMPARISON OF THE CORRELATION, REGRESSION AND DEPEN DENCE COEFFICIENTS OF THE COMPOSITE INDEPENDENT VARIABLES AND THE DEPENDENT VARIABLE 348 TABLE XXXVIX COEFFICIENTS OF TOTAL INDIRECT EFFECT FOR EACH OF THE COMPOSITE VARIABLES 351 TABLE XL DIFFERENT PATHS OF INDIFFERENT EFFECT ON THE DEPEN DENT VARIABLE FOR EACH OF THE COMPOSITE VARIABLES 352 TABLE XLI DEPENDENCE ANALYSIS USING THE BASIC INDICATORS OF THE LEVEL OF ECONOMIC DEVELOPMENT INSTEAD OF THE COMPOSITE VARIABLE ECONOC 362 (xii) TABLE XLII DEPENDENCE ANALYSIS USING THE BASIC INDICATORS OF THE LEVEL OF SOCIAL MOBILIZATION INSTEAD OF THE COMPOSITE VARIABLE MOBILI 364 TABLE XLIII DEPENDENCE ANALYSIS USING THE BASIC INDICATORS OF THE LEVEL OF POLITICAL PARTICIPATION INSTEAD OF THE COMPOSITE VARIABLE PARTIC . . 366 TABLE XLIV DEPENDENCE ANALYSIS USING THE BASIC INDICATORS OF THE LEVEL OF PARTY INSTITUTIONALIZATION INSTEAD OF THE COMPOSITE VARIABLE INSTIT . . 367 TABLE XLV DEPENDENCE ANALYSIS USING THE BASIC INDICATORS OF THE LEVEL OF GOVERNMENT PENETRATION INS TEAD OF THE COMPOSITE VARIABLE PENETR . . . 368 TABLE XLVI DEPENDENCE ANALYSIS USING THE BASIC INDICATORS OF THE LEVEL OF INTERNAL CONFLICT INSTEAD OF THE COMPOSITE VARIABLE CONFLI 369 TABLE XLVII COMPARISON OF THE CORRELATION, REGRESSION AND DEPENDENCE COEFFICIENTS OF THE BASIC INDEPEN DENT VARIABLES AND THE DEPENDENT VARIABLE . 370 TABLE XLVIII INDIRECT EFFECT OF THE COMPOSITE VARIABLES THROUGH THE LEVEL OF INTERNAL CONFLICT AS MEASURED BY ITS BASIC VARIABLES INSTEAD OF CONFLI . . 374 TABLE XLIX SUMMARY OF THE MAJOR FINDINGS OF THE DEPENDENCE ANALYSIS .' 383 (xiii) LIST OF FIGURES FIGURE 1 THE ORGANIZATIONAL AND DEVELOPMENTAL MODELS OF MILITARY INTERVENTION 46 FIGURE 2 DISTRIBUTION OF MILITARY INTERVENTION PER YEAR . 198 FIGURE 3 DISTRIBUTION OF THE ANNUAL LEVEL OF MILITARY IN TERVENTION FOR TROPICAL AFRICAN COUNTRIES, USING TWO SCALING METHODS 211 FIGURE 4 CAUSAL MODEL OF THE LEVEL OF MILITARY INTERVEN TION. 334 FIGURE 5 CAUSAL MODEL INDICATING THE MOST IMPORTANT CAU SAL RELATIONSHIPS 347 FIGURE 6 CAUSAL MODEL INDICATING ONLY THOSE DEPENDENCE COEFFICIENTS WHICH ARE .40 353 FIGURE 7 MAJOR CAUSAL PATHS OF THE LEVEL OF ECONOMIC DEVELOPMENT 355 FIGURE 8 MAJOR CAUSAL PATHS OF THE LEVEL OF SOCIAL MOBI LIZATION 356 FIGURE 9 MAJOR CAUSAL PATHS OF THE LEVEL OF POLITICAL PARTICIPATION 357 FIGURE 10 MAJOR CAUSAL PATHS OF THE LEVEL OF PARTY INS TITUTIONALIZATION 358 FIGURE 11 DEPENDENCE ANALYSIS WITH V18 REPLACING ECONOC . 377 FIGURE 12 DEPENDENCE ANALYSIS WITH V19 REPLACING ECONOC . 378 (xiv) LIST OF APPENDICES APPENDIX I DISTRIBUTION OF TYPES OF MILITARY INTERVENTION BY COUNTRY 446 APPENDIX II NAME, DEFINITION AND SOURCE OF THE BASIC VARIA BLES 447 APPENDIX III BASIC INDICATORS USED TO CONSTRUCT THE COMPO SITE INDEPENDENT VARIABLE 456 APPENDIX IV THE BOUDON METHOD OF DEPENDENCE ANALYSIS 463 (XV) WITH A LITTLE HELP FROM MY FRIENDS It is a pleasure to record the many debts incurred in the preparation of this dissertation. I would first like to thank Morris Janowitz who provided the original idea for this disserta tion. Later Michael Wallace, Jean Laponce, Robert Jackson and Kai Holsti provided many suggestions and insights concerning the study. Michael Wallace, in particular, devoted many hours from his homes in Vancouver and Ann Arbor, to the task of supervizing a thesis concerned with Africa, but written mainly in Vancouver, Chicago or Montreal. I also owe large debts to the researchers of the African National Integration Project of York University who provided the empirical data for this study. Financial assis tance made available by the Canada Council also allowed me to complete the research for this study. Without the programming expertise of Ronald Alepin, few of the calculations in this study could have been performed correctly. My thanks also go to Linda Collier, Fern Miller and Paula LaPierre who patiently read over earlier versions of the manuscript and identified some of its most obvious flaws. 1 INTRODUCTION At the time of this writing (June 1972), there have been 26 successful military take-overs and 18 attempted coups among the 32 countries of Tropical Africa since 1960. If only because of their number, such overt military interventions have become an important aspect of African politics. However, even more important than its frequency, is the totally unexpected nature of this phenomenon. Barely a decade ago, no one predicted a political role of any kind for the new African armies. History textbooks completely ignored their existence (Brunschswlg, 1963; Chailley, 1968); nor were they mentioned to any extent in the first bibliographies on the armed forces of developing 1 nations (Blankstein, 1964; Land, 1964; Lissak, 1964). One early study of sub-Saharan Africa suggested that African countries "lack what many new states of the former colonial world have had, namely, an army which. . . could be called in, or could take over (Coleman, 1962, p. 359)." This same study concluded that African armies were unlikely to follow the lead of their Latin American and Middle Eastern counterparts, but would instead limit themselves strictly to their defence responsibilities. As late as 1965, one observer could still conclude that more than any other continent at any time in world history "Africa is led by pen-wielding intellectuals, rather than by the modern equivalent of saber-rattling men on horseback (Van Den Berghe, 1965, p. 12)." In the same vein Gutteridge predicted that African armies would not intervene in politics 2 because they "lack the necessary professional cohesion and have not sufficient technical know-how to be regarded as uniquely capable of running 2 a country (1965, p. 144)." This prognosis was supported by apparently strong empirical evidence. African armies had no indigenous military tradition nor, as in Latin America, did they have a long history of caudillism. Secondly, African armies had played no role in the struggle for independence and thus had not reached a position of political prominence. Finally, many African national armies had formally been created only months before independence was achieved and did not control a significant portion of the national budget. Moreover, in 1960 the African political systems did not exhibit any signs of immediate collapse which would be likely either to necessitate 3 or facilitate a military intervention. Having led their countries to independence, African nationalist leaders appeared firmly in control. If any problems of political leadership were foreseen it was expected that they would come from the authoritarian and charismatic nature of the new nationalists' leadership rather than from any lack of strength. Also political institutions, parties, legislatures and civil service, had all evidently survived both the transplant to African and the transition from colonialism to national independence. They seemed unlikely to collapse 4 in the near future. While many foresaw the very real problems of elitism, corruption, tribalism and underdevelopment, even the most pessimistic 5 observers did not foresee the danger of military intervention. Yet, by 3 1970 the military were in full political control in ten countries and had 6 actively intervened in the political process of eight more. This rapid transformation of the African armies from a strictly military to a political role is the subject of this study. What forces, both within the military organizations and in the economic, social and political environment, brought about this transformation? Can the specific impact of each set of factors be identified? Was the collapse of the organizational boundaries of the African armies already inherent in the characteristics of the colonial African regiments? How important were the intrigues of the former colonial powers? Can the specific factors which determine the level of military intervention be identified? These" are the basic questions which the study seeks to answer. A study of the process of the military's entry to the political arena may also be relevant to a number of more general issues. To begin with, it is hoped that some of the problems connected with the creation of viable African states will be raised in a more systematic context. The concept of development which has served as the major analytical tool for the study of the politics of Latin American and Afro-Asian states has recently come under attack because of its close association with the Western historical experience, and because of the difficulties connected 7 with its systemic and empirical application. The following analysis of military coups, although it does not formally make use of the concept of development, may contribute to this debate by facilitating the 4 identification of those epistemological and empirical conditions under which the concept of development can contribute positively to an under standing of one aspect of the political process in the non-Western world, 8 namely military coups. Second, by offering a systems-oriented analytical grid and by using it to analyze the historical development of African military organizations, it is hoped that a new paradigm for the study of political intervention of the armed forces of developing countries may result. There are now two such competing research strategies. The first applied mainly by sociologists, emphasizes the bureaucratic, professional and ideological aspects of the military as a modern profession to explain its political involvement. The second research strategy, put forward mainly by political scientists, emphasizes those aspects of the social, economic and political environment in which the armed forces operate. Neither sociologist nor political scientists have dismissed the possibility of links between the socio-political and the organizational variables, but the precise nature of these links has yet to be specified. Thirds, we hope that this study will show the feasibility of integrating the traditional approaches of historical and sociological 5 analysis with the most recent breakthroughs in quantitative methodology. To the extent that these questions can be answered there is a fourth area to which this study may contribute, that of the sociology of social organizations. What parameters determine the permeability of an organization's boundaries? What are the consequences for the internal functioning of an organization of a radical modification in that organization's goals? How can an organization use its institutional weight to redefine its environment? This study falls into three parts. The first reviews the political and sociological approaches to the study of military intervention and then suggests a framework within which both approaches can be integrated to allow for a differentiation between those factors which set in motion the process of military intervention and those factors which determine the level at which military intervention will exist in a given African country. Seeking to answer the first of these questions, Part II provides a detailed examination of the African armed forces as organizations and institutions, their colonial origins, their transformation into national armies and the changes undergone after independence. Part III attempts to identify in an empirical and causal fashion those factors in the African political and socio economic environment which determine the level of military intervention existing in each African country. NOTES AND REFERENCES TO INTRODUCTION When simple citations are involved we used the author's name and publication date in the text. The full titles can be found in the bibliography. An exception was made for Debates of Legislative Assemblies which are fully cited in the text. This interpretation was not limited to American or British observers. In a colloquium held at Dijon in December 1962, the French social scientist Leon Hamon expressed the view that African armies were unlikely to rebel: Je crois que les organismes politiques ont pris une avance trop forte dans 1'occupation du pouvoir; par suite de circonstances locales l'armee aura, me semble-t'il, trop peu d'occa sions de se couvrir de prestige et de lustre. (Hamon, 1966, p. 104). Shils (1962) has expressed similar views in a 1959 colloquium sponsored by the Rand Corporation. The Congolese episode of 1960 was seen only as an illustration of the total failure of the Belgian colonial experience, not as the fore-runner of similar situations to come (Hoskyns, 1961; Weiss, 1965). On African one-party rule and the controversy as to its contribution to political development see Coleman and Rosbers (1964), Hess and Loewenberg (1964), Morgenthau (1964), and Zolberg (1964). On the problems of charismatic political leadership see Apter (1969), Mazrui (1967) and Tiger (1964). On western political institutions and their transfer to Africa see Burke (1967), Price (1967) and Riggs (1964). One area where difficulty was anticipated was the new elites being formed in the universities. On this problem see Dubois (1965), Lloyd (1966) and Wallerstein (1965). The first group includes Burundi, Central African Republic, Congo-Brazzaville, Ghana, Mali, Somalia, Sudan, Togo, Uganda and the Upper Volta. The second group includes the Congo-Kinshasa, Dahomey, Ethiopia, Gabon, Kenya, Nigeria, Sierra Leone and Tanzania. For a comprehensive critique of the concept of development as it is used by the majority of American social scientists see Baran (1957). Throughout this study, the expressions "under-developed countries", "Third World" or "developing nations" will be used interchangeably. No specific ideological connotation is attached to any of these terms. The term Africa will also be used to describe what is not properly speaking Africa but Black Sub-Saharan Africa. PART I MILITARY INTERVENTIONS IN THE STATES OF TROPICAL AFRICA: A SURVEY OF THE LITERATURE AND A FRAMEWORK FOR ANALYSIS 8 CHAPTER I MILITARY INTERVENTIONS IN THE STATES OF TROPICAL AFRICA: A STUDY OF THE LITERATURE The purpose of this chapter is to examine the various hypotheses and theories which attempt to explain the occurence of military coups in Tropical Africa. The reader may find that few full-scale theories of African military coups are given any close attention while several minor attempts at explanation of this phenomenon are analyzed in some detail. The reasons for this emphasis are both practical and methodolo gical. An extensive search of the literature revealed that the major theoretical contributions to the study of civil-military relations in developing countries (Janowitz, 1964; Huntington, 1968; Levy, 1962; Shils, 1962) made little or no mention of the African situation. At the same time, because they are a relatively new phenomenon, African military coups have not yet been the subject of systematic empirical studies. Nor have they given rise to any major theoretical analyses. As a result, the available literature consists largely of short case-studies or journalistic interpretations. The orientation of this survey also aims to avoid some of the most common errors of similar surveys. For example no attempt will be made to demonstrate the uniqueness or superiority of the subsequent study. The aim is not to arrive at an exhaustive evaluation of the current state of research on the causes of military intervention, but only to define 9 those coordinates which place this particular study within the frame-work of the wider research effort now in progress. Instead of assessing the empirical validity of selected explanations of African military coups this survey will attempt to show that a close reliance on western models of military professionalism and political development has led to the adoption of two basic, but distorted, explanations of the phenomenon of African military coups. In the process it will become clear that these two explanations would benefit from an integration of some of their elements. One group of researchers have stressed the internal character istics of African military establishments as the cause of their inter vention in the political process. By projection of the organizational format of European armies, African military institutions are presented as modern, cohesive, non-political and professional organizations whose intervention is guided by a concern for the nation and its survival. This approach will be designated as the organizational explanation. The second approach has emphasized the process of social, eco nomic and political development as the cause of African military coups. In this case it has been suggested that African countries have under gone rapid social and economic transformation, putting increased demands on the existing political institutions which have proven unable to fulfill these demands. Military coups are seen as the result of this breakdown or threatened breakdown in the overall process of modernization. This approach will be referred to as the developmental explanation. 10 Among sociologists and political scientists these organizational and developmental explanations are not as mutually exclusive as this survey would tend to indicate. To some extent the differences have been over-emphasized in order to make more explicit and coherent this presenta tion of an ever-expanding field of research. These two interpretations in fact share a common vision of the causal link between African armies and their decision to intervene politically. Both interpretations picture the military as being involun tarily drawn to intervene in order to lead their country back on the road to modernization and internal stability. As we will attempt to show, this vision does not appear to correspond to the historical reality and thus may have given rise to questionable generalizations on the phenomenon of Afri can military coups. The Organizational Approach to Military Coups in the States of Tropical  Africa Accepting the assumption that "compared with other institutions and bureaucracies, the military establishment has a variety of common organizational features (Janowitz, 1964, p. 28)," military sociologists have selected five characteristics of the military's internal structure as predictors of their political behavior: their functional and national definitions, the social background of their officer corps and their corporate and ideological formats. The following sections will examine the interpretations of African specialists as to the influence of each of these characteristics on military interventions in Tropical Africa. 11 1. The orientation to violence of African armies as a cause of coups d'etat Because armies are oriented toward the exercise of violence, either threatened or actual, they are said to possess a close control over the instruments of this violence. According to Janowitz (1964) this control provides the military with both the possibility and the pretext for a military coup. A coup is thus never outside the realm of possibility for the military as it would be for a group of doctors or lawyers, because to succeed in staging a coup, the military does not need to mobilize skills and resources not normally available to it. Janowitz therefore argues rightly that for the military there is nothing extraordinary in thinking about a military coup, since the possibility of such action is inherent in the concern with violence of any military establishment. This perspective has been accepted by a number of students of African coups who suggest that western observers have exaggerated the impact of African military coups because a similar event would be so traumatic if it was to take place in Europe or North America. For example, as Lantier points out: II ne faut pas trop s*illusionner sur des histoires de kepis ou de chapeaux mous qui, en Afrique, n'ont pas grande signification. Les regimes militaires ne different guere des autres que par le gout qu'ont leurs dirigeants pour les uniformes, les dorures et les decorations (Lantier, 1967, p. 171). 12 Although it is always possible to discover reasons which explain the military's actions, these reasons, according to Decalo (1965) and Snyder (1969), are only post-facto reconstructions based on a logic which rarely corresponds to actual events. As noted by Corpierre (1966) the coup in Africa should not be seen as a trauma tic political event but simply as a "routine military exercise (p. 44)." According to O'Connell (1967) in our search for a rational explanation we may have overdramatized a phenomenon which has simply replaced elections as the most efficient and troublesome means of 1 effecting political change. These authors suggest that military coups have taken place in Africa because it is both possible and easy for the army to carry them out. Empirically this hypothesis has received some support because of the absence of any discernable pattern in the instances of military take-over in Africa. Large and small countries, radical and conservative governments, and former French and British colonies have all been the targets of such coups. However, this hypothesis eventually leads to an impasse: if coups are the results of a random process there must still be some explanation as to why they have occurred repeatedly in certain countries and never in others. Because of this fact it would appear that in Tropical Africa control over the instruments of violence is a necessary but not a sufficient condition for the occurrence of a coup. In fact the basic aim of this study will be to show that African military interventions are not the result of a random process but of a set of causes, both in 13 the military organization itself and in the political environment, which will be identified and discussed below. 2. The national orientation of African armies as a cause of coups d'etat The orientation of the military as a national institution is frequently mentioned by military sociologists as a second major cause of military coups in developing nations. Janowitz (1964) and Welch (1960) both suggested that the military's ethos of public service and its identification with the nation-state often prompts it to intervene when the survival or even simply the stability of the political process is judged to be threatened. Similarly, ' Lefever suggests that African armies being "a more vivid symbol of sovereignty than the flag, the constitution or the parliament (Lefe ver, 1970, p. 21)," constitute an adequate and available alternative to corrupt civilian leadership. In African states this definition of the army as a national institution has often been viewed as a particularly important parameter of its decision to intervene because of the presumed weakness of the central government and the absence of any other institution capable of making a similar claim. Not surprisingly this vision of the military as a symbol of national unity and as a repository of national virtues is shared by African military leaders. For example the young army officers who staged the third Dahomey coup in 1968 talked of their "special responsibilities before the Nation and before History (Afrique Nouvelle, Dec. 27, 1967)." In Upper Volta, Colonel Lamizana spoke of the army as the "incarna tion of Upper Voltan nationhood (Afrique, 53, March 1966, p. 12)." In Mali, the army considered itself the symbol of the Malian virtues 2 of courage, dynamism and autarchy. According to officers and mili tary sociologists alike, the army, because it is the symbol of national independence and national unity, is forced to intervene when national unity is threatened by social unrest or when the inte grity of the country is being undermined by outside agitators. As we will note at length below this interpretation of African military coups does not differentiate between the causes of a coup and the justifications which facilitate the implementation of the coup. By identifying themselves with national unity and national independence the new military leaders are not so much expressing a concern for these goals but an uneasiness over the lack of popular support and the lack of legitimacy for their action. The social background of the officer corps as a cause of coups d'etat Because of the importance of social class in the western sociological tradition, the social origin of the officer corps has often been looked upon as the major cause of military coups. Jano witz (1964) believes that the rural and middle-class origin of the officers in developing countries inevitably leads them to adopt a fundamentalist attitude in their vision of social problems. These problems are seen as consequences of the corruption, frivolities and lack of dedication of political leaders rather than as a result 15 of ideological and structural factors. Because they come from a different social milieu, military officers have developed very little empathy for the civilian elites and their way of solving pro blems. Therefore they have little reluctance in replacing them since such a move is rationalized as bringing an end to a decadent and corrupt administration. However, Janowitz's interpretation of the link between the officers' social origins and their decision to intervene is not shared by all military sociologists. Pye (1962a, 1962b) has suggested that it is the similarities, not the differences, in the social origins of the civilian and military leaders that can lead the army to intervene, as in the Burmese coup of 1958. In this case the close social and personal bonds between the civilian and military leaders, led the officers to feel as intimately acquainted with the problems of the country as the politicians. As a result they did not hesitate to replace them by staging a success ful coup. The most elaborate presentation of the "social origin" inter pretation of military coups is Nun's (1967, 1968). He suggests that in Latin America the survival of a landed aristocracy combined with the rapid extension of electoral suffrage confronts the middle class with the problem of competing for power with both the upper and lower classes. In the battle for political control, the middle class cannot depend on the support of either the trade unions, which 16 are predominantly of working-class origin, or of the legislative bodies, which are usually controlled by the aristocracy. They can only count on the army which, because of the middle class origin of its officer corps, will act as "one of the better, if not the best, structured institution of the middle classes (Nun, 1968, p. 176)." Recent empirical findings cast some doubt as to the empirical validity of this hypothesis as applied, although in different ways, 3 by both Janowitz and Nun. For example, at the end of his detailed study of the Brazilian military, Stepan concludes that the socio economic origin of the officers corps is only a secondary cause of military intervention: For most officers so many life experience and career pattern variables intervene between their entry into the military school system . . . and their promotion to colonel or general 25 or 35 years later that the direct impact of socio economic origin has been considerably weakened (Stepan, 1970, pp. 54-5). Stepan found that the recent political experience, duties and life-styles of officers were better predictors of their political behaviour than their class origin.. In Africa, because of a lack of information on the social background of officers the social origin hypothesis has yet to be 4 empirically verified. However, it has been suggested that the tribal and not the social origin of the officers is an important variable in explaining their political behaviour. 17 In the case of the Sierra Leonese coup of 1966, Cartwright (1968) notes that most officers were Mende while the All-People's Congress party of Premier-elect Stevens against whom the coup was directed had its major strength in the non-Mende regions of the country. In Burundi, Lemarchand (1970) has pointed out that the coups of July and November 1966 marked the end of Hutu political domination and the rise of a new leadership of Tutsi politicians and military officers. Mazrui has also offered an explanation linking tribalism to military coups, based on the following observation: For many African states, the golden age of modern politics coincided with the peak effort of nationalism. When the latter declined as a major determinant of political behavior, modern politics also declined as a nationalized phenomenon (Mazrui, 1969, p. 42). This "de-nationalization" of politics has brought with it a strenthening of those tribal or ethnic loyalties which had, in fact, never completely disappeared. Inevitably, this "re-tribalization" has also affected the armed forces which have been drawn into this new arena of conflict either as participants or arbitrators. Convincing as it may appear Mazrui's thesis has yet to receive empirical support. Unfortunately until statistics on the ethnic composition of the African armed forces are available, this support 5 will be lacking. But already a preliminary investigation of the Nigerian coups of 1966 provides an alternate explanation to what are surely the most often cited examples of tribal coups. Having 18 pointed out that the Ibos were over-represented in the middle ranks, Luckham (1969) and Miners (1971) suggest that the February coup, because it had its origins in the professional grievances of the middle ranks necessarily involved Ibos as conspirators and non-Ibos as victims. In the case of the coup of July 1966, where the roles of Ibos and non-Ibos were reversed, Luckham maintains that tribal reasons were important only to the extent that they "aggregated external poli tical conflicts with the internal organizational strains of the army (Luckham, 1969, p. 224)." In both instances it is suggested that corporate and not tribal grievances were the immediate causes of the coup. Here again then, empirical evidence of a direct causal link between the officers' social origin and their tendency to stage a coup is lacking. The lack of specific information makes it almost impossible for Africanists to apply some of the hypotheses which have been tested in other regions to African events. 4. The corporate format of African armies as a cause of coups d'etat Organizational features of the African armies are often mentioned as a cause of coups d'etat. According to Pye (1961) the army is both a modern organization and a modernizing agent for its members. This organizational modernity of military organizations originated first in the need of the colonial authorities for updated, professional colonial forces which could be integrated within the 19 overall imperial defence structure. Second, because armies are in herently competitive institutions whose "ultimate function is the test of one against each other (Pye, 1961, p. 78)," they must try to meet the highest international standards if they are to survive on the battle-field. Finally, the modernity of African armies is said to be insured by the presence of foreign-trained officers who could implement western military models without opposition since African armies, like most armies, are free from day-to-day tests of efficiency. The recruitment policy utilized within African armies is supposed to extend this process of modernization at the level of individuals. Military induction moves recruits out "of the parti cularistic relationships of traditional life and into the more impersonal and universalistic relationships of an industrialized society (Pye, 1962, p. 80)." Both Pye and Levy (1962) believe that this process of acculturation is carried out more thoroughly and rapidly within the army than by usual channels of education and urbanization. At the same time, because of the army's isolation within the social system, this acculturation takes place within a more stable and secure environment. Contrary to school boys or job trainees, military recruits do not have to return to the family household every evening. They are not subjected to the contradictory experiences of both a traditional and modern environment. 20 According to military sociologists this combination of indi vidual and organizational modernity supposedly makes for a clearly-defined allocation of power and responsibility at all levels. It also favors a degree of rationality and of efficiency which is not present in other organizations (Levy, 1962, p. 588) and the develop ment of a "skill structure which combines managerial ability with a heroic posture (Janowitz, 1968, p. 128)." This hierarchical infra structure with its centralized command, its sense of discipline, its esprit-de-corps and its well developed system of internal and exter nal communications is believed to encourage the military to want to impose some coherence on an unstable political order. Military sociologists therefore conclude that the military stands out as "the only effectively organized element capable of competing for 7 political power and formulating public policy (Pye, 1961, p. 84)." This vision of the army as a model of organizational effi ciency and modernity has been widely used to explain African mili tary coups. According to Rivkin (1967) outside of the ruling poli tical party in one-party states, the African army is the only other organized, identifiable and functioning institution or group. Newbury believes the African army has "a professional cohesion, an esprit-de-corps which makes it unique" and goes on to point out that in times of political chaos the African army benefits from "consi-derable administrative skills, a logistical potential, a communica tions system and a well-integrated hierarchy of command which has 21 few rivals (Newbury, 1967, p. 220)." Feit (1969a, 1969b), sees the small size of the African armies, far from being a deterrent to political intervention, as actually having contributed to the number of coups since such small forces are necessarily more coherent ideo logically and less preoccupied with their traditional defence role than larger military establishments. Markovitz sees the African army as the technological organization par excellence and the mili tary coup as "the logical, if not final, culmination of the tendency 8 towards technology (1966, p. 11)." Recently Lefever has given the idea of the organizational superiority of the African army its fullest expression. African armies tend to be the most detribalized, westernized, integrated, and cohesive institutions in their respective states. The army is usually the most disciplined agency in the state. It often enjoys a greater sense of national identity than other institutions. In tech nical skills, including the capacity to coerce and to communicate, the army is the most modernized agency in the country. It is the best organized trade-union (Lefever, 1970, pp. 21-2). According to Lefever, this organizational superiority inevitably leads to a coup "when a regime is too weak, corrupt, or arbitrary to govern" and must be replaced by the military so as to "avert disaster or 9 effect reform (1970, p. 198)." 5. The ideological format of African armies as a cause of coups d'etat Military sociologists have also suggested that the ideology of military professionalism encourages coups d'etat in developing 22 nations. According to Janowitz, military professionalism presents the officer with an ambivalent self-image: On the one hand, the officer's conceptions of honor, purpose and human nature lead him to assume that he is a standard bearer, who embodies the superior virtues of men, yet, at the same time, he finds it expedient and necessary to present himself as a representative man, not different from other men, and part of the mainstream of contemporary society (Janowitz, 1960, p. 229). In developing areas, particularly in countries with a long history of national independance, this ambivalence is further increased by the juxtaposition of the officers' desire to follow western military models and his desire to be faithful to his own military traditions and "protect himself against foreign principles (BeEri, 1969, p. 353)." In the case of Turkey, this ambivalence led the military to see itself both as the successor of the Ottoman military tradition and as the equal of the other NATO forces. According to Ozbudun (1966) and Weiker (1963) this dual image was influential in the Turkish army's decision to intervene in the political arena in 1960. Similiarly Price suggests that in Africa the officer corps is caught between its self-image as the incarnation of national values and the "reference-group identifications with the officer corps of the ex-colonial power (1971b, p. 404)." According to this interpretation, this identification with and commitment to a set of foreign traditions, symbols and values affects the officers' 23 perception and interpretation of the performance of civilian leaders in the immediate, post-independance period. Whatever their actual performance, the civilian leaders will be judged incompetent by officers whose standards of excellence lie outside national bounda ries and who see themselves as the real repository of their nation's values. The officer corps of developing countries although ambivalent in their own self-image, tend to share a self assurance as to their own capability in directing their country's affairs. Pye (1961) believes that because the day-to-day problems to which the military are confronted are more clearly structured and because the ranges of alternatives is narrower, the military will tend to overestimate 10 its capacity to find solutions to their nations' problems. In Upper Volta and Dahomey, the military justified their interventions by referring to the apparent incapacity of the civilian governments to make decisions in situations of social conflict. Once in power, the officers usually announce a series of quick decisions so as to establish their reputation as men of action rather than thinkers. These decisions often appear to have little relevance to the immediate crisis situation. For example in The Central African Republic, on seizing power Colonel Bokassa issued five edicts aimed at setting up a new moral order: an obli gation for all government officals to participate in weekly reli gious services; the abolition of polygamy; a limitation on early marriages for girls; an obligation for all government employees to act as examples of Central African virtue; and the prohibition of 24 tam-tam playing between sunrise and sunset, except on week-ends and holidays. In Upper Volta, the new military government immediately announced the suppression of fiscal privileges for ministers and civil servants, the suppression of television broadcasting and the closure of certain diplomatic missions abroad. Along with ambivalence and self-assurance, a third character istic of the ideology of military professionalism which has been linked with military intervention is the officers' distaste for 11 politicians and partisan politics. After seizing power in 1965, Colonel Soglo claimed that the Dahomean army had intervened to put an end to the political in-fighting of the civilian leaders which 12 was ruining the country. Political activity is often seen as a dubious necessity in which the army is forced to participate on a temporary basis. For example, in 1966 Colonel Afrifa announced that the Ghanian army "is not there to govern but instead to prepare the way for a lawfully representative Government of the People'(Evening News, Sept. 1, 1966)." Similar declarations are so widespread that many observers have concluded that the military's animosity towards politics and politicians is an important motivation for intervention (Lusignan, 1969; Schneyder, 1965; Tixier, 1966). Thus it is not only the organizational characteristics of the army's corporate infra-structure which is said to favour military intervention but also the ideological super-structure which accom panies it. According to military sociologists, ambivalence, self-assurance and hostility towards politics and politicians may all 25 support, if not actually motivate, the military's decision to intervene. This brief survey has outlined the five characteristics of the military organization which may influence military interventions. Basic ally, military sociologists explain African military coups on the grounds that the African armies are strong, cohesive, nationalist and modern forces, and that their interventions come about through a desire to prevent politicians from drowning their countries in sterile political debates. Since the military is a modern organization, it feels it cannot allow the country to slip back into tribalism because of the corruption and indecision of politicians. When applied to Africa this organizational explanation of military coups suffers from a number of obvious drawbacks to which we have already alluded. To begin with, it is based almost solely on "inferring what the consequences would be of transferring the organizational format of the military in western countries to transitional societies (Price, 1971b, p. 400)." Until we possess for African armies the kind of detailed information accumulated by Janowitz in The Professional Soldier (1960), it will remain impossible to test the validity of this organizational model for the African context. Second, the organizational model fails to distinguish between the capabilities, incentives and causes of a military coup. Although a monopoly over the instruments of violence is a necessary prerequisite for efficiently staging a military coup, it does not constitute a suffi cient reason for initiating such an operation. 26 Similarly, a high degree of discipline, a hierarchical structure and an efficient network of internal communication may facilitate the planning and execution of a coup but they do not cause such an interven tion. Thus most factors which military sociologists associate with military interventions are at best probabilistic rather than causal statements. Third, although Janowitz had emphasized their importance, most other military sociologists ignored the links existing between the military organization and its socio-political environment. One exception is Luckham' study of the Nigerian army (1969) where the activation of primordial commit ments in the military and its subsequent decision to intervene in the poli tical process is seen as a result of changes in the political environment. But even then, Luckham is more concerned with the immediate impact of the organizational variables than with their interaction with the environment. Fourth, by focusing on organizational variables, military sociologis have been led to an exaggerated concern with the formal structures of mili tary institutions. While military professionalism and bureaucratization may both be present in the American and Togolese armies, they are never theless present at a different level and in a radically different form. Finally, certain findings derived from the study of western armies do not seem to be verified in the case of all developing nations. For example Janowitz (1964) on the basis of his study of European, Asian and Latin American armies, has suggested that military interven tion increases with its size and sophistication. But in studying Latin America, Putnam (1967) and Alexander 27 (1958) found no relationship between size and the propensity to intervene; nor did Gutteridge (1969) and Zolberg (1970) in examining Tropical Africa. 13 Also Feit (1969) found the relationship to be a negative one. Similarly it has been suggested that a high level of unity within the military establishment is conducive to coups (Levy, 1962; Pye, 1962a). Nevertheless Fossum (1967) and Putnam (1967) have found that Latin American military regimes are the targets of a larger share of military coups than civilian regimes, thus suggesting that divisions within the military can make a 14 distinct and positive contribution to the number of coups d'etat. Ultimately, the major contribution of the organizational model is that it forces an examination of the sociological variables most closely associated with those individuals and institutions which consti tute the major protagonists in any military take-over. It assumes that institutions, like individuals, have a singularity of their own which guides their behaviour. The organizational model consistently accepts the existence of an entity known as either the "military" or the "officer corps" which is supposedly capable of evaluating judgement and rational behaviour. Partly because of these shortcomings but also because of a different tradition of research, this sociological approach has come under criticism from political scientists concerned with the problems of development. Rejecting the organizational model they examine the political and socio-economic environment in which armies operate for an explanation of their intervention in politics. As a result their vision 28 of the military is profoundly modified. Far from appearing as an auto nomous and privileged protagonist the military is now considered simply as one organization among many, whose role is mainly determined by the changes taking place in its specific socio-political environment. The Developmental Approach to Military Coups in the States of  Tropical Africa One exponent of the developmental approach is Finer who in The  Man on Horseback (1962),suggests that the military's propensity to intervene and the nature of this intervention are related to the level of political culture in a given nation and not to any organizational feature of its armed forces. According to his view, in a mature political culture where "the public involvement in and attachment to civil institutions is strong and widespread (Finer, 1962, p. 87)." military intervention tends to be limited to an attempt to influence the civilian authorities through legitimate channels. In those countries with a less highly developed political culture, where there is some question as to the legitimacy of the procedure to transfer political power, the military will resort to blackmail. Finally, in countries with an immature political culture, military intervention will either take the form of displacement, for example, the replacing of one leader with another, or of supplantenent, replacing civilian control with a military regime. Finer's approach has the merit of placing the military coup within its societal context. There are, however, a number of limitations 15 inherent in his use of the concept of culture. For one thing, the 29 concept includes too many variable components - legitimate leadership, viable mass organizations, an adaptable constitution and recognized procedures for the transfer of power - to be readily applicable. Secondly, as an explanatory variable "political culture" is almost useless. In non-crisis circumstances a well-developed political culture may restrain the military from intervening but when conflict threatens the very foundations of this politically mature society, as was the case in France in the 1950's, the military will be under heavy pressure to intervene to save this so-called mature society from total collapse. Thus a mature political culture does not at all mean that the military is unlikely to intervene but that it will intervene only when the stakes are higher. Finally, the link between a given level of political culture and certain types of military intervention are more tautological than causal. For example, Finer first defines a mature political culture as one which is not susceptible to extra legal forms of government change such as a military take-over. Yet he also concludes that the maturity of a political culture does inhibit military coups. Huntington (1968) approaches the problem in a similar way by using the concept of politicization. Rejecting the attempts of sociologists to explain coups d'etat through a study of the internal features of the military, he suggests examining military interventions as just one manifestation of a more basic phenomenon which affects all developing societies, namely the general politicization of all social forces and 16 institutions. Because of the absence in developing societies of 30 effective political institutions, specialized groups tend to become involved in politics and deal with general problems "not just issues which affect their own particular institutional interest (Huntington, 1968, p. 194." In such societies, where the political structures according to Huntington, lack autonomy, coherence, and adaptability, the military, like the trade unions and the student organizations^will inevitably become involved in politics as a "response to the escalation of social conflict by several groups and parties (Huntington, 1968, p. 211)." On the whole, Huntington believes that military coups tend to occur in those societies with a high level of political participa tions but with few effective political institutions which can cope with this increased participation and absorb its energy. In the same vein, Lissak (1967) considers that most developing nations cannot maintain a balance within the various insti tutional sphere during the modernization process. The resulting crisis in legitimacy strengthens counter-elites such as trade unions and the Communist party which have long claimed political power, or brings to the surface new elites, such as the military which then act to protect their own corporate interests. Lerner (1960) and Robinson (1960) suggest that the Turkish military coup of 1960 was the result of such an imbalance in the levels of modernity reached by the civilian and the military sectors, the latter through its participation in NATO and American military assistance, had become more dynamic than the civilian leaders who were more interested in their regime's short-term survival than in exercising political leadership. This gap "tended to make the 31 army's programs dysfunctional" and as a result its "satisfactions turned into frustrations (Lerner, I960, p. 41)." As a result, by 1960 a military coup had apparently become inevitable, in order to re-establish an equilibrium between the military and civilian sectors. A number of other studies have tried to establish more limited causal links between selected aspects of the political environment and military intervention. Ozbudun (1966), looking at Turkey, and Alexander (1958), at Latin America, have stressed the importance of a stable and 17 well-organized party system in limiting military coups d'etat. According to this view, not only will strong and effective political parties inhibit the military from intervening, they will also render such intervention unnecessary. As long as political parties are there to aggregate and articulate the political demands of the population and as long as they constitute an efficient vehicle for meeting these demands by implementing decisions, the political system will survive 18 the revolution of rising expectations and successfully adapt to change. Political participation that is the level of citizen involvement in national politics and in the political activities of interest groups, constitutes a second component of the political environment which has often been associated with the occurrence of military coups. Finer (1962) and Johnson (1962) both suggest that increasing popular interest and participation in the political process decreases the likelihood of military intervention. Almond (1963), Black (1966), Hopkins (1969) and Rustow (1967) also believe that a high level of political partici-32 pation constitutes the best safeguard against military intervention by 19 making such intervention unnecessary and difficult to achieve. These same authors also insist on an effective performance by the political leadership as an efficient deterrent to military coups. Almond (1963) and Apter (1971) believe that military coups occur when the nation's wealth and resources are not fairly distributed among the population. For Eisenstadt (1966) coups are the result of a sudden decrease in the governments' capacities to reach out and distribute public goods to the citizenry. Criticizing Finer and Huntington as well as Lissak for their exclusive preoccupation with the political environment, Perlmutter (1969) suggests that military intervention is the result of both social and political conditions. Four social and political conditions are singled out: a low degree of social cohesion; the existence of fratricidal social classes; a non-consolidation of the middle-class; and a low level of recruitment and mobilization of resources as a result of the lack of common political symbols. These socio-political conditions in turn tend to cause a low level of political institutiona lization and eventually encourage the military to intervene in the political process. Because neither ideologies nor political parties can unite the various elements of the body politic, a gap develops 20 between the "center" and the "periphery". This gap often degenerates into open social and political conflict which in turn contributes to 21 military coups (Fitzgibbon, 1960; Roberts, 1958; Fossum, 1967). 33 Other socio-economic factors have also been associated with military intervention. Economic growth^ especially industrialization, has been said to diminish the likelihood of military coups (Alexander, 1958; Fossum, 1967). The rationale behind this theory is simple: as an economy develops, prosperity increases and in a prosperous economic climate social conflicts diminish as the general standard of living rises. This optimistic vision is challenged by Huntington (1968), Putnam (1967) and Hoselitz (1960) who suggest that a high rate of economic development which increases the population's expectations can in fact aggregate latent social conflicts, and will increase rather than decrease the likelihood of military intervention. This developmental approach, first applied to the Latin American and Middle Eastern scene, has been applied to African military coups by several scholars. For example Lee believes that it is more important "to analyze the nature of the society involved (1969, p. 160)." than to make comparative speculations "on the basis of school background or sense of professionalism between different officer corps (p. 154)." Zolberg states that any understanding of the behaviour of African armies "must be firmly rooted in the understanding of the unique quality of these societies (1969, p. 201)." Gutteridge suggests that any sound explanation of African military coups is "likely to be as much concerned with the political behavior of the whole society as with that of the defence forces in particular (1969, p. 141)." As in the case of the explanation offered by military sociologists this concern with socio-political environment has given rise to another 34 general explanation of African military coups, based on a theory of modernization and political development rather than military professionalism. According to this approach African societies are involved in an irreversible and accelerated process of modernization. This process involves a break with traditions since it imposes a high degree of social mobility and well-developed associational and occupational systems. Individuals are up-rooted, old values no longer make sense, while new values are still not firmly internalized so as to provide a guide for action. Urbanization, education and industrialization all create a climate of rising expectations with which neither the politicians nor the political system can cope. The outcome is often disequilibrium and uneven development for the various sectors of society, leading to individual frustration and tension. ~ At this point the danger of systemic collapse is so great that the military will decide to intervene to prevent a further deterioration of the modernization process. This standard explanation has received alternative formulation in the literature on African military coups. According to Lantier the military have taken over under pressure from the progressive and modern sectors of the population which grow tired of inefficiency and corruption: C'est sous la pression des forces populaires, decidees a mettre fin tant a la gabegie et a la corruption qu'a la crise economique et sociale que les militaires ont accepte de prendre provi-soirement la direction des affaires courantes . . . 35 les politiciens eux-memes, las d'un carrousel les placant tour a tour en prison et au pouvoir avaient ete les premiers a la souhaiter (1967, p. 172). Similarly Lusignan (1969) maintains that the African military, because of their special regard for efficiency and progress, have always felt obliged to answer the call of those "who stood for obedience, discipline, honesty and authority (p. 365)." O'Connell attributes military coups not only to a clash of expectations and the resulting frustration but also to the quality of the political class since those who could have improved the quality of this class opt instead for "the public service ... in order to replace expatriates and keep the administration machine running (O'Connell, 1967, p. 187)." According to this view, the skills of nationalist leaders in rallying support are of little use when applied to the pro blem of governing a country. The military prove better suited to this task and, according to O'Connell, their intervention should be seen as a sort of waiting-period until a new generation of politicians is deve loped. A somewhat similar view is expressed by Said who suggests that African armies have assumed political significance by default because the civilian authorities have "preferred facile ideological speculation to the wise allocation of social values, reconciliation of conflicting interests and performance of efficient administration (Said, 1968, P. 41)." 36 In the same vein Skurnik in his analysis of the Upper Voltan military coup of 1966, emphasizes the lack of political skills of Pre sident Yameogo: Apart from political obtuseness, he seems to have lacked a fundamental sense of balance in his relations with others. For one thing, his channels of communi cation were so clogged up that he apparently ignored several signs of trouble brewing (1966, p. 7). Hippolyte (1968) believes that the major cause of military intervention is not the inefficiency of the political leaders, but rather the global situation of under-development, for which no political leader, however dedicated and skilful he may be, can find an easy solution. A coup d'etat he maintains is possible and almost inevitable when this under development reaches the crisis stage: Un coup d'etat semble possible la ou la situation economique et sociale offre un terrain propice, la ou il y a carence des autres forces organisees, la ou l'armee est toute puissante ou seule puissante, la ou 1'opinion publique est contrainte ou tentee d'avaler l'entreprise des militaires, la ou finalement les grandes puissances disposent de moyens d'action (1968, p. 48). The most comprehensive formulation of this developmental approach is that of Zolberg (1968a, 1968b, 1969). He argues that in the immediate post-independance period the new political leaders gained additional prestige from the sudden creation of various new political offices and the expansion of state-directed economic activities. At first African politicians who could justify their claims to power because of their roles during the struggles for independence had no difficulties in putting together viable political coalitions by a 37 by a careful distribution of these new benefits. However, in time, distrust developed as the gap between the leaders' promises and their capacity to deliver became apparent. The politicians soon began to rely increasingly on force and coercion to remain in power with a result ing decline in the value of political bargaining and a corresponding rise in that of the police and the army. According to Zolberg the politicians' increased reliance on force was a useful initiation for the armed forces who were responsible for its implementation. The military "learned how much weaker were the African rulers than their predecessors (Zolberg, 1968a, p. 81)." They also learned that "control over even a small, ill-equipped, poorly trained body of men was crucial (1968a, p. 81)." They lost respect for the political leaders and began to question their legitimacy. At the same time, the politicians became less and less adept at "discriminating the real danger signals (Zolberg, 1969, p. 443)," and tended to use all their reserves of force to put down minor civil disorders. According to Zolberg, this increased reliance- on force affects all African politics and therefore military interventions are likely "to occur anywhere in the region" so that it is "impossible to specify variables which distinguish as a class countries where coups have occurred from others (1968a, p. 72)." Whether or not the armed forces of a given country intervene at a particular time is simply related "to highly specific and circumstantial features of that country's current political situation (Zolberg 1968a, p. 72)." 38 and not to any specific corporate variables. Zolberg sees African armies as no different from other social institutions: African armies share the structural characteristics of other institutions. The solidarity of their leadership, the control they have over their own organizations, and the degree of leverage they can exert over the society are not likely to be much greater than what existed in the government they replaced (1969, p. 444). According to this interpretation, African armies are certainly not the organizational ideal types described by Pye (1962a) or Levy (1962). Like Zolberg, Welch sees the political and socio-economic environ ment as the basic cause of military interventions but offers no integrate theory as to their origin. The events leading to African military coups are so complex that they defy causal analysis: Many political systems were involved, each with distinct heritages and problems. To assume that 'popular dis content' or 'economic stagnation' or 'neo-colonialist interference' brought about the coups d'etat does not do justice to the unique combinations of circumstances. Rather than search for a sole cause, we must examine a series of factors, the salience of whose components differs from one African state to another (1970, p. 17). He does suggest five factors which are related in some way to military intervention but does not specify the nature of the causal link or any of the antecent inter-relationships. These five factors all involve a partial breakdown of the political and socio-economic development process: a decline in the prestige of the major political party; an incapacity of the political leaders to adapt themselves to changing political realities; internal social conflicts; economic difficulties; and the corruption and inefficiency of the political leaders. 39 A number of authors have criticized Zolberg and Welch for being too limited in their approach to African military coups. They contend that the phenonmenon cannot be understood only in terms of the political and socio-economic environment of the nation-state in which a coup occurs, but that the continental and international situation must also be taken into account. For example Gonidec (1969) and Corpierre (1966) suggest that the colonial system has so "synchronized" Africa that a coup in one country is quickly imitated in the neighbouring countries. Mazrui calls this process "the inter-territorial demonstration effect", that is, "a capacity to see each other as relevant examples which varied in pro portion to the degree of transnational integration between specific African States (1969b, pp. 94-5)." Other authors suggest that African military coups are not the result of internal factors, or of a contagious effect but conspiracies by foreign governments; thus Ikoko (1970) holds western capitalism and the British government are directly responsible for the 1966 Ghanian coup. His evidence is limited to the presence in Ghana of a West-German journalist working for the Confederation Internationale des Syndicats Libres, a Europe-based trade-union association, and to the British government's insistence that President N'Krumah travel to North Vietnam. Yet, Lewis (1966) holds the American C.I.A. responsible for the same coup. His evidence is a rapid increase in the staff of the American embassy and to the sudden departure, immediatly after the coup, 22 of an information officer from the same embassy. 40 Marxist scholars offer more theoretical interpretations of the role of international factors in African military coups. A recurrent theme in the Marxist view is to see African military coups within the framework of the international capitalist system and its relations with the dependent states at the periphery. In a challenging article Murray (1966) suggests that ideological divisions in the Communist camp and the inner dynamics of advanced capitalism have rendered colonialism inefficient and outmoded as an instrument of control over primary producers. He points out that the military coup is less costly and less flagrantly imperialist than the old system of colonial administrators. Lachenal (1966) sees military coups as reactionary enterprizes carried out to prevent more progressive forces from seizing political power. He believes Colonel Lamizawa of Upper Volta took over from President Yameogo to prevent the striking trade unions from claiming political control for themselves. In the Congo-Kinshasa General Mobutu presumably stepped in when it became clear that President Kasa-Vubu, no longer content to be a tool of capitalist governments, was relying increasingly on the progressive forces within the country, including the rebel 23 forces. For Lachenal, military coups are simply a continuation of imperialism by other means: On a aujourd'hui, cet imperialisme qui n'a pas change de nature, mais doit tenir compte d'une situation nouvelle, s'efforce sous d'autres formes de poursuivre sa politique de brigandage au lieu de s'employer a reparer le mai qu'il a fait. Les chaines sont moins visibles, le resultat est le meme (1966, p. 81). 41 The empirical evidence for interpreting African military coups as instruments of neo-colonialism is contradictory. For example, in Congo-Brazzaville and Burundi, the military regimes have both adopted more progressive attitudes than the civilian regimes they replaced. Also military coups have taken place both in countries which have attempted to free themselves from neo-colonialism (Ghana and Mali) and in countries with very close relations with their former colonial rulers (Gabon and Upper Volta). As is clear from this brief survey of the developmental approach the one common theme which unites all these widely divergent explana tions of African military coups is the impact of modernization on African society. They all see Africa as being caught in the turmoil of political and socio-economic development. As African countries develop economically and reach new levels of collective welfare, an increasing number of citizens become interested in national politics. These desires for increased political participation often cannot be channelled by the existing political institutions, notably the political parties, which have often grown inefficient when confronted with the realities of nation-building. As a result of this inability, social conflicts are exacerbated and threaten to engulf the entire social system. The intervention of the armed forces, often at the request of the civilian leaders themselves, is simply the final step in this process. 42 From this perspective the modernity or non-modernity of the African military organization is of little importance. When society is threatened by general breakdown, if a military institution exists at all, it acquires de facto political importance as Bienen suggests: No matter the degree of professionalization within the officer corps, the level of technology, the organizational format of a particular military force, an army in Africa is always a potential political factor because it can exert some strength in what is essentially a domestic power vacuum (1968, p. 37). Therefore there is actually little validity to the hypothesis suggested by military sociologists of a causal relationship between the existence of a modern professional force and the occurence of African military coups. Such a relationship, they contend, is a spurious one. The real causes of military interventions are to be found beyond organiza tional characteristics. The developmental approach, like the organizational approach discussed above, involves several conceptual distortions. First, it is preoccupied with the most overt and dramatic form of political inter vention by the military, the coup d'etat. But coups, are only one type of military intervention. For example, in both Gabon and Chad which have experienced no such overt military intervention, there is still a high level of covert military involvement in the political process. By focussing solely on the coup, observers like Hippolyte (1968) and Lantier (1969) confuse the immediate events leading up to the coup with the causes of military intervention. Second, the view that coups are an indication of a temporary breakdown in the process of political development has led to circular 43 reasoning. A military coup, by definition, always involves an abrupt modification in the existing political arrangements. Therefore a coup cannot at the same time be considered as a result of a societal break down and be defined as the cause or the symptom of such a breakdown.. Third, according to the developmental approach, coups are often seen as an epiphenomenon brought about solely by environmental factors, either a breakdown in the modernization process or the operation of the international economic system. For example, the idea that African mili tary coups have occurred because of a what is called an "inevitable" crisis in political leadership (Said, 1968), contributes little to an 24 understanding of why a coup has occurred in Ghana and not in Guinea. Fourth, although the developmental approach recognizes the influence of the socio-economic and political environment in any explanation of African military coups, it still fails to establish clear causal links as to which processes in this environment affect the military's decision to intervene. The developmental approach only recognizes that coups are part of the development process and occur in situations where this process has broken down or is under serious strains as a result of internal imbalances. It does not identify the causal paths which links the various dimensions of this environment with the military's decision to intervene. For example, why should the military intervene in those countries where, as Huntington maintains, the level of political institutionalization is low, when in fact one of the first gestures of most military regimes is usually to abolish political parties and dismiss the elected representatives? Unless the specific 44 impact of changes in the political and socio-economic environment on the internal features and the institutional posture of the military organi zation is taken into account, it is impossible to identify precise causal links between the developmental process and military coups. Finally, there is also little empirical evidence for the view that African armies intervene to prevent social collapse. In the Sierra Leone coup of 1968, the military intervened to prevent the peaceful 25 transfer of political power following a valid national election. This intervention, far from establishing an era of political stability, was followed by two additional military coups as a result of internal military rivalries. There are also repeated examples where the army failed or refused to seize power in spite of imminent violent conflict but without any systemic collapse resulting from such non-intervention. In 1964 a successful coup in Gabon was crushed at the last moment by the arrival of French paratroopers without any catastrophic consequence 26 for the country. On November 21, 1966 the Togolese political and social environment exhibited many of the conditions usually associated with military coups: an unstable and confused political situation marked simultaneously by the resignation of two Cabinet ministers, an attempt by President Grunitzky to fire his entire Cabinet, the forced resigna tion of the Minister of Labour and the rift between the President and Vice-President Meatchi; the revival of latent tribal rivalries with the return of Kutuklui, heir apparent to former President Olympyio; the seizure of the national radio station by a group of mutinous soldiers; 44a FIGURE I THE ORGANIZATIONAL AND DEVELOPMENTAL APPROACHES OF MILITARY INTERVENTION ORGANIZATIONAL APPROACH Functional definition. Nationalist orientation I Social background' Corporate format Ideological format' Military intervention DEVELOPMENTAL APPROACH Political culture Politicization Political leadership-Socio-economic development Modernization breakdown Inflation of force. • Ne o-c oloni allsm jMilitary intervention 45 and a protest march of 5,000 Togolese and the threat of further violent political manifestations. Nevertheless the army chose not to intervene and the immediate political crisis was resolved without a general collapse of civil order. Until the main variables associated with the developmental model of African military coups — economic development, social mobilization, political institutionalization ~ are empirically tested it will remain impossible to ascertain the validity of the 27 explanation suggested. As we said at the beginning, the purpose of this review of the literature was not to arrive at an exhaustive summary of the explanations of military coups in Africa, but to illustrate that most explanations follow one of two basic approaches to African military coups: the organizational approach which stresses the internal characteristics of the military organization as the cause of coups, or the developmental approach which stresses the impact of modernization and development on all African societies. Figure 1 summarizes the major causal relationships suggested by the two approaches and reveals that the organizational and the developmental approaches make use of the same factors but in a different causal sequence. Leaf 4.6 omitted in page numbering. 46 does not exist 47 In the end the major criticism levelled at both the organizational and developmental approaches are surprisingly similar. First, in both cases there is an absence of empirical evidence. Models tend to be transposed from military and development literature without any extensive consideration if they are applicable to the realities of African armies and African military coups. Except for the studies by Luckham (1969) and Miners (1971) there are no detailed sociological analyses of the African military than that of Nigeria. No cross-national study has tested the hypothesis, of a link between corporate characteristics of African armies and their propensity to intervene. Nor has there been any cross-national empirical study aimed at testing the existence of a causal link between selected features of the African socio-political environment and their impact on a country's likelihood of undergoing a military coup. Both give a rather static and deterministic view of African military coups. African armies are depicted as pre-determined to inter vene in the political process, either because of certain internal organi zational features or because of a political climate which encourages such intervention. Rarely is the interaction between the organizational and environmental factors taken into consideration. Students of African military coups have paid too little attention to the historical ante cedent of African military organizations. The latter are usually taken as facts whose role is that of independent variables. This neglect of history and empirical scrutiny has led many scholars to accept a 48 distorted view of African armies and of their reasons for intervention. In both approaches, although this is more evident in the case of sociologists, African armies are seen as modern, strong and cohesive forces drawn into the political arena either because of their organizational superiority in the face of threatening civil chaos or because of the pull of a development process which ultimately politicizes all social groups. To meet these criticisms, we will attempt to incorporate elements from both the organizational and the developmental approaches, since it seems reasonable to assume that African military coups are in some way linked to the specific characteristics of African armies and the socio-political context in which they operate. Our own approach is to suggest that in the case of Africa, the problem of military inter vention in politics is in fact made up of two distinct, although closely related questions. First, there is the question of the existence of military coups as such. "Why have African military organizations, which were modelled along the lines of European professional armies, suddenly decided to abandon their role of servant of the state to intervene directly in the political process?" This is the question which military sociologists have attempted to answer by stressing the organiza tional characteristics of African armies. Our concern here will be with the process of military intervention and with those factors which lead the military to decide to intervene and which enable it to implement this decision. 49 There is also another dimension to the phenomenon of military intervention in Tropical Africa, that of its increased occurence since 1965. Not unlike those who share in the developmental approach we see the 26 successful military coups which have occurred in Africa since 1960 not as a succession of individual events which can be explained through the characteristics of the actors involved but as social facts whose re peated and generalized occurence can best be explained from the perspec tive of the more global social system. Thus at the core of our approach is the belief that the process and the level of military intervention are two separate questions which both call for a different research procedure. In the next chapter, we present an analytical grid which differentiates between the organization al and institutional role of the military. This differentiation allows for the reconstruction of the process leading to the military's deci sion to intervene and for the empirical testing of those socio-political factors which can best explain the level of military intervention existing in the various African countries. In Part II, an attempt will be made to apply this grid to a study of the development of African military organizations over an 80 years period. The objective in this section will not be so much to identify all the possible causes of military coups as to reconstruct the histo rical process which leads the military to their decision to intervene politically. 50 In Part III we attempt to verify the empirical validity of a set of propositions derived from the literature reviewed in this chapter and which attempt to explain the level of military intervention in the states of Tropical Africa. 51 NOTES ANTJ' REFERENCES TO CHAPTER I 1. Gutteridge expresses a similar view when he cautions against attri buting long-range causes to African military coups: "the fact that the military coups in question have had little impact on the fun damental economic and social structure of those states indicates that they are not ... to any degree an expression of the depri vation of any sections of society in basic terms (1969, p. 150)." 2. Colonel Sekou Traore, in L'Essor, Jan., 5, 1968. General Mobutu also described the Congolese army as "le creuset du nationalisme congolais (La Voix de L'Armee Nationale Congolaise, June 15, 1966)." 3. For a critique of Nun's theory, see Vandycke (1971). Mosca was the first to suggest that the integration of the officer corps within the ruling class was a necessary condition of their subordination to the civilian authorities. This hypothesis was further developed by Andreski (1961). 4. Lantier (1967) has asserted that unlike Latin American armies the military establishments of Africa had no specific class component. 5. To date such statistics are only available for the Nigerian army. 6. This acculturation process is also seen as providing the recruits witn their first opportunity of systematically advancing themselves and with their first training in citizenship (Pye, 1967, p. 84). 7. According to Lissek (1967) the army is the only institution capable of quick decisions and of enforcing law-and-order. 8. This technological superiority of the army in a society "hungry" for technology is also suggested by Lissak (1967). 9. This vision of the African army follows closely the self-image of the military. For example, the Upper Voltan army is described by its commander, Colonel Lamizana, as a prototype of the modern organization: "L'Armee est disciplinee, l'armee est solidement structuree, elle ignore les regionalismes, elle a le sens des responsabilites et la notion du Devoir (Afrique, 53 March (1966), p. 15)." In the Congo Kinshasa, General Mobutu considered the army as "la seule force disciplinee, la seule autorite aux prin-cipes moraux inebranlables, la seule institution nationale qui ait fait les preuves de son integrite morale et de son efficacite dans son rOle de diffuseur des interets generaux (Mobutu, 1966, p. 33)." 52 10. According to Shils (1962) "if the civilian political elite is self-confident and forceful the military will be less inclined to intrude into the civil sphere (1962, p.40)" than if they appear demoralized or bewildered. Janowitz sees this self-assurance as "a defence against doubt and lack of self-esteem (1960, p. 230)." 11. This repugnance for politicians is most forcefully expressed by Gamel Nasser: "Every leader we came to wanted to assassinate his rival. Every idea we found aimed at the destruction of another . . Personal and persistent selfishness was the rule of the day. The word 'I' was on every tongue. It was the magic solution of every difficulty and the effective cure for every malady (Nasser, 1959, pp. 33-4)." 12. "Considerant qu'apres deux ans, ces responsables politiques par des luttes d'influence nefastes aux interets superieurs du Dahomey, ont demontre leur incapacity de conduire le pays vers des lendemains meilleurs (Colonel Soglo, Afrique Nouvelle, Jan. 5, 1966)." 13 Putnam found a strong positive correlation between the size of the Latin American military budget and the tendency to intervene. Fossum (1968) has explained this correlation by the fact that low military budgets pushes younger officers to stage military coups. 14. There is also confusion as to the exact nature of the link between military professionalism and military coups. According to Sayeed (1968) and Pye (1962a) the Pakistani and Burmese armies have been led to political intervention because of their high"degree of professionalism. On the other hand, Fossum (1968) and Johnson (1964) have suggested that professionalism is not related to the military's tendency to intervene but that it simply modifies the style of the political interventions. Thus professionalism often means less violence in the execution of the coup and more reliance on a military junta as a privileged form of military government. Finally Huntington (1968) suggests that military professionalism will decrease the tendency to intervene. 15. The following discussion is based on Luckham (1969). 16. "The effort to answer the question: 'What characteristics of the military establishment of a new nation facilitate its involvement in domestic politics?' is misdirected because the most important causes of military intervention in politics are not military but political and reflect not the social and organizational character istics of the military but the political and institutional structure of the society (Huntington 1968, p. 194)." A similar approach is shared by Needier who states that it is scientifically more rewar ding to explain coups d'etat "functionally rather than genetically, in terms of factors external to the military rather than in inter nal characteristics of the military establishment (Needier, 1968, p.60)." 53 17. The importance of political party institutionalization in the poli tical development process has been a recurrent theme of the develop ment literature. Olsen (1968) mentions a strong party organization as a determinant of development. Similarly Black (1966), Cutright (1965), Hopkins (1969), Pye (1966) and Ranis (1969), all insist on the existence of specialized and differentiated political institu tions, including political parties, as pre-conditions of a dynamic and balance development process. For a cross-national survey of the role of political parties in political development, see LaPalombara (1967) and Weiner (1967). 18- In his empirical study of Latin American militarism Putnam (1967) finds that interest aggregation by political parties and party stability were strongly and negatively correlated with military coups. However, interest articulation by parties and groups did not inhibit or encourage military intervention. Putnam's findings partly confirm Snow's (1966) hypothesis that interest aggregation by political parties is a more important pre-condition of political development than interest articulation. 19. In the case of political participation, Putnam (1969) shows that it is not correlated positively or negatively with military coups. On the importance of political participation on the development process see Lerner (1958), Saicedo (1968), Scalapino (1968) and Steiner (1968). 20. This dichotomy between "center" and "periphery" was first used by Shils (1962). 21. The link between social conflicts and military intervention is also suggested by Germani (1961), Liewen (1962) and Needier (1963). But Putnam (1967) finds no empirical link between domestic violence and military coups in Latin America. 22. In almost every African military coup a foreign conspiracy has been suggested: the French have been held responsible for the Malian coup of 1968 (Murray, 1966); President Houphouet-Boiny of the Ivory Coast for the 1966 coups in Upper Volta and the Central African Republic (Murray, 1966); and President N'Krumah for the first Nigerian coup and for the 1967 Sierra Leone coup (Barrows, 1968). 23. Throughout this study the name Congo-Kinshasa is used rather than Zaire. 24. This view of military coups as inevitable and automatic is well expressed by Young (1969). 25. On the events surrounding the Sierra Leone coup of 1968, see Barrows (1968) and Cartwright (1968). 54 26. On the Gabon attempted coup see Gabon (1964, 1968). 27. There are only two empirical and cross-national studies of African military coups (Dacks, 1968; Mapp, 1970) both unpublished. Both are unfortunately hampered by methodological problems which limit their usefulness. Studies concerned with the role of the military in developing areas have tended to by-pass Africa because "of the well-known statistical effect of artificially inflated relation ships when a large number of cases cluster at one or both ends of the regression line (Cutwright, 1963, p. 255). 55 CHAPTER II THE MILITARY AND ITS ENVIRONMENT: A FRAMEWORK FOR ANALYSIS The basis for this framework is to conceptualize the military in a dual manner: first, as an open system continuously interacting with its environment and second, as a social institution participating in the power structure of the society. My aim is to show that military coups occur when transformations in the political and socio economic environment modify the exchange process between a military organization and other sectors of the social system. As a result of these transformations, the military may be forced to modify its institutional role, notably by resorting to a coup in order to restore the stability of its exchange process with the environment. In addition, the framework suggests that the level of military intervention existing in a given country will depend on the nature and the intensity of the environmental changes taking place in the economic, social and political sectors. The Military Organization as an Open System and an Institution ' It is only recently that the military has been studied from the 1 perspectives of organizational and institutional analysis. Social scientists were originally most concerned with the character istics and motivations of recruits. This orientation was predominant in the research conducted up to World War II when Stouffer's massive study of American soldiers led to the conclusion that the concepts of 56 morale, attitudes and motivation were too limited to allow a full under standing of combat behaviour. We are forced to the conclusion that personal motives and relationships are not uniquely determinate for organization in combat. . . . Officers and men must be motivated to make the organization work, but not all of them have to be so motivated. ... To put it another way, the best single predictor of combat behaviour is the simple fact of the institutionalized role: knowing that a man is a soldier rather tnan a civilian. The soldier role is a vehicle for getting the man into the position in which he has to fight or take the institutionally sanctioned consequences (Stouffer, 1949, p. 101). Stouffer goes on to suggest that military sociologists apply the concept of organization to the study of members of the armed forces. This concept of organization was not new to the social sciences, but since Weber's work (1957) organizations have been defined principally in terms of the pursuit of an "activity that must be continuously and consciously coordinated (Perrow, 1969, p. 167)." As a result organi zation theory has been concerned principally with the formal anatomy of organizations, that is the structural arrangements which facilitate 2 the pursuit of this activity. This structural approach has been successful in delineating some of the major aspects of the internal functioning of an organ ization, for example the interactions between formal and informal channels of communication or the patterns of bureaucratic control. But by neglecting the economic, social and political environment in which organizations operate, it may have distorted our understanding of a phenomenon such as organizational growth. For example, the 57 success of an organization has traditionally been credited to the quality of its executive, the efficiency of its managerial control and the extent of its communication network. In fact, the success of an organization is often simply the result of a particularly favourable environment (Dill, 1960). A firm which enjoys a monopoly in the production and distribution of a commodity in high demand is less dependent for its growth on finding the most efficient and effective ways of manufacturing its product than one which competes in an active and open market. It can afford a certain degree of flexibility and still remains profitable. In this study, rather than focusing on formal characteristics such as bureaucracies and communication networks, an organization will be defined as a goal-oriented system interacting with its environment through an exchange relationship. Organizations, including military organizations, tend to be goal-oriented. They are usually set up for the purpose of reaching some specific goals and most of their activities are connected in one way or another with meeting these goals. Etzioni (1964) points out that goals provide an organization with its orientation by depicting a future state of affairs. Specific goals also establish guidelines for the operation of the organization and provide observers with some criteria by which to judge the effectiveness of the organization. The existence of goals also provides the organization with legitimacy. As Parsons (1970) suggests, an organization, by striving to attain its goals, also performs a set of different functions for the larger 58 social system. For example, the production of doctorates may constitute one of the immediate goals of a university, but from the perspective of the overall social system the attainment of this goal also contributes 3 to the production and reinforcement of cultural norms. The legitimacy of an organization's goals, although taken for granted by most organ izations, is in fact conditional on the coincidence of these goals with a societal function. When this legitimacy is threatened the organiza tion itself is in danger. In the case of business organizations, market mechanisms, by creating a demand for the organization's product, automatically provide this legitimacy. But an organization (such as the tobacco industry) may still have to invest time and energy convincing the rest of society of the legitimacy of its product. Goals and legitimacy are not inherent to an organization. Few organizations have their own existence as their major goal or as the major source of their legitimacy. An organization's goals are usually located outside the organization, in some need of the overall social system. This "outside" is the organization's environment. The environment has not been completely ignored even by organization theorists concerned mainly with formal and bureaucratic structures. Weber (1957, 1958) has suggested that the emergence of capitalist and bureaucratic forms of economic organization is linked to three transformations taking place in the social environment: the rise of the Protestant ethic, the generalization of the money economy and the consolidation of the nation state. However, a majority of organization studies, particularly Schumpeter (1950) and Simon (1958) 59 see the environment as an "out there". For many observers, it serves as a catch-all category for any important but indeterminate factors not immediately connected with the formal structure of the organization. Even Von Bertalanffy (195o), a pioneer of systems analysis and its application to the study of organizations, did not foresee that processes taking place in the environment could themselves effect conditions of 4 exchange within the system. In this study it will be assumed that the environment is not I merely a residual category or a gigantic container (translated from Sales, 1970, p. 112)" in which an organization floats, but a concrete 5 and differentiated milieu animated by its own internal dynamics. The complexity and dynamism of this environment have a direct bearing on all aspects of an organization's life. Seen in this perspective, any organization, including the military, can best be described as an "open system" where the behaviour of the various components is inter related and where the system as a whole exists by virtue of a "trans action or transposition of energy or materials across its boundaries to and from the environment (Hutton, 1969, p. 31)." This transaction or exchange process involves two distinct 6 operations. First, every organization in the course of reaching for its goals "produces an identifiable something which can be utilized 7 in some way by another system (Parsons: 1970, p. 17)." This "identifiable something" or output may take different forms: material goods in the case of an economic enterprise, regulatory decisions in 60 the case of a government agency or intellectual goods in the case of a cultural organization. However, in all cases something is produced whose absence would modify the behaviour of some other organizations and would eventually affect the state of the global system. Second, the organization, in order to produce this "identifi able something", needs to mobilize and transform resources acquired from the environment. In other words, it needs an input. These resources are of two types: material such as land, labour and capital and non-material such as knowledge, legitimacy and support. The fact that both types of resources are limited places the organization in a competitive situation with other organizations which need these resources for their own input. Should either the input or output of an organization be modified by an outside force, its goals and internal configuration as well as its position within the environment will be modified. This concept of the organization as an open system interacting with its environment can also be applied to the military. However, the military, because of the special nature of its goal, that of finding the most efficient way of maximizing the potential or actual use of force, is in a somewhat different position than other organizations with which it has to compete for resources. Contrary to a business enterprise, individuals rarely join together for the purpose of establishing an army. Nor are military establishments created through an ad hoc process. Increasingly, 61 modern military establishments have become the conscious and exclusive creation of the modern nation state which also claims exclusive jurisdiction over its use. Thus the raison d'etre of the military organization is defined from outside the organization itself, usually by the government. As part of its task the military can be called upon to inflict death on individuals and damage to other organizations. This possibility also sets the military apart from civilian organiza tions. In the contemporary world the use of military force is at best considered a necessary evil, forced upon nations by conditions within the international system. Few other organizations operate in what Parsons (1970, p. 48), defines as an "ethically ambivalent" atmosphere. The "identifiable something" produced by the military is not like the automobile for the automobile manufacturer or education for the university. Armies are organized for the purpose of winning wars but they are not always at war. In fact when a war breaks out it is often assumed that the military has failed in its task of "producing" security. Because its product is not easily measured and is often defined in purely negative terms (security being the absence of war) the military organization has to live with a permanent question as to its utility, especially in time of peace. Yet, at the same time the survival of the entire social system may depend on the extent to which the military meets its goal and produces security. Few other organiza tions have such a vital role. No other organization, even other security-oriented organizations such as the police, functions in a similar context where there is a permanent threat to the survival of 62 its individual members. In no organization is the possibility of loosing one's life so central to the role of the organization. In fact this possibility is so important that it is the object of detailed planning and preparation at all levels of the organization. Because of the special nature of its ethical basis and its goals, the military is easily transformed into a multi-purpose organization formally set up both to prevent and to stage war against rival military establishments. An army can also be used in such disparate tasks as assisting the police in directing traffic or helping maintain civil order; it can also serve as a ready supply of skilled and unskilled manpower to be used in a variety of civilian-oriented tasks. Many features of the internal organization of the military are the result of all these special characteristics. A rigid and hierar chical internal stratification, a high level of discipline, an exten sive network of internal communications and an authoritarian distri bution of power and status are all the result of the military's 8 orientation to danger and violence. It should be stressed that these features are not the result of a conscious decision or choice by the military organization itself but are imposed on it by the nature of its goal. If the military is to survive as an organization dedicated to the maximization of the potential or actual use of force it must have a high level of discipline and a good network of communications. This imposition of certain organizational features determines in part the nature of the exchange process existing between the military 63 organization and its environment. First, an army requires a great many resources, both material and non-material, to carry out its task. Its organizational survival and indeed the survival of the entire social system may depend on the quantity and quality of its equipment. Because it is impossible to predict when it will be needed, an army, if it is to be useful at all, has to be maintained at a high level of readiness, making its financial support a heavy burden for the state. Second, an army will tend to look to its eventual opponents for guidelines as to the quantity of resources it will need and the most efficient way of transforming these resources. For example, interception fighter planes are only necessary if the potential enemy possesses an air force. The military continually has to adjust its needs according to its evaluation of the capabilities and intentions of its opponents, and to balance these against the capabilities and requirements of its client, the state. Third, while it requires a great many resources, the military can count on only one source, the state, to provide it with its most necessary resource financial support. Unlike a business firm, an army generates little or no income to insure it against sudden withdrawal of financial support. Because of the potentially catastro phic consequences of failing to support their army, governments are usually willing to allocate large amounts of money to it. Still, the size of this support is matched only by its precariousness and its remoteness from the military itself. 64 In terms of its legitimacy the military is also entirely dependent on the state. The military organization cannot generate its own legitimacy because of its ethically ambivalent nature; it must depend on the state's legitimacy. This second-hand legitimacy offers its only justification for its monopoly of a large share of the nation's financial resources. Thus,in terms of goals, legitimacy and internal functioning, the military is not an organization like others. Its goal is crucial for the survival of the overall social system, but it is also ethically ambivalent and a potential menace for society itself. Its legitimacy is second hand but it guarantees the state's own legitimacy. It exhibits a high level of coherence, rationality and efficiency in its internal functioning but it can easily be disturbed by outside factors. In fact the position of the military organization in relation to its political environment is highly paradoxical. On the one hand the military constitutes the supreme guarantee of the state's survival, but on the other hand it is entirely dependent on financial support from the state for its own continued existence. The army normally ensures the state's survival but cannot insure its own, since its goals, legitimacy and resources have to come from outside. There is an obvious discrepancy between the importance of the output of a military organization and the uncertainty of the resources with which society provides it. However, any organization, especially the military, is not only an open system at the mercy of its environment, it can also 65 influence that environment. According to Touraine (1969), an organiza tion should not be defined exclusively by its exchange relationship with its environment, but also by its active penetration of the 9 political system. Organizations such as the military are not only concerned with the most efficient and effective way of producing an "identifiable something" but also constitute key elements in the overall structure of economic and political power. According to Touraine, the organization participe au systeme du pouvoir economique et par consequent interprete les orientations culturelles d'une societe en fonction des interets de la classe dominante. L'entreprise a un pouvoir et cherche a. imposer en elle et hors d'elle des conduites sociales et culturelles conformes a ses interets de classe (Touraine, 1969, p. 254).10 As organizations, the firm, the university or the military, are vulnerable to changes in their environment. But as a social insti tution, they help shape this environment by participating in its struggles and its conflicts. The institutional importance of an organization is partly determined by its importance for the maintenance of the overall social system, and partly by its usefulness to those groups and classes competing for political control of this system. As the institutional importance of an organization within society increases, a distinction between its internal administrative management and its institutional management emerges. The internal managers on the one hand, are concerned with an effective and efficient mobilization of resources to meet the organization's goals. Their preoccupation is with the internal workings of the organization. 66 The institutional leaders, on the other hand, are mainly concerned with the role played by the organization in the political and socio-economic environment. Under their leadership the organization becomes a poli tical agent in its own right. This institutional vision of the organization stresses the "political" rather than the "managerial" dimension of the organization. But the two are never completely disassociated. The fact that organiza tions are also institutions not only involves them in outside conflicts it can also introduce conflicts within an organization itself. Eventually the institutional role of an organization may come to threaten the corporate survival of the organization, for example, when a business enterprise or a university sisi associated with the unpopular side in a political controversy. The major responsibility of the insti tutional leaders is to prevent such misjudgements while using the insti tutional weight of the organization to protect or even increase its share of input resources. As an institution, the role of the military is a crucial one. Unlike a business firm, the military is not only one element in the grid of economic and political power but also the umbrella under which this grid is developed and maintained by the groups and classes who benefit from it. In Poulantzas' terminology (1969) the army is the most important of all the "appareils d'etat". The army is never a neutral and benevolent spectator of the exercise of political and economic power. In fact political life or what Easton (1965) calls the "authorita tive allocation of values", always takes place in the shadow of the army's presence. Because it exists and because it is 67 a multi-purpose organization, the military can be used by those groups which control the allocative process. This possibility alone makes the army a vital institution. The fact that the military is perhaps the most vital of all social institutions only accentuates a paradox mentioned earlier. As an organization, the military is more dependent for its input on its environment than a business firm or a sports club. It cannot itself generate the legitimacy and resources needed to achieve its goal. However, an army can always make use of its institutional weight to redefine or increase this input from the environment while a sports club cannot. Because of its importance as an institution, the military organization benefits from a range of possibilities not normally available to other organizations which arejless vital as institutions. The military coup is one such possibility. These possibilities compensate for its dependency on the environment for financial resources and legitimacy. Furthermore, the military institution, because of some of its organizational features (esprit de corps, discipline and communications), can easily transform its institutional weight into effective action to modify its environment at will. This organization-institution perspective suggest that the relationship existing between the military and its environment is not as simple and one-sided as the organizational and developmental approachest-discussedlincehapterJlwwould indicate. The military is neither an all powerful organization, completely isolated from the 68 rest of society, but which intervenes in the political process to save society from a total breakdown; nor is it only an agent of those political and social forces which shape this environment in such a way as to make military intervention inevitable. Both aspects are present and both contribute to the complexity of the relationship between the military and its environment. Military Intervention as an Institutional Response The organization-environment exchange relationship, like all exchange relationships, is highly unstable. The goals of an organization can easily be affected by forces outside of the organization. For example, the discovery of a new medical cure can deprive a volunteer 12 organization of its raison d'etre. Similarly, the decision of a government to finance new social services can have an immediate impact on organizations which have previously provided these services. In some cases such environmental changes may lead to the disappearance of the organization, but in most cases the vested interests in maintaining an existing organization are so strong that new goals will be devised to give the organization a reason for continuing to exist. Changes in the environment can also affect the internal functioning of an organization. Because of a new technological development or because of changes in the political arena, an organization may be forced to rearrange its internal functioning if it is to preserve its legitimacy. This is especially true in the case of business 69 firms which continuously have to accept new regulations as to the remuneration and working conditions of their personnel. Failure to comply with these rules may bring a withdrawal of legitimacy or of human and material resources necessary for the functioning of the organization. When confronted with changes which can affect the stability of its transaction process with the environment, an organization must adopt a strategy which allows not only for its continued existence but also for the preservation of its autonomy. Thompson and McEwen (1958) suggest that the first task of every organization should be to recognize environmental changes accurately and quickly enough to allow time for the formulation and implementation of an adaptive strategy. The next task should be to evaluate the external support on which the organization can count to implement this strategy. Other organizations operating in the same environment constitute one possible 13 source of external support. In some cases, the government, in its legislative capacity, offers another source of support. Bargaining for an alliance with other organizations is often preferable to open competition as a strategy. Before contracting any such alliances, an organization must make sure that by increasing its reliance on other organizations, it will not decrease its own margin of autonomy. Another possible 70 strategy is to modify the organization's goals, to take into account a new technology or a new set of public demands. Still another is simply to allow the organization to fade away, since its product is no longer in demand. In some cases an organization will resist any modification of its goals, since it is felt that any change will destroy the nature of the organization. Instead, the organization may adopt a competitive strategy and attempt to minimize or even reverse these environmental changes. In the case of a military organization, the coup d'etat constitutes such a strategy. When its corporate survival is threatened by changes taking place in its environment, a military organization, specifically the officer corps, may use its institutional weight to modify this environment. By doing so, the military extends beyond its traditional organizational boundaries and creates a new role for itself as an institution. The format of military interventions varies. In some cases the military may want to gain full control over all aspects of the environment, not just the political area. In this militaristic type of intervention, the armed forces will attempt to infiltrate and reshape all sectors and 14 all levels of civilian life. However, more usually, the military, after seizing power, limits its control to the political sector. As a result, political parties and the legislative branch are either frozen into inactivity or abolished. The existence of a military regime is acknowledged, but usually only on a temporary basis. A third type of 71 military intervention leads to a civilian-military regime, where the coup d'etat brings about the replacement of the current government elite by a military junta or by a combined military and civilian executive which administers the country on a temporary basis. Occasionally, a military coup will be staged as a veto against the accession to power of a civilian group considered undesirable or the implementation of a policy unacceptable to the military. Often the military will intervene in the political process without staging a military coup simply by bringing sufficient pressure to bear on the civilian leaders to force them to conform to the military's demands. Whatever the form of military intervention, the major reason for such a move is the military's desire to alter its environment. Military coups are not a quasi-automatic result of the existence of an all-powerful military organization^ as the organizational approach taken by some military sociologists would imply. Nor are they necessarily the consequence of a breakdown in the political process, as suggested by many political scientists. Military coups tend to occur when changes in the political, social and economic environment threaten the stability of the exchange process between the military organization and its environment. In such situations the military will use its position as a social institution to redefine the environment and re-establish what it judges to be a normal exchange process with its environment. 72 However, if environmental changes are a necessary condition for military coups, they are not sufficient. They fail to explain why some military organizations have resorted to coups while others have not, or why coups have been more frequent in certain countries than in others. In other words, environmental changes, especially in Tropical Africa, are a generalized phenomenon which affect all military organizations but do not always produce the same result with the same intensity. The decision by a military organization to use its institutional weight and the propensity with which it will do so depend on two sets of parameters. A first set of parameters determines the vulnerability of the military organization to environmental changes and as such influence its decision to intervene actively in the political arena. Four such organizational parameters are singled out: the nature of the organization's output; the stage of development reached by the organization; the permeability of its boundaries; and the level of integration of its sub-units. These four parameters can be considered as the major components of the process which leads the military to intervene. Two other parameters — the complexity and the nature of environmental changes — are more immediately linked with the facility with which the military can carry out its decision to intervene and with the level of military intervention existing in a given country. As we can see these two sets of parameters are closely associated with the two dimensions of military intervention (process and level) singled out in Chapter I. 73 Military organizations will not always resort to military coups even if their corporate survival is equally threatened by changes taking place in their environment. There must also be a set of conditions which make possible the successful staging of such a coup. These favourable conditions are often created by the same environmental changes which threaten the survival of the organization in the first place. These changes not only affect the exchange relationship between the military organization and its environment and ultimately force it to redefine its institutional role, they also create a set of conditions which may facilitate the application of this new role. For example, a period of economic stagnation can deprive a military organization of financial support and thus prompt it to use its institutional position to reshape its environment. However, this same economic stagnation may create a situation of social crisis which facilitates the execution of a coup. In short, we are proposing to distinguish those factors, mostly organizational, which explain the process of military intervention from those factors, mostly environmental, which explain the level of military intervention. The Organizational Parameters of the Process of Military Intervention 1. The nature of the organization's output A first parameter is the nature of the "identifiable something" produced by the organization. The more intangible this output and 74 the harder it is to measure objectively, the more difficult it becomes for the organization to evaluate accurately the impact of environmental changes on its product and to adapt itself to new conditions. For example, it is far easier for an automobile manufacturer to assess and adapt to environmental changes such as a set of new anti-pollution regulations, than a university or a military organization. Because of this uncertainty, an organization which produces an intangible output will find it difficult to distinguish between those environmental changes which constitute a real threat to its corporate survival and those which are only peripheral to its continued existence. Consequently, it becomes vulnerable to even the most inconsequential changes in its exchange process with the environment. 2. The stage of development of the organization A second parameter which needs to be taken into account is the stage of development reached by an organization. Organizations usually possess boundary-spanning structures (Thompson, 1962, p. 309) which facilitate the transaction process between the organization and its environment and are an important means through which the organization receives and adapts to environmental changes. Organizations may even develop special adaptive sub-units whose responsibility it is to predict the direction of environmental changes, their probable 75 impact on the organization and the best organizational response. Research and development or public relations units are normally concerned with such tasks. During its earlier stages, an organization is more vulnerable to environmental changes since it is unlikely to have developed and experimented with those specialized roles and sub-units which serve to institutionalize relationships with the environment and to predict future changes. A military establishment with a long tradition of service will often be less vulnerable to environmental changes. The more important the role the armed forces has played in the history of a country, the more difficult it is for the political authorities to modify its status and its role. For example, an army which played a crucial role in a struggle for national independence is less vulnerable than one which was created after independence was achieved. On the other hand, armies which have evolved from guerrilla bands have not usually developed any military tradition of their own and as such will often be the first victims of the post-independent political stabilization process. 3. The permeability of the organization's boundaries The extent to which environmental changes affect the legitimacy, goals and functions of an organization also depends on the nature of the organizations's boundaries. The boundaries 76 can be identified by the set of procedures and beliefs which differentiate organizational from non-organizational elements. An organization's boundaries act as a filter through which environmental changes are channelled and interpreted to the organization. For example, recruitment is one of the most important channels through which the environment influences an organization. Often, organizations such as the military, which attempt to ignore certain social and political problems affecting their environment, for example, racial or political tensions, find these same problems introduced into their organization through recruitment. Organizations differ in the degree to which their boundaries are permeable. These differences, in turn affect the impact of environmental changes on different organizations' structure and behaviour. The permeability of organizational boundaries varies from those organizations with extremely rigid boundaries, such as military organizations, to organizations with very open boundaries, such as the Canadian Liberal Party. In the first case: there are specific and rigid procedures governing the entrance and exit of members of the organization (induction, training and discharge procedures). On the other hand, it is characteristic of a cadre political party such as the Canadian Liberal Party that it has very loose criteria for entrance to and 16 withdrawal from the party. 77 Naturally the extent to which the original boundaries of the military organization have been permeated will influence its vulnerability to environmental changes. A military establishment with a well-developed sense of corporate identity and where the recruitment and training programmes have not been contaminated by political considerations will adapt more easily than one whose boundaries have been blurred by the inroads of civilian considerations. While allowing the input-output exchange to take place, an organization must also maintain its integrity as a system by preserving a capacity for autonomous decision. Its boundaries must not be so permeable that it loses all control over the exchange process with the environment. On the other hand, if its boundaries are too rigid, it will become isolated from its environment. Such isolation may prove beneficial for a short time but in the long run it can only be detrimental to the organization's capacity to assess the significance of environmental changes correctly and thus select the proper response. 4. The integration of sub-units in the organization According to Lawrence and Lorsch, the successful organization is one which succeeds in maintaining a balance between internal differentiation and internal cohesion which is "consistent with the 78 diversity of the parts of the environment and the required inter dependence of these parts (1969, p. 134)." An organization with a high level of co-ordination and integration among its various sub-units will usually be more successful in interpreting and adapting to environmental changes. Information about the environment can be gathered and assessed more efficiently in highly integrated organizations. Because its sense of purpose is not undermined by a lack of internal communication or by contradictory reactions from different sub-units, the organization is less vulnerable to rapid environmental changes. In short, we are suggesting that an organization with an intangible output, with a recent history, with permeable boundaries and a low level of internal integration is likely to be affected by environmental changes to a greater extent than an organization with a well-identified product and a long history which has provided it with highly institutionalized boundaries and a high level of internal integration. The Environmental Parameters of the Level of Military Intervention Turning to those factors which affect the level to which an organization will make use of its institutional weight to redefine its environment, we have identified two such factors: the complexity of 79 the environment and the nature of the environmental changes. 1. The complexity of the environment Not only is the environment a distinct entity from the organization, it can also take on different meanings depending on how it is perceived by an organization. Tagiuri and Litwin (1968) suggest that there is a subjective and objective environment whose perspection depends on the cognitive map used by each organization to interpret this environment. In fact the environment alone makes little sense for an organization unless it is perceived in a particular way. As Barnard points out, the environment to be intelligible "requires a basis for discrimination, for picking out this and that as pertinent and relevant (1968, p. 196)." This process of interpretation identifies those aspects of the environment relevant to the organization "depending upon conditions internal to the organization and upon variations in the environment itself (Barnard, 1968, p. 14)." However, Terreberry (1967) points out that- the more reactive and turbulent the environment, the more difficult it is for an organization to evaluate correctly the scope and significance of environmental changes. As a result, every organization faces a dilemma in deciding to what extent it should adapt and to what extent it should stay put in the face of environmental changes. These decisions are 80 particularly difficult to make in the case of organizations like African armies, with little organizational history, few efficient boundary-spanning structures and a very turbulent environment. They are more likely to make the wrong decisions and over-react to environmental changes. As a result they tend to intervene repeatedly in the political process to modify an environment which becomes increasingly complex and turbulent as a result of their interventions. These frequent interventions only serve to accelerate the rate and complexity of environmental changes which increase the strain on the boundary-spanning structures for a correct interpretation of the 18 new environment. Consequently it will become more difficult to differentiate between those changes which are important for the organization and to which it must adapt, and those sudden but temporary modifications arising out of a turbulent environment. 2. The nature of environmental changes The most important factor in explaining the level of military intervention is to be found in the nature of the specific environmental changes rather than in the complexity of the entire environment. Naturally, not all changes in the environment affect an organization with either the same intensity or in the same way. In the case of military organizations, changes in the technological environment are often those with the most immediate and through impact. Janowitz (1970) 81 suggests that in developed countries, the distinction usually-made between civilian and military organizations has been increasingly blurred as a result of a new weapons technology "which vastly expands the size of the military establishment, increases its interdependence with civilian society, and alters its internal social structure (Janowitz, I960, p. 12)." As a larger percentage of the national income is spent on the military and as weapons of mass destruction equalize the risks of warfare for both the civilian and military population, considerable pressures are exerted to open a nation's military and defence policies to general political debate. As wars become more expensive, and the civilian population more directly concerned, the military ceases to be isolated from the arena of partisan politics. The net result of these technological changes has been a deterioration of the military's organizational boundaries. As its need for competent technicians and managerial leaders increases, the military has become more dependent on the quality of the education system and the level of technological development of the civilian sector. In the past, the boundaries of the military organization were protected by the distance which the military maintained from the rest of society. Military camps, military tribunals, schools and uniforms all served to isolate the army. Today these barriers are no longer adequate in the face of the military's need for new skills and new technology. 82 A sudden modification in the configuration of international politics is another environmental change which has profound repercussions on the military organization, especially with regard to its goals. The institutional importance of the military is in part determined by the army's importance as an instrument of foreign policy. The increase in arms control agreements and the decreased importance of numerical military strength has reduced this insti-tuional importance. When the state operates in a friendly inter national environment, or at least one in which there is opposition to the use of force as a means of settling disputes, political leaders tend to question the value of an expensive army. In these circumstances, pressure will develop for a redefinition of the goals of the military organization, moving towards internal security or civic action projects. Changes in the internal political and socio-economic environment of a nation can also have an effect on the military. For example, participation in an unpopular foreign war, such as the French-Algerian conflict, can raise the question of the army's legitimacy and make it into a target for political debate. Similarly, a decision by the political leaders to make alternative use of the army, manpower and skills, for example, to build roads, would influence the army's image of itself and also its image among the general population. In 83 addition/ , a decision by the political leaders to use the army as a supplementary police force in cases of civil conflict will again modify the image of the army and also activate latent ethnic and political conflicts within the military organization itself. In other cases, a government dedi cated to rapid and extensive social change can result in a decrease in the government's financial support for the armed forces, or a redirection of its goals to more civilian lines such as road building or customs duty. The environmental changes which can have an impact on the military are not limited to the political arena. Economic and social transformation may also have a disturbing impact on the military organization. For example, a deterioration in the economic situation can force political leaders to reduce the financial support of the military. Alternatively, economic stag nation can compel a government to increase the size of its armed forces to relieve unemployment. In states where a high level of .tension exists between two ethnic communities, political leaders can impose a policy of equal represen tation of the two groups within the armed forces which in turn influences the military's internal cohesion. In developing countries, a sudden change in the technology of warfare, like the advent of tactical nuclear weapons will have little impact on the military organizations. In developing states, particularly in Africa, military organizations exhibit little vulnerability to technological innovations because of their limited use of sophisticated weapons technology. They tend to be more concerned with changes taking place in the socio-political envir onment. As a result, a decline in the country's economic growth will have more impact on the goals, legitimacy and internal arrangement of an African 84 military establishment of 1,000 men than the successful testing of a new anti-missile missile for which an African army will never have any use. Furthermore, because of their small size and short history, Third World armies have not developed specialized sub-units whose responsibility it is to predict and assess environmental changes. On the other hand, a large and modern military establishment, such as the American one, is likely to possess boundary structures and 19 skills to assess and adapt to a changing political environment. Although we have stated that in the case of African armies changes taking place in the social, economic and political environment are more important to determine the level of military intervention than changes in the technological or world political environment, we will not discuss at this stage the specific kind of environmental changes which are likely to increase the military's propensity to intervene. This question will be taken in Part III when we proceed to test the causal impact of six characteristics of the African environment variables on the level of military intervention, namely economic development, social mobilization, political participation, party institutionalization, government penetration and internal conflict. As this approach suggests, the major causes of military coups are not to be found exclusively in the environment, or in some of the 85 organizational features of the military, but in the changing nature of the exchange process taking place between the two. In fact, military intervention can be said to involve a dual, although parallel causal process where both environmental and organizational factors when taken together, can be considered as necessary and sufficient conditions to explain the phenomenon of military intervention. The organizational parameters can be considered as the immediate causes of the military's decision to intervene in politics in the sense that they are the factors most closely connected with the structuring of this decision. Environ mental parameters on the other hand do not act in the same way. For example a low level of economic development cannot be said to cause the military's decision to intervene in the political process in the j same way as a governmental decision to abolish the armed forces. Environmental factors such as those identified in chapter IV contri bute to define at a more systemic level, those conditions in which the military is more likely to take and implement its decision to intervene. This approach has a number of advantages over the two outlined in Chapter I. Not only does it allow for an integration of the organiza tional and environmental factors, it is also more dynamic, since it 86 views military intervention as but one step in an on-going process of interaction between society and the military organization. It also allows for a distinction between the causes behind the decision to intervene and those conditions which facilitate the implementation of this decision and increase the army's propensity to intervene. For example, the military can decide to intervene out of a feeling of despair and apprehension over its own corporate survival and at the same time benefit as an institution from its ability to carry out this decision. There are also several drawbacks to this approach. It assumes that all organizations including military organizations have goals which are distinct from those of the individual members of the or ganization. Even more basic is the assumption that all organizations are goal-oriented. Maniha and Perrow (1965) in their study of the "reluctant" organization remind us that certain organizations have no goals in the first place and that individuals may join an or ganization because they enjoy one another's company or because they want to take part in some activity which is really peripheral to the official goals of the organization. This approach also assumes that an organization can be consi dered "an indispensable manifestation of the historical man (Touraine, 87 1965, p. 181)" and as such should be analyzed as a social actor in its own right. According to its own rationality, this actor is said to defend some particular interests which do not always correspond to the interests of the individual members of the organization. It can be objected that such a functionalist perspective tends to give too concrete an image of the military organization and thus detracts from a clear understanding of military coups. Part II of this study will discuss the historical development of African military organizations. Applying the framework of analysis defined in this chapter, it will illustrate how the changes which took place in Africa during the 1880-1960 period affected the exchange process between African armies and their environment and thus prompted them to use their institutional weight to redefine their environment. This historical reconstruction of the interactions between the organizational and environmental causes of coups should allow for a re-evaluation of the view of African armies as strong, coherent and modern forces, presented by several military sociologists. Part III of this study will attempt to identify and relate those environmental conditions conducive to a high level of military intervention. A formal model will be constructed and tested for its capacity to identify those environmental characteristics which explain the different levels of military intervention in the states of Tropical Africa. 88 NOTES AND REFERENCES TO CHAPTER II See Lang (1962) for an historical overview of military sociology. Blau and Scott (1962) define the organization by the existence of specialized administrative staff. Selznick (1961) also points to the administrative integration of specialized functions as the central characteristic of every organization. For an historical overview of organization theory see Etzioni (1960) , Perrow (1970), Scott (1969) and Wolin (1969). For a structural approach to organization theory see Thompson (1961), Presthus (1962) and Krupp (1961). For example, Poulantzas (1969) and Touraine (1969) see the university as the central element in the society's reproduction system. Barnard (1938), Wiener (1949) and Snyder (1958) have all recognized that environments determine organizational behaviour but they do not specify the link between environmental processes and organizational behaviour. For an extensive review of the literature on organizational environment see Gelbach (1971). On the concept of environment in recent organizational theory see Emery and Trist (1965), Thompson (1967), Thompson and McEwen (1958) and Rice (1963). This concept is also applied in social psychology, see Chein (1954), Forehand and Gilner (1964), and Bates (1968). The literature on the exchange dimension of every social relationship is vast. See especially the works of Blau (1964; 1968), Dahlstrom (1968), Homans (1958) and Thibault and Kelley (1959). This presentation borrows heavily from the works of T. Parsons (1956; 1957; 1958; 1970). Parsons' organization theory is criticized in Black (1961) and Bourricaud (1955). This study will here not deal with these internal features of the military. Janowitz (1960), Huntington (1957) and Levy (1962) provide extensive descriptions of such features. This distinction between organization and institution is elaborated in Allport (1933), Martindale (1966), Selznick (1957) and Taylor (1956). 89 10. Touraine first presented this approach in his Sociologie de  1'Action (1965). Parsons (1951, 1952, 1961) also develops an action oriented sociological theory but in a different direction altogether more functionalist and less critical. See also Gouldner (1954) and Merton (1957). 11. Lowrau (1970) suggests that institutional analysis ("l'analyse institutionelle") is necessary for a full understanding of the organization: "Definir rationnellement une organisation par les services qu'elle rend ou est censee rendre n'est pas suffisant. II faut aussi tenir compte du fait que l'usine, ou la firme, produisent des modeles du comportement, entre-tiennent des normes sociales, integrent leurs usagers au systeme total (Lowran, 1970, p. 13)". See David Sills (1957) for a study of the impact of environmental change on an organization concerned with infantile paralysis. 13. The problems of inter-organizational co-operation and of organizational autonomy are explored in detail in Levine and White (1961) and Dill (1958). Organizational conflicts, not discussed here, are examined in Boulding (1962) and Pondy (1967). 14. Described by Lasswell (1941) in his analysis of the Garrison state. 15. This point was first made by Katz and Kahn. (1966). 16. The distinction between a cadre and a mass political party is taken from Duverger (1967). 17. Naturally, not all the factors which determine the vulnerability of an organization to environmental change are. included in the list of parameters. For example, the elasticity of the demand for an organization's product, the size of an organization, its past record in adapting to changes, are not specifically cons idered. It was felt that these additional parameters are either included in, or secondary to the six. parameters listed above. 18. Ruff (1970) and Hegland and Nylehn (1968) make a similar point although in a different context. 19. Since the concept of military professionalism seems applicable principally to the military organizations of Europe and North America, little use of it has been made in this study. For an overview of military professionalism see Janowitz (1960, 1969) and Van Doorn (1964). 90 PART II THE PROCESS OF MILITARY INTERVENTION: THE ORIGINS AND TRANSFORMATIONS OF AFRICAN MILITARY ORGANIZATIONS 90a CHAPTER III THE AFRICAN ARMED FORCES AS ORGANIZATIONS AND INSTITUTIONS 1880 - 1970 This chapter will make use of the analytical grid developed in Chapter II to examine the relationships between African military organizations and their political environment during the period 1880 -1970. It will try to show that one set of reasons connected with the decisions of African armies to intervene in the political process is to be found in their increased vulnerability to environmental changes. Consequently, military coups will be seen as the result of a decrease in the organizational capabilities of African armies and not of any organizational superiority as it is often presumed. The Pre-Colonial and Tribal Military Tradition in Tropical Africa Although little is still known of the military configuration of pre-colonial Africa, it seems to have lacked any obvious set of militaristic elements which would help to explain the contemporary resurgence of militarism. While certain historical states, like the Songhah, Yoruba and Mossi, had developed permanent forms of military organizations, others, as in the case of Ghana and Mali, relied for 1 their defence on ad hoc military arrangements. Equally difficult to assess insofar as contemporary military coups are concerned, is the existence of an African military tradition at the tribal level. 90a CHAPTER III THE AFRICAN ARMED FORCES AS ORGANIZATIONS AND INSTITUTIONS 1880 - 1970 This chapter will make use of the analytical grid developed in Chapter II to examine the relationships between African military organizations and their political environment during the period 1880 -1970. It will try to show that one set of reasons connected with the decisions of African armies to intervene in the political process is to be found in their increased vulnerability to environmental changes. Consequently, military coups will be seen as the result of a decrease in the organizational capabilities of African armies and not of any organizational superiority as it is often presumed. The Pre-Colonial and Tribal Military Tradition in Tropical Africa Although little is still known of the military configuration of pre-colonial Africa, it seems to have lacked any obvious set of militaristic elements which would help to explain the contemporary resurgence of militarism. While certain historical states, like the Songhah, Yoruba and Mossi, had developed permanent forms of military organizations, others, as in the case of Ghana and Mali, relied for 1 their defence on ad hoc military arrangements. Equally difficult to assess insofar as contemporary military coups are concerned, is the existence of an African military tradition at the tribal level. 91 Certain tribes like the Kipsigis and Shambalai of Kenya or the Nyakyusa of Nyasaland had very little use for wars and heroic virtues. Other groups like the Mende of Sierra Leone, the Kuanyama and the Ngoni of East Africa were originally established as military societies, and soon came to regard war as a permanent state of affairs and fighting ability as an|important aspect of political leadership (Peristany, 1939; Wilson, 2 1951;,Wianas, 1962). Thus evidence as to the existence of a pre-colo-nial or tribal military tradition in African is still not clear. Until new knowledge is uncovered, the distant pre-colonial past can contribute little to our understanding of the post-1960 phenomenon on coups in ' 3 Tropical Africa. The colonial experience is far more productive as an explanatory tool. The African Colonial Forces and European Domination (1880-1945) After the works of Rothberg and Mazrui (1970), Cartey and Kilson (1970) and Ranger (1967), the view that the European conquest of the African continent was achieved because of the Europeans' moral superiori ty coupled with the Africans' secret desire for subjugation can no 4 longer be supported. While it is true that the Africans could not prevent the European conquest (Coison, 1969)^ in a number of instances they did offer strenuous and effective military resistance. For example, it took eight campaigns and 84 years for the British to crush the 5 Ashanti resistance. Equally violent military confrontations erupted between the colonial invaders and the Emirates of Northern Nigeria, the slaving chiefs of Malawi and the various tribes of Eritrea and 92 Western Angola. In the Sudan Samory Ture and an army of 20,000 men resisted for 28 years all French attempts at "pacification". But Samory's large-scale military resistance movement stands alone. Most other cases of violent resistance were limited to small scale operations 6 and ambushes. Instead of fighting, most African chiefs chose to use their military strength and the relative weakness of the European position to insure their own immediate survival and to secure political advanta ges for themselves through the elimination of rivals or the subjugation of neighbouring tribes. Such political considerations help to explain the willingness of many tribal chiefs to contribute men to the various 8 European military expeditions of the late nineteenth century. I These expeditions served to open the eyes of the Europeans as to the military potential of African soldiers. France had made some use of colonial troops in its eighteenth century battles against Britain, but the first systematic use of colonial forces only came with the crea tion of the Tirailleurs Senegalais in 1857. The Tirailleurs were used extensively in the conquest of West and Equatorial Africa where they 9 accounted for 80 per cent of the French casualties. In the Belgian Congo, the Force Publique, created in 1886, at the insistence of the Vienna Congress, initiated numerous expeditions from Katanga to the Nile. All in all 662 Belgian officers were killed during these military operations while the death toll for African 10 recruits reached the 10,000 mark. 93 In South West Africa, the Germans introduced general conscrip tion in 1896 in an effort to stamp out local military resistance. From 1883 to 1907, 383 violent clashes took place between the German forces and the African tribes resulting in the death of 3,348 European officers 11 and more than 15,000 African recruits. Unlike the other major colo nial powers Great Britain tried to keep the military confrontations with African populations to a minimum. It made little use of African recruits during its colonial conquest of Africa. However, when a show of force was inevitable, as in Nigeria, African rather than European 12 forces were used to assert the British presence. Having proved their usefulness in various campaigns for the conquest and "pacification" of Africa, colonial troops were then used extensively to maintain law-and-order in the newly conquered territories. In British West Africa the first African based military forces were raised not so much for conquest purposes but to fulfill police duties after the arrival of the traders and merchants. In the case of the Lagos Constabulary first raised in 1863 and of the Royal Niger Regiment, raised in 1894, the specific mission was "to keep order in the Protecto rate, enforce the Authority of the Government and suppress barbaric customs such as cannibalism, killing of twins (Geary, 1965, p. 105)." During the 1886-94 period alone the Niger Constabulary participated in 13 56 major "peace-making" operations. However, it was in the Belgian Congo that "native" police troops were used most extensively. In 1894 the strength of the Force Publique had already reached more than 10,000 men and received 45 per cent of the budget for the entire colony. From 1891 to 1905, 44,440 Africans were recruited in the Force Publique and 94 in the words of Janssens, the last commander of the Force, their devo tion to the Belgian cause was unquestionable: Jamais les troupes de la F.P. n'ont failli a leur devoir de retablissement de 1'ordre interieur; en aucune circonst&nces elles n'ont refuse de tirer sur la foule soulevee contre la loi (Janssens, 1960, p. 242). In addition to their contribution to the conquest and policing of Africa, colonial troops were also used in the on-going struggle between the colonial powers for a share of the African continent. In fact it was these rivalries which provided African colonial forces with their official acceptance by colonial military planners. For example, by 1897 it had become evident to the British Colonial Office that simple police forces could no longer deter the French who, having conquered Dahomey in 1894, were now moving closer to the Nigerian Protectorate. Plans were drawn up to organize an African regiment of British officers and African recruits. In 1900 the Royal Niger Company and the Lagos Constabulary were disbanded and their members incorporated in the Northern and Southern Regiments of the newly created Royal West African 14 Frontier Force. The first regular African colonial regiment in British Africa was thus born. The creation of African colonial regiments also served other purposes, especially in those territories where active military resist ance had been encountered or where tribal military activities had to be curtailed. In these territories the new colonial forces provided an alternate avenue for "unemployed warriors". For example, after the 95 Ashanti war of 1896 the British established a local military force made up of African recruits and British officers hopefully to prevent the 15 recurrence of further uprisings. Once their initial missions had been carried out, colonial authorities particularly in France, began to consider integrating the African regiments within the metropolitan defence structure. Already in 1900 and 1905 the French National Assembly had enacted legislation allowing for the recruitment of "natives" through a volunteer system 16 with the provision that conscription could be used if necessary. In 1907 a government commission recommended that North Africans be recruited for service in North Africa. Expanding on this recommendation, Colonel Mangin suggested that the French government seek to establish a Force Noire defined as an "African army whose camp would be in Algeria but with its reservoir in West Africa (translated from Mangin, 17 1910, p. 101)." In favour of the Force Noire, it was argued that Africa constituted an infinite reservoir of potential soldiers who could bring the French military strength up to par with the German which had profited from a higher birth rate. Second, France in the aftermath of the Dreyfus Affair, was experiencing a renewal of anti-militarism led by Jaures. This atmosphere made the idea of using Africans rather than Frenchmen as soldiers very appealing. However, the most important justification for establishing the Force Noire was the supposed superiority of the African soldier on the battlefield. 96 le systeme nerveux du noir est beaucoup moins develcjpe que celui du blanc . . . il est certain que nos noirs peuvent figurer sur n'importe quel champ de bataille. Ils y apporteront un calme et un fatalisme precieux dans la guerre moderne ou le grand danger est 1'extreme nervosity des peuples civilises (Mangin, 1909, p. 28).18 There was little opposition to the principle of a Force Noire, only discussion as to its proper use. Jaures, for example, considered the Force Noire as "a serious effort given by the Black masses organized under French discipline (translated from Ly, 1957, p. 35)." His only doubt came from the fear that such a force would be used in France against striking workers. In 1910, the French National Assembly adopted the project of a Force Noire and Mangin, returning from a fact-finding mission in West Africa, reported that anywhere from 8,000 to 40,000 Africans could be recruited annually for service in the Force. By 1912 it had become clear that even this minimum number could not be raised under a voluntary system and conscription was thus authorized. Any African resident of a French colony, between the ages of 20 and 28 could be required to serve for up to four years. Responsibility for providing conscripts was left to the local chiefs who used the opportunity to get rid of any local undesirables. During the 1890-1914 period military forces in the British territories continued to be organized on ad hoc basis with little attention being given to integrating them into the overall defence structure of the Empire. Military operations were kept to a minimum and the general practice was for British regiments to ear-mark specific 97 20 troops and officers for the colonies. It was for purely economic reasons and only on the eve of World War I that the British amalgamated local territorial units, first created from local police forces, into area units, for example the West African Frontier Force and the Kenya African Riffles. When World War I erupted, France already had a military structure equipped to handle a massive recruitment of African soldiers 21 and their rapid incorporation into the French*'military structure. In 1914 alone more than 30,000 African recruits were sent to Europe from French colonies. For the entire duration of the conflict 225,300 soldiers and 107,000 support personnel were recruited in French Equa torial and West Africa (85 per cent of all French colonial forces). For those units which underwent battlefield experience casualty rates 22 of 35 to 60 per cent were not uncommon. Following complaints by Diagne, an African who had been made Commissaire de la Republique, a few privileges were granted to African recruits: partial exemption from income tax, employment preferences, family allowances and priority in medical and agricultural schools. But these were not still suffi-. cient inducements to attract the necessary number of recruits. By 1918 a rising number of Africans were fleeing to neighbouring British territories where no conscription existed. The war effort in the British African territories was not as extensive as that in French Africa, nevertheless 87,000 Africans served as soldiers and close to 800,000 as support personnel with the British forces. Although relatively few Africans served outside of the continent, 98 approximately 6,000 African soldiers and 100,000 support personnel 23 serving in British regiments died during the conflict. During the inter-war period, the status of the French colonial regiments was modified as a result of the new strategic and political conditions created by the Peace Settlement. First, the Versailles Treaty explicitly forbade the organization of indigenous colonial armies (except for police duty) or the construction of military bases and forti fications in colonial territories now under mandate. Second, there was renewed resistance to the idea that African troops could effectively replace Frenchmen in the defence of their country. In the words of Azan, the idea of a mercenary army is "false and despicable. All nations which have used mercenary soldiers have disappeared and most often have perished by them (translated from Azan, 1925, p. 9)." The rise of the Pan-African movement under the leadership of Garvey also accentuated those fears of a revolt by native soldiers which could 24 engulf the Republic. Third, in 1918, Germany could no longer be considered an immediate military threat: its army had been reduced to 100,000 men, its air force and navy sharply curtailed and its colonies re-distributed among the victorious allies. Thus there was less need forf.a massive French army. Fourth, the rise of a new military doctrine which stressed the importance of firepower equipment rather than man power further undermined the demand for large land armies. Even more important was the fact that it was becoming increasingly difficult to meet the already much reduced quotacof African recruits. For example, in 1918 the contingent from Tropical Africa was to be 100,000 men. 99 In 1920 this figure was scaled down to 63,000 and in 1926 to 24,000. However, even this quota could not be met because of African passive 25 resistance. In 1924, 13 per cent of the conscripts failed to appear. In Upper Volta alone the figure for non-appearance reached 44 per cent. Many Africans saw military service as a disguised form of forced labour. A 1926 decree confirmed this impression by allowing the civilian autho-26 rities the use of military reservists for public works. Nor was the possibility for military advancement for Africans^'very encouraging. Until 1926, the African Non-Commissionned Officer (N.C.O.) could not expect to rise above the rank of lieutenant and then only after 15 years of service. After 1926, and under special circumstances, he could attain 27 the rank of captain. Under no circumstances could an African N.C.O. be put in command of European troops or in charge of an administrative unit. In the British colonies the post-war pattern was one of return to the pre-war situation. Both the West African Frontier Force and the Kenya African Rifles were returned by the War Office to the control of the Colonial Office where the main consideration was one of economy rather than military efficiency. As a result the W.A.F.F. went from a total strength of 5,740 soldiers and 152 officers in 1920 to 2,380 soldiers and 86 officers by 1933 (Bartlett, 1956). In 1931 it was decided, partially for financial reasons, to abandon the practice of appointing one native officer per company who was responsible for investigating complaints and administering small punishments. This change further reduced the attractiveness of the military profession 28 in the eyes of Africans. 100 During 1918-39 period the African regiments had little to do. Occasionally they were used on peace-keeping operations: against the Masai in 1918, against the Mad Mullah rebellion in 1919, in Jubaland in 1925 and against Abd-el-Krim in 1925. Strategically their role was seen as a reservoir of manpower for use in mass attacks. According to a directive from the Service Technique des Troupes Coloniales, the use of African soldiers should be limited to two types of situations: Si une mission de sacrifice s'impose, defense sans esprit de recul, pour procurer le temps necessaire a un regroupement des forces. Pour une attaque brutale visant la rupture du front defensif adverse, on peut employer des troupes noires en forte proportion, peut-etre meme en exclusivite (France, 1928, p. 134). In World War II, 170,000 Africans were inducted into the French Army but could not be used extensively because of the 1940 armistice. Even so, between 1942 and 1945, 100,000 Africans joined the Free French Forces and at one point they constituted 50 per cent of its manpower. The precarious military position of the Free French resulted in additional possibilities of advancement for African soldiers and N.C.O.'s. In the British territories 370,000 Africans were enlisted in the British war effort. Close to 40 per cent of those served outside their home 29 territories particularly in North Africa, Sicily and Burma. By the end of World War II the African colonial regiments could no longer be considered to have a particular strategic skill, simply because of their qualities of obedience and blind courage. New developments in warfare had made these qualities obsolete. Only when they were able to use the weaponry of modern war were colonial troops of any military value. 101 When applied to the 1880-1945 period, the analytical grid developed in Chapter II points to certain features of colonial African military organizations which are important for an understanding of their subsequent behaviour. First, these forces are of relatively recent origin. It was only in the late 1890's and early 1900's that they were organized on a permanent basis. The situation is thus radically different from that of Latin America (Alba, 1962; Johnson, 1964), the Middle East (BeEri, 1969; Fisher, 1963) and even Asia (Rakwijitt, 1965), where contemporary forms of military organizations have a much longer historical development. African armies are essentially young organizations and as such they are more vulnerable to environmental changes than organizations with a long historical tradition. Second, African colonial regiments were not developed as in Thailand and Pakistan (Khan, 1963; Wilson, 1962) on the foundations of any pre-colonial military apparatus. As organizations, African colonial regiments are the creation of the colonial powers. They have few or no roots in an indigeneous military tradition. On the contrary they were created and used extensively to eliminate traditional African military organizations. Even their short organizational history is entirely associated with what has come to be seen as the dark period of African history, the period of colonial domination. There is no glorious pre-colonial tradition to go back to; nor were the reasons for the creation of African armies particularly attractive. Raised to help in the conquest of Africa, these forces were later transformed into regular regiments when the need for para-police force became urgent. 102 The establishment and maintenance of European colonial rule over Tropical Africa would not have been possible without them. In the post-1945 age of African consciousness and negritude, the nature of the "identifiable something" produced by these early African military organizations did not provide them with a favourable image among the nationalist elites or even the population at large. Their creation became synonymous with the establishment of European rule. Because of this close association with the colonial presence and their control over the instruments of violence, African regiments were treated with respect by the African populations but they did not enjoy any great popularity. The servitude of conscription and military service, which often degenerated into forced labour only accentuated this indigeneous resentment which expressed itself in frequent revolts against conscription notably in the French Sudan, Dahomey and Volta 30 regions. Military service was not considered an opportunity to further one's status since the ex-soldier, when demobilized, was condemned "to a miserable and timid existence, or if he has acquired a vigorous mind of his own, and the custom of speaking loudly, he does not hesitate to exploit those who surround him (Buell, 1928, p. 19)." Third, not only were the goals and legitimacy of early African regiments provided by the colonial presence, but the internal arrange ments of these forces were also entirely dictated by the requirements of colonial military planners. More attention was paid to the need of integrating these forces within the metropolitan military structure 103 than to the welfare of the individual members or the African populations. No attention was paid to the ethnic and tribal proportions among the various ranks, the geographic origin of the recruits, their level of education, their training or the possibilities for African soldiers acquiring a general education and certain trade skills. In the recruit ment of African soldiers Imperial rather than territorial considerations were predominant, sometimes with curious results. For example, until 1914 army units in British Africa were made up of a mixture of Sikh volunteers, British officers and N.C.Qs, Indian skilled tradesmen and African soldiers. From Nigeria, Kenya or The Sudan as described by Gutteridge, the ideal soldier, according to European officers was "an illiterate, uncontaminated by mission education and from a remote area (1969, p. 9)." In West Africa warrior tribesmen from the Muslin regions of the interior were preferred to the animist or Christian converts of the coastal regions because they were judged to be more loyal to their European officers. Fourth, the regional rather than territorial basis on whicTi African troops were organized prevented them from developing any sense of close identification with any territorial unit or with a specific population group. Furthermore the fact that colonial regiments were administered from London or Paris and were constantly moved in and around Africa made such an identification even more difficult. As a result African regiments rapidly became strangers in their own land. They did not develop any close contacts with the new African elites. Their models for behaviour continued to be found in Sandhurst or Saint 104 Cyr rather than in Lagos or Dakar. Finally, if the African regiments were insulated from the Afri can reality they were also fully dependent on the evolution of European politics and European military technology for the definition of their role and for their internal arrangements. As the need for land armies subsided, the strategic usefulness of the African regiments also diminished. These regiments could be effectively replaced by police forces on exploration and peace-keeping missions. There was"therefore little need for the colonial governments to continue to spend large sums of money on their maintenance. As a result financial support was seriously curtailed and their role was re-defined, at least in the French territories, in the direction of civic action and public works. Also as the African political environment itself changed, the close identification between the colonial military forces and the colonial authority created new problems for these military forces. The global image of African regiments which emerges is one of organizations with little autonomy, and which are entirely dependent for their goals, legitimacy, resources and internal arrangements on extra-African factors such as the French-German rivalry or the arrival of mechanized warfare. Paradoxically this almost entire lack of autonomy allowed for both a high level of organizational modernity and the total absence of any institutional importance. Because of the absence of a powerful, indigeneous military tradition and because of the requirements of Imperial defence planning African military organizations were 105 structured along the lines of the most modern principles of military thinking. The massive presence of expatriate officers and the fact that these forces were not used to fight in Africa itself for long periods of time, prevented the rise of a dual, metropolitan and indigenous, military structure as was the case in India (Singh, 1952). From the start the format of African armies was essentially modern and underwent no adjustment to specific African conditions. There was no combination of modern and traditional elements. In return this isolation from the African scene made for the institutional insignificance of African mili tary organizations. Although originally indispensable for the purpose of conquest and pacification, African regiments became less and less significant. As a result they could do little to modify some of the environmental factors which were beginning to affect them in a negative way. Without any ressources of their own, their legitimacy tied to that of the colonial presence and their internal arrangements modelled on those of European armies, they were badly equipped to affect or even adapt to the changing African environment. Isolated as they were in the African social fabric, they could easily copy European military models but they could not develop any of the boundary-spanning structures necessary to interpret and adapt to the changes taking place in their own socio-political environment. In short African military organizations were ill-prepared to face the post-1945 period. 106 The African Colonial Forces and the Transition to Independence (1945-58) With the post-war defence reorganization of the colonial powers, the African colonial armies' loss of strategic importance was confirmed. At the same time they became more dependent for their financial support and organizational development on the evolution of territorial rather than European politics. Although the French, Belgian and-British differed widely in their decolonization practices, the final impact on the evolution of their colonial forces was quite similar. In all cases it led to noticeable decline in the organizational capabilities of the colonial force and increased isolation from the struggle for national independence. Great Britain emerged from World War II victorious, but finan cially exhausted. Its military responsibilities as a major member of the Allied coalition, its political leadership of the Commonwealth, and of non-communist Europe and its financial custodianship of the Sterling zone all exerted a great strain on its resources. To meet the challenge of the post-war military situation, Britain proceeded rapidly to a reorientation of its defence policy. At the time of his appointment as Chief of the Imperial General Staff in 1946, Field Marshal Montgomery (1958) enumerated the three objectives of this new British defence policy: an increased reliance on regional military alliances, preservation of the freedom of the seas and close control over the Middle East. No mention was made of the Commonwealth or of the British African territories. By 1950 any idea of an Imperial or Commonwealth defence organization along-the lines of the old 107 32 Imperial Defence Committee had been definitively abandoned. The colo nies no longer played any role in a global British defence policy. Their security would now have to come from their own resources and from the pooling of these resources on a regional basis. In Africa two conferences were called to implement this new orientation. In August 1951, delegates from the major colonial powers plus representatives from the Ethiopian, Egyptian and American govern ments met to discuss the problem of rapidly moving troops and supplies to and from Central and East Africa in case of war. The conference recommended an integration of the participants' military communications. Another conference of all British West African administrators was called in 1953. The British government made clear its determination to return to the 1939-45 policy, when the territorial governments had set up and financed some units of the W.A.F.F. The conference finally agreed that steps should be taken in each territory "so as to make it clear beyond doubt that its military forces are the national forces of the territory, established for the purpose of ensuring the immediate defence of the territory in an emergency and the maintenance of peace, order and good government in the territory (Great Britain, 1954, p.3)." This defence reorganization and the recognition that national independence was likely to be an inevitable rapid and hopefully peaceful process convinced the British defence authorities and the local territo rial governments to curtail further financial support of the African 33 regiments. Table I shows the application of this .policy. After 108 •: TABLE I MILITARY EXPENDITURE BY GREAT BRITAIN ON ITS AFRICAN MILITARY FORCES, 1950-59 (IN POUNDS) Year West Africa East Africa 1950 1,315,902 1,208,283 1951 1,30U,292 1,195,926 1952 1,267,082 1,162,U32 1953 2,180,573 1,723,396 195U 2,38ii,565 2,1*85,827 1955 2,6U9,786 2,856,971 1956 2,213,857 2,U62,876 1957 2,125,000 625,182 1958 1^,000 83,833 1959 20 Source: Great Britain, War Office, Army Appropriations Account, 1950-51 to 1959-60. 109 1953 the British contribution to the financial support of the African regiments rose only slightly. Meanwhile, as Table II shows, the territorial governments appeared unwilling to divert any major part of their resources to support these regiments. The British decision was apparently a reasonable one. The African colonial forces were thought to be sufficiently trained and equipped to deal with any incidents that might arise during the process of constitutional adjustment. Also it was highly unlikely that African troops would again be needed or indeed could be used effectively outside of Africa. Finally, since these forces would eventually be transferred to the new African governments there was little incentive for the British government to spend huge sums of money on them. The end of the war did not bring the same radical and immediate re-orientation of French defence and colonial' policies. In many ways the French policy was ambiguous. For example, although the Brazzaville Conference of 1944 had made it clear that independence for the African colonies could not and would not constitute an acceptable solution, the preamble of the 1946 Constitution stated that "France would lead its colonies to the freedom of self-administration and the responsibilities 34 of self-government." In 1946 the French imperial structure was re organized to form the French Union. This change increased the number of representatives from the colonies sitting in the French National 35 Assembly and in the Senate. In addition, the Union was provided with its own Assembly and a High Council where French Cabinet ministers and government representatives from the associate territories were to meet 110 TABLE II MILITARY EXPENDITURE BY LOCAL GOVERNMENTS IN SELECTED BRITISH TERRITORIES, 1955-60 (IN POUNDS) Year Kenya Northern Rhodesia Nyasaland Sierra Leone Tanganyika Uganda 1955 1 ,301,860 *- 3,232,252 12U,867 60U,283 723,032 1956 1,80U,162 1,700,000 3,853,529 13U,286 652,U53 775,508 1957 1,598,705 2,500,000 3,935,263 132,823 661,61;1 809,633 1958 1,171,926 2,600,000 U,229,606 132,823 675,211 727,891 1959 1,586,86? 3,300,000 277,121 657,2U8 756,212 1960 387,7U9 3,500,000 * 810,969 1*82, hho 802,530 Source: Great Britain, War Office, Army Appropriations Account, 1950-51 to 1959-60. * Unavailable. Ill to discuss the Union's policies. This constitutional arrangement could have provided African political leaders with a platform from which to discuss and challenge the military policies of the French government. However, the colonial conflicts in Indo-China and later Algeria coupled with the permanent political instability of the Fourth Republic prevented its effective operation. The new Assembly had only consultative powers and the High Council did not find the time or the incentive to meet until 1951. The African representatives in the metropolitan political structures tended to concentrate on furthering their own political careers or that of the political pa rties with which they had become associated. Since most of them were also members of the Communist or Socialist parties, they did not tend to concern themselves with military questions. Over the years there was very little discussion of military questions in the Assembly of the French Union. During the 1955-56 session, only eleven propositions out of a total of 395 dealt with military questions. In 1956-57 this number dropped to five. Most of these questions concerned the status of veterans, the implications of guerilla warfare and the overall defence plans of the French Union. There was no discussion of the military policy of individual terri tories, the influence of the transition to self-administration on military questions or the recruitment and training of African officers. Throughout the immediate post-war years France continued to view Africa as an important element of its overall military programme. 112 Speakers in the National and French Union assemblies repeatedly asked for a re-orientation of the French defence policy on the basis of an 36 Euro-African strategy. But this recognition of the geo-political importance of Africa was not translated in any preferential financial treatment for the African regiments. On the contrary, because of the two colonial wars, priority for equipment, facilities and training was given to those units which bore the main burden of the fighting and not to those units which stood guard duties in and around the French terri tories. It was accepted that African troops no longer had the military value they had had in the days of Mangin. Nor could they be used extensively in colonial conflicts because of the repercussions that 37 such a decision might have on territorial politics. The decision to use African soldiers in Indo-China, Algeria and during the Suez inter vention encountered strong opposition from several French political parties, particularly the Rassemblement du Peuple Francais which attempted unsuccessfully to rally the verterans organizations to its cause. As long as the African territories were an integral part of Greater France there was no financial distinction made between an African and a non-African regiment, except on the basis of its military usefulness. The general lack of interest of French and British military authorities in the fate of their African regiments had grave conse quences for the organizational capabilities of the African military forces. Without financial support from the mother country, African regiments could not increase their strength or modernize their equip-113 ment and base facilities. Nor could they acquire new types of armaments or launch new training programmes. With a few exceptions plans for the creation of African naval and air forces were postponed indefinitely. In the Federation of Nyasaland and Rhodesia the fighting strength of the regular forces was cut by 25 per cent between 1950 and 1957 for financial reasons. In the annual reports of the Federation's Depart ment of Defence, a lack of funds was held re