The road to war

Political failure and the ‘diamond war’, 1997-1998

When government forces attacked the last strongholds of the União Nacional para a Independência Total de Angola (UNITA) in November 1998, Luanda’s domination was obvious on many levels. The Movimento Popular de Libertação de Angola (MPLA) suffered only one significant failure in many months: that of UNITA Renovada, a group of dissidents surreptitiously supported by the government, meant to be UNITA’s new rival.

The government ambitiously counted on the prospect that its support for the creation of Renovada would cause the rapid demise of UNITA. Once the failure of Renovada became obvious, the hopes of the Luanda regime faded that it would result in the collapse of UNITA. Combined with other political factors, this hastened the implementation of the military option chosen by Dos Santos.

By attempting to take the city of Baïlundo, UNITA’s symbolic ‘capital’, the Forças Armadas Angolanas (FAA) tried to end a political situation unacceptable to the regime at the time.¹ The political problem embodied in Savimbi remaining as the head of the UNITA ‘state’ had to be eradicated.

Since 1994, Angola was, in fact, a country with two leaders, ensconced in two ‘capitals’, observing each other antagonistically. On the periphery, the UN was wearing itself out trying to maintain an uncertain peace. The UNITA ‘state’, politically diminished on the international stage, was reduced to a few isolated pockets, following the progressive expansion of the Luanda administration throughout Angolan territory in accordance with the Lusaka agreement. The government’s military actions, publicised beforehand in Luanda, officially marked the renewal of civil war. These actions resulted from a latent conflict that began to simmer months before, which the UN could not have prevented.

The first confrontations in the third civil war on Angolan territory occurred in the middle of 1997, when the situation between the MPLA and UNITA deteriorated markedly. The cease-fire was regularly violated, albeit on a small scale.² The growing tension was underpinned by economic and political factors of considerable importance.

At the end of May 1997, the FAA launched "an offensive against the armed rebellion in the diamond province of Lunda Norte, resulting in the paralysis of the peace process provided for by the Lusaka agreement."³ Although the remaining UNITA forces were officially involved in a disarmament and demobilisation process, they resisted the offensive and fought against the FAA. The increasing number of armed incidents affected several provinces. Angolans began killing one another again, and in the north-east of the country, fighting lasted for several months. These were concealed by the warring factions and by UN observers who refused to acknowledge that their mission was failing.

These confrontations resulted officially from the unsuccessful negotiations on the expansion of the state’s authority. However, other issues were at stake:

"The negotiations involving certain areas of major economic importance were about the division of diamond concessions between the government and UNITA."⁴

The conflict waged by Savimbi’s forces since May 1997, aimed to prevent the FAA from taking complete control of Lunda Sul and Lunda Norte, strategic areas from where diamonds are extracted. This was a crucial issue for UNITA and Savimbi declared at the time: "UNITA … will not let go of the diamonds."⁵ UNITA’s determination was unrelenting, since the exploitation of diamonds was an important source of revenue for the movement.⁶

Some Angolan army generals participated in the exploitation of the remaining third of Angola’s diamonds. In their opinion, control over the diamond-rich areas to the detriment of UNITA was a priority and, beyond military considerations, both strategically and economically significant.

Removing UNITA’s control over the mining provinces would effectively diminish its political power and its capacity for armed conflict. However, it was also a potential means to become individually wealthy. Faced with this risk, UNITA stepped up its actions, progressively reverting to guerrilla tactics and diversions.⁷ The ultimate goal of this ‘diamond war’ was to destroy UNITA economically by depriving it of the major part of its financial resources.

From military to political victories

Luanda knew how to use the general context effectively to favour its overall action. The frequent victories of the army, the confirmed political will of the presidency, as well as active diplomacy, were determining factors in the government’s new power. Indeed, the government’s military offensive of December 1998 benefited from preliminary military operations, as well as from three strategic operations carried out successfully in neighbouring countries. These military undertakings reinforced the political domination of the Angolan regime, already in a strong position after the FAA managed to force UNITA to sign the Lusaka agreement in 1994.

The external interventions were all undertaken in countries known to be sympathetic to UNITA’s cause. In May 1997, Angolan troops assisted in ousting President Mobutu, a long-time enemy of the Luanda regime. UNITA took part in the conflict, but in the enemy camp, particularly during the battle of Kenge. With this brutal military campaign, the Angolan civil war, which had never truly ended, was relocated to Zaïre.

The intervention of the Angolan army during the civil war that had bathed Congo-Brazzaville in blood since June 1997, was also decisive. The FAA, which participated in the fight on 12 October 1997 in the country, hastened the fall and exile of President Pascal Lissouba and assisted in ending the civil war. President Denis Sassou Nguesso, an ally of Luanda and the MPLA for decades, took control of the country. However, the Pax Angolana, negotiated in 1997, has only led to uncertain peace in Brazzaville.

There was nothing ambiguous about the true objectives of the Angolan army in these confrontations. Luanda intervened in the former Zaïre and Congo-Brazzaville to deprive UNITA of its rear bases in foreign territory. Although this complicated the logistics of Savimbi’s movement, it was to the benefit of the movement’s activities in the north of Angola. The extensive borderline between the two countries played an essential part in the transit of vehicles, troops, currency, gems, medicines, emissaries, weapons and ammunition.

Luanda could finally take revenge on two enemies of the MPLA — Mobutu and Lissouba — who were both associated with UNITA activities.

However, the primary objective of these external operations was strategic, while also closely linked to the situation in Angola. Protecting the oil-rich enclave of Cabinda from potential incursions by Savimbi’s troops or members of the Frente de Libertação do Enclave de Cabinda (FLEC) remained one of the MPLA’s strategic priorities. For decades, the province’s economic resources had been vital to Luanda in its fight against UNITA. Intervention in Zaïre and Congo-Brazzaville was meant to weaken the logistics and capacity of UNITA by denying it free access to its sanctuaries. This was an essential factor in the conflict, particularly during the tense period from 1997 to 1998.

Finally, in August 1998, the imperative to protect Cabinda induced Luanda to intervene once more in Kinshasa, this time to assist in consolidating President Kabila’s regime, itself confronted with a rebellion.

This spate of military undertakings illustrated how Angolan foreign policy was held hostage by its internal policy of which the primary objective remained the destruction of UNITA. Thus, Angolans from both sides showed how difficult it was to compromise in the long term. In the process, heads of state in neighbouring countries had to acknowledge Luanda’s determination and its capacity to mobilise its forces. President dos Santos was clearly able to wield a military tool that could be utilised on a regional scale.

Political domination

In 1998, the MPLA dominated the internal situation after the apparently successful external activities of the FAA, which achieved the diplomatic and political isolation and encircling of UNITA.⁸ The war of attrition waged by Luanda against its opposition was evidently fruitful. The erosion of UNITA’s political, military, social, psychological, logistic and economic capabilities seemed to have succeeded. Since 1994, Luanda had sought to weaken UNITA’s material forces and moral strength, hoping that the movement, worn out and divided, would stop fighting. President dos Santos had cornered UNITA as far as internal and foreign policies were concerned, and his manoeuvres clearly outclassed UNITA at Baïlundo.

Under the pretext of UNITA’s refusal to continue collaborating with Portugal, Russia and the United States — the troika of observer countries, whose impartiality was contested by the movement — the Luanda regime expelled UNITA ministers from the Government of National Unity and Reconciliation (GURN), and suspended its deputies on 30 September 1998. The Dos Santos government recognised the UNITA Renovada "as the only valid negotiator in the peace process" and obtained the official political support of Southern African Development Community (SADC) leaders for this fledgling movement.⁹

Dos Santos effectively neutralised the UN by requesting it to suspend co-operation with UNITA.¹⁰ The MPLA prevented the UN representative, Issa Diallo, from going to UNITA-held areas, arguing that the FAA was unable to guarantee his safety.¹¹ This was an additional strategy to marginalise Savimbi and his movement by forbidding any direct dialogue with the UN.¹² UNITA was placed in quarantine — Savimbi and the movement’s leadership were geographically and politically isolated.

The difficult mission of the UN had ended without accomplishing its goals. Political normalisation in Angola was a failure, and peace was not attained. Luanda did not take long to demand the departure of UN representatives. This cleared the way for the military solution envisaged by the Angolan presidency. It provoked strong reactions in Angola, particularly from the episcopate, who asked that the UN mission should be maintained. Soon after the renewal of large-scale conflict, the Angolan government ended the mandate of the UN Mission in Angola (MONUA).

President dos Santos succeeded in creating or maintaining close political relations with the troika of observer countries involved in supervising the Angolan peace process. Creating such relationships had not been possible for Savimbi’s movement. Since 1992, UNITA accumulated political mistakes with regard to the troika countries. It consistently accused them of partiality towards Luanda and of being more concerned with their own countries’ interests than with establishing a ‘fair’ peace in Angola. Raphaël Marques, an Angolan journalist well-known for his analysis of both the MPLA and UNITA, also questioned the troika:

"how could Portugal, the United States and Russia — who always were the greatest war instigators in Angola by dumping in it weapons, goods, money and everything that was necessary to continue with the slaughter — suddenly sincerely desire peaceful coexistence between Angolans?"¹³

In fact, Portuguese policy remained hostile towards UNITA. Lisbon felt that the movement was responsible for the renewal of the war, which tore the country apart between 1992 and 1994.¹⁴ Portugal, an important creditor of Angola, remained a privileged commercial partner in Luanda. Once the tumult of the independence period dissipated, the links created by cultural and linguistic commonalities, as well as the political affinities between Lisbon and Luanda, enabled friendships to develop, such as financial and commercial networks, particularly in the mining sector with the close collaboration of the military hierarchy. Lisbon was also active in terms of military co-operation and delivered weapons to Luanda. For UNITA, Portuguese ‘neutrality’ was an illusion.¹⁵

The Moscow-Luanda axis, implemented in 1975, had been essential to MPLA power for years. Even when ideological issues were no longer intense, Moscow remained an important weapons supplier to Luanda. The Russian weapons industry was greedy for revenue paid in hard currencies and Luanda allocated 15% of its gross domestic product (GDP) to acquire such equipment.¹⁶ Commercial stakes were a constant element in their relationship: although Moscow always ‘sold’ its services to Luanda, it remained an efficient ally of the government.

Relations between UNITA and the US had changed considerably since 1992. The alliance uniting the second Reagan administration and the later Bush administration with UNITA, did not survive the arrival of Clinton in the White House. In May 1993, he recognised the Luanda government, a major foreign policy success for President dos Santos.¹⁷ The decision of the Clinton administration displeased Savimbi and his movement. At the time, Washington maintained that Savimbi’s failure to accept the obligations of the Lusaka agreement between 1994 and 1998, had made him one of the main obstacles to peace.

Beyond these political considerations, economic and strategic issues were an integral part of America’s support for Luanda. Washington’s interest in Angola was commensurate with the size of the latter’s offshore oilfields. Angola had the potential to become the foremost oil producer on the continent, ahead of Nigeria. The abundance of Angolan oil resources meant that they were geostrategically important in the context of America’s global policy on the control of energy reserves. The US probably imported up to 14% of their total oil consumption from Angola at the time.

The disappearance of UNITA from the Angolan stage, as announced by the MPLA in November 1998, did not elicit emotive reactions from the troika or, for that matter, from most western governments. UNITA became a complete outcast.

Total war
Luanda’s military failures, 1998-1999

When the conflict resumed, UNITA’s conventional war capabilities came as a great surprise. While most observers thought that the movement had already been disarmed and crushed, UNITA succeeded not only in containing the government’s offensives, but also in counterattacking with surprising strength. Contrary to all expectations, UNITA’s tactical successes accumulated. Government troops were defeated in several clashes. During the first part of the war, the easy military victory imagined by Luanda was the cause of costly FAA failures, and rebel troops advanced to within a few dozen kilometres outside of Luanda.¹⁸

Dos Santos vigorously reacted to rectify the political and military situation, which threatened to spiral out of control. He reshuffled the government, increased the number of MPLA central committee members and replaced his opponents with devoted followers. He appointed a prime minister and a director for Endiama, the state-owned diamond company, and arrested UNITA deputies in Luanda.

Angola withdrew some troops from both Congos. Mercenaries began arriving in Angola to fight on the FAA’s side. Allied contingents of Zimbabwean and Namibian troops, as well as Cuban officers reinforced the FAA.¹⁹ Under the cover of anticipated military service, the authorities began enlisting conscripts from 15-30 April 1999. During this period, 30 000 recruits joined FAA garrisons, further supplemented by press-ganging.

Under an emergency procedure, Luanda received Russian tanks and purchased significant quantities of weapons from Eastern European countries and Brazil. The conflict became internationalised, and its outcome remained uncertain until November 1999.

A government-in-arms ready to go to extremes

UNITA survived the initial shock of the FAA offensive. The movement benefited from strategic advantages that had enabled it to survive for decades, and made it a continuous danger for the MPLA. UNITA’s structure is collegial and pyramidal, acting within a strong political hierarchy, with decisionmaking power concentrated in the hands of Savimbi and his leadership. It is also a militarised political organisation, united and led by a determined and homogeneous élite, convinced of their rights and extremely distrustful. This is probably why it has not yet collapsed, despite severe setbacks that have led, on occasion, to significant losses.

Years of war made UNITA "a war machine that believes first in the strength of weapons."²⁰ UNITA waged its liberation war while completely deprived of equipment and isolated politically and geographically, with no outside help, in the middle of rural populations generally unaffected by Portuguese cultural influences. It therefore developed characteristics appropriate to the movement. UNITA is "a power in arms",²¹ for which political matters always took precedence over military matters within the movement.

While UNITA leaders officially adopted a Maoist ideology, the strategy of the movement was both political and military. Although Savimbi was a war leader, he was first and foremost a political leader exercising "extremely punitive and tyrannical" power.²² Confronted with the MPLA, "UNITA built a totalitarian power, closed and impermeable" in the areas under its control.²³ The population in these areas and its troops were consistently indoctrinated. The majority of UNITA troops were ‘political’ soldiers who fought for what they regarded as a ‘cause’.²⁴ Military officers and soldiers were seasoned by years of fighting. Apart from their own physical survival, the objective of these ‘troops of steel’ was to take power, and to follow in the footsteps of a leader who remained the object of a strong personality cult.

Dissidence and internal opposition were perceived as a threat to cohesion that, from time immemorial, undoubtedly guaranteed survival in a hostile environment. Dissidents were repressed without mercy. UNITA was thus a movement that had "developed an ideology of survival and legitimacy with regard to the means necessary to this end."²⁵ As far as UNITA was concerned, human rights were luxuries in a movement permanently at war since 1966. For UNITA, human rights and democracy were at best accessories and, at worst, dangerous:

"Neither UNITA nor the MPLA were convinced that ‘democracy’ was the best system for governing. A large portion of UNITA’s leaders, including Savimbi, were certain of the contrary, and all their experience made them resistant to it."²⁶

Angolan movements knew that power was won and maintained by force. UNITA regarded the MPLA as the descendant of colonial power:

"There are more Whites, mulattos and Blacks of the same category belonging to the [property] owning class … If the first two groups took advantage of the colonial order … the three categories have been using the colonial order administered by the MPLA to become wealthy … at the expense of the black natives."²⁷

One thing is certain, however: between 1976 and 1990, the MPLA had excluded the Ovimbundu élite from power. As a result, they joined UNITA en masse.²⁸

The past remains an open wound for UNITA. Its leaders fear the repetition of history’s tragedies: the failure to respect agreements made at Alovor, Bicesse and Lusaka,²⁹ attempts to end the movement, and the so-called ‘final’ offensives repeated since 1976, in which UNITA was nearly destroyed on many occasions. The massacres of members in 1975, 1976 and 1992 were seen as extreme acts of treason and perceived as far beyond the violence expected in conditions of war. Atrocities were perpetrated by both sides and both certainly felt a deep-seated hatred towards the other.

Confronted with Luanda’s attrition strategy after the Lusaka agreement, UNITA’s distrust became even more intense. Disarmament, negotiated in the agreement, had been imperfectly executed. UNITA did not hand over all its weapons, but retained part of its fire power. The limited disarmament of UNITA reflected the complete lack of trust between the parties to the conflict. The political weakness and ‘partiality’ of the UN reinforced UNITA’s belief that it was the victim of a troika and MPLA plot in which the UN was only an instrument.

UNITA denounced the fact that the actual disarmament process was aimed only at its troops. President dos Santos had explicitly refused to proceed with the disarmament and demobilisation of the surplus strength of the FAA, in complete contradiction of the commitments made by Luanda.³⁰

Although UNITA troops had participated in the demobilisation, it was clear that several thousand men had been hidden in Angola or across its borders.³¹ In addition, when the government started to prepare for a new offensive, UNITA began mobilising its forces again in August 1998.³²

Against the increasing power of the FAA that was revealed in its external military expeditions, rebels successfully rearmed themselves in almost absolute secrecy.³³ Faced with continued weapons acquisition by the government and the frequent declarations by FAA leaders opposed to the Lusaka agreement, suspicion and fear increased in UNITA. General João de Matos, the Chief of Staff of the army, declared that only the defeat of Savimbi could bring peace, and openly regretted that the political authorities rejected this option.³⁴ General Higino Carneiro, a close relative of the president, declared in November 1998 that the Lusaka agreement should be "nullified."³⁵

Another symptom of the increasing seriousness of the situation was the complete failure to integrate former UNITA soldiers into the FAA, well before the peace process faltered.³⁶ These elements all contributed to the next armed confrontation. There was thus nothing surprising about the measures taken by a movement obsessed with survival. Its leaders were used to face such difficulties alone and knew that they could rely only on themselves. They mocked the scorn of the international community. Rearmed, UNITA awaited the offensive with some serenity: "President dos Santos declared war … We shall see on the field what will happen."³⁷

The conventional victory of Luanda

Over a period of nine months, the government launched three major offensives, with the last beginning on 14 September 1999. The FAA launched operation Restauro with the aim to neutralise the rebels militarily.³⁸ In the course of six weeks, government forces repeatedly attacked UNITA. After the failure in Baïlundo, FAA forces needed an effective victory on conventional level. Luanda’s efforts reflected its hopes to eradicate the rebels. The result was an undeniable success for President dos Santos, although Luanda exaggerated its impact.

Lessons were learned from previous military setbacks. Operation Cacimbo, launched at the president’s personal request, was a strategic victory after fierce fighting. The determination of Dos Santos bore fruit and his exceptional tenacity was rewarded, while the army, careful after previous failures, would have rather envisaged stabilising the situation with more limited action.³⁹

The government’s offensive reclaimed the areas which had taken the rebels six months to conquer. Besieged towns such as Malange or Kuito, full of hungry refugees and pounded by rebel artillery, were freed from the deadly stranglehold.⁴⁰ By chasing peasants off their land, the rebels forced them to converge on towns. Counting on the hardship of these victims, UNITA was hoping for "a victory by social explosion" that would force Luanda to start negotiations. This cruel strategy proved to be a failure in the short term.⁴¹

At the end of October 1999, after announcing that Baïlundo and Andulo had been taken, General de Matos urged "his forces to continue the war until the rebels were completely ‘eliminated’."⁴² On the central plateau, the government army advanced victoriously, killing and capturing UNITA troops. Only a few pockets of minor resistance remained. At the beginning of November, the FAA seemed to be winning and, by mid-November, General de Matos declared that "although the war against UNITA rebels [was] not completed, the end [was] close." He calculated that the FAA had destroyed 80% of UNITA’s conventional military capacity.⁴³

By December 1999, the FAA had captured Jamba, the historical UNITA capital of the 1970s and 1980s. UNITA also lost the main diamond areas in the north-east. The rebels’ exploitation of this resource was thought to have stopped altogether. Several thousand UNITA troops surrendered to the FAA between November and December 1999.

In December 1999, former Angolan prime minister, Lopo do Nascimento, announced that victory was irreversible. He maintained that the rebels were broken and that they would be kept away from areas considered to be ‘useful’.⁴⁴ Rebels would be eliminated within a few months of clean-up operations. Do Nascimento’s analysis was confirmed by the declarations of General de Matos in mid-December 1999.⁴⁵ As far as most of the FAA generals were concerned, the army had put the rebels to flight. UNITA had suffered so many losses that observers believed the war was over, a feeling that also seemed to prevail among the population in areas under government control.

The final part of the government strategy, as announced by General de Matos, was to destroy Savimbi: "[Savimbi’s] whereabouts [was] known and the army [was] following him, because he [was] moving around with a personal guard of less than 20 men."⁴⁶ Special units of the Angolan army were sent in pursuit of Savimbi and his leadership.

There were many reasons for the conventional victory of the MPLA. Supported by new equipment from Eastern Europe and other western countries, the offensive severely damaged UNITA.⁴⁷ Crucial information, supplied by American and Brazilian sources, revealed the location of UNITA troops to the FAA, enabling them to anticipate their movements. Rebel communications were intercepted by professionals from across the Atlantic. The efficiency of the radar screen implemented a few months before by Israeli specialists, also contributed to deprive UNITA of airborne supplies.

Lack of fuel paralysed and incapacitated mechanised rebel units. The opportune defection of General Malaquilla — alias ‘Implacable’ — UNITA’s chief of intelligence, and a portion of his staff in the middle of the plateau fighting, had been prepared by the government. It gave an advantage to the FAA, which had sustained many losses among its own ranks. Savimbi had 19 officers and generals killed for ‘incompetence’ or ‘treason’. Many others disappeared during the fighting and retreat. Confusion reigned within the ranks, and the movement seemed to have been crippled by the government’s thrusts. The general secretary of the movement, Paulo Lukamba ‘Gato’, ordered troops to apply guerrilla tactics throughout the country, but the order went unnoticed.

Shock wave of the civil war

The confrontations and change in the balance of forces in Angola had repercussions in neighbouring countries, particularly Zambia. Before the war had resumed openly, Luanda had already accused Lusaka of supporting UNITA.⁴⁸ At the beginning of December 1999, while UNITA was losing its military capability, Angola put intense pressure on the Zambian regime. Angola threatened Lusaka that it would cross its borders to pursue UNITA rebels who were withdrawing and trying to escape the onslaught of the FAA.

The Angolan army wanted to destroy UNITA’s logistic infrastructure in Zambia, as it had previously done in the Democratic Republic of Congo (DRC). UNITA bases and refugee camps had existed in the country for many years. The FAA knew that weapons could be found in these camps and wanted to seize them.⁴⁹ Luanda had to prevent UNITA from recovering and reorganising itself behind safe borders.

Luanda reproached President Frederick Chiluba for providing logistic support to the rebels. With relative discretion, UNITA converted certain areas of Zambia into sanctuaries through which supplies could transit to the movement. The movement of diamonds and UNITA emissaries to foreign destinations also took place through Zambia. In addition, UNITA could use Zambian airfields and fuel was delivered directly from the oil refinery in Ndola.

Tension between both capitals resulted on 28 February 1999 in a series of bombing attempts on Lusaka. Supposedly organised by the Angolan secret services, this sounded a spectacular warning to President Chiluba’s regime.⁵⁰ Ten months later, two fighter bombers of the Angolan air force struck the refinery in Zambia on 18 and 19 December 1999. The air raid signified that President dos Santos was no longer going to tolerate the supply of fuel to the rebels. With the second blow, Lusaka was also meant to understand that the Angolan government found its policy towards UNITA intolerable when victory was so close.

At the end of December 1999, with the arrival of the FAA near the Angolan-Zambian border, the first armed incidents took place.⁵¹ The Zambian army could not resist the FAA.⁵² Zambia was therefore held hostage by the power of the rebel forces established on its territory or nearby. President Chiluba was caught between two camps and his entire country, threatened by the Angolan war, appeared to have granted sanctuary, more or less voluntarily, to UNITA.

This uncomfortable situation continued. In 2000, as the winter drew near, Lusaka accused the FAA of launching military operations into Zambian territory. The two provinces in the western and north-western part of the country were hit. Although Zambian authorities seized an aircraft transporting freight to rebel areas, this was not enough to appease Luanda’s anger.⁵³ During the summer of 2000-2001, tensions continued, exacerbated by the influx of Congolese refugees, some of whom were armed, from the DRC.⁵⁴

Namibia’s president, Sam Nujoma, tried to mediate between the two presidents, an exceptionally delicate task. President Chiluba met President dos Santos in Angola in an attempt to end the tension on the countries’ common border. Luanda still wanted from Lusaka what Windhoek had offered in 1998: the opportunity to pursue UNITA soldiers on its territory. Such an agreement filled Lusaka with fear. Windhoek’s participation in the Angolan civil war, although the context was different, was not without negative consequences for Namibia.

The South West Africa People’s Organisation (SWAPO) was a long-time ally of Luanda.⁵⁵ Windhoek, with its moral debt towards the MPLA and President dos Santos, had clearly decided to throw its weight behind Luanda.⁵⁶ Namibian involvement in the Angolan conflict increased substantially. Namibian forces were fighting at the FAA’s side in the DRC, and a Namibian contingent arrived in Angola to assist the FAA in 1998.

The Namibian government was simultaneously warding off an attack, since August 1999, by armed rebels in the Caprivi Strip. Windhoek proclaimed a state of emergency and started to repress the small-scale insurrection.⁵⁷ UNITA denied any involvement in the rise of this separatist movement.⁵⁸ Yet, the possibility of a link between UNITA and the Caprivi separatists was a hypothesis that, for want of verifying its authenticity, appeared logical.⁵⁹ Geography, regional history and political stakes pointed to the fact that this was indeed the case. Did UNITA, on this occasion, implement an indirect reprisal strategy against Windhoek?

In July 1998, Luanda asked its Namibian ally for authorisation to use its air bases in the east of the country and in the Caprivi Strip, from where its troops were able to mount combat missions against UNITA. This was an important advantage for the Angolan air force that could thus keep Jamba under its control.⁶⁰ President Nujoma also authorised Angolan troops to pass through Namibian territory in order to bypass UNITA positions in Cuando-Cubango. Windhoek agreed to supply important logistic support, such as airports and other military bases, to the FAA in their offensives against UNITA in the south-eastern regions of Angola.⁶¹ By the end of December 1999, a Namibian contingent was fighting UNITA alongside the FAA in southern Angola.

Namibia became even more directly involved in the Angolan conflict after suffering reprisal actions by UNITA groups in its own territory. Rebels crossed the border from Angola and launched attacks on civilians. Cattle theft, the kidnapping of civilians and ambushes frequently occurred in Namibia at the time, resulting in dozens of deaths among Namibians. The north-eastern population was terrorised and vehicles had to move in convoys. UNITA placed landmines randomly on roads and attacked Namibian civilians, soldiers, police officers and defence force bases.⁶²

Co-operation was extended when several thousand FAA troops conducted search operations in Namibia. The FAA pursued UNITA troops who had found refuge in the country. On this occasion, the FAA looted and stole from shops and residences, exasperating the local population who had already suffered under UNITA raids. The population resisted and confrontations took place, with deaths on both sides. The Kavango provincial authorities demanded the departure of these troops as a result of their ‘unacceptable’ behaviour towards local people.

A significant part of the FAA contingent was indeed repatriated, but war revisited Namibia after only ten years of peace.⁶³ In January 2001, following a skirmish on Angolan territory with a 300-strong UNITA detachment that was in the process of withdrawing after a raid, the Namibian Defence Force uncovered weapons and landmines from China, Russia and South Africa. Peace in Namibia also became tenuous.

Luanda’s impossible victory

The illusion of a military solution

UNITA, yielding to the military pressure of the government army, had no choice but to scatter its forces and renew its guerrilla war. The military strategy adopted at the end of 1999 by the rebels was mostly dictated by circumstances. The conventional victory of the FAA was so complete that avoiding direct confrontation was the only option for a movement on the brink of despair.

On the political level, the apparent disappearance of UNITA was potentially disastrous. Yet, during the last weeks of 1999, and again in early February 2000, the rebels took the initiative away from Luanda and renewed the war.

Reviving the warfare methods mastered during the past 35 years, UNITA rapidly recovered.⁶⁴ Its forces used guerrilla tactics and, sufficiently trained and well-armed, were calculated and efficient in this type of confrontation. Inspired since its creation by Chinese ‘popular war’ principles, UNITA considered that it could wear down the enemy as long as its troops limited their activities within the country to a minimum.⁶⁵

Indeed, the conflict spread throughout Angola. Ambushes on roads and the sabotage of operations increased.⁶⁶ A series of surprise attacks were launched, allowing them to avoid any reaction from the FAA. Although UNITA was scattered, it could concentrate its units and strike hard at the weak points of its opponent. Loyal to the basic principles of guerrilla warfare, UNITA outdid its more powerful opponent with these tactics for a time. The material superiority of the FAA, with its greater fire power, could not be used in this low intensity war.

The FAA controlled an important number of strategic points, but their numbers, even if greatly increased, could not have enabled them to control the whole country. The inefficiency of classic military operations against an opponent adept at evading engagement, no longer had to be demonstrated. The FAA was burdened with the responsibility to protect important towns, infrastructure and communication routes. The guerrillas, free from such tasks, forced the FAA to concentrate large numbers of troops for its defence. The rebels systematically attacked the outstretched communication routes of the army, and waged a campaign as mobile as it was tenacious.

The wave of strikes by the rebels was overwhelming.⁶⁷ Several elements showed that the military situation was much more varied than Luanda pretended. The rebels struck randomly, including targets close to the capital and the presidential palace, Futungo de Belas. In the north near the DRC, fighting took place continuously. In the east, UNITA troops deployed tanks, combat vehicles and heavy artillery, and besieged government garrisons.

FAA aircraft paid a high price for participating in counterattacks on guerrilla forces. Transport aircraft were prime targets for ground-to-air missiles. Another proof of the intensity of guerrilla attacks was the fact that several FAA generals were killed in action after UNITA was pronounced ‘crushed’. Four died in ambushes and one was seriously injured when his aircraft was shot down over Huambo.

On various occasions, UNITA took control of symbolic towns: Andulo, Baïlundo, Narhéa, Catete, Soyo and Munhango, among others. It attacked many others, including Negage, Uige, Kuito and Quibaxe.⁶⁸ On the main road between Benguela and Huambo, several FAA convoys were ambushed, sustaining considerable losses in one incident.

Luanda either contested the veracity of these assertions or kept these events quiet. The facts showed, however, that the war was far from over and this invalidated the government’s claims.⁶⁹ Although the scope of the war was limited, it remained damaging for both camps, as the FAA also did not remain passive. Guerrillas sustained losses in these confrontations and occasionally surrendered in large numbers.⁷⁰ The FAA showed that they would not allow UNITA any respite.⁷¹ However, months of operations, ambushes, searches and losses indicated that this new phase in the war was not going Luanda’s way. The FAA showed signs of weakness, disciplinary problems, and the morale of the troops was low. In addition, the resort to foreign specialists, pilots in particular, had exposed its tactical limits.

At present, UNITA lacks neither men, nor armament or means. The movement has made light of the controls, sanctions and the embargo of the international community. It was able to accumulate enough weapons in Angola and neighbouring countries, as well as elsewhere on the continent and beyond, to fight for several years. The movement’s diamond ‘diggers’ and garimpeiros have resumed their work.⁷² Other diamond concessions established in the DRC have not ceased to produce gems, thus continuously sustaining the movement’s war chest. UNITA has also established appropriate alliances with new actors on the African political stage and has maintained former relations.

There should be no illusions about the strength of the Angolan army. On a military level, Luanda’s victory over the rebels seemed improbable, especially in the long term. The vastness of the territory and the perfect conditions for guerrilla warfare, as well as the country’s lengthy borders are elements mitigating against even a partial armed solution to the Angolan conflict.

Angola’s recent military history shows that, since 1976, control over towns is not sufficient to pacify the country or to impose Luanda’s order. The FAA does not have adequate economic, technical or human means at its disposal to expect a victorious outcome. It should triple its numbers, deploy huge logistics and different technologies for its long-term campaigns. This is difficult as the Angolan military budget depends very much on the oil price.

The fact that UNITA is rooted in the heartland of the Ovimbundu in Angola, the most important ethnic group, is another factor closely related to the rebels’ ability to resume war. Even though its numbers had decreased, UNITA is still very much alive one year after the fall of Baïlundo. Rebels are still cleaning out the rural areas and people are fighting throughout Angola. The country remains ungovernable.

From a military point of view, the civil war is at a complete stalemate. The history of 20th century conflict shows that few large-scale guerrilla forces have been defeated by the use of force, especially when borders remain permeable. Pretending that war is bound to end in peace is a further analytical error. Because any attempt to solve this war by military means is bound to fail, belligerents will have to negotiate a political solution in the long run. However, it will not be an easy task to find an acceptable political compromise while the respective political positions are so rigidly set.

Inability to find a political solution

Both parties are locked in a dynamic of fratricidal confrontation that is out of control. War did not resume accidentally, but because the Angolan political situation was deadlocked, and the problems causing the outbreak of the conflict were unresolved.⁷³ War feeds upon itself, but the problem remains essentially political. What is at stake in Angola is absolute power.

This war is only a symptom of the political and social turmoil inherited from the colonial era. It is a war of undecided succession and an endless competition for power, both poisoned by decades of confrontations. The two rival movements, the MPLA and UNITA, have shown through their leaders that they are incapable of compromising on the running of the country.

Currently, these two politico-military forces, led by competing élite, are stubbornly bent on remaining in or gaining power. Neither seems able to envisage the prospect of sharing power equitably, or of a peaceful political changeover.

After 1992, UNITA’s withdrawal from the political process, as well as the contempt and fear of Luanda towards groups constituting the major part of the movement, reinforced a culture of civil war in which the enemy was to be crushed.⁷⁴ Luanda felt that, as long as UNITA existed, it would constitute a threat to its power. Luanda’s choice was either to eliminate UNITA or risk its own demise. Luanda could not rally the support of these groups to its side. Amnesty offered to UNITA members who surrendered failed to show convincing results.

The current crisis in Angola can therefore only be understood by referring to the MPLA’s commitment to govern the country alone, and to UNITA’s determination to take power away from the ruling party. Experience and the political models offered to local nationalists created an ideological framework that could only lead to an authoritarian political culture in the country.⁷⁵ Between 1975 and 1991, the newly independent country lived under a one-party system that was supposed to prevent the internal political scene from imploding into ethnic strife. It was only under the military constraint and pressure of UNITA that the MPLA agreed to adopt a multiparty system. However, the government immediately committed itself to find ways to oppose "all potential obstacles" offered by what it labelled as ‘uncontrolled’ associations.⁷⁶

Limited contacts, although unsuccessful, had taken place in Southern Africa between high-level officials of both parties under the auspices of the South African and Mozambican authorities. The aim was to negotiate a framework within which more viable relations could be established. UNITA was not particularly enthusiastic about a cease-fire, while a delicate sharing of power is no longer under discussion.

The future of Angola remains unclear. Confronted by UNITA, the Luanda government refuses to envisage even a hypothetical compromise with the armed opposition. The official presidential election, originally planned for 2001, cannot be held under the present unsafe conditions. Neither the government nor its armed opposition wishes to hold elections soon. The present situation is therefore likely to persist.

Yet, several elements offer hope for progress. The government and UNITA frequently speak of renewing an electoral process. Reference to democracy is thus maintained, and elections will take place in the long run. The question remains whether these will be held in appropriate conditions and will ensure a partition of power that could satisfy both.

Secondly, aspirations towards peace in Angola have become noticeable. Civil society has recently started to make itself heard. It vigorously demands the end of the war through negotiation and the democratisation of the regime. The Angolan episcopate has also made its position clear. During an ecumenical council meeting in Luanda in July 2000, representatives of the episcopate demanded peace and reconciliation.⁷⁷ Journalists, among them Raphaël Marques, politicians, lawyers and other actors in Angolan society, without any links to UNITA, openly questioned the policies of President dos Santos, denouncing the lack of democracy, corruption and the bad management of the government.

MPLA dissidents and members of other political parties criticise the government as much as they do UNITA. Demonstrations took place on several occasions in a tense atmosphere. Holden Roberto, the former chief and founder of the Frente Nacional de Libertação de Angola (FNLA), accused the president of corruption linked to oil resources. Global Witness alleged in a report that the regime was affected by large-scale corruption. In published letters addressed to the president, some retired and active FAA generals seemed unhappy about the absence of a military solution to the ‘eternal’ war.

The government and UNITA have to take account of this trend, which could end the polarisation of Angolan political life in time. The new attitude of civil society has weakened and undermined the position of the MPLA.

A fragile regional power

There is no final military solution to the Angolan conflict. Close to 35 years of confrontations have proven this. UNITA stands the chance of surviving. For President dos Santos, disorganising UNITA for a time and forcing the movement’s leadership to go underground, represent at best a partial success. For UNITA, though the political situation worsened after its defeat, this was not decisive.

The impact of Luanda’s external intervention policy also seems paradoxical. FAA action in the east against Zambia had a limited impact and tensions remain to this day. The Angolan army has partially moved its civil war to Namibian territory, leading to reprisals by UNITA rebels against Nujoma’s government. In the DRC and Congo-Brazzaville, on the contrary, the presence of the Angolan army has become an important factor in the survival of the current regimes.

Angola is generally seen as a regional player, though its power remains fragile. Its economy depends on oil prices, and its external policy is the prisoner of its obsession to destroy UNITA. Luanda still does not rule the whole country after a quarter of a century of independence, even though Angola’s regional superiority reflects its ability to mobilise a large army, its financial means, the very strong political will of its president and the weakness of neighbouring countries.

The idea of peace in the near future in Angola fluctuates between the improbable and the impossible. It has often been suggested that both leaders should retire from the scene. Savimbi has always refused to relinquish what he began in 1966 and to leave the political stage. His absence some time ago did not prevent UNITA from functioning smoothly. Many potential successors can be found within the movement. Would they be more conciliatory? Would the fundamental political issues at the root of the conflict be settled by another leader? If President dos Santos, weary of being the head of the country or because of ill-health, decided to withdraw from power, his successor would probably follow in his policy footsteps.

Civil society has demanded the immediate democratisation of political life, maintaining that this would inevitably lead to peace. But implementing such a political system cannot be done on command, and establishing a western-type democratic regime remains unrealistic in today’s Angola. Moreover, democratically elected regimes have plunged into civil war before on the African continent.

For Angolans to resume dialogue and to stop favouring extreme solutions remain extremely complex tasks. The international community has already tried, in vain, on two occasions between 1991 and 1998 to promote this outcome.

Are Angolans capable of living together within the same nation? The political, economic and human fiasco created by the leaders of the regime and the armed opposition makes Angola an example not to be followed. Of nearly 12 million, 3.8 million people are internally displaced, 500 000 are out of reach of aid and many have found refuge in neighbouring countries. Food shortages, hunger and epidemics continue. The population remains the neglected, powerless victim of the civil war. Only time and new political solutions will allow the country to escape from the routine of civil war.

Notes