The Forgotten Sons of the State: The Social and Political Positions of Former Government Soldiers in Post-War Mozambique^*

Nikkie Wiegink¹

PhD candidate at the Department of Cultural Anthropology, Utrecht University is currently finishing her dissertation on life trajectories of former combatants in central Mozambique. She was trained as an anthropologist, but also holds a MA in Conflict studies and Human Rights. The data presented in the article were gathered in the context of ethnographic fieldwork for her Ph.D. dissertation. The research project was funded by the Dutch Organization of Science. Most recent publications: Wiegink, Nikkie. 2013. "Why Did the Soldiers not Go Home? Demobilized Combatants, Family Life and Witchcraft in Postwar Mozambique", Anthropological Quarterly 86(1): 107-132. Wiegink, Nikkie. 2011. In collaboration with Hugo de Vries. "Breaking Up and Going Home? Contesting Two Assumptions in the Demobilization and Reintegration of Former Combatants", International Peacekeeping 18(1): 38-51. Email: N.Wiegink@uu.nl

ABSTRACT

This article shows that 21 years after the signing of the peace accords, the Mozambican state continues to struggle with the social and political niche of the more than 90,000 demobilized soldiers. The article departs from the observation that the memory construction of the past liberation and civil war done by the FRELIMO government was highly influential in creating and sharpening categories of veterans, determining which "type" of veteran has access to privileged resources and which is excluded. Based on ethnographic fieldwork, the problematic relationship between ex-combatants of the government army and the Mozambican government is explored. For veterans, this relationship is characterized by frustration, hope, patrimonialism and the use of the threat of violence. This plays out in negotiations involving material aspects (particularly pensions), but also symbolic aspects of being regarded as worthy veterans.

KEYWORDS

Former combatants, state, memory politics, pensions, violence

Los hijos olvidados del Estado: posiciones políticas y sociales de antiguos soldados del gobierno en el Mozambique de posguerra

RESUMEN

Este artículo muestra que después de veintiún años de la firma de los acuerdos de paz en Mozambique el Estado de ese país sigue luchando contra el nicho social y político de los más de 90.000 soldados desmovilizados. El artículo parte de la observación de que la construcción de memoria sobre la anterior guerra civil y de liberación, realizada por el gobierno del Frente de Liberación de Mozambique -FRELIMO-, fue altamente influyente en la creación y agudización de categorías de los veteranos, determinando qué "tipo" de veteranos tiene acceso a recursos privilegiados y cuál tipo resulta excluido de ellos. Este trabajo explora la problemática relación entre los excombatientes del ejército gubernamental y el gobierno de Mozambique con base en un trabajo de campo etnográfico. Para los veteranos esta relación se caracteriza por la frustración, la esperanza, el patrimonialismo y el uso de la amenaza de violencia. Lo anterior se despliega en negociaciones que involucran aspectos materiales (en particular las pensiones), pero también aspectos simbólicos relacionados con su reconocimiento como veteranos valiosos.

PALABRAS CLAVE

Excombatientes, Estado, políticas de la memoria, pensiones, violencia

DOI: http://dx.doi.org/10.7440/colombiaint77.2013.03

Recibido: 31 de julio de 2012 Modificado: 22 de noviembre de 2012 Aprobado: 4 de febrero de 2013

Introduction

In January 2010, at the beginning of his second term as the President of the Republic of Mozambique, Armando Guebuza installed a new ministry: the Ministry of the Combatant. It replaced the Ministry of the Antigo Combatente, which dealt only with issues relating to the veterans of the liberation war (1964-1974). The Ministry of the Combatant was to accommodate the concerns of veterans from the liberation war and demobilized soldiers from the civil war (1976-1992). It was presented as part of President Guebuza's campaign for national inclusion, but the founding of the Ministry is foremost seen as a concession to the demands of more than a dozen associations of demobilized soldiers of the civil war. Since 2006, these associations have been negotiating with the government about the legal position of demobilized combatants, more specifically the right to pensions and other benefits. Some of these associations have been threatening violent demonstrations out of frustration over the neglect of demobilized soldiers by subsequent governments. To better understand the marginal position of the demobilized soldiers one must first understand the awkward relationship existing between the state and the demobilized soldiers of the civil war. This article shows that 21 years after the signing of the peace accords, the Mozambican state continues to struggle with the social and political niche of more than 90,000 demobilized soldiers. Based on ethnographic fieldwork in Maputo city and in two rural districts, Magude and Maringue, this article explores the problematic relationship between ex-combatants of the Forzas Armadas de Moçambique (Mozambican Armed Forces, FAM) and the Mozambican government, and argues that the reintegration of former combatants is an ongoing process of negotiating the place of ex-combatants in history and society.

Former soldiers have a specific position within a state and its history construction. They may be glorified as heroes of great victories, but may also be a reminder of failure, guilt and shame that is rather (selectively) forgotten (Buckley-Zistel 2006; Cohen 2001), or something in between. Civil wars pose particular questions to the position of veterans, as there have been (at least) two armed groups fighting in the same territory and both have committed atrocities. How a government deals with the former soldiers who fought in a certain (civil) war depends on the type of peace agreement made to end the war, on the resources present, but maybe even more on the politics of historical interpretations of the war, which are often ignored in studies of reintegration and peace building (Kriger 2003, 20-23). As Metsola (2010, 590) points out in the case of Namibia: "Such historical interpretations are not merely a matter of national imaginary. They are highly significant to current socio-political relations, defining terms for inclusion and exclusion." It is through procedure and policy that the state generates certain categories, which determine the category's influence on the state's actions and access to the resources of the state, shaping everyday life of the subjects to a certain degree (Broch-Due 2005, 14).

By analyzing the official memory construction in Mozambique, this paper shows how certain categories of veterans have emerged, determining which "type" of veterans have access to privileged resources and which are excluded. Apart from material aspects related to these categorizations (particularly pensions), the "resources of the state" are also related to more symbolic aspects of being a worthy veteran or, in the analogy of many veterans, to be regarded a son of the state. I argue that the relationship between the FAM veterans and the state is one of mutual dependency, yet this relationship leaves little room for maneuver for veterans. To understand this relationship, a distinction is made between national and local government. It becomes clear as this article progresses that the national government sets the terms of the narrative and the pool of resources, yet local administrative and political dynamics seem to determine whether former combatants obtain access to these resources.

To fully understand the position of former soldiers in relation to the state, there is a second important aspect that needs addressing: the former combatants' embodiment of a threat of violence and instability. Even years after the war is over, former soldiers can become a threat to the authority and monopoly of violence of the state, as has been documented in the 1990s in Zimbabwe, where war veterans have become a powerful force in politics more than a decade after the liberation war ended (Kriger 2003; McGregor 2002; Alexander et al. 2000). In Mozambique, the veterans have not become a comparable political force, but protests by veterans in 2000, which turned violent and resulted in over a hundred casualties, formed a strong reminder of the threat of violence and instability that veterans can embody (Schafer 2007, 1). Recently, the negotiations between veterans associations and the government were influenced by this threat of violence, as a particular association of desmobilizados pressures the government by demonstrating on the streets.² As I will argue in this article, the recent changes in policy regarding demobilized soldiers (such as the Ministry of the Combatant) must be predominantly understood through this dynamic. Thus, while the policies regarding demobilized soldiers are predominantly shaped by the memory of war, the recent changes in policy can be attributed to the threat of violence of certain organized groups of veterans.

To build this argument, I will start with an exploration of the dominant narrative of the civil war in Mozambique, followed by an analysis of how this shapes current policies regarding former FAM soldiers. Then, I focus on the relationship of war veterans with the state, addressing material and symbolic aspects. Subsequently, I explore the role of a specific veterans' association and the threat of violence in negotiations with the government. The final reflections address the significance of veterans in the politics of memory and in post-war politics, stimulating further thinking about the social and political integration of veterans. First, though, there is a need to briefly contextualize the FAM veterans in Mozambique's recent history.

1. Reintegration after War in Mozambique

The ex-soldiers on whom this article focuses fought in Mozambique's recent armed conflict between the Frente de Libertação de Moçambique (Mozambican Liberation Front, FRELIMO) and the Resistencia Nacional de Moçambique, the Mozambican National Resistance (RENAMO).³ This civil war devastated the country between 1976 and 1992 and is often characterized as one of Africa's most brutal wars. It is estimated that 600,000 to one million people died, and over five million people were displaced, either internally or to neighboring countries (Hanlon 1996).⁴ Both RENAMO and the FAM forcibly recruited young men and women into their ranks, often teenagers, who on average fought for nine years for one of the belligerents (Schafer 2007; Pardoel 1994).

In 1992, after the General Peace Accords were signed between RENAMO and the FRELIMO government, over 90,000 combatants were to be demobilized. This was organized by the UN Mission for Mozambique (ONUMOZ) and turned out to be one of the most expensive DDR processes to date (Alden 2002, 354). The combatants were gathered in assembly areas from which they (and their dependents) got free transportation to a place of their preference. Additionally, they received a demobilization allowance for 18 months, a relatively large amount of money in a society with limited cash flow. The peace accords set out the guidelines for the disarmament and demobilization of combatants, but provided little instruction for the post-demobilization policy making (Schafer 2007, 123). The policies regarding veterans were initially captured under the concept of "reintegration," for which the state depended on large funds from ONUMOZ and international donors. But when payments of demobilized soldiers ended in 1996 and employment creation programs were completed in mid-1997, the reintegration process was declared successful by the major observers, and the international donors turned its attention to more developmental issues (Alden 2002, 342; Schafer 2007, 130-138).⁵ From that moment, the FRELIMO government set the agenda regarding the demobilized soldiers. The main issue of negotiation between the government and demobilized soldiers became pensions. As shown in subsequent sections, this became a politically sensitive topic because the government's position toward the former government soldiers was ambivalent and, in relation to RENAMO veterans, rather hostile (Schafer 2007, 124-125).

The reintegration of ex-combatants is often mentioned as one of the factors contributing to the successful Mozambican peace process (see e.g. Cobban 2008). The successful integration of ex-soldiers into civilian life has been attributed to the immense amount of money the UN mission for Mozambique spent on the DDR process (Alden 2002), but even more to community-based integration and purification rituals performed by relatives and community members (Honwana 2006; Granjo 2007, 126-127; Nordstrom 1997, 145-146). However, the focus on these community-based rituals has failed to address the post-war political context and the position veterans have within this context (see e.g. Schafer 2007). This article aims to contribute to filling in this research lacuna, offering recent data and analyses on the uneasy relationship between the FRELIMO government and the veterans of the FAM.⁶

2. veterans and memory Politics

The memory construction of past atrocities is highly important in shaping the identity and legitimacy of a certain group, whether political, ethnic or religious, as these memories contain sensitivities and emotions and implications of blame, of victory, of pride and of shame (Malkki 1995; Alexander et al. 2000; Buckley-Zistel 2006; Shaw 2007). While memories are always univocal and in shaped and re-shaped in the present (Nora 1989, 8; Antze and Lambek 1996, XII-XIII; Nazarea 2006, 325), it is the state that is often in control of evoking memory through symbols, media and textbooks. This way, governments largely determine what is memorable and what is eliminated from the national narrative, and how certain events are remembered (Cohen 2001, 243).

FRELIMO has been ruling Mozambique since gaining its independence in 1975. Initially, Mozambique was a socialist one-party state, yet the end of the civil war was also the beginning of a multiparty democracy. FRELIMO won the country's first democratic elections in 1994 and has won every national election since. Consequently, the party maintained the power over the mechanisms of official memory construction and reconstruction; therefore, it is the FRELIMO version of the past that has been the dominant narrative concerning the wars in Mozambique (Igreja 2008, 544-545; Pitcher 2006). It is this narrative that is most influential in shaping current policy, especially regarding war veterans.⁷ To understand why the FAM veterans feel "forgotten" and marginalized, we need to understand the "FRELIMO version" of Mozambique's recent history.

Similar to other countries that were born out of liberation wars, such as Zimbabwe (Alexander et al. 2000, McGregor 2002) and Namibia (Kossler 2007, 362; Metsola 2010), FRELIMO's leaders build their legitimacy and national identity around the credentials of the liberation war (Schafer 2007, 124). In the latest national elections in 2009, the slogan of the FRELIMO party was "FRELIMO é que fez, FRELIMO é que faz," freely translated as "it is FRELIMO that did, and it is FRELIMO that does."

"Did" refers to the liberation war and presumably also the (albeit uneven) development of the country after independence. Such statements enforce the claim by FRELIMO as the "owners" of Mozambique and provide legitimacy for authority (Pitcher 2006, Schafer 2007).

Not surprisingly, the liberation war has a prominent position in FRELIMO's construction of the nation's history. The liberation struggle enters almost daily in the public sphere through national holidays, such as Heroes' Day, the day the first bullet of the liberation struggle was fired; road names, such as Eduardo Mondlane, as well as other African leaders such as Kenneth Kaunda and the socialist "heroes" like Vladimir Lenin and Mao Tse Tung; impressive murals; and names of schools. These are current reminders that it was FRELIMO that overthrew the Portuguese and achieved national independence, and are part of what Kossler (2007, 362) called "the glorification of history of the liberation warfare."⁸

The veterans of this war, the antigos combatentes, are seen as national heroes and are rewarded as such. Jose Ranqueni, Maringue's district leader of the Association of Antigos Combatentes, informed me that antigos combatentes receive a pension ranging from 3,370 to 5,120 MTN (approximately 80 to 120 euros) depending on one's rank. Taking into account that the minimum wage in Mozambique is more or less 1,800 MTN, these are quite significant earnings. In addition, these veterans receive a range of benefits, such as funeral and medical costs, which cover most expenses people worry about besides food. The antigos combatentes are generally seen as being "in the hands of the government," as being "taken care of" by the state. Ranqueni remarked:

They [the government/FRELIMO] know who have been fighting for the people of this land. We were fighting for the liberation of men and land. It is very different from this war that stopped just now. It even has a different name, how is it again? The destabilization war.

Ranqueni refers to the civil war here, often called "the destabiliza-tion war" due to the interference of the neighboring apartheid states of Southern Rhodesia (present-day Zimbabwe) and South Africa. As Ranqueni said, the war was "very different," implying that it was fought for less glorious causes and that consequently the veterans of this war are different too. In brief, one could say that this war was a dirty war, in which Mozambicans fought other Mozambicans, albeit involving foreign interest. The war did not end with a (moral) victory, but rather with the destruction of the country (Hanlon 1996). In its aftermath, the war has been silenced (Igreja 2008), partly due to the peace accords that promoted silence rather than "dealing" with the war, but also to the war's uneasy character. There was no investigation in war crimes, but while RENAMO was generally held responsible for the greater share of destruction and killing, the FRELIMO party, like the rest of the country, was well aware of the atrocities committed by the army (Schafer 2007, 124-125).

These two different wars, the liberation war and the civil war, have for several reasons resulted in (at least) three different "types" of veterans, due to differentiating institutional economic policies. There are several practical and administrative reasons for the difference between these policies, but they are strongly related to the memory construction of the different wars. These policies privilege antigos combatentes as they represent FRELIMO's legitimacy for governing Mozambique. Even more, Mozambique is practically ruled by antigos combatentes, who form the elite core of FRELIMO and occupy the most important governing positions at all levels of the government structure.

In contrast, the civil war was an awkward war, ending not in a victory, but in an uneasy peace. It resulted in two "types" of veterans: first, RENAMO veterans, who were not (explicitly) considered by the FRELIMO government in post-war policy making. Even though RENAMO was accepted as a political party, its military past was deemed illegal by the FRELIMO government, as were its fighters (Schafer 2007, 124). FAM veterans make up the other "type" mentioned above. The government's position toward this group is, however, more ambivalent for various reasons. There are several practical reasons for these differences: first of all, as the country's economy was in ruins, the FRELIMO government did not have the funds to reward the demobilized soldiers of the government forces. Second, FAM veterans, while not demonized like the RENAMO combatants, did acquire an unsavory reputation (Schafer 2007, 124-125). Finally, they represented a war that was rather forgotten and, even if possible, it would not have been appropriate to reward the FAM veterans as national heroes. The FAM veterans were supposed to "fade away" (Alden 2002).

3. The Politics of Pensions

When the ONUMOZ mission ended and the official reintegration process was branded successful, the FRELIMO government was to set the agenda regarding the demobilized soldiers. The main issue for the demobilized soldiers became pensions, a politically sensitive topic. As I have pointed out above, there was little funding and little political will to address the issues of FAM veterans, yet the government also felt the need to please the demobilized soldiers. There had been some violent incidents and threats by FAM veterans in the early post-war years, which the government aimed to contain (Schafer 2007, 129). Therefore, the issue of pensions was kept relatively quiet. The government gradually released funds to keep the veterans' hopes alive, but at a frustratingly slow pace (Schafer 2007, 125) and benefitting particularly demobilized soldiers of higher ranks.⁹

After the initial demobilization allowance in 1994, the majority of demobilized soldiers did not receive money from the state, because, as it turned out, only a small percentage of the veterans could receive pensions. Decree 3/86, issued in 1986, by which the retirement and pensions of veterans were organized, only covered those men and women who served in the military service for ten years or more (Taju 1998, 50). During the war, the government implemented a compulsory military service of two years, but as the war dragged on, most soldiers ended up staying longer, though often no more than ten years. These demobilized soldiers were, thus, excluded from pensions. The Decree also excluded those ex-combatants who were recruited below the age of 16, as this was an illegal practice that was not supposed to happen and for a long time denied by the FRELIMO government.¹⁰ While RENAMO was excluded from this Decree altogether, Taju (1998, 95) noted that the same Decree was used to govern the pensions of disabled soldiers, handicapped during military service, which also included RENAMO ex-combatants. In other words, the legislation concerning the pensions was far from straightforward.

The application and reception of pensions was and continues to be arranged through veterans' associations. The main association is the Associação Moçambicana de Desmobilizados de Guerra (AMODEG), which was founded in 1991. Initially, it was an association for veterans of the FAM, but after the peace accords in 1992 it changed its statues to include all demobilized soldiers of the war, opening the doors to RENAMO veterans (Schafer 1998). The other main organization is Associação de Deficientes Militares e Paramilitares de Moçambique (ADEMIMO) for deficient ex-combatants who were handicapped during military service. Through this association, disabled ex-combatants can apply for a pension and get paid according to the extent of their disability, between twenty and one hundred percent (the latter for example applicable to the loss of two legs or blindness).¹¹ Although both organizations could not tell me how many members they had and how many of them were receiving a pension, it was generally observed that there were more disabled combatants benefitting from a pension than non-disabled veterans.¹²

While these are associations active at the national level, the dealings of the veterans with these associations happen at the local level. To apply for a pension, a demobilized soldier has to deliver his or her documents (demobilization card, identity document and, in case of disabled veterans, a medical declaration) to AMODEG or ADEMIMO at the district level, where the application is then sent to the provincial level, and then to the central level in Maputo, where it is handed over to the Ministry of the Combatant (previously to the Ministry of Women and Social Change). Then, it goes through a similar number of stages before a veteran's name appears on a list and he or she can collect the pension at the district administration. As with most dealings with government bureaucracy, this is a rather inefficient and foggy process, highly prone to corruption and with an unclear structure of responsibility (Schafer 2007). Indeed, many demobilized combatants waited years for their pension and others never got it.

Jose Sande fought for the FAM from 1979 to 1989. I met him at the train station in Magude, but he was originally from Caia (Sofala province). As he served ten years in the military, he was entitled to a pension, which, to his frustration, he did not receive. The moment he mentioned this during our conversation, my research assistant, Alexandro, also a delegate of AMODEG in Magude, intervened: "You should fill in the forms and hand them in to us [AMODEG]!" Jose sadly shook his head. "I have tried so many times. They [the government, the associations] must have my documents and everything. But it costs money and I don't have that. It will not work. I will make it on my own." Alexandro could not get his head around this and tried convincing Jose in vain. When we left Jose at the train station, Alexandro said, "There are people who do not know the ways of the government and they don't get anything. That is why they should be members of associations. The way to the government is through associations."

Indeed, it seemed that those former combatants close to the associations often did have access to pensions. The representatives of the associations, individuals such as Alexandro, probably know how to arrange a pension, even for someone who has served less than ten years in the military (which was the case for Alexandro). Connections within AMODEG, the FRELIMO party structure or local government institutions seem crucial to the application process. As I will elaborate below, patrimonial logics at the local (district) level play a key role in the distribution of pensions to former government soldiers, speeding up the requests of some, but consequently slowing down others, keeping them dangling in the daunting bureaucracy. Another factor that contributes to a successful application is, as Jose implicitly mentioned above, bribing a civil servant who may transfer the documents to the right place and the right people. But veterans like Jose, who are struggling to make a living, do not have the financial means for such an undertaking.

The government support of the veterans' organizations varies between provinces and districts, determining to a certain extent the capabilities of the associations (Schafer 2007). In theory, AMODEG established delegates in all districts, which seemed to be the case in Maputo province. In Sofala, however, several districts did not have an AMODEG delegate in 2008, especially the district associated (historically) with RENAMO support, such as Maringue and Cheringoma. As Tafula Quamba, the National President of AMODEG said: "There is a reluctance to go to Maringue. There, RENAMO has the power. A place where there was a cobra is difficult to visit again." Consequently, veterans in such places have less knowledge about the existence of pensions and do not apply for these state resources. This reluctance apparently dissolved in summer 2009, when AMODEG's delegate in Beira (capital of Sofala province) told me that the government had asked AMODEG to make an inventory of the demobilized FAM soldiers at the district level. The promise of pensions was more or less implicitly added to this request. "This is what is going to change our lives!", he said, while waving a pile of registration forms. Also, in Maringue, the FAM veterans were organizing themselves. There were probably only 25 of them. This registration process was generally seen as a strategy within FRELIMO's electoral campaign to obtain more votes during the 2009 national elections. Batista, responsible for the registration of demobilized government soldiers in Maringue, said frankly: "We ask every régulo [chief] to call the desmobilizados. Then, we explain and then they have us in their heads when they vote."¹³  This raised the veterans' hopes for pensions; however, it is unlikely that they ever received any. This for at least two reasons: first, those veterans who fought less than ten years still do not have the right to a pension and, second, there seems to be little political will to really pay out pensions. As Jacinta Jorge, Director of ProPaz, a veterans' association, remarked bitterly: "They [the government] promise many things at election time, but then we see nothing." Furthermore, as Xaderique Paulinho, the National Director of the Ministry of the Combatant, said, there are simply not enough funds for all ex-combatants to receive pensions.¹⁴

In all, the government's position vis-à-vis FAM veterans' pensions is shaped by a limited commitment to change policies and a reluctance to spend government funds on pensions. The issue has been used to gain political support for the national and local political FRELIMO campaigns, but veterans encounter legislative and bureaucratic problems in the process of actually seeking pensions. The difference between the national and the local state is telling in this respect. While the national government sets the terms of the narrative and determines the pool of resources, the local government and political dynamics determine whether veterans have access to funding or not. For many veterans, this is a highly frustrating process, which will be addressed in the next section.

4. The Forgotten Sons of the State

In the wake of war, the former FAM soldiers had great expectations of the government. Where did these expectations come from? And how is the veterans' current disappointment formulated? Here, I discuss the perspective of rurally based FAM veterans from Magude, a rural district in the southern province of Magude. Below, I focus on a group of urban-based FAM veterans. Veterans from both groups expressed grievances in relation to the state and the FRELIMO party, but their position towards the state and FRELIMO was very different.

The veterans who participated in the research in Magude were predominantly men. They were recruited by the FAM in their late teens and twenties and fought between two to eleven years. I spoke to veterans linked to veterans' associations, but also to those who were not affiliat-ed.¹⁵ Almost every ex-FAM soldier I spoke to in Magude, women and men, expressed some sort of disappointment in relation to how the government and the FRELIMO party had been treating them. They had fought, risking their lives and limbs, for the government ruled by FRELIMO and afterwards felt that the government should have taken care of them.

Evaristo had been fighting in a special unit of the government forces deployed to execute demanding missions such as attacks on RENAMO bases. After telling me about the fights and the fear, he expressed his disillusionment with his current life:

I want to tell you about more suffering. Even today our lives are like in the troops: they are precarious. It is normal to find an ex-combatant on the streets. There are those who say, 'It would have been better if the war had not ended, at least we had [something] to eat in those days.' Ex-combatants are considered nothing. Now they confirmed the new Ministry. But I don't know if this will change things for better or worse.

Evaristo refers to the state of poverty many ex-combatants find themselves in. Similar to most Mozambicans, many ex-combatants live below any imaginable poverty line. Often ex-combatants attribute their poverty to the education they missed out on because of their participation in the war. While this is a tragedy, one must ask to what extent this is related to their history as ex-combatants. That is, the education system, especially in the rural areas, broke down in the entire country because of the war (Hall and Young 1997). Furthermore, job opportunities were and still are limited in Mozambique (Hanlon and Smart 2008, 15). When the ex-combatants demobilized, they joined the ranks of the unemployed and sustained themselves like most Mozambicans: by small-scale agriculture.¹⁶

The poor and formally unemployed ex-combatants in particular complained about the government's conduct toward veterans, and it was they who demanded pensions with the most tenacity. Those who had paying formal jobs were less negative about the government and FRELIMO. They were a small minority, though. This seemed to coincide with the ranks ex-soldiers had during the war. Higher-ranked veterans were more likely to have gained a pension and had more job possibilities after the war because of their connections, particularly with FRELIMO party officials. The overwhelming majority were, however, rank-and-file soldiers (Pardoel 1994). It is these ex-soldiers who today feel disregarded by the government and who see the possibility of receiving a pension and other state resources as a way out of their misery.

However, this demand for pensions is not only a material issue; it also involves certain symbolic aspects that must be tied to a specific historic, political and cultural context. I want to highlight four aspects that shape veterans' relationships with the state (and with FRELIMO): feelings of entitlement; the comparison ex-combatants make between themselves and the antigos combatentes; the conceptualization of the relationship as that of a father and a son; and the logic of patrimonialism.

First, the ex-soldiers regard themselves as distinct from non-veterans, as they feel entitled to some kind of consideration from the Mozambican government and the FRELIMO party. This consideration is material in the sense of pensions, but they also feel entitled to a more symbolic kind of consideration, namely the acknowledgement of their existence and the value of their service during the war. To illustrate, I recall Fernando, an ex-FAM soldier, who was not yet receiving his pension, but was quite sure his name would appear soon. After the interview, he wanted to clarify one more thing:

I do not feel well seeing our government. I left children here when I went to war. The government did not do anything, not a little thing. I feel let down. I am very angry. In wartime, we were taken care of and now not even a 'good morning.' As if we never existed.

It is clear from Fernando's words that he feels that he sacrificed his family life for the government, but got nothing in return. He does not only want to be "taken care of," but also needs to be recognized for his existence and his service during the war. This resonated across many interviews. Referring to the recent FRELIMO campaign slogan "It is FRELIMO that did, and it is FRELIMO that does," Jose, a disabled ex-combatant said: "They may say it was them that did, while it was actually us that did! We defended against the banditry [RENAMO]." Jose conveyed a certain pride in his (or the army's) wartime achievements, but he also says that this is not acknowledged by FRELIMO. In fact, in Jose's opinion FRELIMO uses the war against RENAMO as campaign material, while it was not the FRELIMO leadership or elite that fought in that war, but rather Jose and his fellow soldiers.

Similar feelings of bitterness are found among groups of former fighters who were not even incorporated into the national DDR process, like former militias. They were local men (and occasionally women) who were trained and armed by the government army and fought side-by-side with the government soldiers against RENAMO. They participated in the same 'operations' and were wounded, killed and traumatized just as much. Yet, they were not incorporated into the DDR program of ONUMOZ and, as far as the government is concerned, they are not veterans. Even more precarious were the cases of ex-militias who were disabled during the war. Aparacio Matavele, a local government official who took up arms against RENAMO, was on his way from his village to the district capital Magude to report a victory of his militias over a group of RENAMO combatants, when his car hit a landmine and he lost his two legs. Since then, he has been confined to a wheelchair, offered to him by "the Germans" (probably the German Cooperation, GTZ).

But the government is not giving anything. I reach 100 percent of deficiency, but according to the Ministry of Defense we were under the rule of the local administrator, so he has to pay my military pension. I was a chefe de posto [government authority of a locality], I had to enter as a soldier, I was a great leader, and now, look at my house!

He was referring to his small cabana where he and a woman were living. Even though Aparacio feels pride in taking part in a certain episode of history, he also feels betrayed by the government he was fighting for. The militias are an example of the blurring of civilians and combatants during war, but also how the governments and DDR programs, like the one ONUMOZ executed, develop certain categories, on which they base their policies, including some and excluding others (Broch-Due 2005, 14).¹⁷

Secondly, I want to underline again the importance of the image of the antigo combatentes for the demobilized FAM soldiers. The veterans entered the military when the socialist state was seen as the "deliverer of well-being" (Taju 1998, 54), but their expectations were fuelled by the privileged position of antigos combatentes, who received great benefits from the state. They soon found out that their position was a rather different one. Demobilized FAM combatants would often make a comparison between them and the antigos combatentes, almost always referring to the economic benefits that the antigos combatentes were enjoying and they were not. Such benefits do not go unnoticed in poor rural communities like Maringue and Magude.

Similar to the antigos combatentes, they feel they have fought for their patria. "The weapons that the antigos combatentes used were the same as the ones we took up," said a former FAM fighter bitterly. This is not only a claim for material benefits, but also for recognition. The liberation struggle was seen as a just and glorious war against the Portuguese oppressor and its veterans are seen as "good soldiers." The civil war, for many reasons, was a more awkward war and is generally not discussed. The veterans from this war are not in any way remembered. This is especially frustrating for a cohort of FAM veterans who entered the army voluntarily right after independence. The antigos combatentes were their heroes, and in the rush of the revolution they also wanted to defend their country. However, the FAM soldiers ended up in a completely different war, often unequipped, underfed and unprepared for RENAMO's guerrilla warfare and terror tactics (Schafer 2007). After the war, they found themselves not the heroes of the day, but rather as an unwanted and unappreciated category. The presence of the antigos combatentes is, for the demobilized soldiers, a reminder of what they lack (material benefits), but also of what they are not, namely national heroes. Yet, simultaneously these antigos combatentes form a promise of what the FAM veterans may possibly get in the future.

A third aspect of the veterans' relationship with the state is the image of the government, or FRELIMO (often used interchangeably), as the father and the ex-combatant as the son. As Jose, who I quoted above, said when I asked him about his political affiliations: "FRELIMO is like a father who should take care of his sons. Where else would we go?" This is a more common metaphor for the relationship between soldiers and the state, especially in the African context. This suggests that the state is seen as a "care-taker" as suggested by Taju (1998). But there is more to it, as Schafer (2007, 156-157) analyzed using Michael Schatzberg's (2001) understanding of the distribution of rights and responsibilities through the use of paternal and familial metaphors in an African context. Schatzberg (2001) noted that the father-son dynamic in relation to states and citizens in Africa implies that the father may "consume" the services/work of the son, but only to a certain extent and in return for protection and caretak-ing. By doing so, the father/government obtains legitimacy as an authority figure. The desmobilizados' claim for "nurture and nourishment" from their "father" may, thus, not only feel a particularly strong entitlement, (Schafer 2007, 156), but also undermines the legitimacy of the state, as it does not fulfill its responsibilities of a caretaker.

This is closely related to the fourth and final point I want to make: the veterans' relationship with the (local) state is characterized by a relationship of mutual dependency. This can be understood through the logic of patrimonialism. Following Weber (1947), Pitcher et al. (2009, 129), patrimonialism is characterized by "systems in which political relationships are mediated through, and maintained by, personal connections between leaders and subjects, or patrons and clients."¹⁸ As I have mentioned above, the success of an application may depend on a veteran's connections with veterans' associations, the local government and the FRELIMO party. Local political and patron-client dynamics are crucial in determining whether veterans may obtain access to funding. Some of these relationships are more personal and may be more "profitable" than others. Yet, these relationships all stem from the same precondition: FRELIMO membership.

With no serious political opponent, FRELIMO has practically established a party-state and is, thus, in control of dividing subsidies. While it is at the national level that decisions are made about pensions and more, the process of application happens at the district level where one needs to "be with FRELIMO" to get things done. Therefore, ex-combatants would not dare to lose their FRELIMO membership, hence the words of Jose above: "Where else would we go?" Besides all their frustration, most demobilized soldiers would not abandon their party or even criticize it, as this would preclude all future possibilities to state resources. In other words, as Pitcher et al. (2009, 138-139) point out, there is a voluntary compliance of subjects with the authority presupposing the degree of legitimacy these networks enjoy, as both patrons and clients are aware of their mutual dependence and reciprocity (see also Chabal 2009, 105).

Indeed, the dependency is mutual; FRELIMO's (local) leaders need the desmobilizados as their local supporters. This also holds true for the party at large. As I have shown above, demobilized soldiers are a specifically targeted group during electoral campaigns, when they are made promises, particularly concerning pensions. One could argue that this hope is kept alive by the government in return for support, well-aware that veterans have little room to maneuver the situation to their advantage. All four dynamics mentioned here involve a lack of recognition by the state, but simultaneously feed into the veterans' hope to receive a pension one day. A great share of the veterans in this article remains inside the structure of the relationship with the state and veterans' associations, characterized by frustration, hope and mutual dependency. Some veterans managed to make this relationship "work" for them by establishing patron-client relationships. But most remained in a rather frustrating and limited social and political position. I want to end the discussion by describing a group of veterans who did find a niche in which to pressure the state through the threat of violence.

5. "We will go to the Streets": The Embodiment of Threat

The veterans' attitude toward FRELIMO in Magude was quite different from the attitude I observed among the members of the Foro de Desmobilizados de Guerra de Moçambique (Forum of Demobilized Soldiers of the War of Mozambique, FDGM) in Maputo. The FDGM consisted of a group of urban-based former FAM soldiers who had previously participated in other veterans' associations, but who had not been able to make this involvement work to their benefit. They did, however, establish their own forum, bringing veterans' issues to the table in a novel, but also employed more aggressive tactics. Due to their previous involvement in other associations, FDGM also managed to participate in the negotiations between the veterans' associations and the government.

The FDGM was profoundly discontent with the state and the other veterans' associations. During a group conversation with members of the FDGM, I made the mistake of referring to them as "ex-FRELIMO fighters." One of them vigorously corrected me:

We did not fight for FRELIMO! We were fighting for the state. We were protecting our patria, our land, our state, the national line. Not the party. The problem is that nowadays the party and the state are together. It is us who suffer, not those old men who are enjoying themselves [ gozar] because of the war.

These FDGM members were rather exceptional in their specific and open criticism of FRELIMO, especially the party's elite. FDGM's leader, Herminio dos Santos, said:

Our politicians are bandits. Corruptos! Look at Chissano [former President of Mozambique], a bandit, and now he received a price, a price that is based on our peace. That money should be ours. It was not he who brought peace. It was us who stopped shooting en masse in 1992. We should get a price like that. If we start shooting, that is the end of the price.

In 2007, former President Joaquim Chissano won the Mo Ibrahim prize of five million USD for African Leadership, partly for his role in the peace negotiations.¹⁹ This enraged the FDGM leadership. They feel the war was fought and the peace was achieved over their backs, while they were left with nothing. Literally, as they did not receive any pension, but also symbolically, as they were never rewarded for their share in Mozambique's peace process. The anger of the FDGM leadership is directed at the government and political party they were fighting for. Highly different from the veterans in Magude and the other veterans' associations, the FDGM publicly accuses the FRELIMO government of neglect and corruption, and openly threatens violence. Herminio Dos Santos said in an interview with a local newspaper the following: "It's not an effort for the desmobilizados to put in danger the democracy or the peace."²⁰ Such messages are met by the government, the other veterans' associations and the media with a mixture of ridicule, but also fear.

Their aggressive attitude makes FDGM an unwanted participant in these negotiations in the eyes of other veterans' representatives, but their presence seems strategic as well. FDGM voices certain discontent that other veterans are afraid to openly talk about. Organizations such as AMODEG, mentioned above, are partially paid by the government and, therefore, probably reluctant in their criticism. Publicly, AMODEG's leaders urge FDGM to take a more peaceful approach. At the same time, FDGM may provide other associations with leverage in their negotiations with the government in two ways. First, the militant FDGM may legitimize the existence of veterans' associations like AMODEG and PROPAZ, which may take up the role as mediator between FDGM and other enraged "dangerous" veterans and the government. During and in the direct aftermath of the DDR process in Mozambique, veterans' associations used seemingly uncontrolled former combatants to pressure the government and international organizations to the veterans' (association) advantage (Schafer 1998, 2007, 126). It may be argued that FDGM currently occupies this role.

Second, FDGM's most effective strategy is, however, to threaten to take desmobilizados to the streets: "We sent a card to the government [to request permission to demonstrate]. But they denied our demonstration two times. They don't want the international community to discover that the desmobilizados are discontent. It is a secret. They pretend everything is all right." Since then, the FDGM claimed several times to be organizing a demonstration across the entire country, which never took place. It is not clear if FDGM has the capacity to organize nationwide demonstrations, but the government and the other associations of demobilized soldiers are not willing to find out, as they are afraid that demonstrations will turn violent, even though they are peacefully intended by FDGM. The most recent demonstration was planned in February 2011, but was called off after the government and the representatives of the organizations of demobilized soldiers signed an accord that would lead to a new law, obliging the government to pay pensions to demobilized combatants.²¹ An analysis of the relations between the demobilized soldiers and the government over the last three years shows, however, that this is a recurrent dynamic of the demobilized combatants, who threaten to organize demonstrations while the government makes concessions and promises to dissuade the ex-combatants from mobilizing.²²

The fear of demobilized soldiers going to the streets became clear after the riots that started the 1^st of September 2010, lasting for three days. The main roads heading out of the cities of Maputo and Matola were blocked, often by burning tires; and shops were sacked, because of the rise in the price of bread. It was reported that 13 people died by police fire during clashes with demonstrators. One of the rumors that spread after the riots was that desmobilizados had organized them. While this was not true, in a newsflash on TVM a month later on the 15^th of October, a woman said the following regarding the new statues concerning the demobilized soldiers: "I hope there will be no more manifestations now," implicitly linking the "strike" to the demobilized soldiers.²³

The position of FDGM members raises several questions: Why do they not fear criticizing the government while other veterans are afraid of losing their pension? Do they create particular momentum? How seriously should they be taken? It is not my aim to answer these questions, but merely to address an emerging phenomenon in Mozambique. I do think a case can be made to see FDGM as entering a niche where Mozambican veterans once more exercise agency through the threat they embody. The recent changes in policy regarding demobilized soldiers must not be based on a revision of the country's history or a re-evaluation of the demobilized soldiers or a claim for due compensation. Rather, as I argue, it has been the threat of violence that former combatants embody that has shaped the negotiations between the associations of demobilized soldiers and the government.

Conclusions

Studying the position of ex-combatants in relation to the government in Mozambique reveals much about the politics in this country and how they are influenced by past wars. Even though there were no official ways of dealing with the civil war, this article shows that this does not mean that this episode does not influence current official memory processes.

The governments' policies regarding pensions for veterans demonstrate how historical narratives interact with procedure and policy, generating and sharpening categories of people. Such interactions determine the state's actions and people's access to the resources of the state, shaping to a certain degree everyday life of the subjects. Parallels can be drawn between the struggle of the veterans in Mozambique and in neighboring Zimbabwe, where, as Alexander et al. (2000) and Kriger (2003) noted, veterans of the national liberation war have been struggling for access to state resources, criticizing government officials for living in luxury, while "the forgotten fighters were languishing in poverty" (Kriger 2003, 256). However, there is one main difference: while the veterans in Zimbabwe had been fighting in the liberation war, which was defining for the government's legitimacy (Alexander et al. 2000, Kriger 2003), in Mozambique the status of the war the demobilized soldiers fought is more ambiguous.

The national policies and attitude of the government and FRELIMO party have been sources of dissolution and discontent for FAM veterans. The veterans' grievances and negotiations revolve particularly around pensions, which are not only a very welcome material asset, but also a symbolic recognition of their status as war veterans and the work they did for the Mozambican nation, in particular for FRELIMO. The veterans see themselves as the "forgotten war heroes." Narratives of the past form one factor among many, shaping the veterans' relationship with the state, which is also formed by an image of the veteran based on the antigo combatente, and local dynamics, such as Mozambique's daunting bureaucracy and patrimonial logics in local politics.

In this article, I showed that most veterans in Magude remain inside the structure of the relationship with the state and veterans' associations, characterized by frustration, hope and mutual dependency. But I also explored the role of a militant veterans' association moving into a niche where Mozambican veterans seem to exercise agency through the threat of violence. The recent changes in policy regarding demobilized soldiers must not be understood in the light of a revision of the country's history or a re-evaluation of the demobilized soldiers or a claim for due compensation. Rather, as I argued, it has been the threat of violence that former combatants embody. This underlines a more general claim that former soldiers do not "fade away" (cf. Alden 2002). Instead, they have a distinct and changing position in a country's history and society.

Comentarios

^* The findings presented in this paper are part of a larger research project on long-term reintegration of veterans in Mozambique. The data presented in this article are based on over eighty interviews with former combatants, members of veterans' associations, civil servants and policy-makers, during eighteen months of ethnographic fieldwork in Maputo and two rural districts, Maringue (Sofala province, central Mozambique) and Magude (Maputo province, southern Mozambique), conducted in 2008, 2009 and 2010. These rural districts represent highly different histories of war and politics. They were chosen not as representative districts for Mozambique, but rather to compare and contrast certain findings and signalize different processes in relation to veterans, local history and politics. In Maringue, the focus of the research was on RENAMO veterans, although several FAM veterans were interviewed as well. In Magude, data were gathered on FAM veterans. It is this data that is predominantly used in this paper.

¹ The research for this article was facilitated by funds from the Dutch Organization of Science (NWO). I want to thank Prof. Dr. Ton Robben, Dr. Yvon van der Pijl and the fellow PhD candidates from the Department of Cultural Anthropology at Utrecht University for their helpful suggestions on initial drafts of this article. I also want to thank two anonymous reviewers for their insights and comments.

² See: Noticias, 12/10/08, "Governo Busca Soluções para os Desmobilizados"; http://comunidademocambicana.blogspot.com/2010/10/desmobilizados-de-guerra-ja-anunciaram.html, Canal de Moz 14/10/10; CanalMoz, 16/08/09, "Desmobilizados de Guerra Lançam Campanha de Manifestações"; Canal de Moçambique, 06/10/10, "Adiamentos na Resolução dos Problemas dos Desmobilizados Ameaçam Paz no Pais"; Manifestação Divide Desmobilizados" MAGAZINE, 27/10/10; Hanlon, Mozambique 178, News and Clippings, 10 March 2011; http://www.canalmoz.co.mz/hoje/18843-governo-faz-cedencias-para-evitar-manifestacoes-dos-desmobilizados-de-guerra.html, 22 February 2011, accessed on 10 March 2011.

³ In this article, I do not touch on the position of former RENAMO combatants, as this is a distinct group in terms of policy and memory politics.

⁴ The vast majority of people died of causes related to the war, such as hunger and disease. Food shortage was, as far as I know, also the main reason for displacement. It is estimated that RENAMO killed between 50,000 and 200,000 people (Hanlon 1991; Hultman 2009); however, exact figures are absent, as statistics about the war are imprecise and prone to exaggeration (Nordstrom 1997, 48; Igreja et al. 2008).

⁵ There were other programs in place at the time, such as income-generating projects for veterans from GTZ and OIM, but these programs did not achieve long-term employment or income-generating capacity. I refer to Schafer (2007, 123-138) for an overview of the external assistance to demobilized soldiers in the 1990s. Some of the demobilized soldiers I spoke to had participated in vocational training courses during their time in the assembly area or just after demobilization; however, these had little impact on their lives.

⁶ In this article, I do not touch on the position of former RENAMO combatants, as there are very different issues at hand here. These will be explored elsewhere.

⁷ The construction of a nation's history is, of course, always a selective process. For an overview of these debates, see Cohen (2001, 243) and Antze and Lambek (1996).

⁸ To be clear, the liberation war did not end with a victory by FRELIMO or the Portuguese military. Rather, it ended due to the carnation revolution in Portugal and the following change of regime that ended all Portuguese wars in Africa (Hall and Young 1997). Nevertheless, FRELIMO does present the end of the war as a Mozambican victory in moral and military terms.

⁹ Still, Schafer (2007, 125) noted that 91 percent of the USD 5.8 million of social assistance went to ex-combatants.

¹⁰ Interviews with Jacinta Jorge, Director of ProPaz, Maputo and an interview with Xaderique Paulinho, the National Director of Reinsertion of Desmobilizados, Ministry of the Combatant, Maputo.

¹¹ The former RENAMO combatants formed, however, a minority within ADEMIMO. In Maringue, most ex-RENAMO combatants had some scars from bullet wounds, but only a few of them were actually disabled. It may well be that the government forces were better able to deal with the soldiers who were (severely) wounded.

It is my informed guess that soldiers of RENAMO died more easily and there were little resources to deal with the disabled veterans, possibly quickening death.

¹² Schafer (2007, 143) estimated that a small minority of the ex-combatants choose to join an association of demobilized soldiers. The leaders themselves have no clear idea about the number of members. Several times I have heard the number 33.000, but this seems to be the total potential number of members, as it may be the number of living ex-combatants in Mozambique.

¹³ See also newspaper Savana, 09/10/09: "FRELIMO 'atira' Responsabilidade aos Funcionarios do Estado".

¹⁴ This must be situated in a context of a country with extreme needs and limited funding, constrained by donor earmarking (Renzio & Hanlon 2007, 21).

¹⁵ Initial contact with veterans in both Magude and Maringue was made through AMODEG and PROPAZ. A snowball method was used to establish contact with other veterans outside the network of the veterans' associations.

¹⁶ It can be argued that former FAM soldiers had even a slightly advantageous position in the job market. In Magude, I observed that many FAM veterans found work in the private security sector, particularly in the large sugar factory near the district's capital.

¹⁷ Another specific group was the approximately 16.000 government soldiers who demobilized before the signing of the peace agreement. They felt deprived of the allowance and other benefits that those demobilized by ONUMOZ received. As they do not differ much from the soldiers who demobilized in 1994, I did not elaborate extensively on this group.

¹⁸ This is reflected in the distinction Weber (1947) made between: legal domination and patrimonial domination. While the first is based in systems of legal bureaucracy and impersonal ties, in the latter authority is based in personal relationships.«« Volver

¹⁹ Website of the Mo Ibrahim Foundation, section of prize laureates, accessed on March 10, 2011. http://www.moibrahimfoundation.org/en/prizelaureates/the-ibrahim-prize/prize-laureates/the-ibrahim-prize-winner-2007-joaquim-alberto-chissano.html

²⁰ Quoted in Canal de Moçambique, 06/10/10, "Adiamentos na Resolução dos Problemas dos Desmobilizados Ameaçam Paz no Pais".

²¹ Hanlon, Mozambique 178, News and Clippings, 10 March 2011, see also http:// www.canalmoz.co.mz/hoje/18843-governo-faz-cedencias-para-evitar-manifestacoes-dos-desmobilizados-de-guerra.html, 22 February 2011, accessed on 10 March 2011.

²² See newspaper articles in: Noticias, 12/10/08, "Governo Busca Soluções para os Desmobilizados"; Canal de Moz, 14/10/10, accessed January 23 2012, http:// comunidademocambicana.blogspot.com/2010/10/desmobilizados-de-guerra-ja-anun-ciaram.html,; CanalMoz, 16/08/09: "Desmobilizados de Guerra Lançam Campanha de Manifestações"; Canal de Moçambique, 06/10/10: "Adiamentos na Resolução dos Problemas dos Desmobilizados Ameaçam Paz no Pais"; MAGAZINE, 27/10/10: "Manifestação Divide Desmobilizados".

²³ Mozambican sociologist Carlos Serra pointed out that the participants in the manifestations were predominantly young unemployed men (but also women and children) whose objective was not to 'stop' the state, but to make a statement against a "distant" government and the dysfunctional public sector. In: Noticias, 21/10/10, "Conselho dos Desmobilizados Desaconselha Manifestações".

References

1. Alden, Chris. 2002. Making old soldiers fade away: Lessons from the reintegration of demobilized soldiers in Mozambique. Security Dialogue 33 (3): 341-56.         [ Links ]

2. Alexander, Jocelyn, JoAnn McGregor, and Terence Ranger. 2000. Violence and memory: One hundred years in the 'dark forests' of Matabeleland, Zimbabwe. Oxford: James Currey.         [ Links ]

3. Antze, Paul and Michael Lambek. 1996. Tense past: Cultural essays on trauma and memory. London: Routledge.         [ Links ]

4. Broch-Due, Vigdis, ed. 2005. Violence and belonging: The quest for identity in post-colonial Africa. New York: Routledge.         [ Links ]

5. Buckley-Zistel, Susanne. 2006. Remembering to forget: Chosen amnesia as a strategy for local coexistence in post-genocide Rwanda. Africa 76 (2): 131-150.         [ Links ]

6. Chabal, Patrick. 2009. Africa: The politics of suffering and smiling. London: Zed.         [ Links ]

7. Cobban, Helena. 2008. Accountability, remorse and reconciliation: Lessons from South Africa, Mozambique and Rwanda. In Pathways to reconciliation: Between theory and practice, eds. Philipa Rothfield, Cleo Fleming and Paul A. Komensaroff, 127-138. Aldershot: Ashgate.         [ Links ]

8. Cohen, Stanley. 2001. States of denial: Knowing about atrocities and suffering. Cambridge: Polity Press.         [ Links ]

9. Granjo, Paulo. 2007. Limpeza ritual e reintegração pós-guerra em Moçambique. Análisis Social 123-44.         [ Links ]

10. Hall, Margaret and Tom Young. 1997. Confronting leviathan: Mozambique since independence. Athens: Ohio University Press.         [ Links ]

11. Hanlon, Joseph. 1991. Mozambique: Who calls the shots? London: James Currey.         [ Links ]

12. Hanlon, Joseph. 1996. Peace without profits: How the IMF blocks rebuilding in Mozambique. London: James Currey.         [ Links ]

13. Hanlon, Joseph and Teresa Smart. 2008. Do bicycles equal development in Mozambique? London: James Currey.         [ Links ]

14. Honwana, Alcinda. 2006. Child soldiers in Africa. Philadelphia: University of Pennsylvania Press.         [ Links ]

15. Hultman, Lisa. 2009. The power to hurt in civil war: The strategic aim of RENAMO violence. Journal of Southern African Studies 35 (4): 821-34.         [ Links ]

16. Igreja, Victor. 2008. Memories as weapons: The politics of peace and silence in post-civil war Mozambique. Journal of Southern African Studies 34 (3): 539-556.         [ Links ]

17. Igreja, Victor, Béatrice Dias-Lambranca, and Annemiek Richters. 2008. Gamba spirits, gender relations, and healing in post-civil war Gorongosa, Mozambique. Journal of the Royal Anthropological Institute (NS) 14: 353-71.         [ Links ]

18. Kossler, Reinhart. 2007. Facing a fragmented past: Memor y, culture and politics in Namibia. Journal of Southern Africa Studies 33 (2): 361-82.         [ Links ]

19. Kriger, Norma J. 2003. Guerrilla veterans in post-war Zimbabwe: Symbolic and violent politics, 1980-1987. Cambridge: Cambridge University Press.         [ Links ]

20. Malkki, Liisa H. 1995. Purity and exile: Violence, memory, and national cosmology among Hutu refugees in Tanzania. Chicago: University Of Chicago Press.         [ Links ]

21. McGregor, JoAnn. 2002. The politics of disruption: War veterans and the local state in Zimbabwe. African Affairs 101 (402): 9-37.         [ Links ]

22. Metsola, Lalli. 2010. The struggle continues? The spectre of liberation, memory politics and "war veterans" in Namibia. Development and Change 41 (4): 589-613.         [ Links ]

23. Nazarea, Virginia D. 2006. Local knowledge and memory in biodiversity conservation. Annual Review of Anthropology 35: 317-335.         [ Links ]

24. Nora, Pierre. 1989. Between memory and history: Les lieux de mémoire. Representations 26: 7-24.         [ Links ]

25. Nordstrom, Carolyn. 1997. A different kind of war story. Philadelphia: University of Pennsylvania Press.         [ Links ]

26. Pardoel, Ton. 1994. Demobilization in Mozambique: Socio-economic profile of the group of 92,881 demobilized soldiers. Maputo: UNDP.         [ Links ]

27. Pitcher, Anne M. 2006. Forgetting from above and memory from below: Strategies of legitimation and struggle in postsocialist Mozambique. Africa 76 (1): 88-112.         [ Links ]

28. Pitcher, Anne, Mary H. Moran and Michael Johnston. 2009. Rethinking patrimoni-alism and neopatrimonialism in Africa. African Studies Review 52 (1): 125-56.         [ Links ]

29. Renzio, Paolo de & Joseph Hanlon. 2007. Contested sovereignty in Mozambique: The dilemmas of aid dependence. Working Paper 25, Managing Aid Dependency Project. Oxford: Oxford University.         [ Links ]

30. Schafer, Jessica. 2007. Soldiers at peace: Veterans and society after the civil war in Mozambique. Basingstoke: Palgrave Macmillan.         [ Links ]

31. Schafer, Jessica. 1998. 'A baby who does not cry will not be suckled': AMODEG and t he reintegration of demobilized soldiers. Journal of Southern African Studies 24 (1): 207-222.         [ Links ]

32. Schatzberg, Michael. 2001. Political legitimacy in middle Africa: Father, family, food. Bloomington: Indiana University Press.         [ Links ]

33. Shaw, Rosalind. 2007. Displacing violence: Making pentecostal memory in postwar Sierra Leone. Cultural Anthropology 22 (1): 66-93.         [ Links ]

34. Taju, Gulamo Amade. 1998. The Social Integration of Demobilized Ex-Combatants in Mozambique. MA Thesis. University of Witwatersrand: Johannesburg.         [ Links ]

35. Weber, M. 1947 [1922]. The Theory of Social and Economic Organization. Translated by A.M. Henderson and Talcott Parsons. New York: The Free Press.         [ Links ]