Trouble for Portugal in Africa began with the granting of independence to its colonial neighbours and the consequent establishment of potential sanctuaries for nationalist movements in the ideologically friendly adjoining countries. With respect to the tiny enclave of Portuguese Guinea, now Guinea-Bissau, the French granted independence to the Republic of Guinea in 1958 and to Senegal the next year. For Angola, Belgium granted independence to the Congo in 1960, and the British gave independence to Northern Rhodesia, renamed Zambia, in 1964. In the case of Mozambique, Britain granted independence to Tanganyika, which then became Tanzania with its absorption of Zanzibar, in 1961, and to Nyasaland, now Malawi, in 1964. Thus within a short six years from 1958 to 1964 Portugal found that the political landscape in Africa had shifted and had quickly developed a hostile stance to its colonial empire.

 

In the case of Angola, the focus of our examination, trouble actually began with the way in which independence was granted to the Belgian Congo, hence we will quickly review the background to 1960 and then follow subsequent events there to show their influence. The Belgian Congo had prospered, since its transfer to the Belgian government from the estate of its King Leopold II in 1908; however, despite all of this apparent progress and its attendant social benefits, there remained a strong undercurrent of resentment within the local population.^[1] Even though the system of forced labour had been abolished and the right of the Congolese to own and work their lands had been restored, no civil rights for the Congolese were thought necessary, and racial segregation was standard practice. When independence came in 1960, there were fewer than 30 African university graduates. Of the 5,000 management-level positions in the civil service, only three were filled by Africans.^[2]

 

Subsequent political developments in the neighbouring French and British colonies, forced Belgium in 1956 to permit the same political freedom in the Congo.^[3] By late 1959 the activities of these various political movements had matured to a point of demanding independence, and the demands for immediate independence were reinforced by rallies and strikes, many violent and bloody. This mayhem intimidated the Belgian masters and created a sense of urgency within the Belgian government that caused it to rethink its colonial policy of a five-year transition to independence culminating in 1964. At the February 1960 Round Table Conference in Brussels it was resolved to grant the Congo its independence in June, a mere four months hence. This lack of a Belgian appetite for violence would drive Holden Roberto to believe that the same weakness existed in the Portuguese.

 

In that short period the Belgian government attempted to overcome the enormous obstacles of accelerating the formation of a new Congolese government, a patently impossible task. The seeds of immediate trouble began during the election campaign, in which candidates made wild promises, took the opportunity to vent long-repressed hostilities toward Europeans, and stirred long-standing tribal animosities.^[4] In the 23 June elections, the party of the charismatic postal clerk Patrice Lumumba outpolled that of the pudgy civil servant Joseph Kasavubu; however, neither side could form a parliamentary coalition. Consequently, as a compromise measure they formed an uneasy partnership with Kasavubu as president and Lumumba as prime minister. The mood rapidly degenerated and turned ugly during the first week of independence. Public services deteriorated with the wholesale departure of Belgian civil servants and a lack of trained Congolese to replace them. When the new civil servants were not paid, there were strikes that turned into riots when the 24,000-man Force Publique attempted to intervene. All semblance of order, however, vanished when the Force Publique itself mutinied. The soldiers went on a rampage, attacking their Belgian officers, raping officers’ wives, and looting. On 6 July, Lumumba, in an attempt to quell disaffection in the Force Publique, replaced the Belgian officers with Congolese. This move did not stem the tide, and the mutinies spawned a widespread wave of violence. Blacks attacked whites and each other, as they vented their anti-European hatreds and intertribal rivalries.

 

The spread of disorder prompted Moïse Tshombe, the newly-elected premier of Katanga Province, to ask Lumumba to appeal to Belgium for troops to restore order. Fearing a return of Belgian rule, he refused. Tshombe then in an attempt to address his executive responsibilities and preserve his province, declared mineral-rich Katanga an independent state in close association with Belgium and seceded from the Congo on 11 July. The continuing riots prompted Belgium to deploy paratroops to protect its nationals, and it seems, they landed principally in Katanga, where they helped to sustain the secessionist state by rounding up and expelling the mutinous elements of the Force Publique. Tshombe then raised his own force of gendarmerie, of which we will hear more later and which will prove to be an important lever in keeping the revolutionary enthusiasm of the future Congo government in check.

 

On 12 July, Lumumba asked the UN to expel the Belgian ‘forces of aggression’ and to compel Katanga to end its secession.^[5] Three days later the first of 25,000 UN troops arrived with U.S. backing, Belgian soldiers began to depart, and Lumumba declared martial law. The political situation then degenerated completely. The people of the lower Congo took their cue from Tshombe and demanded the abolition of the central government in favour of a system of independent provincial governments. This demand was followed by a declaration of independence in Kasai Province. Lumumba’s country was falling apart with good reason, and having no reliable way to enforce his will, he turned to the UN with a request to use its peace-keeping troops to unify the Congo by force. When it refused, he countered by appealing to the Soviet Union, which saw an opportunity in the Cold War context and obliged with a supply of sufficient arms and airlift capability to enable Lumumba to march on Kasai and pour the proverbial gasoline onto an already incendiary situation. In what Mike Hoar describes as a one-sided battle, Lumumba’s forces massacred Kasai tribesmen by the hundreds.^[6] He then turned his force toward Katanga, where it was stopped at the border. Frustrated by this stalemate, elements of the Force Publique turned to brutalizing the population.

 

At this point Lumumba was a spent force in any attempt to unify the Congo politically. While he was a persuasive orator, he proved to be an unstable diplomat and an incompetent head of state. On 5 September, Kasavubu dismissed the prime minister. In turn Lumumba dismissed the president. Kasavubu prevailed by placing Lumumba under house arrest with the support of the former sergeant-major, Joseph-Désiré Mobutu, now a colonel and head of the army. All pretence of a functioning government seems to have ended. On hearing of Lumumba’s arrest and the ensuing confusion, Antoine Gizenga declared the secession of Oriental Province, Lumumba’s power base with its capital in Stanleyville. Gizenga’s forces attacked Mobutu’s garrisons, and a new bout of the most brutal savagery raged through the rain forests.^[7] Mobutu was unsettled by this anarchy and reacted by forming a ‘college of administrators’ on 14 September. He assumed the day-to-day running of the government, ‘neutralizing’ all politicians.^[8] One of his first actions was to close the embassies of the Soviet Union and Czechoslovakia and banish their presence from the Congo, thereby putting himself on the side of the West in The Cold War. The situation stabilized for a time; however, in November, Lumumba left Léopoldville, where he had enjoyed UN protection, to travel to Stanleyville, where his supporters had control. En route he was caught by Mobutu’s troops on 2 December, and on 17 January 1961 was flown to Katanga, where he was reluctantly received by Tshombe in an apparent near-death condition.^[9] Later in the month he was shot by Congolese or Katangese troops while allegedly trying to escape from a mud hut in which he had been held prisoner.^[10] Lumumba’s death caused an international outrage and raised the man to martyr-hero status. Mobutu returned the government to Kasavubu, as his suspected hand in Lumumba’s death became known, and in turn Kasavubu made him commander-in-chief of the armed forces.

 

Eventually the UN intervened and returned Katanga to the control of the central government in January 1963, at which point Tshombe went into exile in Spain. It was at this point between the middle of January and the first of June that the old gendarmerie, loyal to Tshombe, crossed into eastern Angola at Teixeira de Sousa to fight for the Portuguese. Portugal had openly encouraged the Tshombe government and from eastern Angola supported it against the UN forces, hence the gendarmerie knew who its friends were. In June 1964 the UN withdrew, and Tshombe was unexpectedly invited to return as Prime Minister in a government of reconciliation. Unfortunately Tshombe could not bring harmony to the situation, and it degenerated into a civil war. Faced with a weak and demoralized Congolese National Army, he took the controversial step in September 1964 of hiring white mercenary troops to assist his army in regaining order. Mobutu, the army commander, along with selected army units, mercenaries, and covert South African support quelled the rebellion.^[11] South African involvement turned out to be a Soviet bloc lightening rod in The Cold War context. The Portuguese again established strong ties with Tshombe, and this relationship served to dampen insurgent activity through 1965. Indeed, Tshombe harassed the nationalist movement UPA (União das Populações de Angola, or Union of Angolan Peoples) and its leader Holden Roberto, and ‘Rebel activity on the northern border of Angola deceased and morale collapsed in the terrorist camps across the border’.^[12] In August 1965, Jeune Afrique carried an article with the headline, ‘Holden complains bitterly’ and ‘Tshombe is making his life tough’.^[13] In the autumn of 1965 Mobutu decided he wanted the Congo for himself and overthrew Tshombe in a coup, and again he left for Spain. With the Tshombe exit the eastern border of Angola became vulnerable. In 1967 Mobutu suspended the new constitution and the National Assembly and prevented Tshombe from returning to the Congo.

 

The agony of the Congo and its chaos at birth influenced the opening strategy of the UPA and Roberto in his attempt to wrest control of Portuguese territory. When the Belgian Congo became independent on 30 June 1960, its government began to give Roberto practical assistance, including permission to establish a radio station and a training camp within its borders. This sanctuary was an important facet of UPA operations in its early years. Roberto had witnessed the long series of Congolese crises that had begun with the violent political rioting on 4 January 1959 and had led to the accelerated Belgian push toward Congo self-government and independence in eighteen months. By December 1960 he believed that just as the Belgians had quickly grown weary of armed conflict, so would the Portuguese when it was initiated. This belief would set the stage for the 15 March 1961 attacks by the UPA in the north of Angola and end the relative tranquillity in Portuguese Africa. Roberto’s multi-pronged attack was launched with a flood of 4,000 to 5,000 armed men, which is thought to have been increased to 25,000 through forced recruiting.^[14] The area of the attacks is depicted in Appendix 1. Approximately 700 European farms plus additional trading settlements and government posts were overwhelmed.^[15] This mob laid waste to whatever was in its path and killed men, women, black, white, young, and old. It was a senseless act of violence with only a misplaced political aim rather than a military campaign with a political goal. Roberto did not understand the difference and was later quoted as saying, ‘This time the slaves did not cower. They massacred everything’.^[16] Roberto’s attacks also created an estimated 250,000 refugees who crossed the border into the Congo. Alex Vines even estimates the figure to be as high as 400,000.^[17] All of Portugal was shocked at the horror, and Roberto’s belief that Portugal would act as Belgium did proved false.

 

When this gambit did not achieve its designed goal, the new Congo state worried about Portuguese retaliation and gave serious consideration to expelling the UPA from Congolese soil or at least forcing it to cease all activity.^[18] This fear quickly passed, and the greatest threat to the UPA then became its destructive conflict with the other Angolan nationalist movements for revolutionary leadership. This cleavage would persist throughout the war and provide the Portuguese with more that one opportunity to sow confusion in the enemy ranks.

 

Here we will identify the weakness of each of the three nationalist movements and their vulnerability to Portuguese action. The three that sought to challenge Portuguese ownership of Angola had their origins in the 1930s, when the emergence of modern-day black opposition to colonial rule began. The repressive practices of Prime Minister António Salazar’s Estado Novo toward any form of dissent, particularly political, extended from the metrópole to the colonies. Resistance began cautiously, as there was a practical barrier to any such opposition in the ethnic and social fragmentation of the overseas non-white community. Without strong leadership there would be no nationalist movement able to gain the necessary momentum in reconciling these divergent viewpoints and crystallizing resistance to the Salazar regime. Local African grievances were long-standing and had come to the fore during the early twentieth century with the influx of white settlers and abusive labour practices, making any apparent calm illusory.

 

By 1956 the young Marxists of the Angolan Communist Party contributed to the formation of the MPLA (Movimento Popular de Libertação de Angola, or Popular Movement for the Liberation of Angola). The MPLA developed roots among Luanda's urban and largely radical intellectuals, among its slum dwellers, and to a lesser extent, eastward from the capital among the Mbundu, Angola's second largest ethnolinguistic group, and the Chokwe people. These urban roots were composed largely of mestiços or mixed-race peoples, who controlled the party. The movement had little in common with the rural peasants of the east and south of Angola and made little effort to gain their true devotion. In December 1956 the initial MPLA manifesto was openly published in a direct frontal assault on the government. Predictably the national police, known simply by its acronym PIDE (Policia Internacional de Defesa do Estado), reacted adversely, and a number of the MPLA leaders were forced to flee into exile. From 1957 onward PIDE action was so successful ‘that the nationalists were not able to maintain more than the most rudimentary organization inside the colonies and could not communicate with those cells that did exist’.^[19] The parties were forced to conduct their affairs from neighbouring states and were deeply influenced by their foreign connections. As the PIDE systematically wrecked the MPLA organization primarily in Luanda, it became progressively weaker and isolated from its leadership that was now abroad and, as it had no strong constituency or bases elsewhere among the rural population, withered within Angola.

 

The MPLA in exile established itself initially in Léopoldville and aligned itself not only with other independent African nations and their socialist philosophy but also with the communist bloc, including the Italian and French communist parties. The MPLA found that it was in competition with the other prominent Angolan nationalist group at the time, the UPA, for acceptance as the leading representative of the Angolan people. In 1962 the MPLA formed its military wing, EPLA (Exército Popular de Libertação de Angola, or Popular Army for the Liberation of Angola), to project its influence into Angola. This nascent force numbered between 250 and 300 young men who had undergone military training in Ghana and Morocco. The EPLA sought to expand the conflict across northern border of Angola with this force and penetrate the entire country, publicizing the MPLA manifesto. Recruiting proved to be difficult because of ethnic rivalries, and military action was thwarted by the competing UPA. The UPA and specifically Roberto, who established a political alliance with Mobutu by divorcing his first wife to marry Mobutu's sister-in-law, prevailed on Mobutu to force the MPLA out of Léopoldville in 1963 and into its new headquarters in Brazzaville. Mobutu may have also done this to reduce his internal security problem in the Congo. From the Brazzaville sanctuary it was difficult to conduct a campaign across an unenthusiastic Congo and into a now politically and physically distant Angola. As a result northern Angola proved to be barren for the MPLA, and it was not until 1966, with the opening of its second front from Zambia, that the promise of success would come. The MPLA thus had little in common with the larger population of Angola, and this lack of empathy would be exploited by the Portuguese.

 

The UPA was formed in the mid-1950s from a number of small groups with conflicting goals by Barros Nekaka, who in 1958 passed leadership to his nephew Holden Roberto. UPA strength rested in the rural populations of the Bakongo ethnolinguistic region of Angola. These people straddled the border between the Belgian Congo and Angola and extended into Cabinda and the old French Congo, the boundaries of the ancient Kongo kingdom. An ardent anticolonialist, Roberto had been born in São Salvador, Angola, and been educated in the São Salvador Mission School. His parents moved to the Belgian Congo to take advantage of the developing prosperity, and he worked as a finance clerk in the Imperial Belgian Government between 1941 and 1949. Northern Angola was an area that had become more politically aware in the 1950s through white settlement, Baptist missionary influence, and an easy access to the developing political activities of the Belgian Congo. It was also the centre of the cotton and coffee-growing region. Roberto thus felt a close kinship with the peoples immediately across the border. The UPA was able to develop a following there because of the relatively open frontier, and this loyal cadre became the basis for the uprising in March 1961. Portuguese presence in this area took the form of chefes do posto (heads of posts) and administrators, as opposed to PIDE, and these officials were so sparse that it was physically impossible for them to maintain anything but the most casual control over their districts.^[20]

 

While Roberto was relatively well educated, he was a member of the Bakongo ethnolinguistic group, was not a mestiço, and consequently did not share their more European cultural perspective. He was also tribally oriented in contrast to the non-tribal declarations of the MPLA. As a result, the personality and leadership philosophy of the UPA contrasted clearly with the MPLA and its sophisticated mestiço leadership, which was left-wing, intellectual, and acculturalatively Portuguese. Funding and support also glaringly contrasted, the MPLA actually being linked with the Eastern bloc. The UPA received financial support from the American Committee on Africa, from various African governments, preponderantly that of Léopoldville, and from the CIA, which was hedging its bets on the future of Africa.^[21] Accordingly, they were never able to resolve their differences and join forces effectively. This competition would be ably manipulated by the Portuguese.

 

The UPA formed its military wing, the ELNA (Exército de Libertação Nacional de Angola, or Army of National Liberation of Angola), in June 1961 after the March attacks did not achieve a Portuguese withdrawal. Roberto was its commander-in-chief and its other two leaders were Portuguese Army deserters, Marcos Xavier Kassanga, its chief of staff in Léopoldville, and João Batista, its operational commander in Angola with headquarters near Bembe. This leadership was ineffective. The ‘fiery-tongued’ Roberto was so autocratic that he would accept little more than arms and money.^[22] Indeed, he appeared hard, introverted, and even sinister from behind his apparently ever-present sunglasses, which he seemed to wear even in dark rooms and on cloudy days.^[23] Without the necessary leadership and training, the ELNA ‘set a demoralizing example of politico-military incompetence and indiscipline’.^[24] The South African Defence Force vice-consul in Luanda, Willem van der Waals, noted that the ELNA ‘involved itself in military activities in the narrowest sense...but avoided contact with the Portuguese security forces as far as possible’.^[25] Van der Waals further argues that in the traditional exercise of the informational lever of national power, Portuguese propaganda and social work among the refugees returning to Angola after the March 1961 attacks persuaded most of these displaced people to move into controlled settlements or aldeamentos. This development deprived the ELNA of popular support, and its military action consequently occurred in a human desert. It was thus unable to proselytize the population, as in a classic insurgency, and as an alternative focused on preventing competitive MPLA infiltration. It had also neglected to indoctrinate, organize, and win recruits among refugees in the Congo returning to Angola and had thus missed an important opportunity to undermine Portuguese authority. There was accordingly no ELNA internal political infrastructure in Angola. Portugal gained the upper hand and dominated the human terrain until 1974 in a classic example of successful informational warfare with civil support and population proselytizing.

 

ELNA training was so poor and ineffective that despite expansion of the force to about 6,200 troops, their deportment at such camps as Kinkuzu in the Congo was cause for despair.^[26] Andreas Shipango, South-West Africa People’s Organisation representative in Léopoldville, made an appraisal during a 1963 visit: ‘With representatives from a number of other liberation movements, I visited Holden Roberto's training camps near the Angolan border with a view to sending our young men there. But the atmosphere in Roberto's training camps was very bad, and I could not recommend such a course’.^[27] Roberto had only a weak military program unsupported by political indoctrination, and indeed there was no talk of winning the population to the UPA point of view, which was simply that Angola should be an independent country with Roberto as head of state. The approach was arrogant and naïve and, as it proved, totally ineffective.

 

This lack of direction caused great rifts in the UPA leadership. Despite its reorganization at the behest of Mobutu in March 1962 to include additional political groups, to rename itself FNLA (Frente Nacional de Libertação de Angola, or National Front for the Liberation of Angola), and to establish a government in exile named GRAE (Governo da República de Angola no Exílio, or Government of the Republic of Angola in Exile), little of substance was accomplished. A frustrated Jonas Savimbi, Roberto's ‘foreign minister’ and an Ovimbundu, formally broke with the UPA/FNLA in July 1964, labelling Roberto, a Bakongo, a ‘corrupt racist’, and eventually formed the third nationalist movement in Angola, UNITA (União Nacional para a Independência Total de Angola, or the National Union for the Total Independence of Angola) based on his people, the largest ethnolinguistic group in the colony.^[28] Savimbi had publically announced his break at the 1964 Organisation for African Unity (now the African Union) meeting in Cairo, Egypt, and alleged that Roberto had set up a ‘commercial empire in the Congo’ and that FNLA administrators were ‘wage earners and profiteers who enriched themselves on the money of New York financial circles and other international organizations.’^[29] Within two years, Savimbi had built his meagre 12-man force into a sizeable army, gaining popularity and support as the only leader to work within the country alongside his men in battle against the Portuguese. ‘Leaders must fight alongside the people and not stay abroad, sending “second-class” fighters to face the Portuguese,’ Savimbi proclaimed as he denounced Roberto.^[30] While Roberto was renowned for his aloofness, the bearded Savimbi mixed often and easily with ordinary people as well as his military. The next year Alexandre Taty, ‘minister of armaments’, after challenging Roberto in an unsuccessful coup, defected to the Portuguese in Cabinda with a substantial number of his followers. Further along we shall see how damaging these defections turned out to be in the hands of the Portuguese.

 

Indeed, Roberto seemed more interested in his personal power than in a war of national liberation, and according to U.S. intelligence sources, was ‘subservient’ to Mobutu, who protected him from internal challenges to his leadership.^[31] As the Portuguese weekly Expresso observed in 1974, ‘The FNLA is Holden Roberto and Holden Roberto belongs to Mobutu, to whom he is connected by an umbilical cord.’^[32] Mobutu was playing both ends against the middle in loudly proclaiming his support for the FNLA while discretely cultivating good relations with Portugal, for like Zambia, the Congo depended on the Benguela Railroad, which carried more than half of its foreign trade. There were also dissident elements in Angola that, if unleashed in cross-border operations, could make considerable trouble for him. Consequently he kept a tight control over the FNLA activities both within the Congo and without. He provided just enough political and material support to give it international credibility and to provide the Congo with a stake in Angola should the Portuguese eventually leave.^[33] As for Roberto, after 1966 he started a property business in Léopoldville and was often seen driving a shiny black Mercedes-Benz about the city, just as Savimbi predicted. He clearly became less interested in running a nationalist movement and, according to U.S. intelligence, had not set foot inside Angola since 1956.^[34] As Henry Kissinger noted at the time, ‘The strength of the FNLA continues to suffer from Holden’s refusal to move from Zaïre (the Congo) to Angola to take direct control of FNLA activities.’^[35] Indeed, Roberto spoke French and English far better than Portuguese.^[36] CIA Luanda station chief at the time, Robert Hultslander, wrote later that, ‘This organization was led by corrupt, unprincipled men who represented the very worst of radical black African racism.’^[37] ‘It was a squalid spectacle: a corrupt leader dancing to the tune of a foreign master.’^[38] As for his troops, they have been described as ‘underfed, ragged, and villainous’, and were hardly a credible army.^[39] The Portuguese, it seems, had little to fear from Roberto and his organisation after the March 1961 attacks.

 

The CFB was at this time the most direct and efficient way to move the mineral resources of the Congo and Zambia to the industrialised West and thus represented a very potent economic lever. Robert Williams, a dynamic entrepreneur who had acquired in 1900 a 40 percent interest in the Katanga mineral wealth, was the driving force in the enterprise. Williams had initially approached Cecil Rhodes, who had refused to extend the Rhodesian rail line to the Katanga frontier unless Williams gave him half of his interest. King Léopold likewise refused to help, as he naturally wanted to keep as much Katanga wealth as possible in Belgian hands.^[40] Indeed, he favoured a route that ran exclusively through Congolese territory by rail from the mines to Port Francqui, by lighter down the Kasai and Congo Rivers to Léopoldville, and then by rail to the port of Matadi, as shown as Route 1A in Appendix 2.^[41] Consequently Williams approached the Portuguese with the proposal to build a railroad from the Atlantic port of Lobito across Angola to Katanga, as also shown in Appendix 2 as Route 3. Williams was granted a concession for 99 years on 28 November 1902, and his Benguela Railway Company commenced construction on 1 March 1903. By 1914 and World War I, 500 kilometres (310 miles) of the most difficult section through the coastal mountains had been completed, and for reasons of politics and finance construction did not resume until 1920. The rail connection with the border was finally completed in 1929. While the primary purpose of the project had been to capture the export trade in Katanga minerals, until the Belgians connected with the border in 1931, Williams was left with only the domestic Angolan traffic. In the original plans this was considered to be of secondary importance; however, it developed and supported the line over the years and proved to be its main source of revenue until Congo independence in 1960. It was at this point that a new political and economic calculus enabled the CFB to realise its original goal, and the line jumped from carrying only 24 percent of Katanga minerals to 40 percent.^[42] Williams had paid his first stockholder dividend in 1956 largely based on the domestic Angolan traffic. The combination rail-water-rail route through Matadi built by Belgian interests had carried about 40 percent of the Katanga minerals until independence, and thereafter the primary routes were the more efficient direct rail connections through the Portuguese ports of Lobito and Beira, which were far quicker, more direct, and hence more cost effective. The CFB proved very successful and profitable, and increased in importance, particularly in 1973 after Rhodesia closed its border with Zambia, and mineral exports had to find a new rail route. The Tanzam Railway from Zambia to the Tanzanian port of Dar-es-Salaam would not be completed until 1975, so for Zambia the CFB was its key economic lifeline during that three-year period. This fact meant that a nation espousing an anti-colonial stance and supporting nationalist movements fighting for the independence of Angola was forced to depend on the hated Portuguese for the lion’s share of its prosperity. Portugal would use this economic lever on both the Congo and Zambia to keep insurgents on an imperfect leash.

Northern Rhodesia was granted its independence and became the Republic of Zambia on 24 October 1964, following Tanzania by some three years. It remains a landlocked country, dependent on and strategically vulnerable to the transportation systems of its neighbours to trade with the outside world. It consequently must maintain cordial relations with them. Kenneth David Kaunda, a former schoolteacher turned politician, became its first president. He had become an ardent nationalist in the 1950s, had led the drive for Zambian independence, and like Julius Nyerere of Tanzania, was passionately committed to supporting and providing a home for African liberation movements. He and Nyerere thought alike, as both espoused the single party socialist state and clung to political power through this vehicle. The journalist Chanda Chisala correctly, if emotionally, describes Kaunda as a power-hungry, selfish, greedy dictator who has little respect for human rights and less tolerance for contrary opinions.^[43] In 1972 he bulldozed his new constitution into effect and banned all political parties except his own. Citizens could be and were jailed for such trivialities as not displaying his portrait in their offices. While espousing Christian principles, he was happy to detain dissidents without trial and abuse them with beatings, starvation, torture, and solitary confinement.^[44] He used the term ‘stupid idiots’ to describe his opponents, who eventually became a majority of Zambians.

 

While the tide was with him, he made Zambia the home for numerous nationalist movements. Indeed he hardly met one that he did not like. His landlocked country shared a frontier with no less than six countries, three of which were fighting active insurgencies, Angola, Southern Rhodesia, and Mozambique. This gave him ample opportunity to make trouble for his neighbours. Within two years of coming to power, Kaunda offered sanctuary to the MPLA and UNITA, among others, to pursue their campaigns against the Portuguese in eastern Angola.

 

The supply chain in support of this offensive began with Julius Nyerere in Dar-es-Salam, Tanzania, through which arms flowed from such supplier-states as the Soviet Union, Communist China, Czechoslovakia, Bulgaria, and Cuba. Dar-es-Salam was not only a port of entry but also the centre for training the armed elements of the African National Congress (ANC), South West Africa Peoples Organization (SWAPO), Zimbabwe African People’s Union (ZAPU), Zimbabwe African National Union (ZANU), MPLA and FRELIMO (Frente de Libertação de Moçambique, or the Liberation Front of Mozambique).^[45] From this centre of mischief, arms supplies and trained insurgents flowed to eastern Angola over a lengthy trail that wound through Tanzania and Zambia for nearly 1,000 kilometres to Lusaka. From here there was a northern route of another 400 kilometres that served forces entering the Cazumbo Salient and a southern one of about 300 kilometres that supplied thrusts both across the Cuando River and along the Zambezi into Angola. The Cazumbo Salient is the large, approximately square geographical protrusion that is bounded by the Congo and Zambia on three sides and measures about 240 kilometres per side. It is crossed diagonally by the Zambezi, which ultimately flows through Zambia, over Victoria Falls, and through Mozambique to the Indian Ocean. Because of its shape, exposed borders, and remoteness, it was quite vulnerable to incursions.

 

Frustrated with Léopoldville’s restrictive position on operations in northern Angola, the MPLA resolved at its January 1964 Conference of Cadres in Brazzaville to begin a calculated move to Zambia. From there it could re-establish its offensive. While Agostinho Neto elected initially to remain in Brazzaville, he would put his trusted associate Daniel Chipenda in charge of the eastern front, and from this point, things developed rapidly for the MPLA.

 

In April 1966 following an intense effort to proselytize the population in the east, first in the district of Moxico and later in Cuando-Cubango, the MPLA infiltrated a large group of combatants into the Cazombo Salient.^[46] This MPLA force showed a remarkable improvement over earlier Portuguese experience in subverting the population and employing selective terror when it did not cooperate. Subversion of the population in the east was estimated to be a modest 6 percent in 1965 and had grown to about 42 percent by 1968, the year that the MPLA offensive attained its height.^[47]

 

In October 1966 following the penetration of the Cazombo Salient, there was a surge of activity to the south in the triangle bounded by the settlements of Ninda, Sete, and Chiume, as shown in Appendix 3. These were located next to the Zambian border and about 375 kilometres southeast of Luso in an area called Terras do Fim do Mundo or ‘Lands at the End of the Earth’. This area is more than 1,000 kilometres from the Atlantic Ocean, and the settlement of Luiana in the extreme southeast is equidistant from Luanda on the Atlantic and Beira on the Indian Ocean. It was this remoteness and the fact that the temperate savannah climate of central Angola, as it spreads south into this area with its lower altitude, becomes so harsh and desiccated that it prompted early explorers to give it this apt name. Such an operating environment favoured the insurgent tactically in that he was easily able to detect Portuguese forces at considerable distance and did so generally before they were aware of him. This insurgent advantage made it difficult for Portuguese forces to find and destroy the enemy, and as we shall see, substantially influenced the crafting of a campaign plan.

 

In November and December 1966 some MPLA insurgents penetrated the northeast corner of the Salient near the border settlements of Jimbe and Caianda. Portugal protested, and a fortnight later in reaction to this pressure Chipenda was ordered by the Zambian government to remove all of his forces from Zambia, leaving him with the only option of invading Angola. This somewhat successful pressure had its roots in Zambian dependence on the CFB. The railway always represented a tempting target for insurgents, and thus Zambia found itself caught between the practical need for foreign sales and the ideological desire to support nationalist movements. Several temporary closures in December 1966 and again in the spring of 1967 for attacks or acts of sabotage on the CFB carried a strong message to both Zambia and the Congo. Portugal held the economies of both of these countries in its hand with the CFB. As a consequence, Zambia half-heartedly pushed the MPLA out of its territory during early 1967, and it infiltrated the east of Angola in two primary areas, Lake Dilolo on the western edge of the salient, and a larger area located about 265 kilometres due south of Luso and bounded by the settlements of Sessa, Muié, Cangombe, and Cangamba. By 1968 this surge had penetrated as far west as Bié district, and as a result of such overextension, various MPLA columns found themselves quite vulnerable. Considerable numbers were destroyed or captured, but by 1970 there were still 1,300 men in Zambian bases and an estimated 3,850 running loose in Angola creating mayhem.^[48] There were clearly limits to diplomatic and economic pressure in the face of Kaunda’s strong feelings about African nationalist movements.

 

The MPLA was not alone in exploiting the east. The FNLA and its military wing, the ELNA, likewise saw the east as an alternative opportunity to the stalemate in the north and felt that it was losing politico-military ground to its competitors through inaction. The organization had fielded three battalions by 1970 following a regrouping from the jacquerie of 1961. These were launched separately into Angola from the north, the northeast, and east in a general invasion. The northern group, which was fortuitously discovered in concentration, was destroyed with strong attacks by the Portuguese Air Force.^[49] The northeastern group crossed in force near Santa Cruz, a settlement some 600 kilometres northeast of Luanda, and was destroyed by several companies of commandos in one of the more brilliant operations conducted in Angola, Operação Golpe de Flanco or Operation Flank Attack.^[50] The eastern group crossed from its base at Kizamba through Teixeira de Sousa, a primary border town and station for the CFB, to reinforce one of its groups already near Luacano and soon clashed with MPLA forces. The MPLA had a strong presence in the area and after its earlier treatment by Léopoldville was not about to be displaced by its competition. As a consequence of its defeat, the FNLA lost enormous face and prestige and key international support from the Organization of African Unity (OAU). At the conclusion of 1970, the FNLA had approximately 6,000 men isolated in the north of Angola in the nearly impenetrable Dembos forest region, of whom only about 2,000 were armed.^[51] The northeastern and eastern groups had ceased to exist as effective fighting forces.

 

UNITA had moved to Zambia on Savimbi’s leaving the FNLA and entered the fray in the east of Angola. His first attack occurred in December 1966 on the post of Cassamba. He made another two weeks later on Christmas Eve, when 500 UNITA insurgents attacked the Portuguese garrison of Teixeira de Sousa and suffered a disastrous defeat with 234 dead. UNITA was soon alienated from its vital Zambian support, as Portugal struck at the fragile Zambian treasury by closing the CFB for a week, as described earlier. Again in April 1967 there were two UNITA acts of sabotage on the CFB infrastructure, and again Portugal closed the railway and threatened lengthier closures unless attacks ceased. Thereafter UNITA was unwelcome in Zambia, and the MPLA likewise was put on notice not to interfere with the CFB. Savimbi was forced by circumstances into an improbable posture, claiming that a true nationalist movement should operate from bases within Angola. Perhaps this was a necessary policy, as in July 1967 he was finally and completely expelled from Zambia. Certainly the implementation of such a policy ensured his isolation. By 1969 UNITA was surrounded, and Savimbi counted fewer than 1,500 followers. In order to survive in defeat, he and his men came to an accommodation with the Portuguese, and between 1971 and 1973 UNITA activities were restricted to a prescribed zone at the headwaters of the Lungué-Bungo, and it agreed to cease operations against the Portuguese.^[52] As part of this understanding in what the Portuguese styled Operation Timber (Operação Madeira), UNITA would receive arms and medical support and would be free to engage the MPLA.^[53] By this time Savimbi had a force of about 500, of whom 250 were armed.^[54]

 

Throughout this period regular contact was maintained with Zambia through the British administrators of the CFB in an attempt to improve relations. Portugal had also initiated two diplomatic missions; however, they were unsuccessful, and Kaunda was described by Joaquim Moreira da Silva Cunha, the Portuguese Foreign Minister, as a ‘treacherous’ leader who understood only ‘firmness and force’.^[55] From the Portuguese perspective it was thought that while you may not like the enemy, if you maintain a dialogue with him, at least you know what he is thinking. On a brighter front, Portugal, which had closed its border with the Congo following the attacks of 1961, reopened it in 1970, when the insurgent situation throughout Angola had been brought under control, and allowed unrestricted population movement. This was quite important, as there were about a million Angolans living and working in the Congo.^[56] This resident population, loyal to the Portuguese and uncertain about Mobutu, came and went over the relatively open frontier and had a dampening effect on Mobutu’s behaviour. As Colin Beer has observed, ‘Angola had been virtually pacified while the situation in Rhodesia steadily deteriorated’.^[57] The differing balance sheets of these campaigns were an indication of the degree that the political and military establishments of those countries understood such conflict.

 

If the national movements had planned to rely on classic proselytizing of the population for support, then it would not come willingly. The Angolan people were experiencing new political and economic freedoms and financial prosperity and would hardly welcome what would be tantamount to Napoleonic-type foraging by foreign armies. Then too the basic premise of all three insurgencies had been removed with successful responses to the people’s grievances. In a classic counterinsurgency response, Portugal had successfully neutralized the nationalism argument with robust economic development and psychological support of its success. Nationalism for the sake of nationalism and the political goal of a single party state running a country for the benefit of its oligarchy of black elites or of a chief-of-state, such as Roberto, held little popular appeal. Lastly the greatest enemies of the insurgents were each other, as all three were in competition for the spoils to be mined from governing Angola. There was no thought of a coalition government for the greater good of the diverse population, and Portugal fruitfully sold itself as the hope for a better future in a classic case of successful informational warfare.

 

In May 1964 a young officer, Carlos Azeredo, was assigned to the enclave of Cabinda as part of Cavalry Battalion 682 commanded by Colonel Serra Pereira with the duty of securing it against the MPLA and the FNLA.^[58] Cabinda shares frontiers with both Congo-Brazzaville and Congo-Léopoldville, and were there a serious and sustained assault on it, then it would be extremely difficult to defend successfully. The counterpoise was that such an attack would invite cross-border operations that Mobutu wished to avoid at all costs. As it was, the war was one of mines and ambushes. Because of the thickness of the jungle undergrowth, contact with the enemy was generally at very close range, perhaps ten metres. The jungle canopy reached to about fifty meters and in the shaded darkness gave an additional degree of concealment to the enemy. The MPLA had numerous bases on the immediate northern and eastern borders of the territory, and because Cabinda was only eighty kilometres in length and forty kilometres wide, it was easy to crisscross the enclave, as depicted in Appendix 4. The ambushes were hit-and-run affairs, with the enemy quickly re-crossing the border into a sanctuary country, either Congo-Brazzaville or Congo-Léopoldville, as depicted in Appendix 5. The Portuguese were forbidden to follow into these sanctuaries, but did so nevertheless.^[59] One night Azeredo laid an ambush in Sócoto, a tabanca near the Luango River, and captured a man named Alexandre Ambrósio, a nephew of Alexandre Taty, a prominent Cabindan and the FNLA ‘minister of armaments’. Ambrosio had been operating from a province of the Congo next to the frontier, had crossed the river in a canoe, and had come six kilometres into the enclave. Two days later Ambrósio’s uncle Taty presented himself to Azeredo at the battalion encampment at Lake Massabi and convinced him to cross into Congo-Brazzaville, where he had a base at Mount Chibuete, and they could negotiate.^[60]

 

Taty, who had studied to be a priest and forsaken the cloth for a job in the post office, had a strong affinity for women and drink. He had stolen from his employer to support these interests, and when discovered, fled to the Belgian Congo and ultimately joined Holden Roberto's organization. He next attempted and failed to replace Roberto, and subsequently with Azeredo acting as a go-between with the PIDE, negotiated the forgiveness of his postal robbery and his return to service in Angola.^[61]

 

Eight days after Azeredo’s trip to the Congo, Taty reappeared leading a force of about six hundred FNLA troops who wished to defect with him and managed an equal number who remained in the Congo to gather intelligence.^[62] These troops under Taty were formed into an organization called JMAE (Junta Militar de Angola no Exílio or Military Council of Angola in Exile) and fought the MPLA primarily in this first phase of its existence. While it was ostensibly for the independence of Cabinda, it was for the moment co-opted by the Portuguese, as Taty’s strategy was to help the Portuguese defeat the MPLA and the FNLA, both contenders for the territory, and then to negotiate for autonomy with the victors. Azeredo notes in an interview with José Freire Antunes that his troops and those of Taty always got on well and that he and Taty became great friends.^[63] Taty was well organised but lacked more modern weapons and equipment and the skill in using them. His arms were Soviet cast-offs. Taty and his men proved invaluable in securing the borders of Cabinda with their presence in both Congo-Brazzaville and Congo-Léopoldville, but more was yet to come.

 

In December 1965 there was a grave concern in Portuguese intelligence circles that little was known about MPLA and FNLA forces in the Congo and their disposition. The idea developed by the PIDE director in Angola, Dr. São José Lopes, was to relocate Taty’s following from Cabinda to the northern border of Angola and to draw from his organisation the needed troops to defend the Districts of Zaïre and Uíge. The Portuguese Navy had through its riverine operations closed the frontier along the Zaïre River as far as Noqui; however, insurgents were crossing far too easily further to the east across the land frontier and creating difficulties.^[64] Taty now with his intelligence gathering infrastructure in the Congo could provide PIDE and himself with the needed intelligence to support cross-border operations and keep both the MPLA and the FNLA off balance and on the defensive. This sort of activity would also raise concern with Mobutu, as the Congo in its vastness was difficult enough to govern without foreign troops running around creating mayhem. Mobutu controlled the MPLA and the FNLA by having them canton their weapons under his supervision and use transportation that he controlled between their bases and Portuguese territory. Now a third organisation was operating within his country and outside of his control, and Mobutu’s anxiety level rose accordingly. He soon became far less hostile to the Portuguese chargé d’affairs in Léopoldville and was increasingly willing to rein in insurgent activity.^[65]

 

The officer assigned as liaison for this new project of defending the northern border of Angola was Major Duarte Silva, who operated under the codename Lourenço. This was assigned to him, as his mission seemed to resemble that of Lawrence of Arabia in organizing and directing indigenous forces. Taty headed the JMAE with his charismatic leadership and would remain in Cabinda to direct operations there. His taste for money, drink, and women, the original source of his difficulties with the Portuguese, remained undiminished. His right-hand man, Hermínio Ferreira, the son of a protestant mission director, would become the military chief and operate from the north of Angola under Duarte Silva’s supervision. The troops would be named Tropas Especiais or TEs because of the special nature of their recruitment. The disposition of these troops and their bases are depicted in Appendix 6.

 

Ferreira’s tale of manipulation and betrayal by the Mobutu government and the FNLA is telling in that it shows clearly the competition between the nationalist movements and the favouring of Roberto by the Congolese government. Ferreira grew up in Malange and in ‘the confusion’ of 1961 enlisted with the MPLA to follow a course of instruction in the Comoros that would lead to his becoming an agronomist.^[66] When he believed he was headed for the Comoros, he in fact was taken to Czechoslovakia, where he spent eighteen months in a guerrilla warfare course. On his return to Africa, he was routed through the Soviet Union and then to Léopoldville, where he was assigned to the FNLA. He defected to the JMAE shortly after his return to the Congo.^[67]

 

Cross-border operations were the raison d’être of and routinely practiced by the TEs, who were composed entirely of black troops and carried no Portuguese identification. Their missions were thoroughly planned, and they prepared for them by practicing on full-scale mock-ups of the targets at their camps in northern Angola. They wore insurgent uniforms and carried Soviet bloc weapons and equipment on their numerous raids. The Portuguese authorities were never sure if these missions were fully accomplished, and while sometimes sceptical, felt that the TE operations kept the insurgents occupied defensively and off balance. They also made Mobutu considerably more cooperative.

 

Initially the TEs were irregularly trained and were organized in groups of thirty-one men consisting of a leader and three sections of ten men each. Later they were expanded through Taty's recruiting efforts and organized on the Portuguese pattern into four battalions of sixteen combat groups of thirty-one men each. This force of sixty-four groups operated from Cabinda and the districts of Zaïre and Uíge in northwest Angola When the ZML became active in 1966, a battalion was sent there, and the typical roster of a combat group known as the ‘Andorinhas’ (Sparrows) in the Eastern Intervention Zone (ZIL) is depicted in Appendix 7. Portugal paid and fed the troops and ran the operation with a very low profile to avoid potential criticism of its harbouring and using former insurgents.^[68] In 1972 the TEs numbered about 2,000 and were incorporated into the regular forces.^[69]

 

When in 1963 the old gendarmerie, loyal to Tshombe, crossed into eastern Angola, its members were known as the Fiéis or ‘faithful ones’ and numbered about 4,600 with women and children. Under what was known as Operation Fidelity (Operação Fidelidade), the Portuguese initially screened and selected some 2,300 men, whom they organized into three battalions (Batalhão de Catangueses) of five companies each. These battalions were based each at one of three camps: Chimbila, on the border between the districts of Lunda and Moxico and next to the road between Buçaco and Dala; Camissombo, near Verissimo Sarmento in Lunda; and Gafaria, the old leper colony near Cazombo in Moxico. The Fiéis retained their original command structure with their own officers and sergeants, and were under the general command of ‘Brigadier’ Nathaniel Mbumba; however, they were tasked and administered by the Comando Eventual da Área Fronteiriça de Teixeira de Sousa (COMEVAK) with a base at the headquarters of the commander of the Battalion of Caçadores 261.^[70] Their primary duty was initially to protect the crews building roads in eastern Angola; however, they posed a continuing threat to Mobutu.^[71] The Portuguese exploited this situation to maintain their influence over him and the insurgent activities that he influenced from the Congo sanctuary in much the same way that they had with the TEs in the north of Angola. These troops were ‘fearless soldiers who fought like tigers’; by 1972 they had sustained thirty-one killed in action, thirty-four killed in accidents, and thirty-six seriously wounded.^[72] By 1974 they numbered about 3,000 troops. Mobutu remained responsive to Lisbon, as he did not want to provoke it into unleashing the Katangan gendarmes, and consequently he maintained tight control over FNLA activities both in the Congo and Angola.^[73] While he attempted to neutralise the Fiéis by promising to grant them amnesty, they never trusted him, for after a similar promise in 1967, he had the following year murdered almost 1,000 of them, usually after hideous torture.^[74]

 

At about the same time and somewhat to the south there was a similar Operation Colt (Operação Potro) designed to receive a smaller group of Zambian ANC dissidents. In 1968 PIDE organized them into a company-sized combat group of 45 men and codenamed them the ‘loyal ones’ (Leais Zambianos, or simply Leais) under the same type of agreement as the one established with the Fiéis, that is, to fight against the nationalist movements in return for support in their struggle to overthrow Kaunda’s Zambian regime.^[75] They were stationed at Calunda in the Cazombo Salient, where officially they formed Special Group 600 (Grupo Especial n^o 600).^[76] These troops easily infiltrated the porous border between eastern Angola and Zambia and had good relations with the tribes on the Zambian side of the border, which helped to control the frontier. They had a reputation as good combatants, well trained and disciplined. The Leais along with the TEs and Fiéis were truly unconventional forces and were able to live and fight as the insurgents did. They were given a great deal of autonomy in their operations and methods, and their operations ranged well into the neighbouring countries. They were able to gather valuable intelligence on activities in the Congo and Zambia and to act effectively on the intelligence that PIDE furnished.^[77]

 

In its colony of Angola, Portugal was able to exploit the economic dependence and divided population loyalties of its hostile neighbours the Congo and Zambia as well as the cleavages within the nationalist movements to prevent a crippling insurgency. The rulers of both countries were largely dependent on Angola for trade, and were eager to have the side effects of the robust Angolan economy with its 11 percent annual growth brush off on their own moribund economies. While relationships were never comfortable, the sponsors of revolution were forced to reign in their ‘freedom fighters’ in order to maintain the functioning of their export-dependent economies. While the liberation movements would have liked to ignore their host nation’s leash, it was not possible when one was dependent on it for recruits, arms, supplies, and camps. The divisiveness within the ranks of the FNLA and the lack of leadership in Roberto eventually doomed it to a spent force. The MPLA was stalemated two countries away from its target of Angola in Congo-Brazzaville for the first five years of the conflict, and then forced to operate from Zambia in the barren scrubland and desert in eastern Angola, where the population was sparse and food and recruits were far and few. For the MPLA it was a long and impossible way from Zambia to its lucrative target, the centres of Angolan population and commerce in the north and west of the colony. UNITA was effectively ‘turned’, surrounded, isolated, and enlisted to fight its competitor, the MPLA. Lastly, the wide and diverse use of the ‘turned’ irregular troops was unique to Angola and had no other parallel in the two theatres of Guinea and Mozambique. The life of an insurgent is difficult, as he is expected to travel long distances and fight far from the sanctuary of his base, its food, shelter, and primitive medical care. He is consequently very vulnerable, particularly if his ideology is skilfully undermined. The story of Hermínio Ferreira is a case in point, although there were many almost daily cases of disillusionment with the FNLA and the MPLA.^[78] There was also enormous competition between the three liberation movements, and they all fought each other as much or more than they fought the Portuguese. This too gave rise to a steady stream of desertions by relatively trained insurgents who were happy to join the Portuguese, where the food, pay, and medical care were reliably good. Portuguese leadership also played a vital role, as it was able to bring together diverse elements of national power in a concerted policy for dealing with these enemies, whether they be insurgents or their sanctuary countries. In a country as poor as Portugal was at the time in comparison to other European colonial powers, this policy of leveraging national power and playing on the vulnerabilities of its opponents worked well during its war in Angola and still holds valuable lessons for countries that find themselves in similar circumstances.

 

 

Fig. 1 - Operations of the UPA in 1961

Source: Hélio Felgas, Guerra em Angola [War in Angola]

(Lisbon: Livraria Clássica Editora, 1961), 24

 

 

 

Fig. 2 - Export Routes for Katanganese Minerals

Source: William A. Hance and Irene S. van Dongen, ‘The Port of Lobito

and the Benguela Railway’, The Geographical Review 46 (October 1956): 468.

Route 1A: Rail and lighter via Port Francqui to Matadi

(1720 miles + two transhipments)

Route 1B: Rail and lighter via Stanleyville to Matadi

(2235 miles + seven transhipments)

Route 2: Rail and lighter via Albertville to Dar es Salaam

(1717 miles + four transhipments)

Route 3: Rail to Lobito (1309 miles - no transhipments)

Route 4: Rail to Beira (1619 miles - no transhipments)

Route 5: Rail to Union of South Africa ports (2300 miles - no transhipments)

 

 

Fig. 3 - Eastern Angola

Source: António Pires Nunes, Angola 1966-1974, Vitória Militar no Leste

(Lisbon: Prefácio, 2002), back cover

 

 

Fig. 4 - Map of Cabinda Showing Disposition of Forces

Source: Major Duarte Silva, Lourenço, Angola 1966-67,

diary kept during his duty as liaison for the Tropas Especiais.

 

 

Fig. 5 - Map of Cabinda Showing Disposition of Friendly and Insurgent Forces

Source: Major Duarte Silva, Lourenço, Angola 1966-67,

diary kept during his duty as liaison for the Tropas Especiais.

 

 

 

Fig. 6 - Map of the Northwest of Angola Showing Disposition of Tropas Especiais

Source: Major Duarte Silva, Lourenço, Angola 1966-67,

diary kept during his duty as liaison for the Tropas Especiais.

 

 

Fig. 7 - Typical Roster for Combat Group ‘Andorinhas’ (Sparrows)

of Tropas Especiais

Source: Major Duarte Silva, Lourenço, Angola 1966-67,

 

^[1]  Peter Forbath, The River Congo (New York: Harper & Row, 1977), 384.

^[2]  Adam Hochschild, King Leopold’s Ghost (New York: Houghton Mifflin, 1998), 301.

^[3]  Mike Hoar, Congo Warriors (London: Robert Hale, 1991), 10.

^[4]  Forbath, 388.

^[5]  Hoar, 11.

^[6]  Ibid., 12.

^[7]  Forbath, 390.

^[8]  Hoar, 12.

^[9]  Ibid., 13.

^[10]  Ibid.

^[11]  Piero Gleijeses, Conflicting Missions: Havana, Washington, and Africa, 1959-1976 (Chapel Hill: University of North Carolina Press, 2002), 126.

^[12]  Franco Nogueira, Salazar: A Resistência (1958-1964) [Salazar: The Resistance (1958-1964)] (Oporto: Civilização, 1985), 570-577.

^[13]  Jeune Afrique (8 August 1965): 19.

^[14]  Willem S. van der Waals, Portugal’s War in Angola 1961-1974 (Rivonia: Ashanti Publishing, 1993), 64.

^[15]  René Pélissier, Le Naufrage des Caravelles: Etudes sur la Fin de l'Empire Portugais (1961-1975) [The Shipwreck of the Caravels: Studies on the End of the Portuguese Empire (1961-1975)] (Orgeval: Editions Pélissier, 1979), 147.

^[16]  Alex Vines, ‘Holden Roberto’, The Independent (London), 8 August 2007, http://www.independent.co.uk/news/obituaries/holden-roberto-460658.hmtl (accessed 23 March 2009).

^[17]  Ibid.

^[18]  John A. Marcum, The Angolan Revolution, Vol. II, Exile Politics and Guerrilla Warfare (1962-1976) (Cambridge, Mass.: MIT Press, 1978), 179.

^[19]  Malyn Newitt, Portugal in Africa: The Last Hundred Years (London: C. Hurst & Co., 1981), 190.

^[20]  Douglas L. Wheeler and René Pélissier, Angola (London: Pall Mall Press, 1971), 167. The authors cite as an example the Congo district in 1960. For its 37,000 square miles it had fourteen concelhos (basic urban or semi-urban administrative unit) or circunscrições (basic rural administrative division) and thirty-seven posts, for an average of 725 square miles per administrative division. This presence would hardly be effective in controlling a frontier, as the posts would be dozens of miles apart. Large numbers of people could and did cross undetected.

^[21]  Hélio Felgas, ‘Angola e a Evolução Política dos Territórios Vizinhos’ [Angola and the Political Evolution of the Neighboring Territories], Revista Militar (December 1965): 706.

^[22]  Shola Adenekan, ‘Holden Roberto’, The New Black Magazine, 22 October 2007, http://www.thenewblackmagazine.com/view.aspx?index=1042 (accessed 23 March 2009).

^[23]  Ibid.

^[24]  van der Waals, 96.

^[25]  Ibid., 97.

^[26]  Neil Bruce, ‘Portugal's African Wars’, Conflict Studies, no. 34 (March 1973): 22. Kinkuzu was in the isolated hills some four and a half hours drive in a Land Rover northwest of Thysville and just short of the left bank of the Zaïre River.

^[27]  Sue Armstrong, In Search of Freedom (Gibraltar: Ashanti Publishing, 1989), 71.

^[28]  Adenekan.

^[29]  Ibid.

^[30]  Ibid.

^[31]  Gleijeses, 239.

^[32]  Expresso (Lisbon), 17 September 1974, 18.

^[33]  Gleijeses, 238.

^[34]  Vines.

^[35]  Ibid. Mobutu renamed the Congo in 1971 as Zaïre.

^[36]  Adenekan.

^[37]  Gleijeses.

^[38]  Ibid.

^[39]  Ibid, 295.

^[40]  Ralph E. Birchard, ‘Copper in the Katanga Region of the Belgian Congo’, Economic Geography 16 (October 1940): 429-436.

^[41]  William A. Hance and Irene S. van Dongen, ‘Matadi, Focus of Belgian African Transport’, Annals of the Association of American Geographers 48 (March 1958): 41-42, 61-62, 69; and ‘Railroads in the Congo Free State’, Bulletin of the American Geographical Society 39 (1907): 482-483.

^[42]  Ibid., 395.

^[43]  Chanda Chisala, http://www.zambia.co.zm/articles/kaunda.html (accessed 1 February 2007).

^[44]  Martin Meredith, The Fate of Africa (New York: Public Affairs, 2005), 381.

^[45]  Dalila Cabrita Mateus, A PIDE/DGS na Guerra Colonial 1961-1974 [The PIDE/DGS in the Colonial War 1961-1974] (Lisbon: Terramar, 2004), 283..

^[46]  António Pires Nunes, Angola 1966-1974, Vitória Militar no Leste [Angola 1966‑1974, Military Victory in the East] (Lisbon: Prefácio, 2002), 21.

^[47]  Ibid., 6.

^[48]  Ibid., 24.

^[49]  Ibid., 25

^[50]  Ibid.

^[51]  Ibid., 25.

^[52]  Adelino Gomes, “Exército e UNITA Colaboraram antes de 74” [Army and UNITA Collaborate before 74], Público (19 December 1995), 2-4.

^[53]  Cabrita Mateus, 199-207.

^[54]  Pires Nunes, 19.

^[55]  Joaquim Moreira da Silva Cunha, O Ultramar, O Nação e o “25 de Abril” [The Overseas Provinces, the Nation, and the “25^th of April”] (Coimbra: Atlântida Editora, 1977), 62-67.

^[56]  José Ferire Antunes, A Guerra de África 1961-1974 [The African War 1961-1974] (Lisbon: Temas e Debates, 1996), 2:639.

^[57]  Colin M. Beer, On Revolutionary War (Bromley: Galago Publishing, 1990), 7.

^[58]  Freire Antunes, 1:379.

^[59]  Ibid.

^[60]  Ibid.

^[61]  Joaquim Miguel Mattos Fernandes Duarte Silva, interview by the author, 3 April 1995, Lisbon.

^[62]  Ibid.

^[63]  Freire Antunes, 1:381.

^[64]  John P. Cann, Brown Waters of Africa: Portuguese Riverine Warfare, 1961-1974 (St. Petersburg, Fla.: Hailer Publishing, 2008), 111-112.

^[65]  Mobutu renamed Léopoldville as Kinshasa in 1966.

^[66]  The ‘confusion’ was an euphemism used by the local native population when referring to the massacres of 1961.

^[67]  Freire Antunes, 1:451.

^[68]  Duarte Silva interview.

^[69]  Estado-Maior do Exército, Resenha Histórico-Militar das Campanhas de África (1961-1974), Vol. I, Enquadramento Geral [Historical Military Report of the African Campaigns, Vol. I, General Framework] (Lisbon: Estado-Maior do Exército, 1988) 1: 243-244; and Resenha Histórico-Militar das Campanhas de África (1961-1974), Vol. II, Dispositivo das Nossas Forças Angola [Historical Military Report of the African Campaigns, Vol. II, Disposition of Our Forces in Angola] (Lisbon: Estado-Maior do Exército, 1989) 2: 347.

^[70]  Estado-Maior do Exército, Resenha Histórico-Militar das Campanhas de África (1961-1974), Vol. VI, Aspectos da Actividade Operacional [Historical Military Report of the African Campaigns, Vol. VI, Aspects of Operational Activity] (Lisbon: Estado-Maior do Exército, 2006), 6: 465.

^[71]  Dionísio de Almeida Santos, interview by the author, 30 March 1995, Porto. Colonel Almeida Santos was Portuguese liaison to the Fiéis.

^[72]  Ibid.

^[73]  Freire Antunes, 2:638-640; and Silva Cunha, 39 and 59.

^[74]  Gleijeses, 249.

^[75]  João Paulo Borges Coelho, ‘African Troops in the Portuguese Colonial Army, 1961-1974: Angola, Guinea-Bissau and Mozambique’, Portuguese Studies Review 10 (Spring-Summer 2002): 142.

^[76]  Resenha Histórico-Militar das Campanhas de África (1961-1974), 6: 467.

^[77]  Aniceto Afonso and Carlos Matos Gomes, Guerra Colonial [Colonial War] (Lisbon: Notícias Editorial, 1998, 341.

^[78]  Edgar O’balance, ‘To Turn His Coat-or Not?’, Royal United Services Institute Journal for Defence Studies 118 (March 1973): 85-87.