4 - The Assassination of Eduardo Mondlane: Mozambican Revolutionaries in Dar es Salaam

Summary

Dar es Salaam was a mecca for Third World liberation movements, whose presence in the city was fundamental to its emergence as a ‘Cold War city’. This chapter shows how their activities became embedded in the capital’s political life through the case of the assassination of the president of the Mozambican anticolonial movement FRELIMO. Eduardo Mondlane was a skilful politician who used the city’s international connections to publicise his movement’s cause and canvass for foreign support. However, as FRELIMO sought to draw on Cold War patronage to wage war against the Portuguese, it was gripped by an internal crisis that split the movement’s leadership along ethno-racial and ideological lines. Powerful gatekeepers within the Tanzanian political establishment aligned with Mondlane’s enemies to challenge him in public and undermine his security in private. These schisms facilitated the assassination of Mondlane in 1969 and clouded the waters of subsequent inquiries into the crime’s perpetrators.

At around ten o’clock on the morning of 3 February 1969, Dr Eduardo Mondlane pulled up outside 201 Nkrumah Street. The address housed the offices of FRELIMO, the guerrilla movement fighting for Mozambique’s independence from Portugal beyond Tanzania’s southern frontier. Mondlane was FRELIMO’s president. He collected his mail and drove to the beachfront villa of an American friend in the leafy suburb of Oyster Bay. Mondlane preferred to work there, away from the city’s noise and heat. He sat down with coffee and sifted through his post. Unwrapping a parcel bearing stamps from Moscow, Mondlane saw that it contained a rare French translation of the Russian Marxist, Georgi Plekhanov. He flipped through the pages. It was the last thing he did: when the Tanzanian police arrived on the scene minutes later, they found a room blown apart by a bomb.¹

Who was responsible for the assassination of Mondlane? This chapter offers no clear answer, though it considers a plethora of possible culprits and alleged conspiracies. Instead, it uses the assassination as an aperture onto Dar es Salaam’s liberation politics and the contested spheres of power they involved. The previous chapter demonstrated how Nyerere, as the leader of a sovereign state, maintained tight control of Tanzania’s ‘official’ foreign affairs, despite the efforts of international actors to build their influence in the country via Dar es Salaam’s political networks. The circumstances encountered by the city’s African liberation movements were more complicated. The political status of their senior cadres was far from clear-cut. They were at once diplomatic missions, governments-in-waiting, guerrilla commanders, and political refugees. In Dar es Salaam, the liberation movements and the factions they contained drew on the support of Tanzanian elites, OAU officials, and foreign diplomats. Just as these sources of external support sustained liberation struggles, so too they could pull movements in opposite directions. This dynamic also confounded attempts by outsiders to influence the movements. Even Nyerere found his own authority drowned in this whirlpool of entangled relationships, which brought global geopolitics into the same space as personal and ethnic rivalries.

This chapter builds on emerging work which reconsiders the history of Africa’s liberation movements in transnational terms. Early literature tended to adopt a national (often nationalist) framework in addressing these struggles, both mirroring and bolstering the ‘official’ histories set out by the movements once they assumed power after independence. But as Luise White and Miles Larmer write, ‘[t]he notion of a single one-way journey from tyranny to national liberation has arguably restricted the development of a more open-ended, fragmented and inclusive set of conflict histories in southern Africa’.² In contrast, new histories stress connections between liberation movements, as well as with foreign powers. They also highlight the tensions within them, often by moving the analytical lens from the political to the social in examining the experience of exile. They critically reassess the ways that liberation struggles were shaped by the dynamics of ethnicity, race, ideology, and class. The result is a messier history, rife with shifting factional and personal alliances.³

A common theme in the new historiography on transnational liberation movements is their mobility. In moving beyond a national framework, it shows how organising and fighting wars of liberation literally involved crossing borders. Refugees fled into exile. Leaders flew around the globe on diplomatic missions, to address the UN or the OAU. They drummed up support in Cold War capitals. Soldiers travelled abroad for training and then were infiltrated across African borders to wage war. Rather than pursue these movements, this chapter takes the opposite position. By focusing on Dar es Salaam as a key node in the struggle for the liberation of southern Africa, it examines how the city served as a site of encounter between these transnational dynamics. Instead of foregrounding the liberation movements’ international connections with the superpowers and their allies, it emphasises the significance of local Tanzanian actors in this political nexus.

This chapter begins by exploring the political world of the liberation movements in revolutionary Dar es Salaam. In particular it addresses how the movements became enmeshed in the city’s public sphere, through their relationships with the media, their dependence on the support of the Tanzanian government, and their dealings with the OAU’s Liberation Committee. These relationships were characterised by friction as much as cooperation. It then shows how Mondlane used FRELIMO’s Dar es Salaam headquarters to build an organisation that commanded international legitimacy and support across Cold War divisions. However, FRELIMO was fractured by multiple splits, which gradually undermined Mondlane’s authority. He had enemies as well as allies within the Tanzanian state, who worked with disaffected FRELIMO members against its president. Finally, it looks back at the assassination itself, the multiple conspiracy theories which surround it, and their significance in Mozambique’s memory politics today.

Mecca of Revolution or Mecca of Mice?

As we saw in Chapter 1, supporting the liberation of Africa was the guiding star of Tanzania’s foreign policy. The result was that Dar es Salaam became a ‘mecca of liberation’, especially for movements from southern Africa. These principally involved organisations representing Namibia, South Africa, Zimbabwe, and the Portuguese territories of Angola and Mozambique. They were joined by an array of lesser-known, smaller groups, including movements seeking the independence of the Indian Ocean archipelagos of Comoros and Seychelles. Outside of the ‘official’ realm of liberation movement politics, Dar es Salaam also hosted exiles from neighbouring Malawi and Uganda. Other itinerant Third World revolutionaries joined this vibrant scene, representing liberation causes as diverse as Vietnam, Palestine, and the Canary Islands. From their offices, the major movements’ leaders organised guerrilla wars, liaised with their Tanzanian hosts, and worked the circuit of embassy meetings and diplomatic receptions to canvass support. In the process, they became embedded not just in the capital’s international affairs, but also in Tanzanian political life. These relationships were far from straightforward. Although the Tanzanian state trumpeted its support for the anticolonial cause, the status, activities, and behaviour of the liberation movement leaders were a hot topic in the city’s public sphere.

Dar es Salaam had emerged as a mustering point for exiled anticolonialists even before Tanganyika formally gained independence. By the time the OAU was formed in 1963, it was the obvious location for the headquarters of its Liberation Committee.⁴ The Committee was first based on Ingles Street (today’s Azikiwe Street), before moving shortly after to premises on the nearby Garden Avenue. Oscar Kambona, then Tanganyika’s minister for foreign affairs, acted as its chairman. The Liberation Committee’s secretariat, which was also mostly staffed by Tanzanians, liaised with the institutions of the host government: the Second Vice-President’s Office dealt with refugee affairs; the Tanganyika Broadcasting Corporation arranged broadcasts on Radio Tanzania; the TPDF oversaw training camps and the distribution of arms. All three of the Liberation Committee’s executive secretaries during its lifetime were Tanzanians. Hashim Mbita, who held office between 1972 and 1994, recalled that he worked closely with Tanzanian ministers, the armed forces, the intelligence services, and Nyerere himself.⁵ This meant that Tanzanians, both in the Liberation Committee and elsewhere in the government, occupied vital gatekeeper roles within the day-to-day politics of the movements.

The Liberation Committee played two key roles in shaping the ‘late’ decolonisations in Africa. First, it disbursed funds to the movements, allowing them to maintain offices, pay travel expenses, run training camps, and purchase arms and other war materiel. Second, the Committee had the power to recognise individual movements, which conveyed upon them a sense of international legitimacy as representatives of their respective territories. This was not always a straightforward process. In some cases, like the PAIGC for Guinea-Bissau, FRELIMO for Mozambique, and SWAPO for Namibia, a clearly dominant movement established strong relations with the Liberation Committee. In others, there were multiple, credible contenders to be the ‘official’ movement. In Angola, South Africa, and Zimbabwe, the struggle was characterised by profound divisions between the movements. The Liberation Committee tried to hedge its bets, in some instances recognising rival movements while seeking to reconcile them, with little success.

All of these processes were dogged by controversy. The Committee was accused of being excessively secretive and corrupt. Patronage and personal relationships between the governing elites of independent African states and claimants to the leadership of the competing liberation movements meant that any choices made by the Committee were politically loaded. Given the location of the Liberation Committee’s offices and its close relationship with the Tanzanian state, much of this criticism was aimed at Nyerere’s government. Nkrumah’s Ghana engaged in a particularly acrimonious confrontation with Tanzania over the alleged mismanagement of the Liberation Committee and misappropriation of its funds. These concerns contributed to the reluctance of many of the OAU’s members to pay their dues to the Liberation Committee, which only hamstrung its activities further. In February 1966, the Liberation Committee bemoaned that the majority of members had failed to meet their financial commitments. ‘It cannot be possibly seen how the committee can carry out its work, achieve its objectives, and meet the ever-increasing demands of the various liberation movements without the spontaneous payment of these contributions’, it stated. ‘This reluctance will be sadly reflected on the seriousness of Africa to liberate its occupied territories – a humiliating ignominy that no African state can bear to face.’⁶

The entanglement of liberation politics with local affairs in Dar es Salaam went much further than these instrumental relationships. Far from being temporary guests in the city, several of the ‘exiled’ movements had much deeper roots in Tanzania, a factor which complicated relationships with the local government. As will be explained, the presence of a Makonde ethnic community which lived on both sides of the Mozambique-Tanzania border became a destabilising factor within FRELIMO. On an individual level, a number of well-educated liberation movement members were employed by the Tanzanian state. For example, ZANU’s Herbert Chitepo was Tanzania’s first director of public prosecutions;⁷ the ANC’s Ben Turok worked as a planner for the Ministry of Home Affairs;⁸ and Frene Ginwala, another ANC member, edited the nationalised Standard newspaper.⁹ There was no discrete sphere of liberation movement politics, as often implied by the earlier nationalist literature on the anticolonial struggles. Rather, liberation politics were engrained in local Tanzanian affairs and their transnational connections.

The liberation movement leaders were central figures in Dar es Salaam’s international media networks. The foreign press pack gravitated towards their activities. The Polish journalist Ryszard Kapuściński, for example, wrote evocatively of tracking down Africa’s ‘fugitives, refugees, and emigrants’ at the terrace bar of the New Africa Hotel. ‘We, the correspondents, come by here frequently to pick up something’, he noted. ‘We already know all the leaders, we know who is worth sidling up to.’¹⁰ The guerrilla leaders, especially those well-versed in the art of publicity, exploited these opportunities to gain global exposure for their cause through interviews and press conferences. They also contributed to Dar es Salaam’s subculture of print propaganda, via publications which ranged from glossy magazines like the ANC’s Sechaba to crude mimeographed bulletins, which were all poorer movements could afford. Liberation movement press releases were a staple of the local media, which carried exaggerated figures of enemy casualties. The movements also utilised Radio Tanzania Dar es Salaam’s external service to broadcast their propaganda across Africa, maintaining connections over the airwaves with the people whom they sought to liberate.

However, wars of liberation could not be fought by words alone. Since African states were unable to fulfil the movements’ requirements for financial assistance, material aid, military training, and especially arms, they turned to the wider world. In turn, China, the Soviet Union, and other Eastern Bloc states perceived an opportunity to increase their influence in Africa. Supporting armed struggle against racist regimes bolstered the socialist states’ competing claims to be global leaders of anticolonial revolution. Officials in Beijing and Moscow also hoped that influence among the movements during wars of liberation might translate into influence over future independent governments. But, contrary to the fears of especially Western diplomats, their targets demonstrated significant autonomy in brokering international relationships. Savvy guerrilla leaders knew that they could use inter-socialist competition as leverage for gaining the support of either Moscow or Beijing. However, geopolitical choices also had the potential to divide movements, as the case of FRELIMO reveals.¹¹

The lives of members of Dar es Salaam’s liberation movements were marked by stark contrasts. Many lived in refugee camps in the suburbs, including students who awaited scholarship opportunities to study abroad. Their compatriots in the guerrilla camps in provincial Tanzania endured austere barracks lifestyles and harsh training programmes. Even in the capital, the buildings from which the leadership organised their struggles were hardly polished embassies. One ANC cadre remembered his surprise at the state of his movement’s offices, which occupied the ‘ground floor of a shabby building, jammed between a row of struggling businesses’.¹² However, a small group of leaders maintained relatively opulent lifestyles in the city, leaving revolutionary visitors unimpressed. Che Guevara was disappointed to see that many of the liberation movement leaders ‘lived comfortably in hotels and had made a veritable profession out of their situation, sometimes lucrative and nearly always agreeable’.¹³ Stokely Carmichael attacked the liberation movements on similar grounds, drawing a stinging rebuke from the ANC.¹⁴ While the guerrilla leaders rejected such accusations, there was some substance to the idea that some preferred the high life to running liberation wars. Andreas Shipanga, a member of SWAPO’s executive committee, recalled initially being impressed with Dar es Salaam’s ‘fine hotels’, like the New Africa, but bemoaned their subsequent deterioration. After the Arusha Declaration, he wrote, ‘when the waiter, the barman and the cleaner became the “ruling class”, all discipline went by the board’.¹⁵

The quotidian life of Nkrumah Street and its environs, where the movements worked in close proximity, fostered a sense of collaborative struggle against shared colonial enemies. The leaders ate, drank, and chatted together on a daily basis. Cosmopolitan urban spaces provided opportunities for cementing pan-African solidarities, as some cadres forged affective relationships with local musicians through their involvement in Dar es Salaam’s vibrant musical scene.¹⁶ But their leaders’ apparent preference for talking in bars over waging war against the colonialists drew criticism from their Tanzanian hosts. Nyerere described the Zimbabwean exiles as ‘a few chaps here noise-making in Dar-es-Salaam and living in hotels, they are not the real ZAPU and ZANU. The real freedom fighters are in Rhodesia.’¹⁷ Similar sentiments were freely expressed by the public, drawing on the anti-urban and anti-elite rhetoric of ujamaa, which castigated laziness and profligacy. The Standard remarked that the liberation movement leaders ‘haunt the capital’s hotels scrounging drinks, tour embassies asking for money and free flights to anywhere and continually squabble about leadership. The one thing they do not do is their job … Unless someone acts soon, Dar es Salaam could become known as the Mecca of mice.’¹⁸ One letter to the TANU newspaper, the Nationalist, alleged that the freedom fighters could ‘only make their revolution at the roof-tops of Dar es Salaam … The struggle they can offer in revolution is the sweat they give while dancing to the bombshell music of duke boxes.’¹⁹

Tanzanian unease at the behaviour of the liberation movements was accentuated by the influx of arms from their foreign sponsors, which entered the continent via the port in Dar es Salaam. In 1966, Oscar Kambona told a meeting of the Liberation Committee that arms for the movements should be sent to other countries, since there were too many weapons in Tanzania. He complained that the first the Tanzanian authorities knew about arms orders was when they arrived at the port. Kambona reportedly suggested that the Liberation Committee should be moved out of the city and instead rotated among its members.²⁰ More generally, the government feared the danger posed by the presence of trained guerrilla troops in camps across the country as potential armed support for a challenger to Nyerere. The case of the treason trial of 1969, covered in Chapter 7, exposed the potential links between dissident Tanzanian politicians like Kambona, liberation movement leaders, and their armed rank-and-file. Acting as gatekeepers to the liberation struggle gave the Tanzanian state significant control over the movements’ activities yet handing out weapons to refugee guerrillas came with inherent risks.

The Tanzanian authorities and the movements themselves were justifiably concerned at the potential for the subversion of the liberation cause by agents of the white minority states. Spotty evidence suggests that Portuguese, Rhodesian, and South African informers were planted into and recruited within Dar es Salaam’s political networks and the inland camps. Phyllis Johnson and David Martin, the latter a long-time journalist at the Standard with privileged access to Nyerere, alleged that Tanzania had broken up a large Portuguese spy ring in the capital in the late 1960s. The men under question said that they had been recruited by a man who carried a Belgian passport and claimed to be a shoe salesman.²¹ In 1968, the Kenyan press reported that a leaked list of twenty-two people being held in preventive detention contained six Goans believed to have being spying for Lisbon.²² A Portuguese informer in Tanzania later reported ‘a number of Goans’ had been arrested, but all except one had now been released. Those with Tanzanian citizenship were allowed to remain in the country; those without it were deported to India.²³ Here, concerns about Portuguese subversion blended with anti-Asian sentiment to root out ‘exploiters’ to the nation who might conspire with Tanzania’s imperialist enemies abroad.

For all these reasons, the Tanzanian state’s wholehearted public commitment to the cause of African liberation was not universally shared in private. Walter Bgoya, a former bureaucrat at the Ministry of Foreign Affairs, remembered that ‘some high officials and politicians’ believed that the country was ‘unduly exposing itself to dangers’ and ‘expending financial and other resources it could ill afford’ in supporting the liberation struggle. They did not say so in public, since ‘who would dare question Mwalimu Nyerere?’ Instead, they ‘slowed things down’ and ‘occasionally resorted to calling [liberation movement leaders] CIA agents as a way to discredit them’.²⁴ This discontent was also in evidence among the general public. Dar es Salaam’s workers grumbled when asked to make material sacrifices to the anticolonial struggle. In 1972, NUTA resolved that each worker should contribute a small portion of his or her income to the liberation fund. In any case, this was less than 1 per cent of the worker’s salary.²⁵ But several wrote to the government-owned newspapers to complain about this imposition. One argued that ‘over-taxed workers’ would be ‘digging our own graves’.²⁶ Another stated that he would not contribute a ‘single penny’ while ‘the freedom fighters continue to marry Europeans, sit behind very expensive mahogany desks, and drive the most expensive cars’.²⁷ That such mutterings made the letters page of the state-owned newspaper on an issue as sacrosanct as African liberation suggests wider grassroots unease at the costly burden it placed upon ordinary citizens, especially while the guerrilla leaders were perceived as having a lavish lifestyle in Dar es Salaam.

Nonetheless, by the mid-1960s, it was clear that in championing the cause of anticolonial self-determination, the Tanzanian government had turned its capital into a centre of revolutionary activity that rivalled Algiers or Cairo. Hosting the Liberation Committee gave the Tanzanian state a certain gatekeeping control of the movements’ activities, often to the dismay of other African governments. However, this also brought dangers for Tanzania’s own sovereignty, including the threat of military reprisals from the white minority states and the build-up of guerrilla forces inside Tanzania. More generally, Dar es Salaam became a vital centre for espionage and journalists seeking sensationalised stories. The liberation movement leaders were adept at working these networks. Yet these myriad connections, which blurred the lines between liberation activities and local politics, could also be appropriated in the internal struggles which afflicted the movements, as the case of FRELIMO demonstrates.

FRELIMO in the Cold War World

Eduardo Mondlane was born in Mozambique in 1920. He studied at university first in Johannesburg, from where he was expelled after a year, and then in Lisbon, before moving to the United States. There, he obtained degrees from Oberlin College in Ohio and Northwestern University in Illinois. He married a white American, Janet.²⁸ While working for the UN Trusteeship Council in New York, Mondlane came into contact with Julius Nyerere, then spearheading Tanganyika’s fight for independence. On an official visit to Mozambique with the UN in early 1961, Mondlane was struck by a sense of frustrated anticolonial ambitions. He considered a future role in an independence struggle and was in communication with likeminded exiles. In the United States, he had already cultivated relationships with powerbrokers at the State Department. While a student Mondlane developed a lasting friendship with Wayne Fredericks, who became assistant secretary of state for Africa under the Kennedy administration.

Meanwhile, Dar es Salaam was emerging as the focal point for Mozambican anticolonial mobilisation. The multiple origins of FRELIMO’s liberation struggle are too complex to recount here.²⁹ The movement itself emerged from a fluid political situation involving several factions, each based on networks within the Mozambican diaspora in Africa. The Tanganyikan government, for its part, sought to forestall the influence of Nkrumah’s Ghana over one particular group of Mozambican exiles. In June 1962, Mondlane travelled to Dar es Salaam in an attempt to provide coherence and leadership as these various factions sought to create a common front. In a series of political manoeuvres, Mondlane was overwhelmingly elected president of the new movement, FRELIMO. After resigning from his position at Syracuse University, Mondlane moved permanently to Dar es Salaam with his family in March 1963. Janet Mondlane directed the Mozambique Institute, a school for refugees located in the suburb of Kurasini.³⁰ Like Nyerere, Eduardo Mondlane initially pressed for a peaceful transition to independence. However, in the face of Portuguese intransigence, Mondlane decided that there was little alternative to armed struggle. FRELIMO launched its liberation war in September 1964.³¹

With his transcontinental education and connections, Mondlane was adept at promoting FRELIMO’s cause around the globe. ‘The most notable and refreshing African liberation figure I reported on was Eduardo Mondlane’, remembered a journalist working in Dar es Salaam. ‘He had his own press network and when he wanted particular coverage he would use journalists from outside, to ensure better, more broad acceptance and coverage.’³² Mondlane’s engaging character, articulate delivery, and debonair style drew foreign admirers. An American doctor, who was well acquainted with the Mondlanes, recalled the scene at an Israeli independence day celebration held at the Kilimanjaro Hotel in 1967, in which Eduardo was ‘surrounded by admirers hanging onto his every word … his booming voice and precise rhetoric carried throughout much of the noisy gathering’.³³ Following Mondlane’s lead, FRELIMO’s propaganda sought to present itself as a future-looking organisation through a visual iconography that focused on its state-like operations in Dar es Salaam. Over time, this gave way to a focus on the war and the liberated zones of northern Mozambique. Nonetheless, the initial impression of a modern, urban front was important in claiming legitimacy in the international sphere.³⁴

This public relations strategy paid off. FRELIMO was the only Dar es Salaam-based liberation movement to receive aid from all three superpowers. Initially, FRELIMO’s connections to the communist world were strongest with China. Soon after Mondlane first visited Beijing in 1963, Chinese small arms began arriving in Tanzania, accompanied by military instructors who trained the guerrillas in the south of the country.³⁵ Uria Simango, the movement’s vice-president, was the closest member of FRELIMO’s inner circle to Beijing. His Maoist sympathies were well known, and he was a familiar face at the Canton Restaurant, a short walk from FRELIMO’s Nkrumah Street offices.³⁶ However, as the 1960s wore on, FRELIMO developed stronger ties with the Soviet Bloc. This reflected a growing irritation among African states and guerrilla movements at China’s inflexible approach to bilateral relations (a trend to which the Tanzanian government was an exception). The Soviet Union harboured initial doubts about Mondlane’s ideological position and his connections with the United States. It was more impressed with FRELIMO’s secretary for foreign affairs, Marcelino dos Santos. Like Mondlane, dos Santos had been socialised into cosmopolitan political worlds while studying outside of Africa, in his case among radical intellectual circles in Paris. Dos Santos mixed political activism with poetry, leading to an introduction to the Soviet Union through an invitation to the Afro-Asian People’s Writers Conference, held in Tashkent in 1958.³⁷

After the Soviet Union rebuffed Mondlane’s request to visit Moscow in 1963, a letter from dos Santos to the Soviet Afro-Asian People’s Solidarity Organisation produced a breakthrough. Mondlane travelled to Moscow in 1964 and 1966, returning on both occasions with promises of aid and military training. FRELIMO delegations also received aid from Czechoslovakia and East Germany. Over the course of the 1960s, the feeling rose in the Eastern Bloc that Mondlane was coming over to their side. After Mondlane visited East Berlin in 1966, the GDR concluded that he had moved to the ‘left’, under the steady influence of colleagues such as dos Santos and Samora Machel, an Algerian-trained revolutionary who became FRELIMO’s director of military affairs. Mondlane encouraged this change in perception by drawing on the split within the communist world. He complained to the East Germans about the treatment of a FRELIMO delegation in Beijing and China’s ‘divisive’ intentions in the Third World.³⁸ Meanwhile, China began to support a rival organisation to FRELIMO, the Zambia-based Mozambique Revolutionary Committee (Comité Revolucionário de Moçambique, COREMO), though this alternative movement never proved a serious challenger.³⁹

It might be surprising, given these links to China and the Eastern Bloc, that Mondlane remained well-regarded in the West. FRELIMO’s relationship with Britain and the United States was certainly complicated by Cold War geopolitics. António Salazar’s regime used Portugal’s membership of the North Atlantic Treaty Organization (NATO) as leverage to suppress criticism of its colonial policy.⁴⁰ Nonetheless, Washington provided covert support for FRELIMO. Senior figures in Washington recognised that Mondlane represented the best chance for a negotiated settlement in Mozambique and a counterweight to more radical elements within FRELIMO. In 1963, the CIA channelled $60,000 to FRELIMO via the African-American Institute in New York. The sum of $99,700 followed from the Ford Foundation to the Mozambique Institute. Even after FRELIMO launched its war of liberation, the United States retained a degree of cooperation with Mondlane. Despite FRELIMO’s openness to receiving support from Beijing and Moscow, some Western observers trusted Mondlane’s claims that he obtained arms from the communist powers simply because he could not from the West. The former ambassador to Tanzania, William Leonhart, told the National Security Council that Mondlane was a ‘force for moderation’ and that by supporting FRELIMO, ‘we would reindorse [sic] Mondlane’s pride and affection for the USA, buy some investment in stability for the movement and keep a better watch on the direction of the struggle’.⁴¹

By the mid-late 1960s, FRELIMO was the best-organised liberation movement operating out of Dar es Salaam. FRELIMO’s leaders, especially Mondlane and dos Santos, had used their cosmopolitan connections to build relationships with numerous foreign donors, including the superpowers. Reputations earned via the daily rhythms of liberation movement diplomacy in Dar es Salaam were translated into material support during globetrotting visits to Cold War capitals.⁴² Yet as Mondlane shook hands with foreign sponsors, the geopolitical and ideological choices which these aid agreements implied contributed to widening tensions inside his movement. Just as damaging, this same cosmopolitanism, which set the leadership apart from less privileged and educated Mozambican exiles, led to animosities inside FRELIMO.

Towards the Crisis of 1968

Like many African liberation movements, FRELIMO was an unhappy family. Its early years were plagued by splits, as the leaders of the parties which were subsumed into the unified movement under Mondlane resented their reduced status. This is obscured by FRELIMO’s official history, which projected the image of a united movement, into which Mozambique’s various ethnic groups coalesced into a singular, nationalist cause. The writing of this history began with the publicity activities of FRELIMO’s leadership in its nascent years. Leaders like Mondlane and dos Santos, whose formative years had been spent in urban Lourenço Marques and had travelled far beyond Mozambique’s borders, were inculcated into a world in which the modular European nation-state presented a model for postcolonial government. The self-presentation of a liberation movement as national in scope was essential in gaining recognition from intergovernmental bodies like the OAU and UN, which then conferred onto the movement a legitimacy through which to attract bilateral support from foreign powers.⁴³

These visions of national liberation were not uniformly shared across FRELIMO’s membership. In particular they jarred with the interests of the Makonde ethnic group, which straddled the Mozambique-Tanzania border. Under European colonial rule in East Africa, waves of Makonde had migrated north into Tanganyikan territory in flight from Portuguese taxes and labour demands. As Tanganyikan independence approached, some Makonde workers feared potential discrimination by a TANU government, and so explored the possibilities of returning to northern Mozambique. The violent colonial repression of demands for greater autonomy they encountered there pushed the Makonde into resistance to Portuguese rule, and later incentivised them to join FRELIMO. This meant that they did not all buy into the movement’s national project. Rather, they saw FRELIMO as a vehicle for pursuing more parochial ends.⁴⁴

Lazaro Kavandame emerged as the most powerful Makonde figure in FRELIMO. In the late 1950s, he had led a cooperative of cotton producers in Cabo Delgado province, which challenged exploitive colonial practices and was temporarily banned by the Portuguese. After FRELIMO took up armed struggle in 1964, Kavandame became an essential ally in the warzone of northern Mozambique. He was appointed as the movement’s political secretary in Cabo Delgado. FRELIMO’s early military successes expanded the territory in which Kavandame’s cooperatives were able to function. However, according to the FRELIMO leadership, Kavandame abused his position to extort produce from the peasantry and, with the connivance of the local Tanzanian authorities, take a cut from cross-border trade. He subscribed to a narrower, more racially defined vision of anticolonial liberation, in contrast to the colour-blind approach embraced by the Dar es Salaam-based leadership.⁴⁵ Kavandame also commanded support outside of Cabo Delgado among the Makonde diaspora in Eastern Africa, for example from the FRELIMO branch in Zanzibar.⁴⁶

The ‘People’s War’ and model of social revolution in the ‘liberated zones’ adopted by the FRELIMO leadership clashed with the interests of Kavandame and his Makonde supporters. Mondlane advocated the creation of new structures of government to administer the freed territory, incorporate the peasantry into the revolutionary struggle, and thereby develop the political consciousness for a liberated Mozambican nation. This involved the extension of party control over regions where Kavandame had carved out an economic niche. Kavandame was not a ‘traditionalist’, but a new-era capitalist whose business practices broke with typical Makonde peasant economics. He lacked the cosmopolitan experience of Mondlane or dos Santos and was an anticolonialist rather than a Mozambican nationalist: the liberation of Makonde territory from Portuguese rule was his ultimate goal. The ‘modernisation’ versus ‘traditionalist’ dichotomy, propagated by FRELIMO after Mondlane’s death and accepted by sympathetic historians, is therefore misleading. Instead, the conflict between Kavandame and the leadership was over competing ideas of a liberated Mozambique, fuelled and complicated by other dynamics.⁴⁷

Prominent among these other ingredients in FRELIMO’s divisions were Mondlane’s connections with the United States. These were subject to continual rumour in Dar es Salaam, a city rife with anti-American animosity. Such suspicions were increased by the sensational case of Leo Milas. To cut a long, twisting, and mysterious story short, Mondlane first encountered Milas’ name in 1962. Milas was living in the United States, though he claimed Mozambican descent. At a time when FRELIMO was desperately short of trained staff, Mondlane was impressed with Milas’ academic background and so invited him to Tanzania, where he served as FRELIMO’s first publicity secretary. In Dar es Salaam, Milas proved a controversial figure and was soon embroiled in FRELIMO’s fractious politics. He was eventually expelled from the movement in August 1964, after Mondlane discovered that he was actually an American named Leo Clinton Aldridge.⁴⁸ There was also a racially inflected aspect to this criticism of the FRELIMO leadership. Some of its members were white, like Mondlane’s wife, Janet. Dos Santos also fell under suspicion, since he was a mestiço who had a white South African girlfriend, Pamela Beira, an ANC member who worked for the Swedish aid agency.⁴⁹

Many of Mondlane’s critics, both within FRELIMO and among the Tanzanian population, looked unfavourably on what they regarded as his distinctly un-revolutionary lifestyle. Whereas FRELIMO’s rank-and-file occupied crowded student dormitories or camp barracks in rural Tanzania, Mondlane lived in comfort in Dar es Salaam’s upmarket suburb of Oyster Bay.⁵⁰ While others faced Portuguese bullets or harsh training routines, Mondlane moved in a world of international jet travel and cocktail parties. This drew criticism from within his movement. ‘Mondlane’s dogs eat better than we do’, grumbled one member.⁵¹ In 1968, a group of discontented Mozambican students in the United States attacked Mondlane’s decision to stay in ‘luxury hotels’ on ‘senseless journeys’ to Europe and the United States – ‘places where true revolutionaries like F. Castro would not set foot’.⁵² We might see this friction as a form of ‘class’ politics, reflective of a gulf between the lived material and social experiences that split an elite like Mondlane from both the rank-and-file and more powerful figures who lacked his cosmopolitan background.

The expectations of Mozambican students in Dar es Salaam provided another source of tension within FRELIMO. They did not envisage the Tanzanian capital as the final point in their exile journeys, but rather a stepping-stone for gaining a university scholarship to study outside of Africa following the completion of their secondary education at the Mozambique Institute. For these young students, Dar es Salaam was less a ‘mecca of revolution’ than, as Joanna Tague puts it, a ‘city of waiting’. Mondlane and the FRELIMO leadership feared that the students were more interested in their own educational self-advancement than waging the liberation war. There was probably some truth to this, again revealing the divergences between the leadership’s top-down concept of a national liberation struggle and the diverse aspirations of FRELIMO’s members. These tensions became aggravated in 1967, when the leadership resolved that failing students would be placed in the army and those who had already studied at university would have to serve time in the military before being permitted to study for a master’s degree. The students reacted with dismay.⁵³

Many of these frictions became misleadingly subsumed into a simplified division between ‘northern’ and ‘southern’ members of FRELIMO. Certainly, the political leadership in Dar es Salaam was dominated by southerners. Refugees and soldiers, who made up the vast majority of FRELIMO members in Tanzania, were naturally mostly from the northern provinces nearest the frontier, including the Makonde. In particular, the reorganisation of the armed forces in 1966 appeared to many dissatisfied northerners as evidence that FRELIMO’s leadership were pursuing an anti-northern agenda. Samora Machel became the head of the army, replacing Filipe Magaia, who had been killed in mysterious circumstances. Some FRELIMO members believed that he had been killed on the orders of the leadership.⁵⁴ Foreign powers with a stake in FRELIMO reputedly sought to latch onto these tensions. In mid-1967, the Portuguese secret police, the PIDE, reported that Chinese agents in Dar es Salaam were cultivating an opposition faction to Mondlane among Mozambican workers of Makonde background at the Friendship Textile Factory, a Chinese-funded scheme in the suburb of Ubungo.⁵⁵ The same shorthand of the north-south division has also come to serve as an axis for explaining splits inside the movement in recent histories which are critical of FRELIMO.⁵⁶

Questions of ethnicity, race, ideology, regionalism, and a loosely-defined idea of ‘class’ all undoubtedly fed into tensions within FRELIMO. Yet these issues should not be understood as discrete categories. Simplified dichotomies, as Christian Williams remarks, ‘overdetermine meaning and obscure exiles’ daily lives’.⁵⁷ Used by liberation movement protagonists, diplomats, and journalists at the time, they have been sustained in FRELIMO’s own ‘official’ history. There was not, Michel Cahen advises, a ‘simple crisis’ within FRELIMO, but rather ‘tensions at the crossroads of numerous, varied factors, without the possibility of democratic control’.⁵⁸ Mondlane himself was alert to the problems posed by these entanglements. In the Struggle for Mozambique, he warned that ‘[t]he complexities of motive behind divisive conduct makes it the more difficult to guard against: individual neuroses, personal ambitions, real ideological differences are muddled up with the tactics of the enemy secret service’.⁵⁹ In early 1968, these tensions spilled over into violence.

—

In March 1968, Mondlane was on a publicity and lobbying tour of Britain when he received an urgent cable from Dar es Salaam that brought him rushing back to Tanzania. His hasty return was prompted by trouble at the Mozambique Institute. At the centre of the crisis was Mateus Gwenjere, a Roman Catholic priest, who had fled to Tanzania from Mozambique in August 1967. Mondlane was immediately impressed by Gwenjere, who was fast-tracked into FRELIMO’s leadership. Accompanied by Simango, Gwenjere represented the movement at the UN General Assembly. In New York, Simango informed Gwenjere about rising discord inside FRELIMO with the Mondlane-dos Santos-Machel faction. This stemmed from the heavy-handed suppression of dissent within the army and the presence of white faces among the Dar es Salaam leadership, whom Simango claimed included Portuguese agents.⁶⁰

On his return to Dar es Salaam, Gwenjere began to criticise FRELIMO’s education policy. He fomented opposition to Mondlane by tapping into discontent at the Mozambique Institute regarding the lack of scholarship opportunities to study abroad and the leadership’s insistence that students served at the front. The students called for the dismissal of the Institute’s white teachers – with Gwenjere’s encouragement, according to the leadership. The ensuing stand-off resulted in the temporary closure of the Institute and reached a climax when a FRELIMO party, including Machel, raided the student dormitories on 6 March. This precipitated Mondlane’s return from London.⁶¹ George Magombe, the Liberation Committee executive secretary, and Rashidi Kawawa, Tanzania’s second vice-president, set up a commission of inquiry into the affair. Its report concluded that student unrest was the ‘direct consequence of the interference in the affairs of the Institute by Father Mateus Gwenjere’.⁶²

Within weeks, FRELIMO was convulsed by more violence. On 6 May, a group of Makonde forcibly closed the movement’s offices on Nkrumah Street. After FRELIMO’s leadership succeeded in getting the offices reopened two days later, the Makonde returned, armed with clubs and machetes. The administrative workers inside attempted to flee to the ANC offices nearby. One member of the Central Committee was fatally wounded. The Tanzanian police arrested eighteen men. At the time, Mondlane was in Mozambique with representatives of the Liberation Committee.⁶³ The local press reacted with dismay. ‘The place for using weapons is not the city of Dar es Salaam’, stated Uhuru, TANU’s Swahili newspaper. ‘The place to use pangas [machetes] and sticks is in the wars in South Africa, Mozambique, Rhodesia, Angola, and elsewhere.’⁶⁴

At a press conference on his return to Dar es Salaam, Mondlane tried to re-establish control. He stated that while two of the assailants were former members who had long deserted FRELIMO, the rest were unknown to the leadership.⁶⁵ The Makonde-dominated FRELIMO ‘Council of Elders’ rejected these claims. In a letter printed in the Tanzanian trade union newspaper, Mfanya Kazi, they criticised Mondlane’s ‘contemptuous designs’ over the movement. They also accused Simango of conspiring against Mondlane, but then of shying away from cooperation with the Elders when they sought his help in forcing fresh presidential elections.⁶⁶ According to Mondlane, Gwenjere was also at the heart of this latest disturbance. In a letter to George Houser, the head of the American Committee on Africa, an anticolonial pressure group, Mondlane stated that Gwenjere had lobbied the Tanzanian civil service and the Liberation Committee to shut the FRELIMO offices and order elections. When this proved unsuccessful, so Mondlane claimed, Gwenjere encouraged members of his church, who were mostly Makonde, to first close the offices and then attack the reopened premises. After the fracas, the Tanzanian government arrested a number of Mozambican refugees in Dar es Salaam, among them Gwenjere, with the intention of removing them from the capital.⁶⁷

Meanwhile, Kavandame’s supporters in Cabo Delgado continued to challenge the central leadership in Dar es Salaam. In the aftermath of the unrest at the Mozambique Institute, FRELIMO branch leaders wrote to the Central Committee (quite possibly with Kavandame’s blessing) to express their dismay at the turn of events. They claimed that Gwenjere had been sent to Tanzania by the Portuguese to stir up trouble and accused him of being a ‘lipyoto’ – a Swahili term for a large bird that pollutes water supplies. However, having sided with Mondlane in his opposition to Gwenjere, the Cabo Delgado leaders then turned their fire on their nominal superiors in Dar es Salaam. They dismissed reports that Kavandame was a PIDE agent and defended his economic activities in Cabo Delgado as the operations of a FRELIMO cooperative, rather than private business. It was not Kavandame who was doing the work of the Portuguese by dividing the movement, they alleged, but the leadership in Dar es Salaam. Although the Cabo Delgado leaders continued to speak in the name of FRELIMO, they called for a Congress and fresh elections to the Central Committee.⁶⁸ Mondlane’s rivals were not united, as the denunciation of the ‘traitor’ Gwenjere from Cabo Delgado shows. Yet this only complicated Mondlane’s position further still. He came under attack from multiple angles, with each critic tapping into different sources of discontent.

By mid-1968, Mondlane’s position appeared particularly fragile. Word reached Lisbon that morale among the Mozambican exiles was low. One Portuguese informer in Tanzania predicted that ‘at any moment now, there will be an attempt on the life of Dr. Mondlane to assassinate him’.⁶⁹ In these circumstances, Mondlane bowed to demands that FRELIMO hold a Special Congress in July. Kavandame wanted it to take place in southern Tanzania, where his support base was strongest. Instead, Mondlane decided to hold the meeting on liberated Mozambican soil. Fearing an anti-Makonde plot, Kavandame and his supporters boycotted the Congress, at which Mondlane and dos Santos strengthened their positions. The former was re-elected president, narrowly beating Simango in a secret ballot. The Congress passed a programme that transformed FRELIMO into a centralised ‘vanguard party’.⁷⁰ ‘FRELIMO is really more socialist, revolutionary, and progressive than before’, Mondlane told an interviewer, ‘and now tends more and more in the direction of the Marxist-Leninist variety’.⁷¹ Yet beneath this bravado, there remained fundamental tensions between Mondlane and Simango. Furthermore, the problem posed to the leadership by Kavandame was unresolved. At this point, Nyerere intervened directly. In August, he brought Mondlane and Kavandame together in Mtwara, southeastern Tanzania, at a meeting attended by several FRELIMO leaders, plus officials representing TANU and the OAU Liberation Committee. But Kavandame refused to compromise. Instead, he pushed ahead with his attempt to set up a rival Makonde separatist movement.⁷²

Mondlane’s Enemies in Tanzania

FRELIMO’s divisive politics were not purely a Mozambican affair. They also involved Tanzanians pursuing their own agendas in destabilising Mondlane’s position. The longer the liberation movements were based in Dar es Salaam, the more they became ‘domesticated’ and entangled with local politics. Multiple institutions in the capital had oversight for the movements’ activities: the OAU Liberation Committee, various government ministries, and the Second Vice-President’s Office, plus the police, the TPDF, and the president himself. This placed multiple officeholders, often with contrasting aims, in crucial gatekeeping roles vis-à-vis the crisis inside FRELIMO. Critical commentary in the local press also played a role in aggravating the splits. Yet these were not just institutional divisions. Just like the tussles inside FRELIMO, individual Tanzanians were motivated by interwoven issues of ethno-racial identity, political views, and personal relationships.

In FRELIMO’s early years, Mondlane was able to fall back on his good relations with the Tanzanian leadership for support. In particular, Nyerere and Kambona were powerful allies. The Tanzanian government’s sovereign power to expel dissidents from the country had then enabled Mondlane to secure his own authority within FRELIMO.⁷³ However, after Kambona fled into exile in 1967, those Tanzanians tasked with overseeing the guerrillas’ security were less inclined towards Mondlane. Rashidi Kawawa’s Second Vice-President’s Office was responsible for refugee affairs, which included the liberation movements’ security. Kawawa himself expressed some scepticism about Mondlane’s political credentials. He told a visiting Soviet delegation in 1967 that he did not believe Mondlane would be the leader of a future independent Mozambican state.⁷⁴ But if Kawawa’s stance towards Mondlane was tepid, then the attitude of other figures at the Second Vice-President’s Office was positively hostile.

Chief among these officials was Lawi Sijaona, a minister of state with responsibility for refugee matters. Sijaona built his career through his leadership of the TANU Youth League during the struggle for independence. After uhuru, he served in a number of cabinet roles, including as minister for home affairs from 1965 to 1967. His personal politics set him at odds with Mondlane on multiple grounds. Sijaona had a reputation, like many TYL cadres, for his hostility towards the presence of Asians and Europeans in Tanzania.⁷⁵ Mondlane’s connections with the United States, his white wife, and upmarket lifestyle represented everything that Sijaona stood against. In addition, Sijaona and the TYL were closely associated with the Maoist streak inside TANU. This gave Sijaona common ideological ground with Uria Simango, who was reportedly unhappy with FRELIMO’s gravitation away from Beijing and towards Moscow. Born in Mtwara, Sijaona was also a Makonde and so shared an ethnic background with Lazaro Kavandame and his supporters. A shared animosity towards Mondlane also seems to have pushed Sijaona towards collaboration with Mateus Gwenjere.⁷⁶ The Dutch embassy went as far as to describe the priest as Sijaona’s ‘protégé’.⁷⁷ Finally, according to the American embassy, Sijaona disliked the manner in which Mondlane frequently bypassed him in preference for dealing directly with Kawawa.⁷⁸ In short, for ethnic, ideological, and personal reasons, Sijaona resented Mondlane.

Sijaona and senior bureaucrats in Kawawa’s office actively undermined Mondlane, apparently against the inclinations of the second vice-president himself. After the trouble at the Mozambique Institute, Mondlane ordered the school’s closure and for the students to be sent to rural camps. In Kawawa’s absence, Sijaona countermanded Mondlane’s order – until Kawawa returned and overruled his deputy.⁷⁹ In April, despite opposition from Sijaona, Mondlane won Kawawa’s agreement for a round-up of FRELIMO deserters and dissidents in Dar es Salaam. These measures were never implemented.⁸⁰ After the attack on FRELIMO’s headquarters, Kawawa rejected accusations published by Mfanya Kazi that his office could have prevented the violence by providing adequate police protection. He explained in parliament that the Tanzanian government would not interfere in FRELIMO’s internal affairs, so long as they conformed with the laws of the country: Tanzania could not tolerate violence on its territory. On 29 May, the Tanzanian government expelled three white Portuguese teachers from the Mozambique Institute and gave them three days to leave the country. A FRELIMO official told the East Germans that the decision was again taken in the absence of Kawawa, suggesting the hand of Sijaona. On this occasion, when Kawawa returned, he did not overturn the order, but merely extended the deadline for the teachers’ departure.⁸¹

Mondlane himself identified Sijaona as a problematic and influential figure within the Tanzanian state apparatus. He briefed diplomatic contacts on both sides of the Iron Curtain that Sijaona was scheming against him, alleging that there was possible collaboration with the Chinese involved. To Mondlane’s relief, Sijaona was then removed from the Second Vice-President’s Office. In October, Nyerere used a cabinet reshuffle to move Sijaona into the position of minister for health. Mondlane claimed to the British that this was the result of his own petitioning of Sijaona’s superiors.⁸² A FRELIMO militant also recalled that Sijaona’s replacement was part of an effort by Nyerere to consolidate Mondlane’s authority within the movement.⁸³ These concerns may well have played a part in Nyerere’s reasoning, although the embarrassment caused by Sijaona’s involvement in a raucous demonstration at the Soviet embassy following the invasion of Czechoslovakia also contributed to his relocation to a less politically sensitive role.⁸⁴

This opposition to Mondlane from influential Tanzanian politicians went hand in hand with criticism in the party press. The Nationalist was strongly associated with TANU’s radical wing, including members of the Youth League. Its journalists were ardent supporters of China. Like Sijaona, they too were hostile towards Mondlane’s leadership style and geopolitical orientation. The day after the fight at 201 Nkrumah Street, a Nationalist editorial criticised the lack of democracy within FRELIMO and other liberation movements. It remarked that ‘conferences are never called to allow for members to exercise their right to choose their leaders or to endorse their trust in existing ones’.⁸⁵ At a rally to mark African Liberation Day on 26 May, the president of Zanzibar, Abeid Karume, accused the guerrilla leaders of being more preoccupied with issuing news bulletins than liberating their territory.⁸⁶ The Nationalist hammered home the point. It accused certain unnamed leaders of living ‘luxuriously in air conditioned bungalows in independent African countries at a time when their own people are suffering from untold colonial cruelties’.⁸⁷ Complaints about the behaviour of the liberation movement leaders were, as we have seen, not uncommon in the Tanzanian press at this time. Mondlane had already responded to such criticisms after the violence at the Mozambique Institute in March. ‘In Dar es Salaam I don’t even think you see even one per cent of the people who belong to the movements’, he said. ‘I think it is wrong to judge by the behaviour of two or three people who may do some outrageous things.’⁸⁸ But the press seemed to have made up its mind. Even if this criticism was not as visceral a threat as the armed thugs who ransacked the FRELIMO offices, it nonetheless helped to establish a discursive environment which facilitated challenges to Mondlane’s authority.

The extent of this press opposition to Mondlane was revealed again in November, when the Nationalist reported on a visit he had made to Nairobi. It claimed that at a private dinner, Mondlane had briefed a group of Americans who were in Kenya to attend an ‘American-African Dialogue’ meeting, sponsored by the Ford Foundation. Some of them had connections in the State Department, including Wayne Fredericks, Mondlane’s friend, though he had left his position two years previously.⁸⁹ The Nationalist repeated rumours that the CIA had penetrated FRELIMO.⁹⁰ Mondlane claimed that he had been in Kenya to meet President Jomo Kenyatta and had met the Americans by chance. He also questioned why the Nationalist had referenced his wife’s race and nationality in the report.⁹¹ The Nationalist’s selective use of information was another demonstration of its hostility: while condemning Mondlane’s presence at the dinner, it neglected to mention that several prominent OAU and TANU officials, including President Nyerere’s brother, had also attended.⁹² Nyerere himself seemed unconcerned. He told George Houser that the Nationalist articles were ‘ridiculous’, adding that ‘we don’t censor everything that goes into the paper’.⁹³ But other members of the government were less convinced. The minister of state for foreign affairs, Stephen Mhando, told the East Germans that the Nairobi meeting confirmed Mondlane’s proximity to the United States.⁹⁴

Trouble continued to mount inside FRELIMO in the aftermath of the fractious congress in July. In December, Paulo Kankhomba, a FRELIMO representative sent to implement the reforms agreed upon at the congress, was murdered by Makonde militants in Cabo Delgado. Meanwhile, Gwenjere set about planning fresh elections for a new FRELIMO president.⁹⁵ In response, Mondlane moved against his rivals. According to Gwenjere, on 27 December, Tanzanian officials raided his home; the following day, he was arrested and deported from Dar es Salaam to the northwestern town of Tabora.⁹⁶ Then, on 3 January 1969, the FRELIMO Central Committee suspended Kavandame from his duties as provincial secretary in Cabo Delgado. But Mondlane’s actions brought only temporary respite. Portuguese intelligence in Mozambique reported that Dar es Salaam was ‘swarming with people from all around, completely out of control and causing the FRELIMO leadership serious concerns’.⁹⁷ On 1 February, Mondlane met officials from the Second Vice-President’s Office. He expressed concern about the continued threat posed to him by Kavandame and his Tanzanian supporters, including Sijaona.⁹⁸

Two days later, Eduardo Mondlane was dead.

Who Killed Eduardo Mondlane?

Mondlane was not, of course, the only liberation movement leader to be murdered in the struggle for independence in southern Africa. In The Assassination of Herbert Chitepo, historian Luise White traces debate about the car bombing which killed the ZANU leader in Lusaka in 1975. Rather than attempt to identify the perpetrator, White deftly demonstrates how the various stories told about Chitepo’s assassination represent attempts to shape the history of the liberation struggle as a source of authority in postcolonial Zimbabwe. The assassination of Mondlane in Dar es Salaam six years earlier has been subject to similar dynamics – though with the caveat that whereas multiple figures have ‘confessed’ to the murder of Chitepo, the debate about Mondlane’s killers is marked by accusation and denial. Nonetheless, in both cases, the existence of plural, incompatible accounts of the assassination were only made possible by the circumstances of exile. Lusaka and Dar es Salaam were rumour-filled ‘Cold War cities’ characterised by a political cosmopolitanism that brought together issues of ethnicity, ideology, and personal rivalries. These conditions permitted the advancement of so many competing and intersecting agendas as to make disentangling the causes of the assassinations a near impossible task.⁹⁹

In Dar es Salaam, Tanzania’s Criminal Investigation Department (CID) took up the murder case. It soon identified the Soviet stamp on the parcel as a forgery. The remnants of the device – plus two identical bombs encased in further Plekhanov volumes, addressed to dos Santos and Simango in the following weeks and intercepted by the police – were sent to London for analysis. Through Interpol, Scotland Yard found that the batteries in the detonators had been manufactured in Osaka, Japan, and sold by a firm in Lourenço Marques. The Tanzanian police believed that the bomb had been constructed in Mozambique and then inserted into Mondlane’s mailbag by a FRELIMO member in Dar es Salaam.¹⁰⁰ But after concluding its investigation in May 1969, the CID kept silent for three years. In February 1972, Radio Tanzania announced that the police knew who had killed Mondlane but refused to name him. These details were revealed in the Observer by David Martin, who used insider information from the Tanzanian police to establish the technical specifics involved in the bombing.¹⁰¹

Moving beyond this ‘official story’, we enter the territory of rumour and rumination. Chatter in Dar es Salaam’s political circles considered the potential culpability of a whole gamut of suspects.¹⁰² Both the Soviet Union and China might have had vested interests in eliminating Mondlane, who was perceived to be the moderate tip of a movement that appeared to be taking a more radical direction and who had known connections with the United States.¹⁰³ As we have seen, there were already suspicions that the Chinese had sought to foment discontent among Makonde workers in Dar es Salaam. But would they risk an assassination attempt on a guerrilla leader who maintained the support of Nyerere, one of Beijing’s closest allies in Africa? No hard evidence has emerged that links either communist superpower to the assassination. Talk of their involvement is nonetheless indicative of the degree to which FRELIMO had become embroiled in the Cold War world, as well as the predilection for speculating about the hidden hand of the superpowers among journalists and diplomats in Dar es Salaam.

Cutting through this Cold War ‘noise’, most accounts of the assassination suspect that the PIDE was behind the plot. Yet removing Mondlane was not unequivocally in Lisbon’s interests. A Portuguese Overseas Ministry report concluded that although the turmoil arising from Mondlane’s death represented a short-term benefit, the long-term consequences of a more revolutionary FRELIMO were far more disadvantageous.¹⁰⁴ One Portuguese intelligence source in Lourenço Marques expressed his fear to the Americans that an extremist turn within FRELIMO might lead to the beginning of an urban terror campaign in Mozambique, a strategy Mondlane had ruled out.¹⁰⁵ The South African consul in Mozambique reported that while the assassination had not been met with ‘undue surprise’ there, it was also ‘not necessarily good news’, given the likelihood of Simango taking over as president.¹⁰⁶ As we will see, FRELIMO did take a more explicitly Marxist-Leninist ideological direction after Mondlane’s death, even if neither a turn to terror nor Simango’s triumph ultimately came to pass.

The PIDE’s own archives offer no real supporting evidence. An internal report did not hide the PIDE’s disdain for Mondlane as Washington’s ‘pretty boy’, whose ‘sandcastle’ had been undermined by ‘sly’ Chinese diplomacy. It concluded that the responsibility for his death lay with Beijing.¹⁰⁷ But even if we accept this denial as genuine, the absence of ‘official’ PIDE participation does not preclude the involvement of Portuguese agents. Multiple sources have claimed that the shadowy Aginter Press network attempted to disrupt the activities of FRELIMO in exile. Ostensibly a publishing house, Aginter was connected to Operation GLADIO, NATO’s stay-behind network of sleeper cells in Western Europe. GLADIO was originally intended to coordinate resistance in the aftermath of a Soviet invasion, but later became associated with anti-communist terrorism with dubious links to conservative European governments.¹⁰⁸ Aginter may have been involved in stirring up trouble among FRELIMO during 1968 through the cover of a journalist who held interviews with prominent leaders, including Mondlane, dos Santos, and Gwenjere.¹⁰⁹ Aginter was also connected to Jorge Jardim, a Mozambique-based businessman who had a strong relationship with Salazar. Although Jardim denied any responsibility, the editor of the newspaper which he owned, Notícias da Beira, noted that on the day of Mondlane’s assassination Jardim waited at the newspaper’s office in expectation of ‘important news’.¹¹⁰ Several sources have claimed that the bomb which killed Mondlane was assembled by Casimiro Monteiro, a Goa-born explosives expert and Aginter operative. Monteiro was first named as a participant in the assassination plot by David Martin in 1975. This has been corroborated by two PIDE agents and a Rhodesian intelligence officer, as well as Monteiro’s own son, though there remains some scepticism about the trustworthiness of their stories.¹¹¹

Even if the matter of Portuguese culpability is accepted, the question of Mozambican collaboration remains a fraught political issue. From the beginning of the liberation struggle, FRELIMO’s leadership has maintained a tightly policed ‘official history’.¹¹² This equates FRELIMO with the Mozambican nation itself and casts any dissidents within the movement, like Simango and Gwenjere, in traitorous terms. More recent histories have challenged this official narrative, particularly through the genre of autobiography and biography.¹¹³ The outcome has been a heated debate in memoirs, newspaper columns, and on the Mozambican blogosphere. The assassination of Mondlane forms a critical, dramatic juncture around which many of these competing histories pivot. In the absence of hard evidence about the actual bomb plot, explanations for the tensions of 1968 serve as means for casting aspersions as to who might have carried it out, even if there is no direct link between the two.¹¹⁴

FRELIMO’s supporters usually identify three prominent Mozambicans as complicit in the assassination: Kavandame, Simango, and Gwenjere. When they began their investigation in 1969, the CID’s prime suspect was Kavandame. Its chief for the Coast Region, Gerald Manikam, told the American embassy that, while conducting investigations a week after the assassination, he had encountered Kavandame in Mtwara, where he was being sheltered by the local TANU chairman. In response to Manikam’s questions, Kavandame gave inconsistent and incomplete answers. In March, Kavandame defected to the Portuguese, giving FRELIMO scope to cast him as a bourgeois ‘traitor’ to the national liberation struggle.¹¹⁵ The CID’s other main suspect was Silvério Nungu, an official at FRELIMO’s headquarters with access to Mondlane’s mail. Shortly after the assassination, Nungu was moved to a new role inside liberated Mozambique. According to FRELIMO, he was caught while also attempting to defect and died of a hunger strike in prison. Simango claimed he was executed.¹¹⁶

Suspicion of Simango largely stems from his actions after the death of Mondlane. Under FRELIMO’s constitution, the vice-president should have taken over the leadership. However, doubts about Simango’s loyalty led the Central Committee to establish a ‘Council of the Presidency’ in April 1969, in which he shared power with dos Santos and Machel. The latter two developed into a stronger faction. In November, Simango published a pamphlet entitled ‘Gloomy Situation in FRELIMO’, which accused dos Santos and Machel of murder, tribalism, and nepotism, calling them ‘vipers’ and ‘tools of imperialism’. He blamed the split not on ideological division, but rather on the predominance of southerners in FRELIMO.¹¹⁷ Simango was expelled from FRELIMO and subsequently joined COREMO, the Chinese-backed movement operating out of Zambia. After Mozambique gained independence in 1975, Simango was forced to read a ‘confession’ of his guilt before a kangaroo court. He was sent to a ‘re-education camp’ and eventually murdered in 1978.¹¹⁸

Gwenjere, the third major dissenter, appears to have met a similar end at the hands of FRELIMO. Gwenjere fled Tabora for Kenya in 1972. From Nairobi, he attempted to organise a party of FRELIMO dissidents, including Simango. He moved back to Mozambique after the revolution in Portugal, but then returned to Kenya when it became clear that FRELIMO would assume power. He was kidnapped in Nairobi in 1975 and never seen again.¹¹⁹ In FRELIMO circles, Gwenjere is suspected of being a PIDE agent, who was sent to Tanzania to infiltrate the movement. For example, Helder Martins, a teacher at the Mozambique Institute, places Gwenjere at the centre of the bomb plot, which he alleges was only made possible by co-conspirators inside FRELIMO.¹²⁰ In his memoirs, Mondlane’s former secretary, Sérgio Vieira, recounts a conspiracy in which the parcel bomb was transferred to Dar es Salaam via Portuguese agents in Malawi and Mozambicans in Tanzania, including Nungu and Gwenjere. Vieira also claims that Kavandame and Simango knew of the assassination plot in advance.¹²¹

Since Mozambique’s post-independence civil war came to an end and democratic space for dissent reopened inside the country, there has been a backlash against this ‘official’ history. Biographers of Simango and Gwenjere have sought to exonerate their subjects.¹²² In particular, they blame the FRELIMO leadership’s pursuit of factional agendas for the creation of a rift between ‘northerners’ and ‘southerners’, with the latter working closely with ‘Mozambicans of non-native origin’. These divisions then translated into ideological differences. Gwenjere’s biographer observes that the well-educated ‘non-natives’, together with the likes of Mondlane, ‘displayed radical thinking’ and were ‘entrusted with the task of delineating the movement’s line of thought, thus influencing its alignment during the Cold War [sic] geopolitics’.¹²³ These histories suffer from oversimplified categorisations, selective use of evidence, and chronological inconsistencies (Simango, a ‘northerner’, had his own international connections in the socialist world, for example). Nonetheless, they serve as important reminders of alternative interpretations to the dominant FRELIMO narrative of the Mozambican liberation struggle.¹²⁴

Finally, there remains the question of Tanzanian complicity. These Mozambican ‘memory wars’ focus on the Mozambican protagonists, often isolating them from the entanglements of exile politics. Martins, who was among the white teachers at the Mozambique Institute who were ordered to leave Tanzania in 1968, believes that Sijaona was ‘undoubtedly’ involved.¹²⁵ The integrity of the police investigation has also been called into question by various sources. According to information given to the American embassy by Manikam, the police were assisted in their investigation by the Chinese. Manikam said that Sijaona was under surveillance by the security services, who were amassing evidence against him in connection with his embroilment in FRELIMO affairs.¹²⁶ Janet Mondlane has recalled that a police officer working on the case with her was suddenly transferred to Moshi, as he ‘was digging up things that the Tanzanian government did not want to reveal’. She has never seen the Tanzanian report into her husband’s assassination.¹²⁷

The potential implication of senior members of Tanzania’s state and security apparatus may explain why the CID’s findings have never been released. During both the liberation struggle against the Portuguese and Mozambique’s post-independence civil war, FRELIMO depended on a close relationship with the Tanzanian government. Any evidence of Tanzanian involvement in the death of Mondlane would therefore have been a source of embarrassment for both sides. At the very least, as this chapter has shown, members of the Tanzanian government played a role in undermining Mondlane’s position, even as others continued to support him. CCM today continues to draw political capital for itself as a ‘liberation movement in power’, as well as in promoting relationships with its now-independent southern neighbours. Tanzanians, too, are rightly proud of the role they played in the liberation of southern Africa. But a more critical reading of this period reveals a far more complex story.

Conclusion

1969 represented a low point in the struggle to liberate Africa. Both the guerrilla movements and the governments of independent Africa were split over the way forward. FRELIMO was fractured in the aftermath of Mondlane’s assassination. The ANC and SWAPO held major conferences on Tanzanian soil to address factionalism inside their movements. The OAU Liberation Committee recommended to African states that, faced with the resolute position of the minority states, there was a need for a more comprehensive approach to the anticolonial cause. ‘It might not only be imperative for Member-States to contribute materially towards the struggle, but also to take the concrete measures necessary for rehabilitating the African man and expelling the colonialists from our Continent.’¹²⁸ Yet some African governments seemed to be moving in the opposite direction. A few independent states, such as the Ivory Coast, responded positively to South African diplomatic overtures.¹²⁹ Although this détente with apartheid drew fierce criticism from states like Tanzania and Zambia, they too showed signs of softening their stance. In April, fourteen countries from East and Central Africa released the ‘Lusaka Manifesto’, which appeared to re-open the door for negotiated settlements with the minority states.¹³⁰ The ANC leadership was dismayed: this turn away from armed struggle seemed to undermine its own operations, as well as its position that the apartheid regime was essentially illegitimate, thus precluding any African negotiations with it.¹³¹

These tensions extended to the politics of the OAU’s Liberation Committee. Both the guerrilla movements and member states challenged Tanzania’s control of the organisation. At a meeting in Dakar in July, Stephen Mhando conceded that Tanzania would permit its headquarters to be moved from Dar es Salaam. In fact, he would not even oppose the abolition of the Liberation Committee altogether. But he reiterated that the liberation struggle itself had to continue. ‘The proper place for these disillusioned gentlemen to talk tough is in Salisbury and not in Dar-es-Salaam’, Mhando argued. ‘They must not sit down in comfort in the capitals of free Africa … and then have the impudence and the insolence to insult the governments which make possible the struggle in which some of these leaders are unwilling to play a full and physical part.’¹³² When the Liberation Committee next met in Moshi in January 1970, George Magombe, its Tanzanian executive secretary, was glum. His report bemoaned ‘the continued state of seemingly endless spiral of internecine disputes and ethnic disunity in the rank and file of some of the movements. Much energy and time is dissipated on resolving bickerings and clashes of personality’.¹³³ This nadir prompted Nyerere to restate Tanzania’s commitment to armed struggle in the face of white intransigence in southern Africa. In Toronto in October 1969, he stated that ‘if the door to freedom is locked and bolted … the choice is very straightforward. Either you accept the lack of freedom or you break the door down.’¹³⁴

The choice to take up arms against colonialism might have been straightforward, but its implications were not. Liberation movement affairs were characterised by conflicting and converging personal, ideological, geopolitical, regional, and ethnic agendas. This complicated the efforts of foreign powers to influence the struggles, as well as the Tanzanian government’s attempts to maintain some control over their activities. By hosting the liberation movements in Dar es Salaam, Tanzanian state actors took on powerful gatekeeping roles. But the assassination of Mondlane served as a bleak reminder that Africa’s revolutionaries had dangerous opponents. It was further grist to the mill to those within the Tanzanian state who cited the anti-imperialist threat as necessitating greater vigilance against the county’s enemies abroad and their lackeys within, as we will see in later chapters. For all of this, Dar es Salaam rightfully earned its reputation as a revolutionary capital in Africa. The influence of the liberation movements extended far beyond their Nkrumah Street offices. Their daily activities, press coverage, and speeches at rallies or at the university instilled the city’s politics with a militant anti-imperialism. As protesters took to the streets in Paris and Prague in the late 1960s, Tanzania’s youth were therefore already at the forefront of a revolutionary moment which spanned Africa and the Third World. Their experience of the ‘global 1968’ forms the subject of the next chapter.