In ancient Rome, it was common knowledge that an external actor could intervene to change the course of a battle between two combatants. A favourite pairing of gladiators was between the lightly armed retiarius (the ‘fisherman’) and the much more heavily armed secutor (‘the pursuer’). The retiarius, on the one hand, was armed only with a trident, net and long knife. As he was expected to lose, the retiarius was not given armour or a helmet so the crowd could see his wounds and watch his face as he died. In contrast, the secutor had thick armour on his legs, arms and head. He was heavily armed with a curved rectangular shield and the Roman sword (the gladius). Regardless of the competitors' physical stature, political beliefs, culture, skill, or experience, the retiarius would always attempt to exploit his superior speed, agility and reach, while the secutor would, using his shield for protection, charge his opponent, hoping to finish the fight quickly. Although there was an obvious asymmetry in the capabilities between the two contestants, if the secutor could not finish the fight quickly, he risked falling victim to exhaustion. If the game's convener had not enjoyed the previous match, he had the power to intervene and thereby change the following bout by either giving more arms to one, or both belligerents, or by taking weapons and equipment from them. This action would alter the balance between the gladiators and, subsequently, change the nature of the competition. The techniques used in combat between two retiari or, alternately, two secutores, would have appeared very different from the original fight.
This article submits that the fundamental principles of an ancient duel are little different from those applying to the impact of foreign intervention on warfare in civil war. Foreign intervention has the potential of increasing, or decreasing, the capabilities of the belligerents that, in turn, can radically change the nature of the warfare. There has recently been an upsurge in interest in the nature of warfare in civil war. Following Stathis Kalyvas' seminal work, this fresh line of research is currently grappling with the implications of accepting three categories of warfare in civil war: conventional, guerrilla, and irregular (or ‘symmetrical non-conventional’).Footnote 1 In particular, questions have begun to be raised as to whether the different forms of warfare produce civil wars that are of different duration and lethality.Footnote 2 Although these ongoing research programmes are worthy academic endeavours in their own right, they only become relevant to policymakers if third parties can intervene to change the form of warfare – presumably with the intention of changing the warfare to a shorter and less lethal type. Here, I theoretically explain and empirically demonstrate how this can be achieved.
Dylan Balch-Lindsay and Andrew Enterline observed: ‘[t]o date, the literature on intrastate conflicts underscores the idea that the influence of third parties is often instrumental in shaping the dynamics of these conflicts.’Footnote 3 This article builds upon this observation by asking how foreign intervention affects the warfare in civil wars. While doing so, it makes several important contributions. First, whereas previous studies have sought to identify correlations, such as increased duration, lethality or reoccurrence, this article is mechanism-oriented. In other words, it is interested in uncovering how foreign intervention causes changes in the nature of warfare in civil wars, not simply that it does. Second, it develops a revised approach to the balance of military capabilities. Whereas strategic balances have previously been understood and applied as simple relative measurements, this article develops a model of the balance of capabilities that includes both absolute and relative measurements. This innovation accounts for a belligerent's selection of a strategy being a function of its own physical ability and its strength vis-à-vis its opponent(s). This contribution allows for increased dynamism, a stronger predictive element and a more sophisticated method of analysis overall.
The following is divided into four distinct sections. Section one explains the relationship between foreign intervention, the balance of military capabilities and warfare. Section two develops a revised version of the balance of military capabilities and further explains how it influences the form of warfare that emerges at a particular place and time in a civil war. Section three submits novel methods for observing and measuring the volume of foreign intervention, the state of the balance of military capabilities and the form of warfare that dominates a particular place and time. Although seemingly straightforward objectives, this section highlights that measuring these variables is rife with potential pitfalls. The final section advances these theoretical arguments by testing them against case material from the first period of the Angolan Civil War (1974–1976). This period offers a valuable test of the theory as the pattern of foreign intervention changed three times in only 18 months, thus allowing a sound analysis of the relationship between the specific variables under examination.
Foreign intervention, the balance of military capabilities and warfare in civil war
The dramatic influence of foreign intervention on the course of civil wars has been well documented. In part, this can be explained by the majority of interventions having taken the form of providing assistance to one side in the conflict. Between 1945 and 1994, over 95 per cent of foreign interventions in civil wars consisted of the transfer of money, arms, or foreign troops to a belligerent in the civil war.Footnote 4 Neutral interventions aimed at providing humanitarian assistance or fostering the conditions for a peace settlement have been rare. This article acknowledges this historical pattern by limiting itself to partisan interventions. Foreign intervention is thus defined as the transfer of resources from an external state to a contesting party in a civil war.Footnote 5 Resources are broadly defined as any funds, weapons, equipment, materiel or personnel that have immediate or potential coercive value. This definition captures almost all across border transfers of resources from external state governments to civil war belligerents, and includes everything from currency and food through to weapons and training and even the foreign power's own military units.
Foreign intervention and the balance of military capabilities
It is now possible to begin to construct an explanation of how foreign intervention influences warfare in civil war. My argument contains three variables: foreign intervention, the balance of military capabilities and warfare. As already mentioned, foreign intervention refers to the transfer of resources from an external power to either the incumbent or an insurgent actor. The balance of military capabilities refers to the distribution of the belligerents' collective military capabilities. Finally, warfare refers to the confluence of the belligerents' respective strategies. That is, warfare is the function of two or more strategies. These definitions will be expanded upon below.
A belligerent's military capabilities are intimately linked to its supply of resources.Footnote 6 Incumbents and insurgents will expend resources contesting the civil war and therefore must possess the ability to replenish their stockpiles. These resources can be endogenous (that is, originate domestically) or exogenous (originating from outside the civil war system). Generally, endogenous resources will be a larger contributor to a belligerent's military capabilities than exogenous resources. Despite this, however, exogenous resources can markedly increase a recipient's military capabilities and improve its survivability. Indeed, a recent study found that belligerents that have two or more foreign sponsors are four times more likely to survive through to the end of the civil war.Footnote 7 Further, although exogenous support may also come from an international diaspora, a neighbouring insurgent actor, or a humanitarian organisation, rarely will these inputs match those supplied by foreign states. As such, foreign states are, by far, the most significant source of exogenous resources. Figure 1 illustrates the process through which exogenous resources affect the warfare in civil wars (assuming there is only a single insurgent actor).
Figure 1. Resources, the balance of military capabilities, and warfare in civil war.
Balance of military capabilities: average military capabilities and distribution
The next element in the process is the balance of military capabilities. A belligerent's military capabilities is a conceptual representation of its overall military strength; that is, the quantitative and qualitative sum total of the belligerent's manpower, training, weapons, equipment, intelligence, and logistics. Military capabilities are distinct from power.Footnote 8 A belligerent's military capabilities are its total means of inflicting military losses on an opponent; whereas power is generally understood to be an actor's ability to influence other political actors to behave in a way they would not otherwise. The balance of military capabilities is an aggregate representation of the belligerents' absolute and relative military capabilities. It contains two variables: average military capabilities and distribution.
The first determinant is the average military capabilities. This variable is an absolute measure and is equal to the sum of the incumbent's and the insurgent's (or insurgents') military capabilities, divided by the total number of belligerents. This variable responds to the question of how the number of troops and the amount of weaponry and wealth within the civil war state affects the balance of military capabilities. To reach an ‘average’ measurement, the total military resources within the country is then divided by the number of belligerents. As a rule, the greater the number of belligerents in a civil war, the weaker each side is likely to be. The number of belligerents in a civil war will have a major influence on the military capabilities it is possible for each side to possess, as the number of belligerents roughly corresponds with the number of divisions of territory, fighters, and resources. When applied to the balance of military capabilities, the average military capabilities will range between a low and a high measure.
The second variable that affects the balance of military capabilities is the distribution of military capabilities in the civil war. It is nonsensical to assume that there will ever be a perfectly equal distribution of military capabilities between the belligerents, as the average military capabilities variable would suggest if presented alone. Hence, the division of capabilities must also be considered. In fact, unlike the concept of the ‘balance of power’ in International Relations theory, which is assumed to tend towards equilibrium, the logic of the balance of military capabilities suggests that the ‘natural state’ of the distribution of military capabilities will be in the incumbent's favour.Footnote 9 There are three reasons for this proposition. First, the incumbent is, by definition, partially in control of the coercive instruments of the state, and is therefore in a better position to extract resources. Second, even during peacetime, most governments continue to prepare for the possibility of war, while most sections of society do not. As such, the incumbent will usually be better organised for war than the insurgent, at least at the outset of hostilities. Finally, if the balance of military capabilities shifts in favour of an insurgent, one of three outcomes will usually occur. An insurgent with a greater distribution of military capabilities may: (1) win the war; (2) capture the capital (driving the incumbent into rural areas) or, (3) fractionalise and dampen its strategic advantage vis-à-vis the incumbent. In the last situation, the division of the insurgent will result in a corresponding division of its military capabilities and leave the distribution of military capabilities in favour of the incumbent. When incorporated into the balance of military capabilities, the distribution of military capabilities is a purely relative measure that will range between disparity and parity between belligerents. Figure 2 illustrates how the two variables interact to influence the form of warfare that emerges in a civil war.
Figure 2. The balance of military capabilities and warfare in civil wars.
Warfare in civil wars
The form of strategy a belligerent adopts will be shaped by its own physical limitations and its perceived relative position vis-à-vis its opponent(s). This does not deny the importance of other factors in the formation of a particular strategy. Of course, political objectives, geography, leadership, and military culture also influence strategy.Footnote 10 However, I contend that these factors will not determine its ‘form’ (that is, conventional, guerrilla or irregular); rather they will affect more subtle elements of strategy. For example, during the American Civil War, General Lee's Confederate Army adopted an aggressive offensive strategy of manoeuvre while General McClellan's Union Army pursued a much more conservative attrition strategy. These important differences in the conduct of the war are best explained by variations in the political aims, leadership and endogenous resource bases of the contending belligerents. Nevertheless, the form of strategy that each belligerent adopted was undeniably conventional and arose due to both belligerents possessing substantial military capabilities and there being relative parity in overall military capabilities. In short, the category of warfare that emerges in a civil war will principally be determined by the belligerents' reading of the balance of military capabilities.
At a broad, idealised level, there are three forms of warfare in civil war: conventional, guerrilla and irregular.Footnote 11 This is not yet a common typology in political science, but directly follows Kalyvas' recent and influential studies.Footnote 12 The first form of warfare, conventional, derives from the traditional image of warfare between states. Its guiding principle is to deploy the maximum amount of force at a decisive point and thereby destroy an opponent's ability to resist in a ‘set piece battle’. In contrast, guerrilla warfare is characterised by the incumbent and insurgent actors employing different strategies. The insurgent actor attempts to avoid a direct test of strength, opting to structure its units into small, decentralised bands and striking the incumbent's units and economic installations before fading back into the terrain and civilian population. In turn, the incumbent generally employs a strategy that attempts to isolate the insurgent's forces from the civilian population, while protecting its own infrastructure.
The principles of irregular strategy fundamentally differ from those of guerrilla and conventional strategies and, unlike them, are not well established in the theoretical literature on civil war. Irregular warfare emerges where the belligerents' forces are evenly matched – making guerrilla strategy unnecessary – but also lack the capacity to employ conventional strategies. It has long been assumed by historians that warfare in civil wars cannot always be divided into a simple conventional/guerrilla dichotomy. Historians writing on various civil wars have rejected the dichotomy in favour of triple strategic categories, for instance, by dividing the Confederate Army into conventional, guerrilla and militia units. Irregular warfare has been commonplace since the Middle Ages, when European lords organised civilians into local militias for their personal use and to repel marauding mercenaries and raiders.Footnote 13 The modern-day equivalents of these earlier wars have been seen in Haiti, Lebanon, Liberia, Mozambique, Sierra Leone, Somalia, Sri Lanka, the Sudan, Uganda and the Democratic Republic of the Congo. Thus, historians and other observers have been right to elevate irregular warfare to a prominence equal to that of conventional and guerrilla warfare.
The interaction: foreign intervention and warfare
By considering the average military capabilities and distribution of military capabilities in light of the different characteristics of conventional, guerrilla and irregular warfare, it is possible to formulate some hypotheses about the form of warfare that will emerge at a particular time and place in a civil war. First, conventional warfare will be adopted when the average military capabilities are high and the belligerents mutually calculate that there is relative parity in the distribution of military capabilities. Second, guerrilla warfare will occur with high or low average military capabilities, but when the distribution of military capabilities is disparate. Although the average military capabilities will influence the volume of firepower each belligerent can apply, it will not affect the fundamental nature of the warfare that occurs. Finally, irregular warfare will take place when the average military capabilities are low and there is parity in the distribution of military capabilities. In this case, although there is military symmetry in the civil war, neither side possesses sufficient military capabilities to attain the high force-to-space ratio required for conventional offensives or to secure strong positional defences.
Furthermore, the balance of capabilities will usually be dynamic, being highly susceptible to outside manipulation. By supplying exogenous resources, foreign powers will affect the average military capabilities and the distribution of capabilities between the belligerents. However, the balance of capabilities responsiveness to foreign intervention will vary depending on the size of the intervention and the military capabilities of the belligerents. As a rule, the weaker the belligerents, the more susceptible the balance of capabilities will be to external tampering. A foreign power will need to supply fewer resources to a weak belligerent for exogenous resources to represent a substantial proportion of its overall supplies. Hence, as the military capabilities of the recipient grow, the level of foreign intervention will have to increase proportionately for its influence to remain the same. As such, the responsiveness of the balance of capabilities, and thus warfare, to changes in the pattern of foreign intervention will generally become more sluggish as the military capabilities of the belligerents increase. Empirically, this point will manifest itself by civil wars that are characterised by irregular warfare being more disposed to change in character as a result of foreign intervention than, for example, a civil war being fought through conventional warfare.
Measurement and method
Change can be observed across time or space. In this article's case, longitudinal observations are taken of foreign intervention, the balance of military capabilities and warfare in civil war.Footnote 14 I identify significant chronological changes in the type or size of the foreign assistance to the belligerents; then measure the impact that these changes have on the balance of military capabilities; and, finally, observe any variation in the warfare in the civil war. To achieve this, reliable measures of the three variables are required.
To describe changes in the pattern of foreign intervention, all significant foreign assistance from major foreign sponsors is recorded. To the extent that the available data allows, the size, type, value and timing of foreign assistance to the belligerents is detailed.
The balance of military capabilities is deduced though an assessment of the belligerents' military capabilities. That is, each belligerent's military capabilities must first be ascertained before they can be combined and compared in a judgment of the balance of military capabilities. As a concept, military capabilities encapsulate a belligerent's size, technology and fortitude in a single variable. Hence, in measuring a belligerent's military capabilities for this article, I have employed a composite qualitative approach that is primarily comprised of three factors: troop strength, technology and contemporary reports. I argue that, when combined, these measures provide a reasonably reliable judgment of each belligerent's military capabilities and, equally important, the state of each belligerent's military capabilities relative to its opponent(s).
Some may argue that other factors are central to calculating a belligerent's military capabilities, such as organisational structure and training. However, both organisational structure and training contain an endogeneity problem for the balance of military capabilities. That is, these factors not only make substantial contributions to the balance of military capabilities but are also outcomes of a belligerent's choice of strategy. Hence, organisational structure and training feed into both the independent and dependent variables.
Previous attempts at defining conventional, guerrilla and irregular warfare have been insufficient. Generally, definitions have simply pointed to historical cases (suggesting that ‘we will know it when we see it’),Footnote 15 looked for technological markers (that is, tanks as a sign of conventional warfare), or looked at the relative strength of the contending sides. The first method clearly lacks rigor; the second fails to acknowledge that technologies can be employed in novel ways; and the final method obviously represents an endogeneity problem for this particular study. As such, a fresh approach at empirically observing warfare is required. It is submitted here that the forms of warfare are differentiated from each other by their empirical characteristics. In other words, each form exhibits different operational features. More specifically, this can be observed through a comparison of each type's offensive and defensive operational elements.
Conventional warfare
Operationally, conventional warfare is characterised by a clear distinction between offensive and defensive actions. In offence, conventional strategy generally involves massive coordinated ‘pushes’ or ‘thrusts’ into enemy held territory. The key aim is to decisively engage and destroy the opposition's forces through attrition, blitzkrieg or ‘normal progress’. It requires a belligerent to decrease the force-to-space ratio by concentrating its forces at the most advantageous location before advancing into enemy-held territory. In defence, conventional warfare assumes frontlines as the defining feature. The principal aim of fortified positions is ‘to repel attack, to protect people and property, to hold territory, and to minimise damage by the attacker’.Footnote 16 Although not every inch of territory may be protected, strategically-important locations are prepared to be held against the enemy's assaults for as long as possible. This emphasis on territory, in both defence and offence, is alluded to in the common reference to conventional warfare as ‘positional warfare’.Footnote 17
Guerrilla and counter-guerrilla warfare
In guerrilla warfare, the incumbent and insurgent apply very different strategies. This makes it unique among the types of warfare in civil war. The insurgent wages a protracted war of attrition while avoiding open engagements with the adversary. Offence in guerrilla strategy attempts to hold the initiative by engaging in surprise attacks, raids, sabotage and ambushes, thereby avoiding a pitched battle.Footnote 18 An insurgent applying a guerrilla strategy endeavours to create the impression that its forces may strike anywhere and anytime, which compels the incumbent to spread its forces thinly in order to protect important infrastructure. As a consequence, the incumbent has fewer available forces with which to aggressively pursue the guerrilla force. In defence, guerrilla forces do not attempt to defend territory or population. Instead, and in contrast to conventional strategy, guerrillas attempt to decrease the force-to-space ratio. Defensive guerrilla strategy thus requires the insurgent to withdraw from territory when challenged by the enemy forces, even if this means leaving highly-valued assets like family homes and sites of symbolic value unprotected. For this reason, Clausewitz appropriately likened the strategy to a cloud that parts as solid pressure moves towards it.Footnote 19
Offensive counter-guerrilla strategies adopted by incumbents may attempt to isolate the guerrillas from their social support base and deny them readmission, tempt or force them out of cover and into a direct confrontation, or annihilate the social base from which the insurgency originates.Footnote 20 Although the methods used in counter-guerrilla strategy are unusual within a conventional context, their primary strategic objective is similar: to locate the enemy and destroy it through superior manoeuvre and firepower. In defence, counter-guerrilla strategy requires the protection of important political, economic and military positions. Incumbents are forced to invest heavily in constructing elaborate conventional defences (such as trenches, bunkers and observation posts) to protect important infrastructure.
Irregular warfare
Irregular offensives exhibit different characteristics in urban and rural areas. In urban areas, irregular offensives frequently resemble severe gang violence, with seemingly ‘confused street battles’Footnote 21 and, in more recent times, the increasing use of civilian cars, motorbikes and trucks for swift hit-and-run attacks. In rural areas, unlike the thrusts of conventional warfare or the ‘hit and run’ strikes of guerrilla strategy, irregular offensive operations frequently resemble trickles. Irregular advances often follow the route of least resistance. In Sierra Leone, for example, the Revolutionary United Front (RUF)'s decentralised advance edged forward, following the path of least resistance. A farmer from the Pujehun District in Sierra Leone recounted that: ‘[t]he RUF boys had two-way walkie-talkies. News of the easy advance was passed on. They said, “Come, there's no resistance. The [All Peoples' Congress] is nothing.” Another group entered Pujehum [district] at Suluma, hearing the word.’Footnote 22 Clearly, instead of concentrating its forces for a conventional drive, or using guerrilla infiltration, the RUF captured territory through a third form of offensive strategy, that is, capturing territory and asserting its control over population by advancing into unprotected space. In short, in offensive operations, irregular strategy attempts to achieve two objectives: to capture territory and avoid ‘pitched battles’.
In defence, belligerents that employ an irregular strategy attempt to hold and defend territory and population. In practice, this characteristic is evidenced in a number of ways. One example is the roadblocks and checkpoints that frequently litter civil wars that are fought using irregular warfare. Across cases of civil war, the sturdiness of roadblocks varies dramatically, from concrete guardhouses through to ‘a few stones or empty plastic crates’, as reported in Rwanda.Footnote 23 Regardless of construction, the role of roadblocks remains to assert a belligerent's control over a territory and population. The appearance of roadblocks and checkpoints in a civil war frequently create unmistakable frontlines. However, unlike the frontlines in conventional warfare, these frontlines are relatively lightly defended and frequently appear in large numbers. Indeed, a single city may be divided between several belligerents, each asserting its control over different suburbs; such was the case in Beirut, Mogadishu, Baghdad and Monrovia. In Monrovia, the different militias ‘raced around in pickup trucks on their side of the city, clutching assault rifles, rocket launchers and two prized anti-aircraft guns’.Footnote 24 Although roadblocks are also commonly used to control territory and population in counter-guerrilla, these small obstacles represent easy targets for powerful opponents and, as such, it is generally only in irregular warfare that insurgent actors use these constructions.
Case Study: Foreign intervention and warfare in the Angolan Civil War
The behaviour of external powers not only shaped the conflict in the Angolan Civil War, but was also responsible for its onset. On 25 April 1974, in what became known as the Carnation Revolution, a cabal of army officers overthrew the authoritarian Portuguese Government in Lisbon. The new government announced, on 27 July, that all Portuguese overseas territories would be granted full independence. In Angola, the distant coup instantly transformed the political and military landscape of the anti-colonial struggle between the Portuguese Governmental Authority and the African guerrilla movements. Since the outbreak of hostilities in the early 1960s, competition between the anti-colonial guerrilla movements had always been an undercurrent in the overall war against Portuguese rule. However, the Lisbon coup elevated that rivalry to the principal political and strategic motivation of the belligerents. Thus, with independence imminent, the rival anti-colonial movements trained their limited capabilities away from the Portuguese and onto each other.
At the outbreak of the Angolan Civil War, there were three leading contenders for power: the Movimento Popular de Libertação de Angola (MPLA), the Frente Nacional de Libertacão de Angola (FNLA) and União Nacional para a Independência Total de Angola (UNITA). The MPLA and FNLA had fought the Portuguese, and occasionally each other, since 1961. UNITA was formed by Jonas Savimbi in 1965 after splitting from the FNLA. There were a number of contributing factors to the belligerents' irreconcilable differences. These included their contending political ideologies, their leaders' egos and the fact that they each recruited from different ethnic populations. Clashes between the three belligerents escalated from late 1974 to plunge the country into civil war.Footnote 25
Foreign intervention in the Angolan Civil War was complicated, contradictory and, at times, convoluted.Footnote 26 Although the belligerents had received assistance in their struggle against Portugal, the announcement of Angola's pending independence represented a turning point in the pattern of intervention in Angola. Suddenly, following the 1974 Lisbon coup, the MPLA could now request assistance from Moscow and Cuba to resist the ‘neo-colonial challenge’ represented by its opponents, while on the other hand, the FNLA and UNITA could seek outside help in resisting the ‘communist threat’ posed by the MPLA.Footnote 27 The actors and interests remained constant; however, foreign capitals began to conceive the conflict as a contest between East and West, an impression the belligerents were more than happy to cultivate.
How did foreign intervention affect the warfare in the Angolan Civil War? To answer this question, this section trails the causal relationship between foreign intervention and warfare. It focuses upon the first period of the civil war (November 1974 – March 1976) and is divided into four parts. This first part provides an overview of the initial period of the civil war. The next three parts trace the impact of foreign intervention on the balance of military capabilities and the pattern of warfare.
Overview
Over the span of the first period (November 1974–March 1976), there were three significant changes in the pattern of foreign intervention. The first configuration of foreign intervention extended from November 1974 through to October 1975 and was characterised by very small volumes of external aid to the belligerents. During the second quarter of 1975, however, the pattern of foreign intervention underwent a dramatic change, with all sides gaining increased quantities of sophisticated foreign weaponry, finance and troops. This change catalysed the advent of a second distinct phase in the pattern of foreign sponsorship from November 1975 to January 1976. Finally, in late January 1976, the pattern of foreign intervention changed for a third time. In Phase III the FNLA and UNITA lost their foreign sponsors, while the MPLA consolidated its position with the continuation of significant volumes of foreign assistance.
Corresponding with the three different patterns of foreign intervention, there were also three changes in the balance of military capabilities and the type of warfare. During the purview of Phase I, three, roughly evenly matched belligerents contested the civil war, each with very low military capabilities. As the theory predicts, this configuration in the balance of military capabilities produced irregular warfare. In Phase II, the introduction of large volumes of exogenous resources to all belligerents maintained a relative parity in the distribution of capabilities while significantly increasing the average military capabilities. The result was conventional warfare. In Phase III, the disengagement of the FNLA and UNITA's sponsors, coupled with battlefield defeats, had two impacts on the balance of military capabilities. It propelled the distribution of capabilities in the MPLA's favour and decreased the average military capabilities in the civil war. These changes in the balance of military capabilities led to guerrilla warfare. Figure 3 traces these changes in the balance of military capabilities and warfare.
Figure 3. Variations in the balance of military capabilities in the Angolan Civil War (November 1974–March 1976).
Foreign intervention and irregular warfare, Phase I
In November 1974, a year before formal independence, the belligerents' ongoing political jostling broke out into open warfare. At first, the major clashes were centred in Angola's political and economic capital, Luanda. Initially the People's Movement for the Liberation of Angola (Eng. trans.) (MPLA) stronghold, a ceasefire in October 1974 between the departing Portuguese administration and the anti-colonial movements led the National Liberation Front of Angola (Eng. trans.) (FNLA) to challenge the MPLA's authority in the capital by establishing political headquarters within the city. Almost immediately, street clashes erupted between the respective militias, which on 10 November resulted in almost 50 dead.Footnote 28 The street fighting rapidly escalated and, as is typical of irregular warfare, there was almost daily clashes between the different belligerents' militias without any significant, let alone decisive, battles. Nonetheless, by March 1975, the street skirmishing in Luanda had resulted in the deaths of some 20,000 people.Footnote 29 The National Union for the Total Independence of Angola (Eng. trans.) (UNITA) was considerably weaker than either the MPLA or FNLA and so did not risk competing for the capital.
Meanwhile, outside of the capital, Angola had been carved up by the three leading warlords: Agostinho Neto, Holden Roberto and Jonas Savimbi. Neto's MPLA controlled the coast (including the oil rich province of Cabinda) and incorporated parts of the inland province of Moxico, Roberto's FNLA remained rooted in the north, while Savimbi's UNITA was concentrated mostly in the central highlands and southeastern Angola.Footnote 30 During Phase I there was sufficient buffer space between the different spheres of influence that major clashes in rural areas were infrequent. However, when clashes did occur in the vast areas separating the belligerents' base areas, the low troop density resulted in small groups of poorly armed and inexperienced militia sporadically attacking other like units.
Foreign intervention in Phase I
Foreign powers showed only cursory interest in the anti-colonial war in Portuguese Angola.Footnote 31 Although Moscow had begun providing some military aid to the MPLA following Neto's 1964 Soviet Union visit, the value of Soviet aid up until 1974 nevertheless only totalled in the vicinity of US $63 million.Footnote 32 Scholars generally agree that whatever the precise value of Soviet assistance to the MPLA, it was ‘relatively low, affording minimum operations, but never providing the movement with the means to prevail against the Portuguese’.Footnote 33 It was Cuba that showed the greatest interest in the MPLA. From 1966, most MPLA cadres had received some political and military training from 1,000 Cuban advisors who had been based in Brazzaville to support the revolutionary Congolese government.Footnote 34 When the MPLA members passed through Cuban training schools, they were also supplied with an unknown quantity of light weapons, technical training and financial support. The combined effect of the Soviet and Cuban efforts was to maintain the MPLA as a viable organisation, but not to significantly increase its military capabilities.
The US and South Africa were even less interested in Portuguese Angola. Prior to 1974, Washington transferred no resources to any of the belligerents in Angola. The US broadly adhered to the UN embargo on providing equipment to the Portuguese African colonies. South Africa, like the US, did not support any of the future civil war belligerents prior to 1974. Indeed, South Africa cooperated extensively with the Portuguese authorities in Angola, which allowed the South African Defence Force (SADF) to search for South West Africa People's Organisation (SWAPO) guerrillas in southern Angola. In exchange, the SADF would attack any MPLA, FNLA or UNITA fighters it discovered during its incursions.Footnote 35
The balance of military capabilities in Phase I
When the MPLA, FNLA and UNITA forces began to emerge from the Angolan hinterland in 1974, ‘they were little more than a collection of small guerrilla units that had rarely seen their comrades let alone fought alongside them’.Footnote 36 The groups that exited the jungle were small in number, poorly-armed and ill-equipped. At the outbreak of the civil war, the MPLA could count on only an estimated 1,500 fighters.Footnote 37 However, this belies the fact that it was also the fastest-growing belligerent and, by January 1975, could scrap together some 6,000 militiamen.Footnote 38 In 1975, the MPLA also had several factors operating in its favour. Foremost among these was that the MPLA's forces were concentrated in Luanda and represented the largest militia force in the city. This allowed the MPLA leadership to gain control over its forces more quickly than the other belligerents, whose forces were still scattered in rural areas. Nevertheless, the FNLA's 10,000-strong force was the largest military force in Angola and looked poised to sweep through and capture both Luanda and Cabinda.Footnote 39 UNITA was the weakest of the belligerents, fielding a particularly poorly equipped force of only about 2,000 fighters.Footnote 40 Consequently, UNITA sheltered its forces in the hinterland, content to remain a bystander as the MPLA and FNLA fought for control over Luanda.
The quality of the MPLA, FNLA and UNITA's militias was also poor. Although the commanders frequently had years of guerrilla warfare experience gained through fighting the Portuguese, they typically led untrained and inexperienced militiamen. The MPLA Comandante Ndozi reported that the ‘troops with me are boys taken straight from the streets to the front. They ought to be in school, but we closed the schools in order to have an army, since we have to defend ourselves.’Footnote 41 Not only were belligerents limited to poorly trained, inexperienced troops, but also none of them possessed heavy weapons, armoured vehicles or even adequate small arms. For instance, in 1974 it was estimated that UNITA possessed no heavy weapons and only about 1,500 antiquated rifles.Footnote 42 Along with a general lack of weapons, commanders also complained to journalists of an ‘eternal’ shortage of ammunition.Footnote 43 Thus, at the outbreak of the civil war, there was a low average military capabilities and the distribution of capabilities between the MPLA and FNLA was close to parity.
Irregular warfare in Phase I
During Phase I, the principal offensive aim of the MPLA and the FNLA was to eradicate the influence of their rivals in the capital. Underlining this aim was the belief that the party that controlled the capital at the time of formal independence would be in a strong position to claim sovereignty over all Angola. To this end, the belligerents employed similar violent methods in their attempts to seize power. The fighting in Luanda was disorganised, opportunistic and brutal. The respective belligerents distributed rifles and grenades freely among their supporters – including to teenage boys – which resulted in running street battles between the rival sides.Footnote 44 During these early days of the civil war, a common method of attack involved carloads of gunmen speeding past opponents and spraying them with bullets and throwing grenades.Footnote 45 As the violence intensified, the MPLA began to gain the upper hand over its competitors. By the end of April 1975 the momentum was with the MPLA as the impact of the Soviet weapons deliveries the previous month and the recruitment of 3,500–6,000 experienced Katangese gendarmerie swung the street fighting in its favour.Footnote 46 The fighting between the militias continued in Luanda until 20 July 1975, when the FNLA was forced to withdraw from the capital city. It then regrouped, about 150 kilometres north, in Ambriz, where it immediately began military preparations to return to Luanda before Independence Day and claim power.Footnote 47
The pattern of offensive tactics in rural areas took the form of skirmishes, which essentially involved rival militia groups engaging in ‘displays of firepower’.Footnote 48 Fighting would result when one side attempted to encroach on another's area of influence. Ryszard Kapuscinski – one of the few journalists physically on the frontline – described a typical offensive attack as follows:
the [MPLA] unit moves up close to where the enemy is. We open fire just before dawn. The inexperienced soldier thinks the main thing is to make a big racket. He fires like a man possessed, blindly, because all he cares about is noise, communicating to the enemy how much strength is approaching. This is a form of warning, a way of evoking a fear in the opponent that will be greater than ours. And there is a sort of rationale to it. Because the other side is also unfamiliar with war, unfamiliar with gunfire; surprised by volley, they withdraw and flee.Footnote 49
The effectiveness of offensive operations created a paradox. At the tactical level the warfare was fluid. Journalists reported that travelling through rural areas was highly unpredictable as villages and roadblocks regularly changed hands. But, at the strategic level, no side possessed the military capabilities to make decisive and permanent gains into their rivals' territory.
Early in the civil war, the defences of the belligerents were generally flimsy and unsophisticated. In Luanda, defensive positions were limited to securing buildings and establishing roadblocks,Footnote 50 while outside the capital, defensive positions were mostly limited to roadblocks. The aim of roadblocks was to assert control over, and hamper an enemy's encroachment into, a controlled area. They were generally feebly constructed, using whatever materials were on hand. As one observer chronicled,
[o]n important routes where major checkpoints are found, the road is blocked by colorful barriers that can be seen from a distance. But since materials are scarce and improvisation is the rule, others do the best they can. Some stretch a cable at the height of a car's windshield, and if they don't have cable they use a length of sisal rope. They stand empty gasoline drums in the middle of the road or erect obstacles of stones and volcanic boulders. They scatter glass and nails on the macadam. [Or t]hey lay down dry thorn branches.Footnote 51
Besides being flimsily built, the roadblocks, inside and outside the capital, were usually guarded by poorly armed, equipped, and trained militias. The individuals manning the roadblocks were of ‘diverse professions and ages. Rear-guard soldiers, homegrown militia, and boys caught up in the passion of war, and often simply militia.’Footnote 52 The gunmen defending the roadblocks were lightly armed with a miscellaneous assortment of weapons, including ‘submachine guns, old carbines, machetes, knives, and clubs’.Footnote 53 This weaponry, although sufficient to manage traffic, failed to provide much protection against enemy forces. Thus, the roadblocks were intended to impose the belligerent's control over, and deter rival movements expansion into, an area, rather than repel a determined enemy assault.
Foreign intervention and conventional warfare, Phase II
By the second half of 1975, the strategic situation in Angola had changed considerably. By August, the MPLA had cemented its control over the capital city and emerged as the largest and most militarily capable belligerent in the country. The FNLA, meanwhile, had concentrated its forces to the north of Luanda and UNITA's forces were still scattered in pockets across central and southern Angola. The FNLA and UNITA had also entered into an uneasy ‘alliance’.Footnote 54 Throughout 1975, all the foreign powers gradually increased their respective commitments in the civil war. However, late October and early November represented a clear qualitative and quantitative divide in the pattern of foreign intervention in Angola. Until this point, the exogenous resources received by the belligerents had principally been in the form of military equipment. From late October, however, Cuban, Zairian and South African troops began to appear inside Angola in support of their respective allies.
Foreign intervention in Phase II
At the outbreak of the civil war, the small size and limited capabilities of the belligerents encouraged foreign powers to intervene in the Angolan Civil War. That is, foreign powers could ‘get in the game’ at relatively low cost. Once involved, however, Moscow, Havana, Washington and Pretoria found that they were collectively obliged to continually escalate their respective interventions in response to each other's actions. The overall effect was that by November 1975, vast quantities of exogenous resources were flowing to the belligerents, radically altering the pattern of foreign intervention in the civil war.
Although the MPLA received assistance from a range of sources, its two principal sponsors remained the Soviet Union and Cuba. The dawn of 1975 heralded a new wave of Soviet military aid to the MPLA. Between March and November 1975, Moscow delivered an estimated 30 to 40 cargo planes and 27 shiploads of weapons to the MPLA. By January 1976, it is estimated that Soviet military aid to the MPLA had already reached $200 million, which included conventional assets such as T-34 and T-55 tanks, armoured personnel carriers, 122-mm Katyusha rockets and MiG-17 and MiG-21 aircraft.Footnote 55 The flood of more sophisticated Soviet weaponry to the MPLA created a need for foreign trainers and advisors. On 25 July 1975, it was Cuba, which responded to Neto's appeals and deployed 50 weapons specialists to Brazzaville.Footnote 56 The Cuban commitment to the Angolan Civil War then too proceeded to rapidly escalate. In August, the Cuban training contingent was increased to 480 specialists.Footnote 57 Next, on 4 November, the decision was made to launch ‘Operation Carlota’ which would eventually inject over 30,000 Cuban troops into the Angolan Civil War. The first of these troops – mostly specialist artillerymen and Special Forces – landed in Luanda on 7 November. The first Cuban ships carrying the bulk of the intervention force steamed into Luanda on 27 November, carrying in excess of 1,200 troops and all their equipment.Footnote 58
On the FNLA and UNITA's side, the key intervening powers were China, the US, Zaire and South Africa. The US' intervention in Angola began tentatively, only providing Holden Roberto with $265,000 between 22 January and 14 July 1975.Footnote 59 It was not until President Ford signed ‘Operation IAFeature’, on 18 July 1975, that significant volumes (roughly $31.7 million worth) of American assistance begun to be provided.Footnote 60
It was left to Zaire and South Africa to provide specialist troops to the FNLA and UNITA. In May 1975, approximately 1,200 Zairian troops crossed into northern Angola in support of the FNLA. The South African involvement in Angola was a mix of economic support, equipment and direct military intervention. First, in July 1975, Pretoria sent $14 million-worth of weaponry to the FNLA and UNITA. Next, in September, the South African intervention was expanded to include training camps inside Angola itself, when 12 South African instructors established a camp at M'pupa to train several hundred FNLA soldiers. A second camp was set up at Calombo by 18 SADF instructors to train UNITA recruits into two infantry brigades.Footnote 61 Finally, in October, South Africa launched ‘Operation Savannah’ which represented South Africa's first major injection of troops into the civil war. On 28 October, 1,000 Bushman infantry, 50 armoured cars manned by 250 soldiers, and an additional 750 SADF support troops, reinforced the FNLA recruits at M'pupa to become Combat Formation Zulu.Footnote 62 Similarly, a second battle group, codenamed Foxtrot, was organised from the UNITA and SADF soldiers at Calombo and allocated a squadron of Eland armoured cars.Footnote 63
The balance of military capabilities in Phase II
As weapons, equipment, and combat troops flooded into the civil war, the balance of military capabilities underwent a radical change. Since the outbreak of the civil war, all the belligerents' military capabilities had been steadily increasing, a trend that accelerated sharply during the first six months of 1975. The result, in Phase II, was that the average military capabilities moved towards much higher levels.
Indeed, in this phase, the MPLA had already quadrupled in size, the FNLA had grown by half and the number of UNITA troops had doubled.Footnote 64 The quality of these forces also improved. Large numbers of the belligerents' soldiers received training in conventional military proficiencies by Cuban, Zaire, and South African instructors. Foreign officers, either directly or indirectly, led these new regular formations. In addition, the quantity and quality of the equipment used by the belligerents' forces steadily improved during the first year. From using antiquated rifles, grenades, and machetes, the belligerents had upgraded their arsenals to include armoured vehicles, artillery and, in some instances, combat aircraft. For example, the MPLA acquired some 198 BTR-40, BTR-50 and BTR-60PB armoured personnel carriers, 68 PT-76 light tanks, 80 T-34/85 tanks, 150 T-54 main battle tanks, 100 BM-21 122mm mobile rocket launchers, 40 ZSU-57–2 anti-aircraft guns, and 8 MiG-17 and 24 MiG-21MF/Fishbed-J fighter aircraft.Footnote 65 Finally, in addition to the increasing levels of exogenous resources pouring into the country, the departing Portuguese army abandoned significant quantities of arms that also found their way into the possession of the belligerents.Footnote 66 In sum, hundreds of millions of dollars worth of weapons and thousands of troops pumped into the civil war by foreign sponsors greatly increased the coercive potential of the belligerents.
During this phase, the distribution of capabilities remained relatively balanced. However, it was the MPLA, which benefited the most from its foreign sponsors and, consequently, it began to increase its advantage over its rivals. The MPLA's particularly rapid growth meant that by August 1975 it had overtaken the FNLA as the largest of the belligerents in Angola. By the end of 1975, the MPLA could call upon an estimated 20,000 soldiers.Footnote 67 Furthermore, due to Cubans being the most numerous foreign soldiers in Angola, the MPLA forces were generally better trained and organised than their rivals. A correspondent from the London Times reported that the ‘major advantage’ of the MPLA, ‘which did not possess any significant weaponry superiority’ over its competitors was its ‘manifestly superior organizational and infrastructural capability’.Footnote 68 Organisational superiority, along with greater volumes of external assistance, allowed the MPLA to gradually gain an edge over the FNLA and UNITA.
The MPLA's position was, however, not insurmountable and the FNLA and UNITA continued to pose a serious challenge. The FNLA remained the MPLA's main rival with an army of 15,000 troops in a threatening position to the north of Luanda.Footnote 69 It was reinforced by a thousand Zairian troops, along with armoured vehicles and artillery. UNITA's force strength, concentrated in the south of the country, is more difficult to estimate. Yet, at the end of 1975, UNITA is likely to have fielded about 4,000 troops.Footnote 70 Hence, when combined, the FNLA and UNITA could assemble almost as many troops as the MPLA.
Conventional warfare in Phase II
From October 1975, observers of the Angolan Civil War began reporting dramatic changes in the nature of the warfare. Kapuscinski, for instance, observed that the ‘war in Angola has changed in character. Until recently it was primarily […] fought with light weapons […but] [t]oday it is more and more a war of regular armies and heavy equipment.’Footnote 71 Similarly, a Western journalist reported that the warfare ‘is no longer a bush war of small resistance bands. It is becoming a sort of conventional war of rapidly moving vehicle columns, artillery and projectiles.’Footnote 72 The rapid change in warfare was a direct result of the large volumes of exogenous resources provided to all sides in the civil war.
Over the course of the conflict's first year, offensives changed from disorganised, sporadic, and random assaults on enemy buildings in Luanda and occupied villages, to massed, concentrated, drives towards key cities and strategic positions. From October, large columns of motorised troops, supported by amour and artillery, zig-zagged across Angola following the road system that linked the major provincial towns. On the southern front, as battle groups Zulu and Foxtrot advanced into Angola, they entered a theatre dominated by irregular warfare. Previously, as the belligerents had possessed limited military capabilites, neither side had been able to exploit the weakness of their rival. Thus, as Edward George described, ‘the war in southern Angola had degenerated into a messy stalemate, each movement skirmishing with its opponents whilst making no effort to capture their power bases’.Footnote 73 As such, it is unsurprising that when injected into this environment, Zulu and Foxtrot made easy progress, driving deep into Angola. Their concentrated thrust captured 250 miles in only five days, making quick and easy progress against the MPLA. Zulu captured many of the main southern towns before, on 28 October, the South African force, along with 1,000 UNITA soldiers, captured its objective: the southern port of Namibe.Footnote 74 Meanwhile, Foxtrot was tasked with securing Huambo, Angola's second-largest city. On 5 October, Foxtrot successfully captured the city after defeating the dug-in MPLA-Cuban defenders in a pitched battle to the city's west.
Over the span of Phase II, a clearer distinction between offensive and defensive tactics emerged. In defence, the belligerents prepared fixed defensive positions characterised by trenches, bunkers, dug-in artillery and machine guns. For the belligerents, equally as important as the tools of war that had been supplied by the foreign powers was the technical expertise supplied by Cuban, Zairian and South African specialists, who instructed the belligerents' former-guerrilla fighters and raw recruits in the methods of positional defence. The effect of this training became evident in the MPLA's defence of Luanda.
As the MPLA held the capital and the Cabinda region, it was a greater proponent of defensive tactics.Footnote 75 The moment the first Cuban Special Forces battalion landed in Luanda on 8 November, it set about organising the defence of the city. The topography to the north of Luanda is favourable to a defender. Here, the main northern road passes through the Quifangondo Valley, a natural defile with the Atlantic Ocean to the west and the impassable Panguila Lagoon to the east. Upon landing, Cuban combat engineers immediately began preparing deep bunkers and trenches and digging in the MPLA's artillery batteries along the northern heights of Luanda, which overlooked the Quifangondo Valley. Surveying the MPLA's defensive preparations, Kapuscinski described the frontline as being ‘heaps of earth, lines of entrenchments, artillery, tents, crates’ and on ‘the other side, in the distance, a sunlit hill, the enemy's fortifications’ at Caxito.Footnote 76 After a week of probing, and with less than 24 hours until independence, Roberto finally ordered the major FNLA assault on Luanda. The FNLA offensive began with it laying down a barrage from three South African-supplied 140-mm howitzers on the MPLA's positions, which was followed by a bombing run by three South African Canberra bombers flying from Namibia. 12 armoured cars, followed by a massed formation of FNLA and Zairian troops, led the assault column.
The MPLA and Cuban defenders withstood the opening bombardment and then patiently waited for the entire assault force to enter the valley. At the precise moment when the last FNLA-Zairian infantry had entered the northeast end of the valley and the first armoured car was about to reach the MPLA-Cuban forward-most defensive positions, the defenders opened fire with machine-guns, mortars, artillery and rocket fire, trapping the FNLA's helpless column. Particularly devastating were the Soviet-supplied 122-mm ‘Stalin Organs’ rocket launchers. Reportedly, in only minutes, these weapons fired two thousand rockets onto the massed FNLA troops filling the Quifangondo Valley with sharp splinters of shrapnel and thunder clap-sounding explosions, which quickly turned the FNLA army into a disorganised and retreating mob.Footnote 77 Roberto's troops fled in disorder, abandoning weapons, vehicles, and wounded comrades, never again to pose a serious challenge to the MPLA. John Stockwell concluded that the Russian supplied rocket launchers ‘as much as any one thing, eventually decided the outcome of the civil war in Angola’.Footnote 78
Clearly, in sum, the new pattern of foreign intervention was instrumental in changing the form of warfare in Angola from irregular to conventional. This shift required ‘know-how, leadership and planning on a level which [was] not readily available among Angola's black population’.Footnote 79 Thus, the move from irregular to conventional warfare was, to a large extent, ‘fuelled by increasing amounts of arms received by the three movements from abroad’.Footnote 80 The warfare changed from irregular to conventional because all sides received sufficient levels of foreign assistance.
Foreign intervention and guerrilla warfare, Phase III
By the end of January 1976, the pattern of foreign intervention in the Angolan Civil War had changed for a third time. The MPLA received reduced levels of external assistance, while the foreign support to the FNLA and UNITA was cut altogether. The FNLA and its supporters had been decisively defeated at Quifangondo, while UNITA's principal sponsors disengaged from the civil war, severing its flow of exogenous resources. This deviation was sufficient to produce Phase III beginning from late January 1976.
In Phase III, the majority of the fighting shifted from the outskirts of Luanda and the southern coastal roads to central Angola to the southeast of the country. This was primarily for two reasons. First, UNITA's ethnic base was among Angola's 2 million Ovimbundu people located in the central highlands, who became its main source of support.Footnote 81 Second, the MPLA's political, economic and military centre was in Luanda and, like most weak regimes, its authority decreased over distance.Footnote 82 Hence, UNITA was able to avoid the MPLA by escaping into Angola's remote outer provinces from where it began a guerrilla campaign.
Foreign intervention in Phase III
In early 1976, the pattern of foreign intervention changed for a third time. Somewhat ironically for the Soviet Union and Cuba, the MPLA's victory over the FNLA and UNITA committed them even more heavily to Angola. For Moscow and Havana the price of maintaining an ally in southern Africa was that it could not immediately disengage from the civil war. Nevertheless, they both made attempts to extricate themselves. The Soviet Union honoured its commitment to supply $400 million worth of arms in 1976, but swiftly cut its aid in 1977 to $150 million per year, which remained steady until 1980.Footnote 83 Havana, likewise, attempted to scale back its involvement. However, continued fighting and political disorder within the MPLA resulted in only a slight drop from the 36,000 Cuban troops at the start of 1976 to 34,000 troops in January 1977. This downward trend was reversed in 1979 when the number of Cuban soldiers in Angola climbed to 40,000.Footnote 84
The US and South Africa disengaged from the Angolan Civil War in December 1975 and January 1976, respectively, leaving a wounded UNITA, scattered, demoralised, and without any source of exogenous resources. The growing realisation in Washington that Moscow and Havana were deeply committed to the Angolan conflict caused great anxiety among certain political circles. The US Senate passed the Clark Amendment on 19 December, prohibiting further CIA assistance to any belligerent in the civil war. Among the repercussions of this decision was that South Africa was left politically and strategically isolated. The first South African direct military intervention in the Angolan Civil War ended on 24 January 1976; after the South African Prime Minister announced that he was ‘not prepared to fight on behalf of the free world alone’.Footnote 85
The balance of military capabilities in Phase III
The American and South African disengagement from Angola radically reshaped the balance of military capabilities. The average military capabilities within the civil war experienced some changes. In 1976, significant amounts of military capabilities were removed from Angola, with the defeat of the FNLA's army catalysing the removal of roughly 20,000 FNLA personnel, 1,200 Zairian troops and all their armoured vehicles and artillery,Footnote 86 and the withdrawal of all of the SADF's armoured vehicles, artillery, and personnel to Namibia. In 1977, Cuba withdrew 2,000 troops out of Angola, scaling its commitment back to 34,000 soldiers.Footnote 87 These factors combined to decrease the average military capabilities.
However, in Phase III, the most significant change in the balance of military capabilities was in the distribution of military capabilities. In 1977, the combined MPLA-Cuban army totalled over 56,000 personnel, equipped with Soviet-supplied tanks, artillery and advanced attack aircraft.Footnote 88 On the other hand, neither the FNLA nor UNITA were potent military forces. Indeed, following the battle at Quifangondo Valley, the FNLA had ceased to exist as a coherent belligerent. The UNITA, for its part, could only call upon between one and three thousand poorly armed and trained guerrilla fighters scattered across central and southeastern Angola.Footnote 89 ‘Practically alone against the MPLA-Cuban forces’ observed Guimarães, ‘UNITA was routed and its forces dispersed’.Footnote 90 By September 1976, one journalist summed up the military situation in Angola as follows:
[the] MPLA had gained control of twelve out of Angola's sixteen districts. But their real position was in fact much stronger than this position may indicate. Nourished by a flood of volunteers and led by veterans of the long struggle against the Portuguese, the army of the MPLA is now in control of the whole of Angola except for the two northerly districts […and] two central districts […] and a few scattered points in the far south.Footnote 91
Indeed, in April 1976, the MPLA had captured N'Giva, which had been the last remaining provincial capital outside of its control. Clearly, the distribution of capabilities between the MPLA and UNITA was extremely wide. The MPLA's distribution of capabilities was almost total, with UNITA's military capabilities representing a tiny fraction of the total capabilities present in the civil war system.
The interplay of the decrease in the average military capabilities and the shift towards disparate distribution had consequences for the course of the civil war. Only three days after the South African evacuation, UNITA abandoned its political and operational capital at Huambo and retreated into Angola's eastern bushland.Footnote 92 Most observers predicted that the ill-trained peasant force would quickly disintegrate.Footnote 93 However, the military capabilities of the MPLA were not sufficient to impose control over all Angola and as a consequence, UNITA was able to escape annihilation and adopt a guerrilla strategy. Savimbi disclosed that since ‘early January my feeling had been that the conventional war was over’.Footnote 94 He argued that the change in the pattern of foreign intervention was the cause of UNITA's adoption of guerrilla strategy, insisting that it ‘was not possible for us to continue because we were not being given the right kind of arms, and even if we did begin to get them we did not have enough people trained to use them effectively […so I gave] the order to our people to disperse into the bush so that we [could] save men and arms’.Footnote 95
Guerrilla warfare in Phase III
In 1976, guerrilla warfare emerged as the dominant mode through which the civil war was fought.Footnote 96 Unlike previous evolutions in warfare, however, where the belligerents' style of warfare had changed in similar ways, this time their tactics evolved differently. On the one hand, the MPLA attempted to hunt down the remaining UNITA fighters by conducting large, multi-brigade, conventional sweeps through central and southern Angola. On the other hand, UNITA attempted to avoid contact with the MPLA's forces while striking at its economic base. In other words, the warfare had evolved into a classic guerrilla pattern.
The MPLA dubbed its operation against UNITA as the ‘Fight against the Bandits’.Footnote 97 Between April and November 1976, Cuban soldiers spearheaded four search-and-destroy operations in the Moxico, Cuando Cubango and Cunene provinces against UNITA guerrillas. Until the early 1980s, such joint MPLA-Cuban search-and-destroy operations became the regime's primary form of aggressive action against UNITA. Villages would be surrounded by the MPLA-Cuban forces which then ‘searched for hidden supplies, ammunition and any valuables, all of which was confiscated (and a great deal looted by the troops) before the [village] was set alight and the troops withdrew’.Footnote 98
For UNITA's part, in early 1977, Savimbi declared that his military objective was to bring ‘the Angolan economy to its knees’.Footnote 99 As such, UNITA's early attacks took the form of sabotage and raids on government sources of endogenous inputs. One of UNITA's favourite targets was the Benguela railway, which ran from the Zairian border across provinces where UNITA was active to the Angolan port city of Benguela.Footnote 100 Indeed, sabotaging a section of the 1,000-kilometre track was a relatively easy task for UNITA. Suspended, destroyed, or stolen cargo cost the MPLA some $89 million per annum.Footnote 101 Although UNITA's forces were focused in the southeast, on occasion its guerrilla units struck as far north as Luanda and even in Cabinda. On 30 November 1981, for instance, UNITA attacked and damaged an oil refinery in Luanda.Footnote 102 As a consequence, the MPLA was forced to allocate fixed and permanent garrisons to defend all of Angola's provincial capitals and commercial centres, especially along the Benguela railway and around oil-producing instillations in Cabinda. The task of protection was so large that eventually the majority of the MPLA's troops were assigned to defensive duties. In addition, most of the Cuban contingent was also appointed to defensive positions. Each of Angola's 17 provincial capitals was defended by at least one Cuban regiment and much of the remaining Cubans defended Luanda.Footnote 103 The larger garrisons, such as those at Huambo and Lubango, had up to 5,000 Cuban soldiers assigned to their defence, supported by tanks, artillery, radar and anti-aircraft missile defences.
In defence, UNITA's strategy had two dimensions. First, UNITA's forces refused to stand and fight against the MPLA's larger, more sophisticated and better-trained forces. Savimbi boasted that when the MPLA ‘come with tanks, it is true, we will run away. But we will return when they have passed. They've just wasted petrol.’Footnote 104 UNITA had developed an organisational structure that was highly mobile. Even UNITA's ‘capital’ called Jamba was, in practice, nothing more than a roving command centre. It was reported that Jamba, for security reasons, was sporadically relocated in the southeastern part of Cuando Cubango province.Footnote 105 The second dimension of UNITA's defence strategy was social. UNITA used the forceful relocation of civilian populations as a security measure against ‘information leakage’.Footnote 106 UNITA developed a classification system, which only allowed those deemed to be loyal to UNITA to remain near areas controlled by the MPLA. All ‘others were displaced to areas more difficult to flee, such as the bush’.Footnote 107
Conclusion
As observed at the outset, there has been a paucity of theory that has sought to explain how foreign intervention influences the course and nature of the warfare in civil wars. As such, this article sought to make a modest contribution to filling this void. It was submitted that foreign assistance alters the balance of military capabilities between the belligerents, which, in turn, is responsible for influencing the form of warfare that emerges at a particular place and time in a civil war. The case study supported the theory. Variation in the pattern of foreign intervention was instrumental in shaping the type of warfare that emerged in the first period of the Angolan Civil War. Indeed, the significance of foreign intervention on the course and nature of the Angolan Civil War has not been lost on students of the period. Commenting on the change from irregular to conventional warfare, George remarked that ‘Cold War military technology had arrived dramatically in Angola, and the scale of forces engaged at Quifangondo demonstrated how far foreign intervention had escalated the conflict since street-fighting broke out exactly one year before.’Footnote 108 The ease with which the warfare changed from irregular, to conventional, and then to guerrilla was the most striking feature of the opening stage of the civil war.
Since 1945, civil war and foreign intervention have largely replaced international war as the most frequent form of armed conflict. Especially across Africa, leaders have demonstrated a preference towards supporting insurgent groups in neighbouring countries rather than declaring war.Footnote 109 Indeed, in many cases, bilateral agreements to stop supplying insurgent organisations within each others' countries have replaced traditional ‘peace treaties’. This fact pattern shows no signs of reversing and, so, foreign intervention and civil war are likely to remain common phenomena into the foreseeable future. As such, any progress in understanding their dynamics and particular characteristics must be welcomed. As further research emerges on the specific differences between the characteristics of the types of warfare (in particular, data on their respective duration and lethality) the balance of military capabilities approach to conceptualising the impact of foreign intervention will continue to grow in value to commentators and policymakers alike.
