ARTICLE « DHLAKAMA E MANINGUE NICE! " An Atypical Former Guerrilla in the Mozambican Electoral Campaign Michel Cahen2 Despite significant problems with urban criminal gangs and highway robbers, Mozambique is a country at peace since 4 October, 1992.3 Under the auspices of the Italian religious congregation of Santo Egidio,4 Joaquim Alberto Chissano, president of the Republic and of the Frente de libertagao de Mozambique (FRELIMO), once the 'Marxist-Leninist' single-party of the country, and Afonso Dhlakama, president of the Resistencia national de Mogambique (RENAMO), the former guerrilla group and longtime recipient of aid from South Africa, signed the 'General Peace Accord' on that date. This agreement has never seriously been called into question. During the first 15 days following the accord there were a number of incidents. For example, it contained provisions for maintaining each party's administration in their part of the country, but did not specify the zones belonging to one party or the other. As a result, a number of RENAMO regional commanders sought to occupy areas that they had not previously controlled, the largest one being Angoche in Zambezia. It appears that they had told Afonso Dhlakama that they already held these areas, and following the peace accords, they urgently sought to gain real possession of the territory in question. In general, however, nobody made any mistake about the situation: when Dhlakama ordered his guerrillas to stop fighting, the fighting stopped. The country as a whole understood what was happening and immediately came back to life: bus lines that had not functioned for years started right up again; people went back to cultivating fields at the outskirts of cities; reconciliation ceremonies began. Although many observers, influenced by FRELIMO's total monopoly on domestic and international news coverage, believed that RENAMO was little more than a constellation of groups of largely autonomous 'armed bandits' acting for their own interests and scarcely controlled by Dhlakama, it became clear that TRANSFORMATION 35 (1998) ARTICLE CAHEN RENAMO was in fact a disciplined and strategically very centralised army. For two years, there was not a single military incident involving RENAMO soldiers, even though their quartering conditions were absolutely disastrous and they could have legitimately felt abandoned.5 RENAMO's capacity to enforce the cease-fire among its own soldiers was in itself a very important political fact. This certainly does not mean that there was anything less than profound mistrust between the two parties. The initially very slow and only later brutally accelerated rhythm of the demobilisation illustrates this distrust, as does the state of quasi-isolation that characterised a number of RENAMO-controlled zones for a year after the peace agreement. Concerning the matter of isolation, the guerillas were not the only ones to be distrustful: often the local populations themselves did not wish to see the return of the comrades a term that designates by analogy with the Zimbabwean soldiers who had cruelly attacked and bombed them, anyone coming from the world of FRELIMO, from the state, or from the city. Also included under this rubric were activists from the smaller emerging parties as well as businessmen perceived as belonging to the same world, sociologically speaking. Back to Life! There is no doubt that the people's will for reconciliation, which almost amounts to a collective desire for amnesia, has been the manifestation of a very powerful instinct of personal and social survival. Everybody agreed: some had been forced into the military service for the state, others had been shanghaied by RENAMO, and everyone had 'done their part'. 6 Following 15 years of civil war, all testimony supports the remarkable observation that with few exceptions, the society does not blame RENAMO for the war, or at least no more than it blames FRELIMO. Nor for that matter has there been any rejection of 'armed political parties' as such, which could have opened a political space for the new emerging parties. Indeed, this reveals a great deal about the nature of the war. Although an authentic, albeit atypical, civil war (see below), it was without doubt a war against civilians, with an equal share of atrocities committed by both sides; largely a private war between two armies, imposed upon the civilian population who neither had nor have any real stake in the war. The war has not been forgotten, far from it, but a will to survive and a desire for reintegration on the part of those snapped up by the two warring parties (including those guilty of massacres against their own community) have dictated that the war, as a theme, be spontaneously rejected. People conducted the necessary ceremonies in order to appease the spirits of the unburied dead. The Christian churches, notably the Catholic Church, organised large public prayer TRANSFORMATION 35 (1998) CAHEN ARTICLE services for peace and reconciliation. Even FRELIMO, who planned to cite the 'past' and the war as its principal campaign themes, found its Mozambican and Brazilian public relations advisors counseling it against doing anything of the sort.7 Dhlakama, for his part, brought up the subject in all of his meetings in a way that almost legitimated it: 'Quern e que nao matou? Foi a luta!' ('Who hasn 't killed? It was war!'). This process of popular reconciliation must be distinguished from pressure in favor of political consensus and national union between the parties. The former was totally spontaneous, and further studies will have to be done to measure its intensity and its frequency. While reconciliation is indispensable to the democratisation of the country, an imposed political consensus could be harmful.8 The Elections of Silence9 The desire for reconciliation and the tremendous will to revive the society1" did not signal an end to the fear and apprehension that the war would start up again. On the contrary, one was acutely aware that the goal of these elections was not simply to choose new deputies and a new president, but above all to restore peace. People expressed this yearning when they spontaneously responded that they wanted to vote for both parties in order that the two would get along. All testimony agrees that, as in South Africa, the population was aware of the importance of the historic moment it was living. The rate of voter absentention, less than 13 per cent, shows this and signifies that practically everybody who could vote did vote despite two major impediments that prevented many from voting. First of all, the electoral law stated that a person could only vote in the district of residence at the time of the last census. Following the war, many people returned home and thus no longer lived in the district that had them on the voter rolls. Secondly, the scarcity of polling stations in rural areas prevented a portion of the electorate from voting, a factor generally disadvantageous to RENAMO. As in South Africa, democratisation in Mozambique is the product of a fundamental struggle between irreconcilable actors who, on the basis of their unique point in common - their conflict - managed nonetheless to progressively integrate their strategies into a common framework (Darbon 1995). But the resemblance between the April elections in South Africa and the 27 to 29 October elections in Mozambique stops there. While in the 'country of apartheid' tne politics that marginalised the majority of the population had the paradoxical effect of creating a series of highly structured frameworks for the different South African 'civil societies', permitting on the political level the progressive affirmation of a true civil society from 1976 (Darbon 1995), in Mozambique no TRANSFORMATION 35 (1998) ARTICLE CAHEN real political space, no civil society, was created. The civil war expressed, in a way deformed by the intermediary of a military structure initially introduced from the outside, the marginalisation of a large portion of the society by the modern state, and, indeed, the RENAMO zones constituted regions of severe autarchy. The period of peace has not at all been one of revitalisation of social activism, with the very specific exception of the movements organised by veterans and disabled soldiers." Despite the wave of strikes that occurred in January, 1990, n at the introduction of political pluralism but still in the midst of the war, the period 1990-94 saw very little union activity. The only possible exception to this was the 'birth' of certain, self-proclaimed independent unions from the crisis-ridden Organizagao dos Trabalhadores Moqambicanos (OTM) (the only legal Mozambican union). There were, moreover, very few strikes (the teachers' strike being the only important one) and very few collective actions. After 19 years of single party rule and 15 years of war, the society found itself in a state of disintegration: exhausted, fragmented, and reduced to a lumpenproletariat. Under these conditions, theft (notably within companies) and other money-makjng schemes were much more socially feasible than organised collective bargaining." In any case, what does it mean to demand a 100 per cent salary increase when the bare minimum calls for 1 000 per cent? It has been ages since one actually survived on one's salary: to a certain extent the working and wage-eaming classes have been physically destroyed. Unlike in South Africa, where the conflict began to have an integrating effect near the end of the 1970s, in Mozambique, the conflict asserted the dissidence of large marginalised social groups or 'marginalises'. 14 The cessation of hostilities in 1992, however, provoked no modification in the practices of the state, nor any move towards integration. Besides, in a country without a nation," without a domestic market, and without transportation, granting freedom to the society does not necessarily induce a movement towards integration: each individual may simply prefer to go home to his own world, to his own identity. In accepting the cease-fire in Rome in October, 1992, FRELIMO accepted RENAMO before the international community, and to a much lesser extent, before the people of Mozambique. At the very least, FRELIMO began referring to RENAMO by its name and no longer by the de-politicising and illegitimating epithet of 'armed bandits' that was created by Samora Machel to deny RENAMO any socio-political dimension.16 If, however, FRELIMO was henceforth willing to consider the guerrillas individually as 'misguided brothers wandering in the bush', FRELIMO's conception of RENAMO as an organisation did not budge. TRANSFORMATION 35 (1996) CAHEN ARTICLE In its rhetoric and, above all, in its political practices and habitus d' elite (in the sense of French sociologist Pierre Bourdieu), it accorded RENAMO no political legitimacy whatsoever. RENAMO remained a group of fringe elements, illiterates, assassins, and macacos (monkeys).17 FRELIMO, on the other hand, had created the nation, produced and owned the state. The idea of losing the election was inconceivable for FRELIMO. This was clearly due to certain material reasons easily understood on the part of a nomenklatura become solidly middle-class, but also stemmed from FRELIMO's conception of its role in Mozambican society. Its political arrogance constitutes an established fact and it was a shock for FRELIMO negotiators to learn that the delegates sent to Rome by RENAMO were not illiterate 'savages' . One quote from president Joaquim Chissano in 1993 succintly summarises FRELIMO's most audacious positions concerning RENAMO: 'I'm ready to take ministers from RENAMO once they've joined FRELIMO' . It is another example of the old logic of integration that had already been tried at the time of the first negotiations in 1984 and at the time of the amnesty law in 1988. FRELIMO could conceivably imagine letting the marginalised into its world, but could never recognise them as 'Others' who were themselves legitimate political actors. Thus, though there were numerous festive aspects during the public meetings of the campaign, the elections themselves, from 27 to 29 October, 1994, were not the magic act of the previous 'South African autumn'. There was no popular enthusiasm; instead there was a seriousness, even a feeling of resignation, that one had to vote in order to bring peace, though one did not really want to choose. In their state of exhaustion, people wanted peace and wanted to be left in peace. FRELIMO's victory provoked no popular enthusiasm, not even in the southern capital, Maputo, where only the FRELIMO activists were out in the streets. The spectacular victory of RENAMO in Sofala did not spark any greater jubilation in the provincial capital of Beira. Although these elections were undoubtedly a powerful political moment, they were nevertheless the elections of silence. To understand this, we have to work backwards. The Peace Process: democratisation or pacification? The peace process took place under strong international pressure. Nevertheless, it would be entirely wrong to argue that the establishment of political pluralism was simply imposed from the outside, a perspective for analysis very much in vogue to explain the failure of democratisation. First of all, the movement towards a pluralist constitution took place in January, 1990. In other words, the introduction of pluralism took place during the war, before direct negotiations had opened and at a time when FRELIMO was still TRANSFORMATION 35 (1998) ARTICLE CAHEN hesitating about the usefulness of political pluralism. FRELIMO realised that pluralism might very well serve to destroy its opponent by offering certain RENAMO leaders the opportunity to give up the military struggle to form a political party. Above all, however, the country had reached a point where the single party system could no longer function as before. Even if pursuing the war served the interests of a whole class of corrupt military officers - a fact that always kept the military option open - FRELIMO knew that it could not defeat RENAMO, even if the latter were deprived of South African support. Moreover, even though FRELIMO's foreign investors did not apply any pressure, FRELIMO still understood that its own interests lay with the development of these investments (personal profit for FRELIMO members through 'commissions' paid by these investors, for example). More and more numerous were those saying that RENAMO supporters were also Mozambicans. The increasing likelihood of direct negotiations with RENAMO forced FRELIMO to take the initiative in establishing political pluralism to avoid the appearance of acceding to all RENAMO demands. The 5th FRELIMO congress (1989), which had ratified the abandonment of Marxism, had been a disappointment: the 'Mass Organisations' - unions and youth organisations in particular - did not function at all; the Women's Organisation became more and more like an NGO with social goals; the journalists' and professors' organisations expressed their desire for independence. As a result of support for structural adjustment, the effects of laissez-faire capitalism, and racketeering among its leaders, the once austere FRELIMO had fallen into a state of moral decay (illustrated, indeed, by the personal practices of certain of its leaders).18 FRELIMO was no longer a party, but rather a coalition of factions united only by their fear of losing power. The state, weakened by liberalism in war-time and by the invasion of NGOs, resided more and more, not in the party, but in the Presidency. It is clear that even more than the Angolan MPLA, FRELIMO needed (and needs) foreign aid, since it has neither oil nor diamonds, and the natural gas in Pande has yet to be exploited. Was it thus all the more sensitive to foreign pressure? In fact, aid has been sufficiently generous and diversified that, paradoxically enough, it has afforded FRELIMO a certain independence: whenever a given donor threatened to suspend or delay its contribution, FRELIMO always found several other countries or institutions willing to fill the gap. The government exploited this competition between donors very adeptly, furiously seeking to preserve its habitus. An example of this arose during the controversy surrounding the Government of National Unity (GUN). More and more involved in business activities, influenced by the considerable weight of the military, and totally insensitive to the international and regional (notably TRANSFORMATION 35 (1998) CAHEN ARTICLE South African) pressures, FRELIMO rejected the formation of the GUN, thereby expressing its desire to keep the state for itself and maintain its own conception of democracy according to which it had a right to everything once it had 51 per cent of the vote. FRELIMO was, however, sensitive to pressure from the United Nations Operation in Mozambique ONUMOZ (which was incomparably more powerful than its Angolan equivalent) because the ONUMOZ with its 7 000 person staff became a true internal political actor. It obtained a veritable directing role in the peace process, aided by the strong personality of the Italian Aldo Ajello, Boutros Boutros-Ghali's special envoy. The National Electoral Commission (CNE) was truly pluralist, with ten members of the government, seven from RENAMO, three from other parties, and a president, Brazao Mazula, who, despite having a family strongly linked to FRELIMO, was able to establish a truly independent role for himself. The fact remains that the fundamental goal of the international presence in Mozambique was not the democratisation of the country, but rather its pacification and stabilisation. No chance was given to the small parties, even the most serious. The financial aid coming from the Trust Fund (dubbed by many the 'tristofundo1 , the 'sad fund' ) managed by ONUMOZ, though appreciable for those embezzling from it, was very limited for any opposition party aside from RENAMO. In any event, the money was often handed out so late that •campaign materials ordered from South Africa ended up arriving the day of the vote.19 The formation of the new army was done following a strict division of power, starting with an equal number of troops from FRELIMO and from RENAMO. Young soldiers were supposed to re-enlist, despite the fact that many of them had been taken from their homes up to ten years earlier. Indeed, the majority of the troops dreamed only of escaping their life of slavery under arrogant and corrupt officers (for those in the FRELIMO army) or their life of misery as guerrillas (for those in the RENAMO army). The policy obviously failed and was the cause of numerous incidents involving soldiers demanding to be discharged and paid.20 It is very significant that no one envisioned the creation of a new army made up of new recruits having never fought with either army. More generally, however, we can say that the pressure for a political consensus around the major issues, if not for the GUN itself, has been (and still is) antagonistic to the democratisation of the country and leads to ethnic division. Let us stress once again the importance of distinguishing between the popular process of national reconciliation and the political consensus at the top. The latter entails an established and unmodifiable political program, preventing the TRANSFORMATION 35 (1998) ARTICLE CAHEN establishment of small parties and associations since all potential political actors would have to avoid debate on a number of key issues: structural adjustment, economic liberalism, regional free trade (though RENAMO did have certain criticisms regarding this last point), etc. Joao Gomes Cravinho pointed out that, in 1993: There were 17 political parties or partisan projects. Sixteen of them proclaimed their support for social democracy. The differences in the programs of these diverse parties were always minimum. ... The economic part of RENAMO's government program was practically copied from FRELIMO 's own program. The lack of cadres outside of FRELIMO is one of the reasons for that lack of opposition proposals, but other factors do come into play. Finally, even asfar as demagogy goes, the opposition offered little opposition. Why did so many parties and so many candidates all defend almost exactly the same thing? Certain party projects had no political significance, being the result of pure megalomania. This is obviously the case for the PPPM ofPadimbe Kamati, for the PPLFCRM of Neves Serrrano.21... There werealso diverse cases of parties and candidates that sprang forth from serious, but entirely personal differences between members of the same political formations. This is what happened with Casimiro Nhamitambo who left the PALMO to create the SOL22; with Vasco Campira (PACODE) who saw his former PCN disciples withdraw their support for his candidacy23; with Miguel Mabote who founded the Labour Party (mysterious name given that the party is neither socialist nor has any link with the union movement) when he realised that he could not unseat Padimbe Kamati of the PPPM; it's what led to the schism of PANAMO from PADEMO24 ... In summary: ideological divergence does not explain the proliferation of candidacies. The great majority of partisan projects and of candidacies are explained by 'micro-political' factors, that is to say, personal or extremely localised questions. To understand what happened during the elections, one must rely above all on micro-political analyses25 (emphasis added). AH this really does is express the exhaustion and the economic, social, and intellectual fragmentation of the segments of the elite who have been historically excluded from power. This reproduced, in caricature, what had arisen from April to September, 1974, with the blossoming of small political groups later outlawed by FRELIMO. In practice, the parties were tempted to make themselves TRANSFORMATION 35 (1998) CAHEN ARTICLE conspicuous in expressing a point of view that Joao Gomes Cravinho qualifies as 'very localised'. In other words, they sought to express the point of view of certain lineage, ethnic, regional, and even religious groups.26 The opposition, however, have not been the only ones susceptible to ethnic division in Mozambican politics: FRELIMO itself is undoubtedly the principal ethnic party of the country. We will come back to this point. In general, neither the context, nor the process left the least chance for the emergence of a third force, independent of the two armed parties. Such a force, however, is an essential element for the emergence of 'civil society'. Although we may criticise it from the perspective of democratisation, the process in Mozambique has truly been one of pacification. In Mozambique, unlike in Angola, the United Nations has been able to rely on one certain fact: barring a threat to its leaders' security, RENAMO had decided to abandon the war, regardless of the situation. Furthermore, this was a point never called into question despite endless provocations continuing up until the days following the elections.27 The guerrilla leaders understood very early on that their survival could only be political. I was personally convinced of this in my first meeting with Afonso Dhlakama, then still without international recognition, when he came to Paris in June, 1992. I expected to find a classic guerrilla leader, but instead found a dyed-in-the-wool politician. RENAMO succeeded in doing what Angola's UNITA never did: namely, its 'civilisation' in the literal sense of the word: becoming 'civil' in contradistinction to 'military' (see below). Notable in this transformation is that the anti-urban aspect of the movement, which was relatively clear during the war, did not prevent urban professionals from joining following the armistice. An Atypical Civil War We will not recapitulate the history of the war here, which in any case remains largely to be written, but we will refer to the characteristics of the war useful for the present analysis. It is clear that RENAMO would not have existed without Rhodesian support. Nevertheless, we cannot dismiss the organisation, even at its inception, as simply a group of mercenaries or as a Rhodesian puppet group. From the start there was always a 'kernel of dissent' in certain areas and in certain social circles.28 Likewise, one should not underestimate the listenership among black Africans of Africa Livre - a radio station run by former colonists and Creoles hostile to FRELIMO and founded even before the creation of RENAMO. RENAMO did not, however, become a real force in Mozambique until it managed to become a part of the profound social crisis of the country and, in a way, express that crisis. TRANSFORMATION 35 (1998) ARTICLE CAHEN It succeeded in doing this because of the overlap of two processes of marginalisation: the first, inherited from the colonial period and the history of FRELIMO in that context and the second, the result of the single-party's own policies. In order to sketch the history of this first process (Cahen 1991b, 1994b), I would say that the drama of Mozambique takes shape in 1903, the date of the transfer of the capital from Mozambique Island in the north-east to Lourenco-Marques (today Maputo) in the extreme south. This was obviously a major turning point in the history of Portuguese colonisation, the authorities having decided to make the colony profitable with a service economy oriented towards the British hinterland. The result was one of complete economic, social, and political marginalisation of the core of the old Creole elite (ethnically mixed in a variety of ways: blacks from prazos,19 Indian Suaili from the north coast) and the recreation of a new and tiny elite in the extreme south in the very bureaucratic fashion of Portuguese colonialism. Nonetheless, it was this southern nucleus that provided the core of the intial leadership of the nationalist movement. The former elite groups, socially different (without doubt more bourgeois, being linked to commerce and to the plantations), but greatly weakened, had the most difficulty in becoming a part of FRELIMO, even though they had demonstrated a certain vitality during the proto-nationalist phase (and were crushed early on precisely for this reason)." FRELIMO recruited few members from either the centre or the north of the country and had great difficulty keeping them, being essentially a coalition of cadres from the extreme south and a guerrilla mass in the extreme north. None of this, however, was decisive and FRELIMO found itself well received in 1974-75 in those regions of the country where it had no real hold. Later, however, FRELIMO's attitude changed. With the establishment of one-party rule, the creation of a state hostile to African society in which domination by the south could never be discussed, and the imposition of a homogenous nation through the repression of identities, the phenomenon of a 'Foreign State' appeared. Growing sectors of the society felt that the FRELIMO state represented them no longer or not at all, that its language (Portuguese), its discourse - the 'AbaixoF ('Down with tribalism!' , 'Down with feudalism!', 'Down with obscurantism!', in short: 'Down with the people!'), and its policies (communal villages, state farms, pricing favourable to the cities, intense repression, rigid social organisation) were foreign to their most common aspirations and values, despite FRELIMO's populist discourse of Stalinist-Marxist rhetoric (Brito 1991; Cahen 1993a). The abandonment of this rhetoric and the adoption of a laissez-faire capitalist economy during wartime in 1985 did not modify this 10 TRANSFORMATION 35 (1998) CAHEN ARTICLE perception since the social group in power remained unchanged, with its advantages stemming henceforth not only from privileges inherent to the nomenklatura, but also from business deals and from graft. The overlapping of these two processes of marginalisation has thus entailed a spirit of dissidence within large sectors of the population. Were it not for Mozambique's unfavourable geography - bordering on Rhodesia and South Africa - this spirit of dissidence could not have become militarised and would simply have led to the classic phenomenon of the 'uncaptured peasantries' . For various sectors, peasant or otherwise, however, the introduction of a military structure in the country gave rise to the hope and in some cases to the reality of withdrawing from the modern state thanks to the protection of the guerrillas.31 So was it really a civil war? The debate about the 'civil war' has been obscured for moral reasons that FRELIMO has adroitly exploited in its effort to depoliticise and illegitimate RENAMO. Indeed, saying that the conflict was a true civil war meant recognising that South Africa was not the only cause, that FRELIMO was at least partially responsible, and that the problem was thus political. As a result, FRELIMO and its numerous intellectual sympathisers abroad qualified as 'pro-apartheid' all analyses stating that the war of aggression had undergone a transformation into a civil war (Cahen 1994c). This domestic war was clearly not the result of a popular revolt: it was a Rhodesian and South African war of aggression, relying on a minimum of local political realities from the beginning, which permitted the militarisation of a spirit of dissidence and the 'reactionary' will to withdraw from the modern state. This militarisation transformed the conflict into a truly civil war, albeit an atypical one. It is evidently impossible to define a precise date for the passage from a war of aggression to an atypical civil war. And in any case the moment would be different for different zones. Nevertheless, it seems clear to me that, at the national level, the process of transformation took place between 1982 and 1985, the period when the war extended to the entire country. The simple fact of stably occupying significant portions of the country (probably a bit less than 20 per cent of the surface area in general, though more in 1986-87) and of conducting guerrilla operations over a much larger area had two major implications for the guerrillas. On the one hand, this implied that they had to manage these territories by putting in place some form of civil administration, however limited it may have been: perhaps only a few hundred people, but of a fundamental importance today. On the other hand, they had to represent in one way or another, the aspirations of those that accepted them and, in some cases, openly supported TRANSFORMATION 35 (1998) 11 ARTICLE CAHEN them. Otherwise, RENAMO faced the prospect of rapidly degenerating into a simple bandit gang and of fragmenting into local rival groups. As far as the question of aid from Rhodesia and South Africa is concerned, we may ask, is this in itself a factor in denying that this was a civil war? A civil war necessarily has a popular aspect. Can one conduct a 'popular' war with the support of apartheid? Numerous authors have answered 'no' and have thus over-emphasised the 'international relations' aspect of their analysis. In reality, apartheid was a truly remote phenomenon for the people of Mozambique, while the Zimbabwean airplanes that were bombing their villages and dropping napalm were much more tangible. Nor must we forget that many considered South Africa to be a kind of paradise, the choice destination for both legal and clandestine emigration. Finally, numerous guerrillas never saw a single South African soldier in 16 years of war. The question of their alliance with South Africa was obviously a subject that I raised systematically with all of the RENAMO members that I met. None of them denied the fact, but none of them felt the slightest embarrassment either: they needed allies and FRELIMO was worse than apartheid. The relationship to the modern state was a determining factor for the people's position during the war and it is for this reason that we can talk of an authentic civil war. The war became a struggle of two populations: those remaining in the sphere of the state (the cities, communal villages, shops, etc) and those seeking to withdraw from the state by means of a severe and illusory autarchy. Herein lies another reason why the 'community' (and not the ethnic) aspect of the war was the most horrific. RENAMO, for example, had no compunction about killing civilians driving on the major highways linking the cities because their targets belonged to the 'state' population (and it was, moreover, extremely unlikely that those carrying out the order would encounter a relative, however distant). This is also the reason that the FRELIMO army did not hesitate to systematically kill the men found during raids in RENAMO territory. To paraphrase Ernst Bloch, this was a contemporary war, though 'not fought using contemporary forces' . The principal struggle was neither military nor economic (the destruction of infrastructure by RENAMO, for instance), but one for control of the population: 'taking population' by RENAMO was an essential objective targeting not only young men, but also women and the elderly. This explains FRELIMO operations (vasculhas) carried out in order to increase the population of recuperados placed in camps or in 'rural cities' (a later manifestation of the communal villages). These 'rural cities' were organised with the assistance of NGOs (notably Italian) who did not wish to see aid given to anyone other than the victims of 'bandits'. 12 TRANSFORMATION 35 (1998) CAHEN ARTICLE Considering the war as a conflict between two populations also explains the government's unmitigated hostility towards the International Committee of the Red Cross. The ICRC was in fact the only organisation that sought to help both populations from the very beginning. The military obviously seized a portion of the material aid that the ICRC managed to deliver to the RENAMO zones. The government, for its part, sought to take advantage of the slightest incident.32 During the first months following the cease fire, humanitarian aid was used as a weapon very successfully: due to the nature of the country's infrastructure, aid entered through the ports, the airports, and the major highways, all held by the government. Part of the aid disappeared at each step of the way, to such an extent that only a fraction finally made it to the RENAMO zones. The presence of aid in the government zones was an important factor provoking an exodus out of the RENAMO zones. RENAMO itself remained extremely suspicious of any entrance whatsoever into its territory and humanitarian organisations often had great difficulty in preventing the military organisation from controlling the aid.'' The battle for the two populations continued. There can be no doubt that the war was largely one fought against civilians. Aside from those who joined the fight voluntarily, the rest of the population found itself faced with afait accompli and the decision of whether to choose sides or to run.34 I am also convinced that the war was equally savage on both sides, even if the total domination of the media by FRELIMO for the 15 years of the war has led even those most desirous of remaining objective to attribute the majority of the atrocities to RENAMO. The people themselves were not duped: they attributed various acts of banditry and certain massacres to 'RENAMO 1', but others to 'RENAMO 2' - the euphemistic term for FRELIMO soldiers and militiamen acting on their own.35 Does the fact that this war was largely fought against civilians call into question the civil nature of the war? In my opinion, no. What people have ever chosen war? That this was a war against civilians highlights its atypical character, but to conclude from that that it was not a civil war is equivalent to saying that the present conflict in Yugoslavia is not one either. Following the armistice, the battle for the two populations continued in the form of a controversy over the 'two administrations'. The General Peace Accord did not specify the zones assigned to each party, though it did state that in RENAMO zones, the members of the state administration should be chosen 'only from those citizens residing in the zone and possibly belonging to RENAMO' . A national bipartite commission was to manage the relationship between the minister at the national level and the administration in the RENAMO zones.36 In theory, there was only one administration, but this principle was rendered TRANSFORMATION 35 (1998) 13 ARTICLE CAHEN meaningless by the criteria chosen for its application. After 15 years of civil war, anyone residing in a RENAMO zone was necessarily one of the organisation's supporters and the existence of a central commission showed that the local functioning of the RENAMO administration went on without horizontal communication with the FRELIMO administration. RENAMO claimed that there were indeed two administrations. A few of its civil administrators at the district or local level (for the pastas administrativos) were finally integrated into the state administration, but the process came late and turned out to be incomplete.37 Moreover, the administrators chosen for the state administration were to come from the principal city or village in a given area. In many districts, however, these remained in FRELIMO control, even if RENAMO ruled the surrounding bush. This administrative duality continued in many zones up until the elections (and may even exist today): the civil servant administrator from FRELIMO controls the principal town or village and RENAMO maintains control of the surrounding bush. RENAMO also had a foothold in zones that it never dominated militarily and that remained under state administration. This fact was dramatically illustrated during the campaign, then later by the results obtained, and proves without doubt RENAMO's ability to exercise political as well as military power. This brings us to a discussion of the social base of the former guerrilla movement. Maintaining the 'Coalition of Marginalises' What then was RENAMO's political situation when the war ended on 4 October, 1992? Thanks to the military dynamic, RENAMO had succeeded in establishing a 'coalition of marginalises'. By 'marginalises' I mean those social, cultural, and ethnic groups that were historically excluded from the modern state: the Portuguese colonial state and then later the independent nationalist state. These marginalities were diverse in character: the core of the old northern Creole elite (from the Zambesi Valley, from Mozambique island, and Ibo island) marginalised by the change of capital in 1903 and later silenced by single party rule in 1975 following a brief political opening from April to September, 1974; the populations of certain regions where illiteracy was close to 100 per cent up until the very end of colonisation and that have remained unaffected by the great social and economic progress of the 1960s and early-1970s; ethnic groups that were never well represented in the formation of FRELIMO nor later its government (Ndaus, Senas, Macuas); young people expelled from the cities in 1983 by 'Operation Production'; regulos3* and members of religious groups3* humiliated by the government; thousands of victims of reeducation camps, 14 TRANSFORMATION 35 (1998) CAHEN ARTICLE whose numbers have certainly been underestimated by all observers; peasants forcibly regrouped into communal villages, etc. All of these social strata conceived their situation differently and had ideals that were dissimilar and at times opposed (the young man expelled from his urban school environment who dreams of returning to the city is in stark contrast to the regulo). Only the unity imposed by military action against the state preserved the cohesion. Would this coalition be able to survive after the demilitarisation and demobilisation? Would RENAMO, the armed social movement40, be able to become RENAMO, the political party? Two conditions were indispensable: there had to be enough competent civilian party managers who could give shape to this nebulous protest movement and the post-conflict political context had to be favourable. The political context was indeed highly favourable to maintaining the coalition. Political pluralism was instituted in the beginning of 1990 and peace came in October, 1992. ButFRELIMO, despite abandoning its Stalinist-Marxist discourse, in no way modified its governmental practices. A completely exhausted cabinet was maintained, without modification, for four years. The fusion of the party and the state went on - illustrated to point of caricature by the FRELIMO electoral campaign led by the governors and state employees; the 'ethnic overtures' of the state apparatus were homeopathic and did not permit the government to modify its image of a southern-controlled state, foreign and non-representative to the majority of its citizens (administrators needing interpreters, for instance, as in the colonial days). As far as the problem of trained party cadres goes, during my stay I found a situation quite different from the one I had expected. I attended approximately 70 RENAMO meetings and I was able to interview the members of as many local leaderships. To my surprise, not only was it extremely rare to meet anyone from the military in the local leadership (even simple demobilised soldiers), but what's more, these civilian members were generally not recent recruits. Rather they were people who constituted the civilian wing of the RENAMO zones during the war: civil administrators, school teachers, nurses, members of the civil intelligence service. This was particularly clear in Upper . In my previous writings, I had underestimated the size of this civilian wing41 because it did not express itself politically. Unlike FRELIMO, which always totally dominated all that was publicly said or written about Mozambique, RENAMO, the 'unwritten' movement, produced hardly any written documents, the sustenance of any researcher!42 It is this historically important wing that is coming to the fore today to build the civil organisation and which is showing itself capable of bringing TRANSFORMATION 35 (1998) 15 ARTICLE CAHEN together the other components of the movement. These consist, on the one hand, of the formerly clandestine urban members of RENAMO (numerically few, but of considerable political importance) and, on the other hand, of the new recruits (since the Peace Accords) who can themselves be divided into two categories: those who hate FRELIMO for personal reasons related to the arbitrary nature of one-party rule (re-education camp survivors, regulos, peasants forced into communal villages, students discriminated against at school for ethnic reasons, etc) and those who are opportunists. Thus in spite of starting from a position of great political weakness, RENAMO has succeeded in doing what Angola's UNITA never accomplished: the creation of a civil organisation with an emerging network of thousands of local activists and hundreds of administrators to manage approximately one million members." Nevertheless, the dearth of experienced personnel, necessary to direct the movement's domestic and foreign policy, remains patently clear. As time goes on, some new leadership may emerge from the provinces. The two necessary conditions have thus existed not only for the coalition's survival, but also for its reinforcement against a state unrepresentative of the multi-ethnic population of Mozambique, a country without a nation.44 We must be careful, however, to avoid any static conception of this phenomenon that excludes the possibility of evolution. The 'coalition of marginalities', which, I believe, explains the political development of RENAMO, is unstable, and it is clear that it is not a coalition of all the marginalities of the country. Fearing such a vision Joao Gomes Cravinho has criticised the very idea of such a coalition: Michel Cahen (Savana, 11 April, 1994) describes RENAMO's big electoral success as a 'coalition of marginalities'. They would be the survivors of reeducation camps, demobilised soldiers, those uprooted and sent to communal villages, the victims of' Operation Production', the families of the latter as well as a whole collection of individuals who for one reason or another feel marginalised by FRELIMO. Together these various groups are responsible for the approximately 40 per cent of the vote that RENAMO received. The idea is at first seductive, but ends up being unconvincing. It is certain that the marginalised voted RENAMO. It is equally certain that they also voted FRELIMO. Thus if we wish to maintain Michel Cahen's idea (and it seems to me that, in any case, it could be useful), we are going to have to make it considerably more complex. We are going to have to refrain from looking for traits common to the marginal groups, instead, we must look at what is 16 TRANSFORMATION 35 (1998) CAHEN ARTICLE specific to each marginalised group. We are going to have to attempt to understand if the individuals identify themselves with that marginality and how this affected the election. We are going to have to cross one group with another to see what impact that has (for example, refugees returning from abroad, those within the country, and those holders of 'traditional power'). In conclusion, what I have just stated points to the necessity of combining the concept of marginality with that of 'location', in other words only the specific local conditions permit us to ascribe a meaning to the vote. I wish to say as well that we have in no way witnessed a popular revolt of the oppressed by means of the ballot box (emphasis added) (Cravinho 1994). My response to JC Gomes' critique is clear: I am totally in agreement with his remarks! Let me reiterate: this war and the resulting elections are not a popular revolt, but the affirmation of heterogeneous dissident tendencies. It is clearly necessary at times to go down to the 'micro-anthropological' level to understand why, in a given place, a given factor ensured the population's loyalty to FRELIMO, while elsewhere the same factor explains a massive dissent in favor of RENAMO. This must not mask the larger trends: although it is clear that there were those among the marginalised (and even among the marginal) who voted for FRELIMO, nevertheless, one can hardly call FRELIMO the 'party of marginalities'. FRELIMO was, after all, the creator of the independent state and the political vehicle of its Creole elite, a group that constitutes a minority, but not a marginality. And, although it is also clear that RENAMO is far from having joined together all such groups, I would argue that this marginalisation is one of the essential reasons that pushed a given professional (lawyer, engineer, and so forth), a given tribal leader, a given community to dissent militarised by a guerrilla group from the outside. The psychology of a marginalised citizen (and not a marginal citizen) is very revealing when one analyses his or her life-history, political behavior, search for dignity, etc (see below). Moreover, this necessary local level of analysis must not hide the realities which although certainly local, are also true on a larger scale: namely, regional or ethnic feelings that have not always, but sometimes very strongly, affected the vote. We will return to this point. The existence of 17 parties, the potential (and as yet weakly established reality) of the 'coalition of marginalised social groups' permitted by RENAMO, the localised dimension of the political life of the country, all this is precisely the expression of the economic, social, cultural, and intellectual fragmentation of the country. Outside of FRELIMO, where could one find any concentration of TRANSFORMATION 35 (1998) 17 ARTICLE CAHEN organic intellectuals of such a 'plebeian society', a society become a lumpenproletariat, and a beggar in the international community? I've already pointed it out in the beginning of this text: the necessities of the society are defined by the simple need to survive. 'Communism' and 'Identity' Thus we see that the factors - the existence of 17 political parties, the as yet weakly established coalition of marginalities, and the division of the political life of the country into smaller local elements - all express the economic, social, cultural, and intellectual fragmentation of the country. This state of fragmentation explains Afonso Dhlakama's credibility when he denounces, even today, the 'Marxist FRELIMO', the 'communist FRELIMO'. At first glance, this is a surprising criticism, given the context of 'laissez-faire' liberal capitalism. Dhlakama has evidently never read a word of Marx or Lenin. The words 'Marxism' and 'communism' refer in fact to the maintenance of a social group in power, one felt to be ethnically determined, from the radical-nationalist 'FRELIMO 1' to the liberal-entrepreneurial 'FRELIMO 2'. The arrogance of the FRELIMO elite, who have never conceived of losing the elections, who consider the country as their creation and their property, who scorn as illiterates the pro-RENAMO population, exacerbates that perception of the modern Mozambican state. As one militant said, 'Communism? That's everything that's bad'. The 'People' and the 'Masses' A local example may help to clarify this point. In the estuary of the Save River, which forms the border between the Inhambane and Sofala provinces and is the southernmost point of the regions peopled by Chona-Ndaus in contact with the Vatsuas, the towns of Machanga and Mambone have always remained under the control of the state administration despite an early political infiltration by RENAMO (as early as 1977). During the very difficult years of 1983-84, famine ravaged this zone that was already worn out by drought and cut off by war from any overland access to Beira or to Inhambane. Material aid arrived only rarely, but when it did arrive, the 'members' and the 'people' received two sacks of flour per week, while the simple 'masses' received only two liters a week. The 'people', a concrete political concept, consisted of the families of 'members' of the party and of the administration. The 'masses' consisted of the 'unorganised', the simple elementos da populagdo (elements of the population). It was not merely a question of the local nomenklatura's privileges either. Indeed, 95 per 18 TRANSFORMATION 35 (1998) CAHEN ARTICLE cent of the 'members' of the state-party were Vatsuas in that zone whose population was 95 per cent Ndau. Here the social cleavage reflected the local ethnic reality and the result was one of tribal division. It is hardly surprising that in that area of strong historical traditions (Cahen 1.99la) people never talked to me about 'FRELIMO activists' but about 'FRELIMO Vatsuas'. Thus was the perception, exaggerated, certainly, given that a few Ndaus were in fact members of FRELIMO, including the local administrator and a hero of the anti-Portuguese resistance. There was a complete ethnic split in the elections and RENAMO won with a crushing majority, but the story was completely different just a few hundred meters away, south of the river, once inside a Vatsua zone. For the 'masses', the elementos da populagao of Machanga and Mambone, what difference did the change from state ownership to liberalism really make? Isn't the former manager of the state saltworks (who had done everything possible to bring about the nationalisation of the Catholic Mission saltworks) now the private owner of his former company? The enemy stayed the same: 'communism'.45 Identity and Tribalism In addition to the social question, but still inextricably linked with it, there arises the question of identity. There is no doubt that RENAMO has aggressively sought to use the ethnic factor to its advantage. Nevertheless, it would seem to me unjust to qualify their campaign as 'tribalist' (at least in the pejorative sense potentially attributed to this word). The most powerful theme in the discourse of RENAMO's lider lay in the demand for a 'better division of power': perhaps 'political', but above all 'social', with governmental participation by all socio-ethnic groups and a concomitant sharing of the benefits. Dhlakama's most subversive remarks were not those made in Nampula when he promised them a Nampulense governor nor those made in the rural zones of Nametil/Mogovolas when he said that the local administrator would be Macua.46 Dhlakama's most subversive rhetoric was much simpler: he spoke very rarely of 'Mozambicans', and he did not begin his speeches with 'Mozambicans!', but rather with 'MacuasT or 'Muanes!' and so forth. In other words he called people by their ethnic identity, by what had been barred from the nationalist discourse for years. I myself was astonished by the effectiveness of this tactic. Although anthropologists and linguists have always identified a large Macua-Lomue41 ethno-linguistic group it was far from certain that in 1974 there really existed a Macua ethnic group, since ethnic consciousness was very weak at the level of the entire group, identification being much stronger on a smaller scale: the clan or lineage. Moreover, Macua is not the name that members of the group give TRANSFORMATION 35 (1998) 19 ARTICLE CAHEN themselves, but rather an unflattering 'hetero-denomination' that inhabitants of the coast give those living inland. It is a name of the sort: 'those who go naked', 'those who eat rats', and so forth. The effectiveness of this ethnic appeal in the Macua areas remains to be verified. In all the areas I visited, including the coast, the tactic did indeed seem to work. The concept of Macua ethnicity was very likely created by none other than FRELIMO itself, by its de facto disdain (albeit denied) for those groups outside of the sphere that it historically controlled. Dhlakama thus systematically made use of the ethnic question, though, as I said, not in a way that I would call tribalist. Often, he remained more moderate than his grass-roots supporters. In Chimoio, when speakers from the audience denounced 'Maputo which is talcing all our corn', Dhlakama responded, 7 live in Maputo, I'm not going to defend Maputo, but I am a nationalist leader. The truth is that in Maputo the people have nothing, absolutely nothing. It's the ministers who eat well' - and received an ovation. Whether this was real or simply formal is not the question for the moment. What is important to note is that Dhlakama always positioned himself for a better or more just division of power; that is to say, he criticised the tribalism of a FRELIMO accused of exclusively favoring the southern clans. In attacking FRELIMO, he mimicked their own anti-tribalist discourse, with unquestionable success. 'Filho desta terra'... True, but Which Land? At times, the situation on the campaign trail made using such themes of identity very complicated. In Cabo Delgado, for example, RENAMO had named Vicente Ululu as the head of the list because he was a native of that province, 'umfilho desta terra' ('a son of the land1). Ululu is the secretary general of the party, whom Dhlakama always introduced as the 'number 2 man' in the party.48 In the meetings held in this northern province, however, there was a very palpable aati-planalto sentiment: the local delegates accused FRELIMO of having development projects only for 'the highlands' in order to make 'the coast' dependent. This was clearly a geographical euphemism used by the Macuas, Muanes, and Suailis speakers to denounce the Macondes - inhabitants of the highlands and founders of FRELIMO. Vicente Ululu is indeed a native of Cabo Delgado, but he is also a Maconde. RENAMO believed that it could destroy the FRELIMO myth of the 'old liberated zones' of the Portuguese war era in which the Macondes spearheaded the independence movement by putting a Maconde at the head of its own list and thereby encroaching on FRELIMO's 'turf. However, Cabo Delgado is a province whose population is more than 70 per cent Macua and in which the Macondes have hoarded the majority of the appointed 20 TRANSFORMATION 35 (1998) CAHEN ARTICLE positions thanks to their political power. In thus reviving the feelings of resentment dating from the pre-colonial period by putting &Maconde at the head of the list (even if the vast majority of other candidates were indeed Macuas, Muanes, or Suailis), RENAMO created more problems for itself than it solved. During the major campaign meeting in Pemba, the provincial capital on 13 October, 1994, Dhlakama let Ululu address the audience first since he was heading up the local list. In front of a Macua audience, however, Ululu could clearly not deliver his speech in Ximaconde, but he didn't know how to speak Emacua, either. He ended up speaking in Portuguese, a solution that did not please the audience. (He could have spoken Kisuaili, which is a lingua franca, but apparently he did not think of it.)49 Being a Maconde made his campaign no easier on the planalto, either. As for his village of Muidumbe-Nhangololo, in the south of the plateau, he had left years earlier and although he was accepted as an individual at the time of his return, he was not at all accepted as a leader of a party whose guerrillas had, upon arrival in 1986, stolen all the inhabitants' pigs.50 This was the only overtly hostile meeting that I attended, during which it was not the small groups of young people organised by FRELIMO who were throwing stones, but rather the population itself. Dhlakama came out of it profoundly affected, and Vicente Ululu was not retained in Cabo Delgado to continue the campaign as had been planned. Maputo, the capital, presented RENAMO with another ethnic problem. Overall, RENAMO realised that the south constituted essentially hostile territory, even if they did not foresee their electoral disaster in the four southern provinces. RENAMO organised only one public event in the south: a parade going through most of the urban and suburban parts of Maputo on 22 October. It was a fiasco: the parade attracted almost no one, save onlookers standing on the sidewalks yelling 'Nos queremos presentes, queremos camisetes, capulanas'. Above all, the event did not bring the demonstration of support hoped for. In addition, the Mozambican journalists accompanying me (by no means necessarily supporters of FRELIMO) pointed out to me that when the RENAMO activists spoke either Portuguese orXironga their accent immediately identified them as coming from the north, even if they were born in Maputo or resided there. Even in the capital, with its more cosmopolitan tradition, RENAMO apparently could not break through the ethnic barrier of the southern groups. Knowing that the best it could hope for in the south was a weak showing, RENAMO chose an Indian notable, A Jaffar, to head up the list and chose General Herm'nio Morais, a mulatto native of Maputo, as second on the list.51 The result was less than modest, with Jaffar being the sole RENAMO candidate elected and FRELIMO picking the other 17 seats. Nevertheless, this blow was not due solely TRANSFORMATION 35 (1998) 21 ARTICLE CAHEN to ethnic reasons, which FRELIMO did aggressively exploit. It also stemmed from RENAMO's incapacity to find concrete, palpable, and popular themes for its southern campaign, and its only specifically urban campaign. One thing, nevertheless, is certain: even if RENAMO does not call into question the role of Portuguese as the language of national unity, Afonso Dhlakama has always delivered his speeches in at least two languages: Portuguese and the local language. Not only was the audience warmer when he spoke in a Bantu language, as we might expect, but his speech itself was stronger, more colorful, and even more poetic.52 Even if Dhlakama has no problem 'speaking' in Portuguese, he is clearly more at ease 'speechmaking' in a Bantu language. His choice of language was thus not a demagogical ploy, but rather an obvious pleasure. FRELIMO's Odd Brand of anti-Tribalism We must not be led to believe that RENAMO was alone in making use of the question of identity. FRELIMO, despite its radical anti-tribalist tradition, has made extensive use of this theme. There again, FRELIMO's direct or indirect domination of the media made itself felt: when the initial polls were published, the newspapers described the pro-RENAMO tendency in the centre and north as an 'ethnic vote' while the pro-FRELIMO vote in the south remained purely 'national'. Before the campaign, FRELIMO's secret service had played upon the Sena-Ndau dispute about the language used in Catholic mass. The Episcopal Summit of Mozambique had decided that in the provincial capitals priests would celebrate mass only in the language that was traditionally the most widely spoken in the area. In Beira there was no doubt that historically speaking it had always been Cindau. Nowadays, however, Sena immigration to Beira has been so massive that Cisena is as widely spoken. Archbishop Jaime Gonc,alves, himself a Ndau who played an important role in the first contacts with RENAMO, declared that mass would be said in Cindau. Although we may question the wisdom of the Episcopal Summit's decision, in practice it posed no problem as the majority of the Sena immigrants spoke Cindau. Following this decision, however, churchgoers who had never been seen before started coming to church with what seemed to be less than entirely spiritual motives, demanding that Cisena be used for services. They provoked sufficient agitation that the archbishop was forced to reconsider his decision and accept both languages. At the same time, FRELIMO supported at arms' length the emergence of an organisation named 'Sotemaza', which despite its simple geographic designation, ('Sotemaza' takes the first two letters of the provinces of Sofala, 22 TRANSFORMATION 35 (1998) CAHEN ARTICLE Tete, Manica, and Zambezia), is in fact an exclusively Sena and almost explicitly anti-Ndau organisation. Armando Guebuza is said to have had an important role as an advisor for this association. Sotemaza was able to rely on the linguistic quarrel fomented by FRELIMO's secret service in Beira as well as the realities of a war in which, according to several sources, RENAMO had been more violent towards the Senas than towards the Ndaus. Nevertheless, one cannot help but note the failure of this particular FRELIMO policy of ethnic provocation in the centre of the country since a majority of Senas as well as of Ndaus, Monicas, and Niungues voted RENAMO. All evidence suggests that the problem was indeed one of identity, but more a question of regional than ethnic identification: all of the central region was opposed to FRELIMO. The situation was similar in the four southern provinces of the country, but there the FRELIMO discourse, with its tribalist overtones was clearly successful. The reasons for RENAMO's disaster in that region of the country go back to FRELIMO's history, the history of the civil war, and the events of the two years since the cease-fire. To cite only the most immediate causes of this phenomenon, let us note that the southern populations have felt under attack for two years from other ethnic identities reproaching them for their takeover of the state. Even at the lower end of the social scale, one knows perfectly well that RENAMO 'comes from the north' (an immense north that begins at the Save River). Furthermore, RENAMO engenders a certain religious fear.53 This pushed the southern populations to be loyal towards FRELIMO as a defensive reflex. FRELIMO exploited these fears by setting off a massive disinformation campaign 15 days before the vote. What happened? At a meeting at Pemba, Cabo Delgado on 13 October, Afonso Dhlakama was eager to respond to a preceding FRELIMO disinformation campaign having to do with the Macondes. RENAMO was apparently going to 'expel them to Angola' in the event that it won. Dhlakama responded as follows: This isfalse. The one who will be expelled is Chissano himself and since he is Changane and since the Changanes are of South African origin, he'll be sent to South Africa (laughter) (Cahen 1994d). Chissano's 'expulsion' obviously referred to his hoped for electoral defeat. But that very evening, Maputo television reported these words, immediately developing its own interpretation: all Changanes were going to be expelled. The daily paper, Noticias, went even further and included an unsigned leading article dedicated to the issue. Although, there are only 40 000 television sets in Mozambique and only some 10 000 people read the daily papers, that suffices to diffuse such boatos (rumours). The national leaders of FRELIMO, when TRANSFORMATION 35 (1998) 23 ARTICLE CAHEN questioned by journalists, responded that Dhlakama had indeed said this, if only by inference (Chissano is Changane, Chissano will be expelled, thus all Changanes will be expelled). The local FRELIMO leadership did not even bother to mention that Dhlakama's statement was by 'inference'. I have written elsewhere, however, why RENAMO's disaster in the south cannot be explained by this one FRELIMO campaign (Cahen 1994e). Let us simply conclude here that from the Sena affair in Beira to Sotemaza to the 'expulsion' of Macondes and of Changanes, the 'anti-tribalist' FRELIMO certainly did not hesitate to provoke ethnic hostility in its quest for votes. Demagogy and the 'Desire for a Master' All the parties led very demagogical campaigns, but RENAMO's campaign had an aspect that was entirely disarming. During his speeches, for example, the former rebel lider said: 'In the hospitals there is no medicine! FRELIMO says that RENAMO has stolen everything, has burned everything!' Then, anxious to show that the people of the south had also suffered, he immediately added: 'My friends, I live in Maputo. In Maputo, there was no war. It's not RENAMO who has been stealing, but in the central hospital of Maputo, my friends, there is nothing! No sheets! No towels! Mothers lying on the ground with their little ones just born! My friends, what misery!' Finally, he concluded: 'RENAMO has solutions! RENAMO is going to get sheets, is going to buy medicine, is going to hire good nurses, good doctors, is going to raise their salaries!' In the immense meeting held at the end of the campaign on 24 October in the Ferroviario de Manga Stadium near Beira,54 Dhlakama said that it was dishonest to promise too much and then added, 'I'm going to promise just one thing: Dhlakama in power means an end to unemployment, an end to misery'. One is forced to recognise, however, that as far as demagogy is concerned - and here I come back to the remarks by JG Cravinho cited above - RENAMO was not politically relevant, especially in the south. Although the party could not play on the regional or ethnic issues there, it could have taken advantage of the social issues, notably in denouncing the rapid accumulation of wealth by FRELIMO ministers. In Gaza, the head of the FRELIMO list was Armando Guebuza, the man responsible for the forced migration to communal villages in the Limpopo valley in 1978 where the intense conflicts over land have never subsided and for 'Operation Production', that sent thousands of people from the south to their deaths. Despite these opportunities, however, RENAMO did no campaigning around social issues. This illustrates not only the extent to which RENAMO was foreign to the urban world, but also its ignorance of the recent 24 TRANSFORMATION 35 (1998) CAHEN ARTICLE history of numerous Mozambican communities. This brings us back to the history of the war and to the history of the rebellion. The history of the rebellion is the history of a group completely centralised around its leader. When its first commander, Andre Matsangaissa, died on 17 October, 1979, RENAMO almost failed to withstand the blow. This hyper-concentration of power in the RENAMO command could not avoid creating problems during RENAMO's process of 'civilisation'. It also raises questions about the current nature of the movement. Indeed, in many meetings the popular affection for RENAMO went far beyond simple curiosity or political sympathy. This affection expressed what Christian Geffray, studying paternalism in Brazil, has called the 'desire for a master'.55 To a certain extent comparable to the ego's desire for a father, the desire for a master that 'We' (as a collective group) experience relates to the social need for a dominant figure whom we fear, but can love. This figure is opposed to the capitalist who exploits the free labour market, a capitalist whose imaginary archetype was the Jew, a dominant figure who is not a Master. Clearly this desire is in no way democratic, but it is perfectly understandable in the context of a society whose extreme disintegration recalls the situation at the worst moments of the slave trade. This desire for a master, combined with the defense of 'traditional power' and of the authority of the family,56 with the promotion of 'African authenticity', and with crowds mechanically raising their hands to respond to the cheers, does all this define the beginnings of a black fascism? No, for two reasons. First of all, the atmosphere has always been one of joyous disorder, and laughing families instead of the iron discipline of the Nazis, Mussolini, or even Salazar. And above all, the people came to hear a defense of their ethnic identity, to support their own traditions, but they also came to support the call for democracy, for justice, for liberty - all highly progressive themes. There is always the suspicion that RENAMO is insincere about these themes, but that does not alter in the slightest the fact that people showed up to voice their own progressive aspirations. RENAMO thus comes forth as a great conservative populist party, in a what may be called an incredible reversal of situation. At the risk of oversimplifying, one may say that FRELIMO has to a greater and greater extent appeared as the party representing the most modern and most capitalist sectors of the Mozambican bourgeoisie and RENAMO as the party of the poor, of the have-nots?57 RENAMO's Structural Weaknesses RENAMO's programmatic weakness was not a central problem in the campaign. The movement published a small pamphlet that contained what the embassies wished it to contain. The disproportion of material resources presented TRANSFORMATION 35 (1998) 25 ARTICLE CAHEN a much more serious problem. RENAMO had at its disposal, towards the end of the campaign, money coming from the Trust Fund managed by ONUMOZ and probably from private sources as well;58 but it was all highly concentrated on Dhlakama's personal campaign59 and was, in any event, considerably less than what FRELIMO had. For 15 million inhabitants, FRELIMO ordered six million t-shirts, millions of capulanas (wraps), baseball caps, and thousands of bicycles. Dhlakama tried with some success to make his own lack of means an issue in the campaign. He called on his supporters to ask for five shirts when offered one, 'since it was the people's money that FRELIMO stole for 20 years' .m Upon his arrival in Nampula, tens of thousands of people yelled, 'Sent camisetejdganhou! Sent camisete, jd ganhou!' {'Without a t-shirt, he's already won!'). The effectiveness of such handouts is not direct, but it is significant: people don't vote FRELIMO because it gives them a wrap (though in cases of extreme misery the gift itself may be important), but by ostentatiously spending money, FRELIMO emphasises that it's still in charge. And naturally it is best to be on the side of the leader, especially when his administrators cruise through the bush in cars marked 'Estado' and don't hesitate to threaten 'bad' voters. RENAMO was certainly capable of the same attitude, but it didn't have the state. On the other hand, it did know how to express its discontent effectively: during the September meeting of its National Council, a majority declared itself in favour of abandoning the electoral process so long as the money (from the Trust Fund) had not arrived while a minority (generally provincial leaders with a strong political base) argued for continuing the campaign, whatever the cost, even doing it on foot. Dhlakama had to personally (and theatrically) 'overrule' the decision of his National Council in order for RENAMO to come back into the process. ONUMOZ was anxious enough to speed up delivery of RENAMO's funds. To my knowledge, even if these two positions did in fact co-exist within the National Council and reflected different activist practices (leaders from the provinces as opposed to advisors in Maputo), there was never a majority in favor of a complete break with the process (which, in any case, did not mean a return to the war). It was all for show. RENAMO's poor campaigning, of which we have already seen several aspects, was, in my opinion, RENAMO's greatest frailty and it comes in large part from the highly centralised and personalised aspect of the party. Maintaining a Militaristic Mentality I have stated before, and still maintain, that RENAMO accomplished what Angola's UNITA never succeeded in doing: its demilitarisation and its 'civilisation'. The organisation would not have succeeded if it had had to recruit 26 TRANSFORMATION 35 (1998) CAHEN ARTICLE its civilian membership from scratch; a core group was indispensable and RENAMO had that. It is important, however, to avoid having an idealised image of what was admittedly a successful process of demilitarisation. It was not a well-oiled and smoothly executed plan. On the contrary, studies in progress61 show that as far as the demobilisation went, the RENAMO leadership was often playing it by ear. They frequently halted the process and found themselves forced to go through with it because of dangerously mounting pressure from their soldiers eager to go home and not wishing to enlist in the new army (numerous former guerrillas integrated into the new army had already deserted or had simply not shown up). Undoubtedly well advised by the ONUMOZ, Dhlakama saw his ability to command called into question and decided to let go. RENAMO ultimately demobilised more soldiers and integrated into the new army fewer soldiers than originally planned. Despite the fact that RENAMO's beginnings were almost exclusively military, it nevertheless fought an indigents' war with light arms62 where the military hierarchy, although authoritarian, was socially much more fragile than its UNITA equivalent. In addition, the RENAMO command was aware of this fact. This pushed the RENAMO leadership to understand that its survival could only be political, that it depended on the party's 'civilisation', and that going back to war was unthinkable. Although RENAMO's membership may be overwhelmingly civilian now, the organisation still continues to function with an exclusively military mentality. With the exception of some of the few competent provincial or national leaders, no one takes even the slightest initiative and everyone remains waiting for orders, even concerning such trivial details as what kind of drinks and sandwiches to provide for reporters. The consequences are disastrous and affect all levels of the organisation. RENAMO may be a civilian party, but it is not an exaggeration to say that the political department (headed by Raul Domingos) does not function at all. Moreover, there was almost no general RENAMO campaign, there was an almost exclusively Dhlakama campaign. This engendered increasingly greater problems and induced a loss of momentum from Tete onward, despite the final triumph at Beira. The party did little organising before or after visits by Dhlakama, unlike FRELIMO which did considerable campaign work before and after each appearance by Chissano. As mentioned before, the party secretary Vicente Ululu conducted almost no campaign in Cabo Delgado. Raul Domingos, who headed the list in the strategic province of Sofala, was in the United States for most of the campaign and in any case, almost never spoke, not even during TRANSFORMATION 35 (1998) 27 ARTICLE CAHEN the demonstration at Ferroviario da Manga in Beira!63 This was certainly linked to internal tensions that remain to be elucidated. To a large extent, Dhlakama was a man alone, surrounded by party staff who repeatedly told him that that situation was excellent and that only through a massive fraud could he lose. Despite astonishing energy, he could not do everything, nor react to everything as quickly as required, nor always interpret the facts of the campaign with the circumspection necessary. On the Edge of the Abyss Dhlakama's difficulties notwithstanding, the electoral campaign could not have begun better. RENAMO was truly surprised by the crowds that rushed to see them when they passed through; crowds whose numbers were ridiculously underestimated by the daily press, radio, and television. One can obviously come to a meeting out of curiosity and the evidence shows that this played a role. Unlike FRELIMO, who hired the best groups in the country, RENAMO had no orchestras; they had no parachuting shows, no t-shirts, and no beer at 1 500 meticais (the normal cost being 5 000). RENAMO activists tended to cite their poverty as proof that people attended their events for the organisation itself. There was, of course, an element of hype in all this: 'Dhlakama' the guerrilla leader had finally left the bush and - hero or villain - people wanted to see him. Nevertheless, the curiosity factor does not explain everything: if in the same location, a few days apart, there are 20 000 people for Dhlakama and 4 000 for Chissano, as happened, for example, in Nametil-Mogovolas in the middle of a rural zone, there must be a reason for the difference." People were truly lending their support to the organisation and expressing their 'desire for a master'. The polls conducted by the Brazilian firm Vox Populi predicted RENAMO as the sure winner in Sofala and in Zambezia , though beaten in Nampula, the heart of the Macuane. I can attest to the fact that the greeting in Macua territory was, however, much warmer (except in a few places such as Lalaua and the important city of Angoche). The Ifder's arrival in Nampula on the 2 October was certainly greeted by at least 100 000 Nampulenses - all estimates are wrong, of course, but all agree that there were as many people that day as there were when the Pope came in 1990. After his success in Nampula, Afonso Dhlakama began to think that he had a chance of winning, and he was not necessarily wrong. Something was happening in the country: the Uder of RENAMO had truly succeeded in changing his image. Certain segments of the population saw RENAMO as a vehicle to express their desire for change. Of course, all the leaders of RENAMO always said that they would win. Dhlakama had been careful, however, to specify that RENAMO had 28 TRANSFORMATION 35 (1998) CAHEN ARTICLE 'already' won on the essential point, that is the establishment of a pluralist system. As such, he provided for an honourable electoral defeat and offered his guerrillas and activists legitimacy and meaning to their struggle; in a word, a certain dignity to assure their post-electoral stability. The tangible sign that Dhlakama personally thought a victory possible appeared when he began to discuss regularly in all public meetings, even in the smallest and most remote parts of the country, the question of state security services and of civil servants in general, a theme totally absent in the first part of the campaign. He attacked FRELIMO for having fired all of the civil servants (police officers, members of the military, administrators, etc) of the colonial state in 1975 and 1976, and he affirmed that RENAMO would not make the same mistake. Of course, it would be necessary to 'slightly modify' the behavior of the police and the secret service and to professionalise them. Switching to a more demagogic tone, he proposed providing them with new cars and radio equipment and increasing their salaries in order that they don't shake down the people(!). He also mentioned that nurses, doctors, and teachers would not find themselves obliged to join RENAMO. It was clear that Dhlakama began to think about how to manage a RENAMO presidency with a security apparatus, and more generally, a state apparatus, created by FRELIMO. Delighted by the excellent welcome in Zambezia and in Nampula, by an encouraging welcome in the Macua, Muane, and Suaili Cabo Delgado, Dhlakama made a significant mistake: he reasoned in terms of demography. And demographically, however, it is clear that the north largely dominates the country: Electoral registration: the population and number of parliamentary seats, by province Provinces Voters % Seats Niassa 282,513 4.59 11 Cabo Delgado 568,169 9.24 22 Nampula 1,365,796 22.21 54 Zambezia 1,270,098 20.65 49 Tete 397,260 6.46 15 Manioa 322,201 5.24 13 Sofala 530,066 8.62 21 Inhambane 471,524 7.66 18 Gaza 398,381 6.47 16 Maputo Province 330,887 5.38 13 Maputo City 459,166 7.46 18 Total 6,148,842 100.00 250 Source: National Electoral Commission TRANSFORMATION 35 (1998) 29 ARTICLE CAHEN Two out of 11 provinces, Zambezia and Nampula, make up 103 of the 250 representatives. Since the vast majority of Cabo Delgado and 60 per cent of Niassa were Macua, a group that has never been sympathetic to FRELIMO, a victory was by no means out of the question if RENAMO was equally successful in Tete, and of course in Manica and in Sofala, the birthplace of the rebellion. This vision of the election assumed, however, not only that ethnicity would influence the vote, but that there would be an ethnic vote. In this case, not only could RENAMO win, but by a substantial margin. The fact that ethnic identity constituted an influential factor for reasons not only cultural, but also linked to the relationship with the state did not in itself imply that a given ethnic group would vote en masse for a given party; particularly when that group was demographically significant and geographically extended. Especially in the case of the Macuas, ethnic affirmative on a global level - in my opinion, beyond dispute in this electoral process - did not preclude other more local affiliations of clan and of lineage (Eratis, Metos, Macas, etc). One level of ethnicity does not necessarily exclude others. Moreover, the local history of the war had a very strong influence in the elections: in the region of Mocuba (Zambezia), rocked by the massive aerial bombardments, FRELIMO had no chance of winning, but in the case of Lalaua (Nampula), where RENAMO had allegedly committed a massacre? In his defense of ethnic dignity and identity, Dhlakama probably intended to provoke an ethnic vote. We will see that although this was what happened in the centre of the country to his advantage, in reality the Macuas split into two groups. As far as the south was concerned, there was a pro-FRELIMO vote so radicalised that it can be labeled as 'tribalist' (ethnically politicised to save the power of the state). RENAMO's legendary disorganisation aggravated their error of prognosis: Dhlakama began almost all of his meetings with considerable delays. In some cases these delays reached up to five days, as in Lalalua and in Mecuburi on 9 October, where the local population ended up waiting three days for him. This seemed to indicate a real lack of respect for the electorate. Because of these delays, he skipped a number of towns and villages altogether, notably in Niassa and Tete. As a result of being constantly rushed, Dhlakama could not spend the time necessary to listen to the local choirs, admire the dances, and talk to the iocal notables. (These would be: in the first row the local political leadership, followed by the regulos, in Muslim regions, the cheiks, and finally the land chiefs). Above all, Dhlakama's own team misled him into neglecting the south. He knew that it was a lost cause in Gaza. He said, and I quote, that a 'strong fight' would be necessary in Inhambane, Maputo Province, and Maputo city to obtain 30 TRANSFORMATION 35 (1998) CAHEN ARTICLE a respectable result. A poll conducted by the Portuguese marketing firm, hired by RENAMO, gave him a score of 25 per cent in the city of Maputo. Dhlakama knew that he was not popular among the Changanes, but he was convinced that he was, I quote, 'very popular among the Chopes and the Bitongas' and that he had 'a certain following' among the Vatsuas and the Rongas.65 It is true that RENAMO had large military bases in the south, but hardly any real 'liberated zones', only small pockets. RENAMO never managed to construct a true 'coalition of marginalities' in this region, remaining instead a marginal group. Once demilitarised, RENAMO found itself 'naked', without the minimal civilian base necessary to organise a real party. It is true that FRELIMO exercised considerable pressure as a political party, as a police force, but also as a society that was overwhelmingly hostile to RENAMO with the exception of a few chiefdoms, notably around Jangamo in the province of Inhambane, and in Mambone on the Save River. The southern party leadership, because they were residents of Maputo or the surrounding areas, had an influence over Dhlakama that was disproportionately strong given their real political base. They assured him that there was no need for him to visit, that they were 'taking care of everything'; in a word, they hid the real situation from their leader. Dhlakama, however, never sensed the danger. He saw his popularity confirmed to him by the great success of the meetings in the province of Manica at Gondola, Vila de Manica, Chimoio (19 to 20 October). The Inhambane meeting of 21 October was a fiasco (1 300 people in total if one includes both those waiting at the airport and those who came to the public meeting), but this failure barely affected Dhlakama. The parade in Maputo, held the following day, was a complete political failure. In the midst of his congratulatory advisors, however, Dhlakama did not understand this: he was genuinely shocked when I pointed out to him at the end of the parade that the popular support had been 'very modest'. Dhlakama was losing his grip on the situation. Finally, the campaign ended on a triumphant note at Beira, the second largest city in the country, on 23 October: the crowd in rags, running barefoot, filling six lanes of the highway for the ten kilometers separating the airport from the city: there were perhaps 150 000 people, and 60 000 more at the closing meeting, united at Ferroviario da Manga in a state of indescribable enthusiasm. Dhlakama managed to delude himself completely. He left that very evening, as though on a presidential voyage, to meet with the heads of state at the Summit of Front Line States in Harare. Two days later, he came back unrecognisable. Although the detailed history of the meeting remains to be written, this much is known: Dhlakama suffered the humiliation of waiting an entire day to be TRANSFORMATION 35 (19^8) 31 ARTICLE CAHEN received. In the end, he did not meet with the Summit, but instead he met personally with Quett Masire (Botswana) and Robert Mugabe (Zimbabwe). Although he never stopped repeating that there was a massive fraud being prepared,66 the African leaders pressured him to abandon that contention, without, however, trying to reassure him (Robert Mugabe apparently said that, 'indeed, fraud was not acceptable')- Dhlakama, however, had an 'Angolan' obsession: not to be another Jonas Savimbi, not to denounce the elections 'after' the vote. He always held - and he did not change his position - that if there were to be a denunciation it would be before the elections. It was indeed 'before', but so last-minute that for all intents and purposes the denunciation came 'during' the elections. It is not particularly important to document the critical moments, hour by hour, leading up to the call for a boycott. In short, Dhlakama remained unreachable in Beira all day on the 26th, despite the fact that the elections began the following day. On the evening of the 26th, a majority of the opposition parties met to demand additional guarantees, without precisely calling for a boycott (even if RENAMO in Maputo tried to make it seem that way). The fact that Dhlakama remained in Beira left the situation uncertain: an active boycott implied a solemn statement from Dhlakama in Maputo. The next day, several RENAMO leaders sent a message that 'RENAMO is not participating in a process sullied by fraud' and demanded a postponement of 15 days. But the elections had already begun. The official radio station started playing up RENAMO's call for a boycott. What did this final loss of control mean? It was obviously a disastrous political error: if RENAMO boycotted, FRELIMO would not have needed fraud to win and the international community would have had to declare the elections 'free and fair'. In my opinion, this error comes back to the great weakness of the RENAMO leadership mentioned before. Surrounded by party members, especially those from the south with the weakest political base but who had profited the most from the process and were panicked at the idea that it would end,67 Dhlakama found himself repeatedly listening to their sanguine reports that only fraud could deny RENAMO victory. In the end, the same man, who in a closed meeting in September, had labeled as stupid anyone predicting victory lost contact with reality. Obsessed by the case of Jonas Savimbi, humiliated by the Harare summit and undoubtedly seeing the summit as part of an international plot, he became radicalised to the point of denouncing the process. It was an incomplete radicalisation, however, since anyone who understands the internal workings of RENAMO knows very well that even his declaration 'not to participate in the vote' would not be taken by his observers in the voting booths as a real order to leave. In order to really pack up and leave, his activists 32 TRANSFORMATION 35 (1998) CAHEN ARTICLE would have had to receive a literal order. In fact, there were only rare cases (in Maputo, in Nampula, a few examples in Zambe'zia) where the boycott was observed. Moreover, as a mere result of simple material difficulties, the rumor of a boycott did not reach many places until the second day of the vote, and by then, the vast majority of the electorate had already voted. By mid-day on the 27th, Dhlakama knew that he had made a serious mistake, but he also had to save face. Long negotiations with a close colleague of Aldo Aiello resulted in a solemn declaration from all the ambassadors of the CSCE (Electoral Supervision and Control Commission) asserting that if they detected fraud, they -would not accept the elections. There was no fundamental change, except a recognition that the 'National Electoral Commission had not succeeded in satisfying RENAMO' as far as the guarantees it sought, but Dhlakama saved face, voting himself on 28 October and obtaining the extension of the elections for a third day. His hedging probably cost RENAMO very few votes, but his own international image did suffer. He appeared unstable and extremely susceptible. In fact, it was above all RENAMO that demonstrated its political weakness. The publication of the initial results exacerbated this problem of image all the more when RENAMO began saying that it would not accept them because of fraud. The government-allied weekly Domingo published a 'confidential document' on 13 November: the minutes of a meeting of RENAMO leadership deciding on the revitalisation of the military structure. The ONUMOZ and the ambassadors of the CSCE, of which several countries (the United States, Germany, Portugal, Great Britain) were included as supporters of this 'revitalisation', reacted very firmly, thus enabling RENAMO to appear as moderate and concerned about stabilising peace in the country.68 The National Electoral Commission had great difficulty establishing definitive results for the elections, after the breakdown and validation of numerous null votes69 and the examination of the claim. Brazao Mazula finally declared that he did not doubt the 'calm, clear, and fair' elections, theoretically speaking for himself as well as for RENAMO members on the CNE, since the commission functioned only on the basis of unanimous votes. A few hours later, Afonso Dhlakama repeated that the elections were not fair, but that he knew that they were the 'only elections possible' and that he accepted them as the 'basis for the exercise of democracy'. TRANSFORMATION 35 (1998) 33 ARTICLE CAHEN Elections of deputies by party and by region Provinces FRELIMO RENAMO UD Niassa(H) 7 4 - Cabo Delgado (22) 15 6 1 Nampula (54) 20 32 2 Zambezia (49) 18 29 2 Tote (15) 5 9 1 Monica (13) 4 9 - Sofala (21) 3 18 - lhambane(18) 13 3 2 Gaza (16) 15 - 1 Maputo Province (13) 12 1 - Maputo Crty (18) 17 1 - MOZAMBIQUE (250) 129 112 9 Results of the presidential elections held 27,28,29 of October, 1994 (candidates listed In the order of appearance on the ballot) Eligible voters 6,148,842 100.00% Abstentions 774,587 12.60% Number voting 5,402,940 87.87% Number of ballots cast 5,405,836 (100.47%)70 Blank votes 312,143 5.78% Null votes 149,282 2.76% Votes accepted 4,941,515 100.00% Maximo D Jose Dias 115,442 2.34% Carlos Jose M Jeque 34,588 0.70% Casimlro M Nhamitambo 32,036 0.65% Mario Carlos Machele 24,238 0.49% Carlos A dos Reis 120,708 2.44% Afonso MM Dhlakama 1,666,965 33.73% Jacob NS Sibindy 51,070 1.03% Padimbe M Kamati A 24,208 0.49% Domingos AM Arouea 37,76 70.76% Vasco Campira MA 58,848 1.19% WehiaM. Ripua 141,905 2.87% Joaquim A. Chissano 2,633,740 53.30% 34 TRANSFORMATION 35 (1998) CAHEN ARTICLE Results of the legislative elections held 27,28,29 October, 1994 (party names listed in the order of appearance on the ballot) Eligible voters 6,148,84 2100.00% Abstentions 773,583 12.58% Number voting 5,404,199 87.89% Number of ballots cast 5,405,611 (100.47%) " Blank votes 457,382 8.46% Null votes 173,592 3.21% Votes accepted 4,773,225 100.00% AP9 3,031 1.95% UNAMO 34,809 0.73% FT 26,96 1 0.56% FUMO/PCD 66,527 1.39% FRELIMO 2,115,793 44.33% SOL 79,622 1 .67% PIMO 58,590 1.23% RENAMO 1,803,506 37.78% PRD 48,030 1.01% PAC ODE 52,446 1.10% PADEMO 36,689 0.77% PPPM 50,793 1.06% PCN 60,635 1.27% UD 245,793 5.15% The Definitive Choice for Peace RENAMO's commitment to the electoral process has always been a source of doubt, and some observers have considered the slightest incidents as 'preparations for war'. I believe that within RENAMO there have always existed tendencies hostile to the electoral process, which does not by any means necessarily indicate a desire to go back to war, but rather an inclination towards some sort of power-sharing agreement without elections. This tendency, however, has never been clearly expressed, coming through only at different moments in the behavior of certain individuals. RENAMO, as an organisation, realised early on that its survival could only be political, its legitimation only through pluralism. There is also a very strong psychological reason for this: for reasons that are individual or linked with their community, the members of RENAMO are marginalised individuals. All their actions centre on the goal of being iet in through the front door' of urban society, of the hated Creole society, that they TRANSFORMATION 35 (1998) 35 ARTICLE CAHEN hate according to the classic scheme of attraction-repulsion. For example, during such controversies as the issue of 'Dhlakama's house' in Maputo (in 1993 he refused several houses that were offered to him and thereby delayed his move to the capital) and the Trust Fund moneys, there was more at stake than simply luxury and money: there was luxury and money, but also the dignity that they confer on those who benefit from them. From this perspective, the elections were an extraordinary victory for RENAMO. Although electorally defeated, RENAMO managed a score that was unimaginable for most observers only shortly before the election. This group once labeled 'armed bandits' and accused of the most unspeakable atrocities, won the vote of four out of ten Mozambicans, the absolute majority in five regions of the country representing a majority of the population and 70 per cent of the GDP. In a word, they won a resounding victory of legitimation. On the other hand, although FRELIMO was the electoral victor, it did not take the election because it had the majority in most of the country, but rather because in addition to obtaining a significant minority of the vote in regions that were mainly pro-RENAMO, it scored a crashing victory in the south (80 per cent of the vote).72 Its victory had an air of defeat: the former party of all the people was suddenly the party of 53 per cent of the people, and would have to fight hard to avoid becoming simply a southern party. The ethnic fraction of the overall vote in FRELIMO's favour appears larger than the corresponding fraction going to RENAMO. There was undeniably a Chona ethnic vote {Monica and above all Ndau) for RENAMO. On the other hand, the large Macua group, the Ajauas, the Nianjes, and the Niungues were divided: in three of the five provinces where RENAMO had a majority, it owed this victory to a vote that was not ethnic. In four of six provinces where it won a majority, FRELIMO owed it to an overwhelming ethnic vote, an embarrassing fact for an organisation having always denied that characteristic. The single party, party-state has already come to an end, but now it is the myth of the party-nation that is falling apart. One party can no longer capture the entire national ideal. In many regions the votes for RENAMO and for FRELIMO were not at all of the same nature: to vote for the party in power always has an aspect of allegiance, while voting for the opposition is, by definition, a vote for mobilisation. In the south, there was without doubt a powerful defensive phenomenon of mass psychology in She vote for FRELIMO, but elsewhere it is obvious that the client networks had a profound role, reinforced by the fact that it was the state that did the party's campaign. 36 TRANSFORMATION 35 (1908) CAHEN ARTICLE The fact that RENAMO obtained notable scores in some zones where it had never been militarily dominant shows that the phenomenon of a 'captive vote' has not been so important. Finally, just as there was no Macua vote, there was no Muslim vote. The country's new electoral geography describes a situation where the centre of gravity of the country has moved closer to where it had been historically. Even if the capital remains in Maputo,73 even if the essential part of the state remains in the capital, the new parliament is totally different from the old: the representatives have truly been elected by the provinces and are no longer party representatives chosen from the provinces. The simple fact that the electoral constituencies are the provinces themselves gives them a considerably greater political weight that will come into conflict with the desire to maintain a highly centralised political structure.74 The end of the 'people's party' is, in itself, a crisis for the Jacobin project of a modern, homogeneous nation based on the European model that has always inspired the ideals of the FRELIMO leadership. The evidence indicates that the democratic process is unfinished: pluralism is a necessary, but not sufficient, condition. Let us not underestimate its importance, as is seen fitting in some circles, by considering it as merely 'formal'. Pluralism is never completely 'formal'. Although one can certainly cite the high percentage of erroneous votes,75 this phenomenon merely indicates the pedagogical importance of the elections themselves, the first in the history of a country whose rate of illiteracy considerably exceeds South Africa's. RENAMO: a party that's here to stay? The question that arises today is whether the disappearance of the party-nation will leave a space for the stabilisation of a 'political RENAMO'. The fact that the principal activists in the organisation succeeded in entering the urban world 'through the front door', with a recognised sense of dignity, does not preclude the danger of a remilitarisation of the grassroots members, as local gangs or highway bandits. RENAMO has been accused of surreptitiously restaffing its bases. The real phenomenon is quite different: soldier-guerrillas, after years of life on a military base in an environment of strict, but nevertheless reassuring military discipline, have found themselves cast back into their native society by the demobilisation. Some have very rapidly reintegrated.76 Others, estranged after years in a military environment, have returned after a few months to the base, the only place familiar to them, and have reopened it, unarmed or sometimes with a few weapons. This hardly expresses RENAMO's will, but rather a difficult social problem demanding urgent treatment. There is, however, TRANSFORMATION 35 (1998) 37 I 1 ARTICLE CAHEN a precedent: why at the time of independence did FRELIMO cover the country with 'former soldiers' villages'? Why didn't these men simply go home? Recruited of their own free will or shanghaied into military service, a portion of the guerrilla army has never been able to return to their childhood homes. The historical precedent established by FRELIMO may provide some solutions to today's problem. It is certain that currently one of the country's most serious problems is 'armed banditry'. The future of the army itself is uncertain: with its numbers reduced and its budget cut, the army will very probably seek to grow over the next two years. There will not necessarily be a cleavage between the ex-FRELIMO soldiers and the ex-RENAMO soldiers, at least if the integration of the latter is done under respectable conditions. The composition of the new government is a good lesson about the nature of FRELIMO. It had every reason to open up and include RENAMO and the other smaller parties in the government. It could have demonstrated a sensibility to the regional pressures in giving governorships to RENAMO in zones where they had a majority. Beyond the mere fact that this would have demanded giving up a bigger piece of the pie, it also would have called into question the deeply held conviction of the FRELIMO elite that this state, this nation-building project, this country belonged to them alone. The government has brought up some young party members, but all the ministers, the vice-ministers, and the junior ministers are from FRELIMO; not even one without party affiliation! The president of the National Assembly went to a FRELIMO member, the former Attorney General, who never finished the inquiry into government corruption. The governors will all be FRELIMO. The only 'power-sharing' appears to be internal to FRELIMO: a group of slightly younger party members and technocrats will control the government, while the parliamentary group belongs to the 'hard liners' (Armando Guebuza, Sergio Vieira, and so forth). It certainly appears that FRELIMO has begun fighting to save its habitus of a monopoly on power, even if it is no longer a case of one-party rule. As a result of this, it opens a space for RENAMO, even though RENAMO will have intense problems of material survival that the control of a few ministries or provinces would have very probably resolved.77 RENAMO, the small military group in 1977 that turned into the important armed social movement of 1982 to 1985, that succeeded in imposing negotiations in Rome in 1990-92, and that ended up winning an impressive victory of legitimation. Has it become a political party? I realise that these terms are themselves grounds for a debate. It is certainly a civilian party, even if a few generals continue to play a generally discrete but important role.78 But does the 38 TRANSFORMATION 35 (199B) CAHEN ARTICLE fact of being 'civilian' define a program, an ideology? This question of'program' and 'ideology' is only interesting to the extent that they define a social policy and provide representation for other different elements of the society. There does not have to be a Eurocentric conception of a political party, and indeed one could ask if our political 'parties' really are. I am entirely in agreement with the remarks of Rob Buijtenhuijs on the parties of Chad: In analysing the political programs and other declarations of the Chadian political parties, we thus arrive at the conclusion that as far as social and material progress are concerned, these parties are not proposing any plans for large-scale societal change. ... We cannot stop there, however, in our quest for such plans. Indeed, it turns out that a lively discussion has developed between the Chadian political 'promoters' in other areas, notably bilingualism and the form of the state.... [When] one criticizes African parties for not having such social projects, one often forgets to say that in certain countries such as Chad, one does to a certain extent all one can to avoid the emergence of such projects.... In denouncing the absence of such projects on the part of African political parties, there is the more or less explicit suggestion that the parties elsewhere in the world would be better off in this area.... As far as the United States is concerned, the two parties do not have any real projects of large-scale social change (Buijtenhuijs 1994). In fact, the questions concerning the partisan nature of RENAMO are exactly the same today as those that could have been asked of FRELIMO: has not FRELIMO become above all a coalition of cliques wielding power? In doing this, it represents a social reality. As long as RENAMO continues to move forth on its 'coalition of marginalities', it too, represent a similar social reality. Dhlakama seems to understand this phenomenon when he announces his intention to reside alternatively in Maputo and in Quelimane, the capital of Zambezia {Savana, 152, 1995:32). REFERENCES Alexander, J (1994) Land and Political Authority in post-War Mozambique: A View from Monica Province, Oxford/Madison: Oxford University/Land Tenure Centre of Wisconsin-Madison University, multigr. Brito, L de (1991) he FRELIMO et la construction de V Etat national au Mozambique. Le sens de la reference au marxisme (1962-1983), Paris: University of Paris VIII, 1991 (Ph.D.). Brito, L de (1995), 'O comportamento elitoral nas primeiras eleicoes multipartidarias em Mozambique, in Mazula 1995: 473-499 (maps). TRANSFORMATION 35 (1998) 39 ARTICLE CAHEN Buijtenhuijs, R (1994) 'Les partis politiques africains ont-ils des projets de soci6te7 L'exemple du Tchad', Politique africaine, Paris: Karthala, 56, December, 1994. Cahen, M (1990) Mozambique, analyse politique de conjoncture 1990, Paris: Indigo Publications, March, 1990. Cahen, M (1991a) 'Les "mutineries" de la Machanga et de Mambone (1953): conflits sociaux, activisme associatif et tension religieuse dans la partie orientale de la "zone vandau"', Bordeaux: CEAN, January, 1991, unpublished paper to Second International Meeting of Social Scientists on Portuguese-speaking Africa, Bissau: INEP, 1991. Cahen, M (1991b) 'Le Mozambique, une nation africaine de langue officielle portugaise?'. Revue canadienne des etudes afiicaines, Toronto, XXIV(3), June, 1991. Cahen, M (1993a) 'Check on Socialism in Mozambique - What Check? What Socialism?', Review of African Political Economy, Sheffield (UK), 57, July, 1993. Cahen, M (1993b) 'Mocambique: a reconciliacab contra a democratizacao?', InformAfrica Confidencial, Lisbon, 54, October, 1993. Cahen, M (1994a) Ethnicitepolitique. Pour une lecture re"atiste de I'identiU, Paris: L'Harmattan, 1994 Cahen, M (1994b) 'Mozambique, histoire gebgraphique d'un pays sans nation', Lusotopie/Enjeux contemporains dans les espaces lusophones, Paris: L'Harmattan, 1 -2, June, 1994. Cahen, M (1994c), 'Le contexte politico-documentaire de la recherche en histoire contemporaine et immediate sur 1'Afrique lusophone', Cahiers d'histoire immediate, Toulouse, 6, Fall 1994. Cahen, M (1994d) 'RENAMO, o grande partido conservador-populista', Savana, Maputo, 4 November. Cahen, M (1994e, 'A verdadeira vit6ria da RENAMO. Resposta ao camarada Paul Fauvet', Savana, Maputo, 18 novembre 1994. Cahen, M (1997) 'Entrons dans la nation'. Notes pour une etude du discours politique de la marginalite': le cas de la Rename- du Mozambique, Politique Africaine, 67, October, 1997 Cravinho, J Gomes (1994) Democracia multipartiddria: Mocambicanizagdo, localizacao, Maputo, 7 November, communication from the Seminar on the Elections, Eduardo Mondlane University. Darbon, D (1995) 'Une transaction de'mocratique', Les Temps Modernes, Paris, 585, Nov.-Dec. 1995 (special issue on "Afrique du Sud: le Cap de Bonne Espe'rance"). Fauvet, P (1994) 'Urn marxista frances nas hostes da RENAMO', Savana, 11 November. Geffray, G (1990) La cause des armes au Mozambique, anthropologie d'une guerre civile, Paris: Karthala. Geffray, G (1997) Le Nom du Maitre. Contribution a ianthropologie analytique, Paris: Arcanes. General Peace Agreement of Mozambique 1992 - Acordo Geral de Paz de Mocambique 1992 (1993), Amsterdam: African-European Institute/AWEPAA. International Red Cross Mission in Mozambique (1994), Activity Report, November. Marchal, R and C Messiant (1997) Les Chemins de la guerre et de la paix. Fins de conflit en Afrique orientale et australe, Paris: Karthala. Mazula, BrazSo (ed.) (1995) Mocambique, Eleicdes, democracia e desenvolvimento, Maputo, 1995. Ministerio das Financas - Gabinete de Reintegracao (1992) Banco de dados dos militares a desmobilizar das FAM no ambito do Programa de Desmobilizacdo e reintegragao para militares mocambicanos. Informacao, explicacao e conclusdes, Maputo (confidential). 40 TRANSFORMATION 35 (1998) CAHEN ARTICLE Oliveira, P (1993) 'Le President et le Transcendant', Politique africaine, 52, December, 1993. ONUMOZ, Cease-fire Commission (1994) Problems/Incidents in Assembly Areas and Other Areas, Maputo: ONUMOZ CCF, 8 September, p.94. Soares, J (1994) 'Mocambique, as eleic5es do silencio', Expresso, Lisbon, 12 November, 1994. Waterhouse, R (1994) '17 Parties Registered', Mozambique Peace Process Bulletin, Amsterdam: AWEPA, August, 1994, special issue. NOTES 1. A slogan typical of the Mozambican Portingles, the mixture of Portuguese and English spoken in Mozambique. In this case, for the slogan 'Dhlakama emuito giro!' - 'Dhlakama is terrific!'. In this article, the names of ethnic groups inside Mozambique are written with the Portuguese spelling; other names have been, wherever possible, translated into English. See also note 47. 2. This is a translated version of " 'Dhlakama e maningue nice!' line gueiilla atypique dans la campagne electorate au Mozambique", originally published in L'Afrique politique, Paris: Karthala,1995,pp.ll9-161. This article was written at the beginning of 1995, after the first multiparty elections in Mozambique, and is published here without any changes (except for a few bibliographical references). Obviously, the domestic situation in Mozambique has changed, but the main tendencies were already clear in late-1994: Renamo, the warrior social body (corps social guerrier) was able to civil-ise itself, to become a civilian apparatus, and to be given a chance to transform into a political party - a transformation not achieved in 1992 by UNITA in Angola. Renamo's survival has been helped by the same reasons which permitted it to survive in the 1992-94 period, after the General Peace Agreement: Frelimo's incapacity to open power, its will to strictly apply the winner-take-all principle, paradoxically permited Renamo to solidify a coalition of marginalised social groups and acquire more legitimacy. But the electoral process had made clear Renamo's great difficulties in dealing with urban realities when the ethnic situation was not favourable to it. Perhaps Renamo's most important evolution, currently underway, is in the realm of a developing urban and social consciousness and the ability to articulate it. This was seen when Renamo published an official press release for Labour Day (1 May, 1996), and above all when it supported, or organised, urban demonstrations in the central and northern cities against the very difficult situation of the great majority of the population, in the first part of 1997. But Renamo today does not only exemplify the phenomenon of the transformation of a warrior social body into a political party. It expresses too the more general problem of the survival of opposition parties in Africa after they have lost elections. Without access to the State and international rent (including that from NGOs), how are they to survive? The next local elections in Mozambique (postponed to May, 1998) pose the problem of a real social (not only political) sharing of power. Renamo is certain to win in some of the biggest cities of the country, which means that Frelimo will lose its political hegemony and complete social ownership of the State. An issue which has to be closely observed is the politico-ethnic situation of the country. The civil war was not an inter-ethnic war: during the 1994 elections, ethnic concerns weighed heavily, but the results were not always along ethnic lines. But what will be the situation in 1998, four years after the elections, when a savage economic liberalism is aggravating regional imbalancies, when the World Bank and IMF order prohibits political parties from TRANSFORMATION 35 (1998) 41 ARTICLE CAHEN having any alternative programmes and objectively pushes them to find their originality in regional or ethnic identities? Is it not possible that a few years of economic liberalism will be able to produce inter-ethnic conflicts - something which a 15 year long civil war did not do? Michel Cahen, 8 October, 1997. 3. This article is the result of a research project conducted in Mozambique from 19 September to 21 November, 1994 for the Groupement de recherche 'Afrique australe' of the Centre national de la recherche scientifique within the program 'Democratic et developpement dans les pays du Sud-Afrique australe et Afrique lusophone*. The particular theme for this study was RENAMO's transformation from military organisation to political party. I wish to thank the RENAMO leadership for being willing to include me with their electoral team, making me the only independent observer to have been able to follow the entire RENAMO campaign. Thanks also to Christine Messiant (EHESS, Paris), Luis de Brito (Universidade Eduardo Mondlane), Eric Lubin, Mich61e de Rosset, and Henri Valot (ONUMOZ) for their help while I was in Mozambique, and Jeremy Grest (University of Natal) for his remarks and help. 4. There is no doubt that the Santo Egidio Congregation acted as the unofficial representative of the Vatican, which, as a state, had diplomatic relations with Maputo and did not wish to be directly involved. 5. Author's Note, October, 1997: See chapters by C Messiant, in Marchal, R and C Messiant, (1997). 6. As far as the draft is concerned, in practice young men did not report for service at all. The process of recruiting consisted of round-ups by militiamen at movie theater exits, religious services, and of schools, in the street, and by encircling villages, and entailed a whole series of abuses, including recruiting children under 16. According to the government's own confidential statistics, in August, 1992 out of a group of 15 682 soldiers being demobilised (out of the 90 000 total being demobilised), there were only 11 soldiers aged 15,16, or 17. The ages at the time of incorporation are quite different, however: 1 945 were recruited at the age of 16 or 17 (12.4 per cent of the total); 687 at 14 or 15 (4.38 per cent), and 340 at 13 years old (2.16 per cent); thus, 1 027 'children' (6.27 per cent) according to UN norms (the qualification 'child' being applicable to those 16 years or younger). The phenomenon of child-soldiers is thus not specific to RENAMO. According to the data of the International Red Cross dealing mostly with former RENAMO zones, in 1994 there were 3 654 war-experienced children living on military bases without parents. Of these, about 650 were true child-soldiers out of a total of about 22 000 guerrillas (thus around 3 per cent). These children were, however, sometimes much younger than the children in the FRELIMO army. Two-thirds of them were able to return to their families. The sources for this information are: Ministerio das Financas - Gabinete de ReintegracSo 1992), p.39, and documentation internal to the International Red Cross Mission in Mozambique (1994). 7. Conversation with Josfi Luis Cabaco, ONUMOZ, Maputo, 27 October, 1994. FRELIMO hired the Brazilian marketing firm Vox Populi to organise its campaign. 8. About this aspect, see Cahen (1993b). 9. Provided by Christine Messiant, the well-titled article by Joao Soares, Portuguese observer and son of the current Portuguese president (Soares 1994: 21). 10. ... at times with dramatic social consequences, such as these extraordinarily young marriages that are becoming more and more frequent in the Islamic regions of the north, in order to 42 TRANSFORMATION 35 (1998) [ CAHEN ARTICLE rebuild the society. I have personally seen cases of nursing mothers, sometimes as young as 11, married to middle-aged men, who are themselves sometimes the only living men. 11. Let us congratulate the exemplary role, at least for once, of an NGO, Handicap International, that was not satisfied with simple socio-medical projects with a budget based on international humanitarian aid or on the charity of western citizens, but sought rather to organise and mobilise civilians and soldiers maimed during the war. These Mozambicans, in turn, created their own organisations. This movement enabled this group of individuals to obtain aid and achieve a certain level of dignity following a war that the UN considered as resulting in deaths and displacements, but no disabilities. 12. As far as the wave of strikes in the beginning of 1990 is concerned, see Cahen (1990). 13. For example: in the Central Hospital of Maputo, an injection done with a single-use syringe costs a Brazilian video cassette. 14. Marginalitis is the term used in the original French article. 15. For more on the concept of a state without a nation, see Cahen (1994a), and specifically chapter 7. Regarding the particular case of Mozambique, see Cahen (1994b). 16. So long as RENAMO was merely a small band, FRELIMO mistook its activities as simply the work of the Rhodesians and designated them as 'counter-revolutionary activities', thus qualifying them as political. The spread of the war to the whole country forced FRELIMO to greater denial of the problem, hence the concept of 'banditismo armado'. 17. Let us not forget that Samora Machel himself sometimes privately used the same charming epithet for his own people. 18. Regarding this case, widely covered in the media, of Joaquim Chissano's membership in the Transcendental Church, see Oliveira (1993). 19. This is in particular what happened to the PCN (Partido da convengao national), one of the parties with the most serious projects; interview with Lutero Simango (leader of the PCN), Tete (Mozambique), 18 October, 1994. 20. An exhaustive list of all the incidents of the demobilisation has been drawn up by the Cease-fire Commission (ONUMOZ 1994). See also Marchal and Messiant (1997). 21. Padimbe Kamati is a somewhat known figure of the Maconde immigration to Kenya. Barely conversant in Portuguese, he directs, since 1958, a ghost-party, the Partido progressivo do povo de Mozambique (he was present in Mueda at the time of the riot of 16 June, 1960). Claiming to be a millionaire, he promised to help the Mozambicans with his own resources. Neves Serrano, thrown off the police force for incompetence (in Mozambique, a true feat!), is the self-proclaimed 'ontological-oxologicaT candidate and has punctuated the program of his Partido progressivo liberal federalista das comunidades religiosas de Mozambique with a completely ridiculous discourse, something that added a bit of humor in a campaign that needed it. His candidacy was ultimately rejected by the Supreme Court. 22. PALMO (Partido liberal de Mozambique) was, along with the questionable UNAMO (Unuo national de Mozambique, a Zamb^zian creation of FRELIMO), one of the first and most important opposition parties and started off with a great deal of energy. Directed by activists trained in Eastern Europe whose professional expectations were unsatisfied upon their return, the party quickly disintegrated amidst mutual recriminations and infiltration by FRELIMO. Casimiro Nhamitambo, a university professor, then created SOL (Partido socio-liberal e democrdtico de Mozambique - 'Sol' means 'sun' in Portuguese). TRANSFORMATION 35 (1998) 43 ARTICLE CAHEN 23. PACODE, Partido do congresso democrdtico; PCN, Partido da convencao national (directed by Lutero Simango, son of Una Simango, former FRELIMO leader, native of Sofala, who abandoned the movement in 1970, without ever going over to the Portuguese. He was arrested, sent to a re-education camp, then shot without trial; regarding the PCN, see above). 24. PANAMO, Partido national de Mozambique ; PADEMO, Partido democrdtico de Mozambique. For more information on all these smaller groups, see Waterhouse 1994. 25. Cravinho 1994:2-3. Joao Gomes Cravinho, researcher at the Institute of Economics and Management, Lisbon, was responsible for the team of European election observers in Tete province (centre of the country). 26. This was the case of PIMO (Partido independente de Mozambique), formed, under the neutral designation imposed by the law, by Muslims and directed by turban-wearing, Hollywood-style, Yacub Simbindy (Jacob Simbindy before the campaign/show began) who recently converted to Islam and is a native of the same region as Afonso Dhlakama. This must not be taken to mean that the Muslim community supported him; far from it. It seems to me quite possible that FRELIMO encouraged the development of this party to make a RENAMO foothold among Muslims more difficult to obtain. The richest Muslim and Indian merchants in Maputo supported FRELIMO, but the Muslim community was everywhere divided. There was no 'Muslim vote'. 27. The government circulated rumors that RENAMO was going to take Beira (the second largest city in the country) and the daily Domingo published the allegedly 'detailed plan' of RENAMO's preparations for war, etc - all manoeuvers that sought to make RENAMO think that FRELIMO was preparing the psychological conditions for a strike against RENAMO like the one in Luanda on All Saints' Day conducted by the MPLA against UNITA in 1992. RENAMO was to be thus pushed to making real defensive preparations, which would in turn justify police action. 28. During my study in the Save River estuary (Mambone, Machanga), for instance, I met civilians who have been members of RENAMO since 1977 for reasons that are very clear and who had never been either to South Africa or to Rhodesia. 29. The prazos were originally fiefs granted for three generations by the Portuguese crown to Portuguese establishing themselves in Mozambique (hence the term prazo, meaning 'time limit'). The phenomenon became profoundly African and gave rise to quasi-states and to a local aristocracy that was often made up of women. 30. For more on proto-nationalism in Manica and Sofala, see Cahen 1991a. 31. See Christian Geffray's now classic case-study of the Erati-Macuane zone (Geffray 1990); various articles by Otto Roesch, Southern African Report, 1991 -94; and the work of Jocelyn Alexander on the Manica region (Alexander 1994). 32. For example, the Red Cross in Germany sent shirts to the RENAMO populations. Problem! When the boxes arrived, the cargo turned out to be surplus military shirts and FRELIMO sought to paralyse the Red Cross operation by accusing them of aiding RENAMO. Apart from a few right-wing missionary organisations, the Red Cross was the only NGO before the cease-fire not to adhere to the government's guidelines, guidelines that were obviously pro-FRELIMO. 33. Testimony of Jean-Claude Legrand, who worked with OXFAM and later with UNICEF and was one of the first to enter the RENAMO zones regularly. RENAMO threatened to kill him for refusing to hand-over aid destined for the civilian population to a military commander. 44 TRANSFORMATION 35 (1998) CAHEN ARTICLE 34. From a political point of view, anyone fleeing ends up remaining in the sphere of the modern state, be it Mozambican or other. 35. Numerous accounts collected by me during my stay, some of them horrifying. 36. Line *d' of number 9 of Title HI of Protocol V ('Das Garantias') of the General Peace Agreement of Mozambique 1992-Acordo Geralde Paz de Mozambique 1992 (1993: 55). 37. In Zambezia, for example, I met RENAMO teachers who, two years after the Peace Agreement, continued to work without pay or recognition. 38. The regulos or regedores were the African chiefs recognised by the Portuguese administration: in some cases puppet leaders, in other cases perfectly legitimate, and their role as community representatives remained fundamental in many regions - something the FRELIMO entirely repudiated as 'tribalism' and 'feudalism'. Starting in the middle of the 1980s, without modifying its official view, the government took a more and more pragmatic and conciliatory approach with the objective of limiting the RENAMO advance; a strategy which did not necessarily undo the harm already done. Within FRELIMO today, there exists a very different tendency, namely to fawn over the 'traditional' chiefdoms as an expression of authenticity, and so forth. 39. The most violently oppressed religious group was the Jehovah's Witnesses with thousands of members deported. 40. The French term used in the original is corps social guerrier. For more on this subject, see Cahen 1990 and Geffray 1990. 41. ... not to mention the numerous commentators who have always denied the existence of any civilian wing within RENAMO, seen simply as an 'armed band'. I had identified the existence of such a wing (Cahen 1990), but underestimated its importance. 42. I analysed the bits of written internal RENAMO discourse that exist in my communication to the Colloque international Lusophonie-Lusographie, Rennes, September, 1994 (Cahen 1997). 43. It's impossible for me to discuss the exact significance of the number of members: the mass is impressive and indisputable, but the electoral results show that, in certain zones, RENAMO had fewer voters than it had members. The RENAMO leaders cried fraud. I'm not at all sure that that was the only reason - this is a classic phenomenon. 44. These paragraphs borrow from the article that I published in Mozambique at the end of the electoral campaign (Cahen 1994d). 45. Numerous interviews in Machanga and Mambone, November, 1994. 46. This nuance merits consideration: in this large capital city of the north, Dhlakama speaks of Macuas but also of Nampulenses, which is not an exclusively ethnic term (numerous are those who, 'ethnically' from the south, live in the north, for administrative, military, commercial reasons and their children born in Nampula can be considered Nampulense). In rural areas, he speaks exclusively of Macuas. Al. Emmakhuwa-lomwe for those well-acquainted with the allegedly international, in fact English, spelling of ethnic names; in fact only of ethnic names designating African and Asian peoples of the Third World - the 'primitive peoples'. Who would ever think of imposing a unique spelling - in every language - for the Scots, the Auvergnats, or the Italian-Americans? 48. This insistence on the part of Dhlakama to introduce Vicente Ululu as the 'number 2 man' has a double meaning: Ululu has never been in the military wing of RENAMO and thus f r TRANSFORMATION 35 (1998) 45 r ARTICLE CAHEN illustrates the 'civilisation' of the party; secondly, his rank clearly indicates that in the internal power struggle, Raul Domingos is necessarily number 3. 49. It is also worth noting that the RENAMO leadership for the Pemba district is made up exclusively of Macuas, with the exception of a Portugaese-Maconde mulatto. 50. The Maconde planalto is a non-Muslim enclave in a largely Muslim region and raising pigs is one element of a strong local identity. 51. Hermfnio Morais, better known during the war by his nom de guerre Bob Charlton (in honor of the English football player!) is a man of humble and timid appearance, but was in fact a dreaded war machine, the chief saboteur, responsible for operations within Zimbabwe. The Zimbabweans did not tangle with him for long and ended up signing a partial peace accord with RENAMO. 52. Based on the testimony of Mozambican and Zimbabwean journalists present. 53. During the displacement towards the south of the Nguni state at the end of the 19th century, the king Ngungunhana forcibly recruited tens of thousands of Ndaus into his army, of which a large portion died unburied during deportation and in combat. The spirits of the unburied dead, however, are vengeful spirits. The Ndau spirits have a considerable importance in the phenomena of possesion in the south of Mozambique. RENAMO was, to a certain extent, seen as the 'envoy' of the old feelings of resentment for coming to combat the FRELIMO state and as mostly Changane, thus Nguni (the ethnic group Changane is an identity recently formed by the partial 'Ngunisation' of the old Ronga core). In this way, some RENAMO massacres (Taninga, for example) have been interpreted, and to a certain extent thus legitimated by the victims themselves, as the price that had to be payed. 54. This stadium is located seven kilometers from the city. The day of the meeting, numerous city trucks broke down. Nevertheless, thousands of people went on foot from all of the surrounding areas. (FRELIMO never let up on this form of petty harassment: for instance, when Dhlakama arrived in a city, the residence rented for him by the local leadership would be equipped with a phone and electricity, but these would tend to break down and be restored only several hours later.) 55. In the original French article the term desir de maitre is used (Geffray 1997). 56. FRELIMO frequently took charge of children - at times sending them to Cuba for years - without notifying their parents. 57. This assessment incited the wrath of Paul Fauvet (Fauvet 1994). This whole paragraph has been taken from Cahen 1994d. 58. The pro-FRELIMO press talked of CIA funds delivered by the United States ambassador in Maputo... 59. Afonso Dhlakama almost always transported his electoral team by helicopter, costing almost $2 000/hour. 60. An order more or less followed by RENAMO militants, including Dhlakama's body guards, who consistantly went after anyone wearing a 'camisete Chissano' and daring to approach the RENAMO convoy. 61. Eric Lubin and Christine Messiant. 62. I in no way wish to downplay the violence of the two groups: knives, kalashnikovs, anti-personel mines, burning houses, villages, convoys, etc - all this certainly caused directly or indirectly, more than 1 000 000 deaths out of a population of 15 000 000. 46 TRANSFORMATI ON 35 (1998) CAHEN ARTICLE 63. On the other hand, in that region RENAMO benefitted from a remarkable agitator and gifted politician, Manuel Pereira, the provincial political delegate and second on the list of candidates for the parliamentary representative from Sofala. 64. I was very impressed by the meeting at Mogovolas on 4 October, in the heart of Macua country. The members of CIVPOL (the civilian police of ONUMOZ) who had attended Chlssano's meeting a few days earlier couldn't believe their eyes and wondered where all these people had come from. I attempted to estimate the crowds at all the RENAMO meetings: my estimates were always half the size of RENAMO's, something that never stopped me from meeting with total incredulity on the part of people I spoke with in Maputo. Dhlakama brought together approximately 600 000 people during the campaign: an impressive figure for a country of 15 000 000 with enormous transportation problems. 65. The Chopes and the Bitongas have historically been the enemies of the Ngunis (and thus of the Changanes) at the time of the wars and invasions of the end of the 19th century and the beginning of the 20th and were, as a result, the allies of the Portuguese. The Rongas, disturbed by their progressive 'changanization'', claim to be the true 'owners' of the capital (the Changanes are located farther north) and a certain reaffirmation of their identity continues today. The Vatsuas are the most northern of the southern groups. 66. RENAMO accused FRELIMO of having disposed of the extra ballots and voter-registration cards (1 900 000) resulting from the difference between the predicted number of eligible voters (8 300 000) and the actual number who finally registered (6 400 000). A circular issued late by the Technical Secretariat of Electoral Administration (STAE), controlled by the state, allowed anyone with a voter-registration card to vote, even if that person's name was not on the voter rolls. This circular was annulled then replaced by another ... with the same number ... with the effect that many polling stations thought the two circulars were one and the same and didn't read the second. RENAMO also systematically decried massive Zimbabwean infiltration, while providing, however, numerically insignificant proof. There were certainly cases of fraud, above all at polling stations where the canvassers were unanimously in agreement. For technical and political reasons, however, it seems impossible to me that they were significant enough to affect the outcome; FRELIMO itself needed internationally accepted elections. Indeed, much more important was the 'extra-systemic fraud' that took place before the elections, the shameless use of intimidation and of the technical methods at FRELIMO's disposition. Likewise, the problem of erroneous votes was much more significant (see above). Finally, at the risk of appearing machiavellian, I do not discount the possibility that the leaks and blunders leading one to believe that FRELIMO was preparing fraud (for example, a semi-true document from Vox Populi describing serious machinations), were in fact from groups close to FRELIMO itself, the idea being to push RENAMO into denouncing the elections and into committing a serious error at the last moment. 67. A deputy's salary is 1 000 000 meticais or about $200 a month, while the Trust Fund allowed for advisors to be paid much more generously. 68. The publication of the Domingo 'document' occured at the same time that rumours circulated about the occupation of Beira by RENAMO. Was a clan within FRELIMO playing a dangerous game to cause RENAMO to commit an error? Or - to continue in the same machiavellian tone of note 66 - was it a pro-RENAMO group that published this document, thereby enabling RENAMO to demonstrate its pacific nature and to sacrifice its concerns about 'fraud' on the 'altar of peace' for the benefit of the country in accepting the results? TRANSFORMATION 35 (1998) 47 ARTICLE CAHEN 69. The validation of more than half the null votes was perfectly justified: indeed, many citizens had marked the candidate's photo (for the presidential election) or the patty symbol (for the legislative elections), rather than the corresponding square. The law, however, considered as valid any 'clearly expressed vote'. 70. This refers the number of ballots in the ballot boxes. The fact that this number was slightly higher than the number of voters does indeed indicate fraud, but on a small scale. 71. Same remark as in footnote 70. 72. At the time of this writing, although the parliamentary seats have been assigned, the detailed results of the elections broken down by province have not yet been published by the.CNE. 73. RENAMO has not called for the capital to be moved either to Quelimane or to Beira, probably to avoid inconveniencing the embassies located in Maputo. The debate about the capital will thus figure among the questions for the new Parliament. 74. For the moment, there are no plans for general local elections, only for the cities (cidades); the towns (yilas) and rural zones are not included. 75. The UD (UniSo Democrdtica), a very small political party without any particularly noteworthy political message succeeded in surpassing the 5 per cent necessary to be represented in Parliament solely because its party symbol was last on the ballot. For the presidential elections, Chissano was last on the ballot and thus many citizens wishing to vote FRELIMO but having remembered only 'vote for the last on the ballot' voted UD in the legislative elections, as FRELIMO was listed 5th (and not last) on that ballot. Likewise, we observe that as Dhlakama's glasses appear very clearly in his photo on the ballot, so all the other candidates wearing clearly visible glasses received more votes than their party's legislative candidates received; for example, Carlos Reis received 2.44 per cent while his party UNAMO received 0.73 per cent. FRELIMO thus lost approximately 4 per cent of its electorate, as did Dhlakama. If one adds up the erroneous votes, blank votes, and the null votes (starting with the hypothesis that in Mozambique there is no tradition of 'political' blank or null votes), it becomes clear that a relatively significant percentage of the population (10-15 per cent after acceptance of more than half the null votes) did not succeed in voting correctly. This is in no way surprising given the complexity of the ballot (numerous parties and presidential candidates, and moreover, situated in different positions on the two ballots). 76. I met several demobilised soldiers who proudly told me, 7've already started work on my fields. I've already rebuilt my house. I've already gotten married1. 77. One may expect that certain party cadres will leave RENAMO for paid work in the private sector. 78. Hermmio Morais, already mentioned, or Issufo Momade, a Muslim Macua, in charge of religious issues... 48 TRANSFORMATION 35 (1998)