Explaining Sudan's economic and political turmoil

Soldier and State in Africa Author(s): Claude E. Welch Source: The Journal of Modern African Studies, Vol. 5, No. 3 (Nov., 1967), pp. 305-322 Published by: Cambridge University Press Stable URL: https://www.jstor.org/stable/158726 Accessed: 02-09-2019 23:54 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide

range of content in a trusted digital archive. We use information technology and tools to increase productivity and

facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

https://about.jstor.org/terms

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Modern African Studies

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

The Journal of Modern African Studies, 5, 3 (I967), pp. 305-22

Soldier and State in Africa by CLAUDE E. WELCH*

DIRECT military intervention, aimed at unseating civilian governments and replacing them with ruling councils drawn largely from the army, is a relatively recent phenomenon in Africa. With the exception of the Sudan, where officers led by General Ibrahim Abboud seized control in November I958, no supplanting of civilian authority by a military junta occurred until 19 June I965 (Algeria). Then, in rapid succession, the Governments of Congo-Kinshasa (25 November 1965), Dahomey (22 December I965), Central African Republic (I January 1966), Upper Volta (4 January 1966), Nigeria (15 January 1966), Ghana (24 February 1966), Nigeria once again (29 July 1966), Burundi (28 November 1966), Togo (I 3January I967), and Sierra Leone (23 March 1967) fell victims to coups d'Ytat. What factors affected the timing of these coups? What salient differences can be observed between the relative quiescence of African armies before mid-I965 and the subse- quent outright seizures of control? Might the forces that helped to topple civilian authority cause a subsequent downfall of military govern- ment, as occurred in the Sudan in October I964? Should we view army intervention with alarm, as inherently detrimental to the best interests of African states, or should we concur with a noted authority who suggested that 'frequent coups are a sign of change and progress' ?

Students of African political change almost totally neglected the role of the military until the uprisings made the omission distressingly obvious. Writings on African armies were practically non-existent.2 The major emphasis was placed, rather, upon 'charismatic leader- ship', 'institutional political transfer', 'mass parties', and similar slogans used by political scientists. Africa was the continent of the 'political kingdom' or the 'primacy of politics', not the continent of the army caudillo or frequent coups. Patterns of political change in Africa made knowledge of the military seem irrelevant. Two factors accounted for this lack of attention: the manner in which colonial territories

gained independence, and the historical heritage of African armies. Most African states gained independence through constitutional * Assistant Professor of Political Science, State University of New York at Buffalo. 1Samuel P. Huntington (ed.), Changing Patterns of Military Politics (New York, I962), p. 40. 2 To take one instance, a U.S. State Department bibliography, 'Role of the Military in

Less Developed Countries, January 1958-February 1964', contains references to four articles dealing with Africa, contrasted with 37 items on Latin America and 33 on the Middle East. There are no references to Africa in the bibliography in Huntington, op. cit.

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

CLAUDE E. WELCH

negotiation, through pressure exerted by party leaders against colonial powers relatively willing to withdraw. Self-government did not come through revolution. Tropical Africa (leaving aside the territories still under Portuguese control) had no Bolivar or San Martin, no Ho Chi Minh or Ben Bella. The so-called 'African revolution' thus differed

from many other great political changes: hegemony was handed over without large-scale civilian uprisings, campaigns of civil disobedience, or other techniques of political violence, in most cases.

African armies did not play a major role in nationalist movements. Standing forces were the creations of the colonial power; in the evolution toward independence, they did not participate directly, although the impact of demobilised soldiers on the growth of political awareness could not be denied.l But this impact was made by individuals, not by the military as a corporate group. Resentment against colonial rule was channelled through political parties, not through military uprisings.

In tracing the types of military intervention in Africa, one can distinguish three stages. The first stage, that of relative passivity and abstention from political interference, was usually confined to the immediate post-independence period. The armies remained under substantial expatriate influence, which precluded (or certainly made more difficult) any meddling in politics. In the second stage, resentment against European officers and African political leaders exploded in mutinies. These outbursts were not intended-at least not directly-to unseat the government in control. They were aimed rather at forcing the government to adopt certain policies, notably higher pay, pension privileges, or immediate Africanisation of the officer corps. Coups d'etat belong in the third level of military involvement. The occupants of presidential palaces were removed, possibly executed; into their offices moved the initiators of the intervention, intent on 'restoring' the country

1 A key political event in Ghana's political history, for example, was the march on Christiansborg Castle, led by members of the Ex-Servicemen's Union on 28 February 1948. Riots broke out after the marchers were dispersed (and two killed) by police gunfire; the outbreak of violence led to the Watson Commission Report, of tremendous significance in the hastening of constitutional progress. See David E. Apter, Ghana in Transition (New York, 1963), pp. I69 if., and Dennis Austin, Politics in Ghana, 1946-1960 (London, I964), pp. 73 if. Demobilised veterans in Nigeria helped carry the gospel of nationalism to the 'bush': more than 0oo,ooo Nigerians served in World War II, 30,000 of them outside their native country. See James S. Coleman, Nigeria: background to nationalism (Berkeley and Los Angeles, 1958), pp. 254, and Frederick A. O. Schwarz, Jr., Nigeria: the tribes, the nation, or the race: the politics of independence (Cambridge, Mass., 1965), pp. 57-8. Also see Donald S. Rothchild, The Efects of Mobilization in British Africa (Pittsburgh, Duquesne University, Institute of African Affairs), Reprint No. 2.

3o6

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

SOLDIER AND STATE IN AFRICA

to 'normal' patterns. To be certain, this categorisation is only approxi- mate, It is meant as a guideline for understanding and initial analysis.

MILITARY NON-INVOLVEMENT: THE POLITICS

OF DECOLONISATION

Independence brought changes in the political and administrative sectors of African governments far more rapidly than in the military sector. The pace of Africanisation differed dramatically. The replace- ment of European senior civil servants by Africans was a first priority for newly-independent states; this task was eased by accelerated retire- ment and pension schemes and by absorption into the domestic civil service of the former colonial power. In the armed services, however, the replacement of Europeans in command positions was relatively leisurely-at least until mutiny or political decision to diminish reliance on foreign officers came about.

No army in tropical Africa, apart from the Sudanese, had a significant proportion of African officers when independence was granted.1 In Ghana, which enjoyed perhaps the highest educational level of the sub-Saharan countries, only Io per cent of the officers were African in 1957.2 The Force publique in the former Belgian Congo had no Africans in command positions: there were but three African sergeant-majors in an army of 24,000 soldiers and non-commissioned officers, 542 officers, and 566 junior officers.3 Not until I959 did an African from British East Africa complete training at Sandhurst.4 The late start in Africanisa- tion can be attributed to the pace of political change in Africa since World War II, which far exceeded the pace envisaged by the colonial administrations.5 Until independence was clearly recognised as the objective-in the short run, not in the distant future-of constitutional evolution, there was little value in replacing Europeans with possibly

'James S. Coleman and Belmont Brice, Jr., 'The Role of the Military in Sub-Saharan Africa', in John J. Johnson (ed.), The Role of the Military in Underdeveloped Countries (Princeton, I962), pp. 366-7.

2 Ibid. p. 370. Gutteridge notes that in January I961, three months after Nigeria gained independence, 8I of about 300 officers in the Nigerian Army were Africans, the most senior being a Lieutenant-Colonel. William F. Gutteridge, 'Military Elites in Ghana and Nigeria', in African Forum (New York), II, I, Summer I966, p. 37.

3 Coleman and Brice, op. cit. p. 379; Catherine Hoskyns, The Congo Since Independence, January I96o-December I96I (London, 1965), p. 59.

4 Coleman and Brice, op. cit. p. 373. 5 On the eve of World War II, the British Secretary of State for the Colonies commented,

'It may take generations or even centuries for the peoples in some parts of the Colonial Empire to achieve self-government'. Clearly he had Africa in mind. Quoted in Kenneth Robinson, 'World Opinion and Colonial Status', in International Organization (Boston), vmii, 3, Summer I954, P. 468.

30o7

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

CLAUDE E. WELCH

less-experienced Africans. The supply of Africans with the requisite educational background was limited. And, at a more subtle level, the loyalty of European officers to their home country could be taken for granted. Control over the armed services was never turned over to African supervision before self-government was granted. The armies of contemporary Africa are the direct descendants of

forces created by the colonial administration. They supplemented police forces in restoring calm-often by shows of force, a reputation for rapacity, and occasionally outright brutality. Soldiers were often recruited from 'martial' tribes, then stationed far from their place of origin, often among traditional enemies. Ill-trained and ill-disciplined in many instances, colonial armies 'were often regarded by the local population with deep distrust, as hated tools of the white conqueror'.1 Few African leaders embarked upon drastic military reorganisation

immediately after attaining self-government. The Africanisation of command positions was accelerated, but it was nearly impossible-and perhaps politically inexpedient-to remove expatriates on the spot. Although politicians supported Africanisation in principle, the uncer- tainties of the post-independence period gave them second thoughts about dropping experienced Europeans. President Nkrumah of Ghana, for example, retained a British chief of staff for more than four years after independence, despite pressures to name an African to the position. Even so mercurial a leader as Patrice Lumumba of the former Belgian Congo felt the need for continuity in the army-that is to say, for the retention of European officers. 'We are not, just because the Congo is independent, going to turn a second-class soldier into a general'.2 Even more forceful was his demand, made to the Congo Executive College two months before independence:

it is essential that the Force Publique, the only force available, stays intact. It must pass under the command of the Congolese Government exactly as it is-with its officer class, its junior officers, its traditions, its discipline, its unique hierarchy and above all its morale unshaken. This is not the moment to disorientate the Force by innovations, for an army in the process of organ- isation or reorganisation is not in a fit state to carry out its duties.3

Continuity entailed limitations on immediate promotion-a constraint many Congolese soldiers were unwilling to accept. The contrast in the pace of Africanisation between the civil service

and the officer corps was striking in many African states. As noted

1 Pierre L. van den Berghe, 'The Role of the Military in Contemporary Africa', in Africa Report (Washington), x, 3, March 1965, p. I7. 2 Hoskyns, op. cit. p. 6o. 3Ibid.

308

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

SOLDIER AND STATE IN AFRICA 309

earlier, a mere I o per cent of officers in the Ghanaian army were Africans

in 1957. Three years earlier, however, nearly 40 per cent of senior officers in the Ghanaian civil service were Africans.' Analogous disproportions in the rate of advancement existed elsewhere, and helped to account for the subsequent outbreak of mutinies. Budgets for the armies of Africa, small in the period of colonial

administration, remained limited after independence. The limited size of pre-independence forces and the retention of European officers may have inhibited lobbying or pressure for increased appropriations. The conception of the army as the former tool of colonial interests made it easy for politicians to skimp military spending. Limited resources were better spent on such popular objectives as education, health, and ex- pansion of other public services. Even after independence, the former colonial power bore many military costs, particularly through providing equipment and training. In fact, until the officer corps was largely Africanised, until the former colonial power had disavowed its willing- ness to intervene in the event of insurrection, and until the aura of unity surrounding the dominant party had been broken, military pressure upon politicians-and consequently the funds devoted to military expenses-remained low. These brief observations point inescapably to the conclusion that

African armies, in the immediate post-independence period, were not 'national' armies. Their organisation and equipment depended heavily on external sources. Most of the trappings of the colonial period were simply carried over, save in states where self-government brought rupture from the former administering country.2 Defence pacts, the dominance of Europeans in the officer corps, and a relatively slow Africanisation of command positions, in contrast to the rapid Africanisa- tion of the civil service, helped to perpetuate the image of the army as expatriate-controlled. Armies were not viewed, then, as institutions in which Africans took pride.3 They were remnants of a time of' national' shame, in the eyes of prominent politicians. This makes an interesting contrast with the states of Latin America, where independence came

1 Austin, op. cit. p. 158. 2 Guinea is an obvious instance. After the Non vote in the referendum of 28 September

I958, which resulted in the country's independence, the French Government 'almost at once began to repatriate Guineans then serving in its army'. All French troops were with- drawn by 30 November 958, and the Guinea Government eschewed any French assistance in training. For a dithyrambic view of the reorientation of the Guinean army, see Victor D. Du Bois, 'The Role of the Army in Guinea', in Africa Report, vmII, I, January 963, pp. 3-5, from which the quotations were taken. 3 Roger Murray, 'Militarism in Africa,' in New Left Review (London), July-August 1966,

p. 36.

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

CLAUDE E. WELCH

Armed Strength and Defence Expenditures of African States1

Regular Defence As % of As % of Country armed budgets total govt. estimated

forces2 ($ooo) expenditure G.N.P.

WEST AFRICA

Chad 900 5,835 I3.5 I 8 Dahomey 1,8oo 4,070 12.0 2 0 Gambia - - -

Ghana 17,000 42,000 7'4 2.5 Guinea 5,000 5,870 8*I 3 I Ivory Coast 4,000 8,825 6-9 2'4 Liberia 3,200 3 I00 6.7 I1 8 Mali 3,500 8,825 2 I2 3.2 Mauritania I,000 4,060 I 79 5'I Niger 1,200 3,650 I o8 I. 2 Nigeria 11,500 54,000 9'9 0'9 Senegal 5,500 2 I,050 i 6 7.6 Sierra Leone 1,360 2,585 4'9 I'3 Togo 1,450 2,757 I3'5 4 I Upper Volta I,500 2,8I9 I4 I 6'I

Total 58,9Io 169,446

NORTH AFRICA AND THE HORN

Algeria 48,000 101,000 I II 4' 2 Ethiopia 35,0oo 3I,I75 I 7o 2'3 Libya 7,000 I4,000 5'8 3'5 Morocco 44,800 62,000 IO05 4'7 Somali 9,500 6,670 I8 I 4.8 Sudan I8,500 40,000 I7'7 4'4 Tunisia 17,000 8, i8o 4' I I'5 U.A.R. 180,000 450,000 I7'4 8-6

Total 359,800 713,025

through armed struggle, and 'the leaders of the revolutionary armies moved easily and naturally into the political vacuum created by the disappearance of royal authority. Thus at the very beginning ofnation-

1 Source: David Wood, The Armed Forces of African States (London, Institute for Strategic Studies, I966), Adelphi Papers No. 27, pp. 28-9.

2 These estimates exclude the police and gendarmerie.

3IO

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

SOLDIER AND STATE IN AFRICA

Regular Defence As % of As % of Country armed budgets total govt. estimated

forces ($ooo) expenditure G.N.P.

CENTRAL AFRICA

Burundi 950 970 6-9 0-7 Cameroun 3,500 5,800 I9'5 4'2 Central Afr. Rep. 600 2,325 7'9 0-6 Congo-Brazza. I,800 3,785 8'9 I0 9 Congo-Leo. 32,000 22,500 I4'5 I.7 Gabon 750 2,540 7'6 5'I Malawi 850 I,500 3'3 II Rwanda 1,500 1,300 9.7 0-7 Zambia 3,000 13,525 5.7 2-5

Total 54,950 64,245

EAST AFRICA

Kenya 4,775 10,200 6-9 9-8 Malagasy 4,000 9,130 8-8 I o Tanzania I,800 7,225 3. 8 0-3 Uganda 5,960 17,025 I0o2 I'5

Total I6,535 43,580

SOUTHERN AFRICA

Rhodesia I6,9o0 6. 6 I9 South Africa 322,000 19'9 3'5

Total 338,900

hood the armed forces [of Latin America] assumed extramilitary [that is, political] functions'9. Only in Algeria, and to a much more limited extent in Morocco, Tunisia, Cameroun, and Kenya did bands of guerrilla fighters carry forward the quest for independence.

MUTINIES AND STEPS TOWARD INVOLVEMENT

Direct military involvement in the political life of African states came initially through mutinies. Their proximate cause appears to have been

1 Edwin Lieuwin, Arms and Politics in Latin America (New York, I961 rev. edn.), p. 19.

3I"

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

CLAUDE E. WELCH

dissatisfaction within the ranks over such internal issues as promotions and pay scales. Though they had a direct impact upon politics, the mutinies were not directly political.' In other words, these revolts primarily represented insubordination to officers, not direct rebellion seeking to displace elected officials. The causes were intrinsic to the army, not extrinsic to the political system. Africanisation of the officer corps had barely started in most sub-

Saharan states when the colonial power withdrew. The European commanders viewed their duty as being to ensure continuity in a period of profound stress. The most candid expression of this view came from General Emile Janssens, commander of the Force publique in the former Belgian Congo. A few days after independence, he addressed a sullen group of officers and soldiers at the Leopoldville military camp. The Congolese troops 'had the impression that independence had passed them by'-and this impression was seemingly confirmed when General Janssens wrote on a blackboard, 'Before independence= after inde- pendence'.2 This remark infuriated the soldiers, who rebelled and dismissed all European officers. The first significant mutiny against an independent African state had started. The self-confidence of General Janssens was more than matched by

the confidence ofJulius Nyerere (then Prime Minister) of Tanganyika. Shortly after the Congo mutiny, Nyerere declared, 'These things cannot happen here. First, we have a strong organization, T.A.N.U. [Tan- ganyika African National Union]. The Congo did not have that kind of organization ... there is not the slightest chance that the forces of law and order in Tanganyika will mutiny'.3 Three and a half years later, after Nyerere had devoted considerable attention to reorganising T.A.N.U. and assuring military loyalty, the First Battalion of the Tanganyika Rifles rose against their British officers and marched into Dar es Salaam, demanding higher pay and total Africanisation of the officer corps. The vaunted strength of T.A.N.U. and its respected leader seemed to crumble when confronted by a handful of mutinous troops.

MILITARY INVOLVEMENT IN POLITICS:

THE SEIZURES OF CONTROL

The military seizures of control that rocked the sub-Saharan area from mid-I965 on cannot be attributed to a single factor. The com-

1 A similar distinction is drawn by A. Mazrui and D. S. Rothchild in their article 'The Soldier and the State in East Africa: some theoretical conclusions on the army mutinies of I964', in Western Political Quarterly (Salt Lake City), xx, I, March 1967, p. 94.

2 Hoskyns, op. cit. pp. 87-8. 3 Quoted in Mazrui and Rothchild, op. cit. p. 82.

312

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

SOLDIER AND STATE IN AFRICA 313

plexity of events belies simple, uni-causal analysis. Many political systems were involved, each with distinct heritages and problems. To assume that 'popular discontent' or 'economic stagnation' or 'neo- colonialist interference' brought about the coups d'etat does not do justice to each unique combination of circumstances. Rather than search for a sole cause, we must examine a series of factors, the salience of whose components differ from one African state to another. Significant factors that helped promote military intervention may

be summarised in tabular form:

I. The declining prestige of the major political party, as exemplified by (a) an increased reliance upon force to achieve compliance, (b) a stress upon unanimity in the face of centrifugal forces, and (c) a consequent denial of effective political choice.

2. Schism among prominent politicians, weakening the broadly- based nationalist movement that had hastened the departure of the former colonial power.

3. The lessened likelihood of external intervention in the event of military uprising.

4. 'Contagion' from seizures of control by the military in other African countries.

5. Domestic social antagonisms, most obviously manifested in coun- tries where a minority group exercised control (e.g. the Arabs in Zanzibar, the Watutsi in Burundi).

6. Economic malaise, leading to 'austerity' policies, which most affected articulate, urbanised sectors of the population (members of trade unions, civil servants).

7. Corruption and inefficiency among government and party officials, a corruption especially noticeable under conditions of economic decline.

8. A heightened awareness within the army of its power to influence or displace political leaders.

Let us consider these factors in greater detail.

I. Declining Party Prestige

Enthusiasm for the dominant party flagged rapidly in many African states, following the initial spurt of organisational enthusiasm, and the jubilation of self-government.1 The waning was in part psychological.

1 Aristide R. Zolberg, Creating Political Order: the party-states of West Africa (Chicago, 1966), PP. 37 ff.

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

CLAUDE E. WELCH

Independence may have aroused unwarrantedly high expectations. The 'golden age' would be ushered in; in some remote regions, self- government was interpreted as an end to taxes, communal labour, and other unpleasant aspects of administrative control. Such hopes obvi- ously could not be fulfilled. Economic stringency also contributed to the loss of party fervour. But the major agent of the declining prestige of political parties was internal. The tactics and organisation of parties appropriate for anti-colonial activities were not necessarily appropriate for the tasks of government after independence. Failure in the structural adaptation of political parties made them considerably less effective instruments in a self-governing African state. Difficulties arose from the limits of the changes that political groups

could effect. Loyalty to a dominant party, to its leader, and to the 'nation' it claimed to embody could not be readily accepted in the face of conflicting, ethnically-based loyalties. For most newly enfranchised Africans, an understanding of politics could not be disentangled from traditional outlooks and antagonisms. The broad, anti-colonial nationalist movements had succeeded, in varying degrees, injuxtaposing what one may call the 'sub-national nationalisms' of ethnic groups. With independence achieved and its novelty gone, the united front showed clear signs of disintegration, of a loss of vigour. Primordial attachments remained strong, despite the success of the nationalist movements in winning independence. Indeed, the extension of political awareness through party activities may have exacerbated group tensions. The greater political awareness created by participation, C. Geertz has noted, 'does indeed tend to lead to the stimulation and maintenance of a very intense popular interest in the affairs of govern- ment. But... much of this interest takes the form of an obsessive concern

with the relation of one's tribe, region, sect, or whatever to a center of power'.l To regain the apparent unity of the anti-colonial struggle was a problem with which most African leaders struggled after inde- pendence. Agreement had to be created once again-and it was to be revived through the achievement of unanimity.

Two steps are involved in this task, according to A. R. Zolberg: erasing all traces of political opposition, and modifying both dominant party and government through the creation of a new institutional order.2 The methods by which opposition parties have been harassed and eventually removed from public attention have been analysed

1 Clifford Geertz, 'The Integrative Revolution: primordial sentiments and civil politics in the new states', in Clifford Geertz (ed.), Old Societies and New States (New York, I963), pp. I I9-20.

2 Zolberg, op. cit. p. 66.

3I4

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

SOLDIER AND STATE IN AFRICA 315

elsewhere in detail.1 Essentially what was involved was civil action against politicians-legal restrictions on their activities, the manipula- tion of electoral machinery and regulations, rewards for 'crossing the carpet'-rather than widespread campaigns of intimidation undertaken against supporters of opposition groups. By removing the spokesmen for 'sub-national nationalism', African political leaders hoped to eliminate detrimental parochial tendencies. However, given the deeply-rooted nature of ethnic identity and its stimulation through political participation, such measures may only have worsened the situation. It is in this context that politicians turned increasingly to the use of force, to attain goals of integration unreachable through persuasion. The attempted elimination of political opposition by the use of force

in turn affected political life in African states. The following points, drawn from Zolberg, illustrate the complexities of the shift from 'power' to 'force'.2

I. Governments may fall into the snare of believing that harsh treat- ment will eliminate disruptive demands. However, the governments may become more vulnerable to other threats, since the limited capital of force is used up-and not readily replenished. In other words, the possibility of a 'run on the bank' within an African political system is heightened.3

2. Increasing use of force enhances the significance of police, military, and other groups capable of its exercise. By contrast, the significance of political parties, civil services, or similar institutions declines.

3. If groups, or individuals, lack the opportunity for legitimate political activity-that is, exercising power to express their demands- they are tempted to use force to press these demands.

2. Political Schisms

The quest for unanimity, sought by means of the dominant party, must be examined in a different context. Maintaining the cohesion of

1 See, inter alia, the country studies and concluding chapter of James S. Coleman and Carl L. Rosberg, Political Parties and National Integration in Tropical Africa (Berkeley and Los Angeles, 1964); Martin L. Kilson, 'Authoritarian and Single-Party Tendencies in African Politics', in World Politics (Princeton), xv, 2, January 1963, pp. 262-94: and Zolberg, op. cit. p. 66.

2 Aristide R. Zolberg, 'The Structure of Political Conflict in the New States of Tropical Africa', unpublished paper delivered at the I966 annual meeting of the American Political Science Association, pp. I3-I4. I have omitted Zolberg's point about the problem of legitimacy: he notes that the shift to force 'enhances the problem of the legitimacy of the rulers in the eyes of those to whom the implementation of force must necessarily be entrusted'.

3Talcott Parsons, 'Some Reflections on the Place of Force in Social Process', in Harry Eckstein (ed.), Internal War: problems and approaches (New York, I964), pp. 58-61.

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

the party entailed a reformulation of party-state relations. In this reformulation, an increasing emphasis upon individual leadership emerged. In few African political parties has collective decision-making prevailed-if it ever existed. The reins of control have been held by a single leader. However, in states where political rivalry could not be contained within the framework of a single party, the likelihood of coups d'etat increased. Splits between top political leaders furnished the pretext for several military incursions into African political life.

Two examples illustrate this point. General Joseph Mobutu's first intervention into Congolese politics came in September I960. Prime Minister Patrice Lumumba and President Joseph Kasavubu had found themselves in a constitutional impasse: Kasavubu had relieved Lumumba of his duties, an action whose legality Lumumba denied. As government activity shuddered to a halt, Mobutu stepped in to neutralise all politicians and assume power, for a period of slightly over three months. He insisted that his action constituted a 'peaceful revolu- tion', not a coup d'itat.1 The government would be run by techniciens, by which Mobutu meant Congolese university graduates and students.

Colonel Christophe Soglo attempted to heal the long-standing split in Dahomeyean politics in October 1963 by temporarily replacing President Hubert Maga.2 However, the long-standing tensions among regionally-based politicians in Dahomey could not be effaced, and the formation of a single party (the Parti dimocratique dahomeen) failed to resolve the unstable situation. To set the government right by elimin- ating the schism and regionalism encouraged by conflicting leaders was a major encouragement for military intervention. In other words, soldiers like Soglo and Mobutu were prompted to seize control to restore a political equilibrium which had been disrupted by personality conflicts. Their goal was 'caretaker' administration-not sweeping change, but minor adaptation. Minor shifts in personnel and policy would be undertaken; the status quo would be restored.

3. External Intervention: flow Real a Threat?

The plotters of coups d'itat must always calculate the possibility of international reaction against their planned takeover.3 In Africa,

1 Hoskyns, op. cit. p. 214. 2 The regional tensions are extensively discussed by Virginia Thompson, 'Dahomey', in

Gwendolen M. Carter (ed.), Five African States: responses to diversity (Ithaca, I963), pp. 207-35. 3 A sound strategy for coups, Major Goodspeed has commented, involves assessment of

the sympathies of the country's armed forces, the state of public opinion, and the international situation. D. J. Goodspeed, The Conspirators: a study of the coup d' tat (New York, I962), p. 2 0. The six seizures of control analysed in this book are European.

316 CLAUDE E. WELCH

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

SOLDIER AND STATE IN AFRICA 317

fear of external intervention to prevent military seizure of control- particularly the fear of intervention by the former colonial power-has been a significant variable in the cases under consideration. The role of European officers has already been examined. It is

reasonable to assume that the presence of expatriate officers moderates the willingness of African officers and soldiers to embark upon coups d'etat, where the expatriates and the former colonial power are obviously committed to upholding the African government concerned. More important, there is the risk of large-scale intervention from the ex- administering country. British forces rapidly quelled the mutinies in Kenya and Tanganyika, upon request; French troops crushed the uprising in Gabon, once again upon request. From the limited evidence available, it appears that the British and

French Governments have been relatively circumspect. They have intervened only on the request of the duly constituted government (though how well-established Mba's Government was is an open question), where a show of force (as by disarming mutineers) would restore calm. The former colonial powers have not intervened in cases of popularly supported revolutionary upheaval (e.g. Zanzibar, the overthrow of President Fulbert Youlou in Congo-Brazzaville), in relatively popular coups in which politicians are removed from office but not executed (e.g. Central African Republic, Dahomey, Upper Volta), or in violent coups carried out with great rapidity (e.g. Togo, Nigeria).

How likely are the Belgian, British, and French Governments to become embroiled in protecting future African politicians from the ire of discontented armed forces? Not very likely, it would seem, save in cases in which very substantial investments and numerous citizens were involved, and in which international political considerations-that is to say, east-west alignments-obtrude. The use of paratroops against Stanleyville in I964 fulfilled these criteria, but a repetition of the sur- rounding circumstances appears unlikely. European troop deployments in Africa have been cut dramatically. 'As far as the major powers are concerned', David Wood has commented, 'on the whole their military involvement in Africa is limited by the fact that the continent is not of prime concern for their own security. Only for France do bases in Africa constitute a major part of her strategic scheme, and she is progressively withdrawing from the area'.l The I6,ooo French troops stationed in Africa at the start of 1966 have been considerably reduced,

1 Wood, op. cit. p. 3.

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

CLAUDE E. WELCH

and bases closed down; should any of the i countries with which France has defence agreements call for assistance, units for intervention would come from the i Ith Infantry Division, generally based in south- west France and Brittany.1 The British Government after March 1966 maintained only 2,000 soldiers in Africa; no Belgian troops are stationed south of the Sahara, although advisers have been provided for the armies of Burundi, Congo-Kinshasa, and Rwanda; almost all American troops in Africa (5,500 out of 6,000) are stationed in Ethiopia and Libya.2 Despite the rapid mobility offered by modern aircraft, and the retention of some French troops, the capacity of the former colonial powers to intervene has declined considerably-as well, perhaps, as their desire to become involved in what they view as domestic political crises. And, further, if a coup is rapid and the removal of politicians complete, intervention by the former colonial power would be nugatory. Leaders of coups learn quickly the value of 'Twere well it were done quickly.'

4. 'Contagion' Three weeks after the assassination of President Sylvanus Olympio

of Togo, Colonel David Thompson, commanding officer of Liberia's National Guard, was arrested on suspicion of plotting a coup d'etat. 'If only 250 Togolese soldiers could overthrow their government, a Liberian Army of 5,000 could seize power easily', Colonel Thompson is alleged to have argued.3 Successful seizure of control in one state may touch off a series of coups. The Zanzibar uprising helped to trigger off the East African mutinies; similarly, the intervention of Soglo in December I965 helped to touch off coups in the Central African Republic, Upper Volta, and Nigeria.

Contagion must be considered on two levels: the personal links among African officers in different countries, and the increasing extent of inter-state ties. Shared experiences in the French army provided the leaders of intervention in the Central African Republic, Dahomey, Togo, and Upper Volta (respectively Bokassa, Soglo, Eyadema, and Lamizana) with potentially significant individual ties. Quite likely the success of one in winning political control prompted the others to con- sider intervention-though no conclusive evidence can be adduced. The international connexions result from increasing interaction among African states. Since independence-especially through the periodic

1 Ibid. p. 26. 2 Ibid. pp. 19, 22, and 27. 3 Helen Kitchen, 'Filling the Togo Vacuum', in Africa Report, vmII, 2, January 1963, p. 9.

3i8

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

SOLDIER AND STATE IN AFRICA 319

conferences organised by the Organisation of African Unity and its subordinate units-contacts among African leaders have multiplied. As Africa has increasingly become a sub-system in world politics, reciprocal influences among African states have multiplied. These reciprocal influences include the germ of military intervention.

5. Unstable Social Situation

Reference has already been made to conditions of social disequilbrium affecting attempts at army seizure of control. When members of the armed forces perceive the government as dominated by members of an ethnic group hostile to the interests of another ethnic group heavily represented within the military, grievances based on tribe or region may quickly develop. Often army intervention may be given impetus by the clash of' primordial sentiments'.

6. Corruption

The puritanical tendencies of military leaders have often been com- mented upon. Their distaste for ostentation has served as a rationalisa- tion-and I suspect as a reason-for intervention. To clear up the affairs of state and restore fiscal integrity and responsibility are themes frequently expounded by the leaders of coups d'etat. General Ankrah, for example, accused President Nkrumah of bringing Ghana 'to the brink of economic disaster by mismanagement, waste and unwise spending'.' Attempts to reduce government spending, undertaken by all the new military leaders under review in this article, may be viewed as a response to a belief in widespread corruption. Similarly, the banning of political parties (another typical step following intervention) was partially intended to cut down opportunities for corruption.

7. Stagnating Economic Situation

Rising government budgets do not indicate increasing prosperity. The dramatic increase in the expenditures of African states in the late colonial and immediate post-colonial period reflected the strong desire of political leaders to spread social benefits widely and rapidly. Univer- sal primary education, for example, was a goal with obvious attractions -but one involving tremendous costs in relationship to total resources. An economic squeeze resulted, which was compounded by falls in the world prices of primary commodities such as cocoa, coffee, and cotton, on which many African countries depended for export earnings. The termination of preferential marketing agreements (especially between 1 Quoted in Africa Research Bulletin (London), III, 2, February I966, p. 4670.

21

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

France and her former dependencies), as well as absolute declines in the quantity of some of the goods exported, further weakened the economic position of African states. A seemingly insatiable demand for consumer goods, manufactured items, and machinery essential for even the rudiments of industrialisation increased the need for foreign exchange. Lacking ready reserves, or a sufficiently growing revenue from exports to cover these needs, African states found themselves in an economic quandary. Few avoided inflation. Deficits were the rule, not the exception.l Governments became involved with 'suppliers' credits', loans of dubious economic validity, and stop-gap schemes for economic betterment.

Yet the demands could not be stilled. The economically most privi- leged parts of contemporary African society-the civil servants, the trade unionists, and above all the politicians-benefited to the greatest extent from government pay rises. Even so, their inflated aspirations could not always be satisfied. Discontent with salaries in Dahomey and Upper Volta helped to ignite the civil disturbances that preceded the coups d'etat; the precarious economic position of Ghana before the anti- Nkrumah coup has been well documented.2

Revolutions are made by the relatively privileged, not the down- trodden.3 The privileged of modern Africa are the educated. They gained most from independence, especially through the expansion of benefits brought by Africanisation and salary increases. When the flow of these perquisites was interrupted through financial necessity, however, the discontent of the educated could not be easily contained.

8. Military Self-Identification

The seizure of control by militaryjuntas normally does not occur when officers recognise civilian control of the armed forces as legitimate. For intervention to take place, members of the military must distinguish between the 'national interest' of the country and the policies espoused

1 In the Congo, to take an admittedly extreme example, the central government in 1962 estimated it would receive approximately 4,000,000,000 francs, while expenditures totalled more than 19,000,000,000 francs. Edouard Bustin, 'The Congo', in Carter (ed.), Five African States, p. I38n.

2 See reports of hearings of 15 commissions of enquiry established by the Ankrah Govern- ment, and the rescheduling of Ghana's external debt of 280 million, in West Africa (London).

3 Brinton's observations on revolutions seem apposite here. In his comparative analysis, he suggests that revolutions occurred in societies 'on the whole on the upgrade economically before the revolution came, and the revolutionary movements seem to originate in the discontents of not unprosperous people who feel restraint, cramp, annoyance, rather than downright crushing oppression. Certainly these revolutions are not started by down-and- outers, by starving, miserable people'. Crane Brinton, The Anatomy of Revolution (New York, 1965), p. 250,

320 CLAUDE E. WELCH

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

SOLDIER AND STATE IN AFRICA 321

by the government in power. The denial of the legitimacy of the duly constituted civilian government is the first---and probably the most important-step in the military seizure of power. The 'disposition to intervene', as S. E. Finer has suggested, is based on a distinction between 'national interest' and the practices followed by the government.' If the armed forces are to remain in their barracks and not attempt to displace the political leadership, they must accept civilian supremacy. The 'disposition to intervene' arises in part from the unique tech-

nological and organisational position of the armed forces. The military pride themselves on their efficiency and esprit de corps. Their centralisa- tion, hierarchy, discipline, and intercommunication-quite apart from their control of modern weapons-set them apart from most institutions within a given society. Recruits into the armed forces generally receive intensive training, instilling in them a sense of the unique heritage and duty of the army. As many analysts of the role of the military in new states have observed, this indoctrination may prove extremely beneficial in implanting 'a sense of citizenship and an appreciation of political action'.2 Yet this indoctrination may serve just as well to implant loyalty to the army as an institution set apart from its political system. There may be the emergence of a strong sense of the honour and duty of the military, including the duty to intervene to correct the misdeeds of politicians. When the military seizure of control is rationalised by coup leaders, in the words of a main plotter against Nkrumah, as 'necessary to save our country and our people', the 'disposition to intervene' is clearly present.3

The preceding analysis has been aimed at presenting the complexities of the 'causes' of military intervention. To pick out a single factor as 'The Cause' of a coup d'etat is nearly impossible. Military intervention resulted from the combination of many factors. In the coups surveyed in this article, economic, cultural, and political reasons were inter- twined; the declining prestige of political parties and the growing con- sciousness among the military of their power obviously played significant roles. The diminished likelihood of external intervention, and the effects of contagion, must be added to a sense of grievance within the army as

1 S. E. Finer, The Man on Horseback: the role of the military in politics (London, I962), pp. 23-60.

2 Lucian W. Pye, 'Armies in the Process of Political Modernization', in Johnson (ed.), The Role of the Military, p. 83.

3 Colonel A. A. Afrifa, The Ghana Coup, 24th February 1966 (New York, I966), p. 37.

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms

322 CLAUDE E. WELCH

part of the background. When and how the coups occurred differed greatly. There seem to be no uniform circumstances under which the military seized control-save that the popularity of the government had noticeably declined among the politically relevant strata of the popula- tion, and that army officers seemed to become aware of a unique duty to protect the 'national interest'. To change personnel and policies was the major objective of coups d'etat; accordingly, they grew out of a back- ground analytically distinguishable from the conditions that promoted the mutinies, discussed previously. By early 1966, it was clear that the armies of Africa had emerged as perhaps the most significant shapers of policy. No longer can our understanding of post-colonial Africa leave aside the political role of the military.

This content downloaded from 128.197.201.233 on Mon, 02 Sep 2019 23:54:48 UTC All use subject to https://about.jstor.org/terms