6 - The Diffusion of Repertoires of Collective Action

Summary

Chapter 6 assesses why ‒ though facing similar stalemates and other structural challenges ‒ two adjacent districts in Zambézia province experienced the diffusion of militias so differently. The chapter shows that communities learned from neighboring communities about how militias formed and “diffusion agents” migrated to spread the message of militia success, which helped initiate militia diffusion. However, “sustained diffusion” ‒ the persistence of militia activity in a district and integration of the militia into the local security apparatus ‒ depended on the cohesion of elites. The chapter explores the validity of the argument by analyzing the diffusion of the Naparama to a district in Nampula province.

Stories and rumors about Naparama’s successes across the region featured prominently in the narratives I heard about the militia. Naparama commanders and initiators in Murrupula, the center of Naparama activity in Nampula province, told me that people came to their district to receive the Parama vaccine because they had heard that the “war did not reach [Murrupula].”¹ They suggested that residents from surrounding districts sought to form their own militia groups because they saw how Naparama was able to ward off the rebels in Murrupula. The same dynamic evolved in neighboring Zambézia. After learning about Naparama’s effective response to the violence in adjacent districts, local residents wanted to form a group of their own. Zambézia’s Naparama leader Manuel António went “on foot if necessary, to ‘wherever the people call me to help’.”² And when Naparama forces sought to mobilize new followers in new places, they pointed to the stories of past successes: “[When] we reached Namacurra, we convinced people that the group that had come and retaken the town was [us, Naparama], so we [said we] want to [mobilize] more men.”³ Thus, much of the growth of Naparama was due to diffusion, the spread of rumors and stories, followed by the spread of a new form of collective action across community boundaries.

The core argument of this chapter is that militias tend to emerge in places that border on communities where militias already exist. This occurs through two mechanisms: (1) learning and (2) migration. Communities hear about and learn from each other’s experiences through the sharing of information. Knowledge about what happens in other communities during upheaval and war is often distributed through rumors and storytelling (Greenhill and Oppenheim 2017), in particular when communities are linked through social networks, such as ethnicity (Larson and Lewis 2017). Ethnic, ideological, cultural, and historical bonds between communities facilitate the spread of information, and thus the diffusion of collective action repertoires, in particular when these bonds are reinforced by population movements and the (temporary) migration of community residents and local elites between these communities.

Although rumors and storytelling facilitate the militias’ ability to spread to neighboring communities, they are not enough to sustain this diffusion. The sustained diffusion of community-initiated militias – that is, their integration into the local security apparatus as well as their ability to influence war dynamics – depends on conditions that I seek to identify and explore in this chapter. I suggest that militias integrate into local institutions when social, political, and military elites are relatively unified. Relative unity among elites has two main consequences. First, it prevents community-initiated militias from becoming the private army of an elite. And second, community residents and local elites have more trust in the ability of the militia to curb violence, and hence they support the new institution, ensuring its survival.

Using the methods of controlled comparisons and process tracing, I develop these arguments with an analysis of militia diffusion to two adjacent districts with similar geographic, demographic, and war-related characteristics in the Zambézia province of the central region of Mozambique. While the Naparama militia virtually replaced the state military in one district, the newly founded force quickly fell apart in an adjacent district. I then explore my argument further with a third case of Naparama diffusion to a district in Nampula province. The case studies show that migration and learning were the mechanisms that facilitated the initial militia diffusion, but that diffusion was only sustained because elite conflicts were absent and community residents and local elites had faith in the militia to improve the local security situation.

6.1 Militia Diffusion in Zambézia and Nampula Provinces

Much of the initial spread of the Naparama militia was driven by the travels of the main Naparama leaders who offered their services to communities, either on their own initiative or on the initiative of community residents requesting the militia leaders’ services. Naparama had two main leaders, one in Zambézia and one in Nampula province, and each was responsible for forming units across their province.

Coming from Nampula, António entered Zambézia province in 1989 in the Alto Ligonha area in Gilé district and first worked in Alto Molócuè district (see map in Figure 6.1) (Pereira 1999a, 82). In Alto Molócuè, António began offering his services to bus travelers seeking protection for their travels on roads often targeted by Renamo ambushes. He then formed small groups to attack Renamo strongholds in the northern part of the district. From a base in Nampevo, Naparama launched offensives into Mocuba, Ile, and Gilé (Pereira 1999a, 86). The base at Nampevo also served as a launchpad for Naparama’s epic attack on Renamo’s regional base at Muaquiua. In Nauela, a former Renamo stronghold, Naparama punished the population for supporting Renamo by pillaging their property. Consequently, Naparama lost the support of the population and António had to move further south (Pereira 1999a, 84). The militia then spread, seemingly along the main N1 highway, into the Ile, Gilé, and Mocuba districts.⁴ In Mocuba, António established his main headquarters, later moving them to Nicoadala. He also formed Naparama militias in the districts of Pebane, Maganja da Costa, Namacurra, Nicoadala, Lugela, Namarrói, Milange, Gurué, and Inhassunge.

Figure 6.1. Map of initiators and the diffusion of Naparama militias across Zambézia and Nampula provinces

Note: This map presents an overview of Naparama leaders and the districts in which they formed Naparama units in the period between 1988 and 1992.

(Cartography by Sofia Jorges)

In Nampula province, the traditional healer Ambrósio Albino from Nampula city claimed that he and his nephew Silva Taitosse Mirasse had brought the Parama vaccine to Murrupula district, the center of Naparama activity in Nampula. His nephew then responded to requests from communities in Mecubúri and Namapa districts to form militias there. Murrupula’s eminent traditional healer, Nampila, became an important Naparama leader in Murrupula and neighboring districts, and mobilized militias in Moma, and presumably also in the coastal districts of Mogovolas and Angoche, at the request of their respective communities. In 1990, communities who had heard about Nampila’s work called him to work in Zambézia, in Pebane district close to Nampula province, where he disappeared in 1991 (see overview of Naparama leaders and the districts in which they formed Naparama units in Figure 6.1).⁵

The first part of this chapter tells the story of how the militia spread to two adjacent districts in Zambézia province. In 1990, Zambézia’s Naparama leader António arrived in Lugela and Namarrói, two districts that had seen much Renamo activity due to their proximity to rebel sanctuaries in Malawi, the deep forests in which rebels could hide, and residents’ historical sympathy for opposition movements. Lugela and Namarrói both lie in Alta Zambézia, a mountainous region at the southern end of the Great Rift Valley populated by the Lomwe ethnic group (see map in Figure 6.2). Lugela and Namarrói districts had about the same number of inhabitants at the beginning of the war – 107,000 and 100,000, respectively, in 1980 – but Lugela had a much lower population density, covering twice the area of Namarrói (2,300 square miles). The districts were prone to political activity by the PRM, or Africa Livre, the opposition movement based in Malawi that united with the rebel group Renamo in 1982 (Vines 1991; Chichava 2007). These districts were also the first to be affected by Renamo’s war in Zambézia province.

Figure 6.2. Map of Lugela and Namarrói in Zambézia province

Note: Cartography by Sofia Jorges

As outlined below, government reports at the time identified Lugela and Namarrói as among those districts in Zambézia province in which the military, political, and social situation was the most critical. In 1986/1987, Renamo occupied both district towns for many months, and the rebels maintained control over about two-thirds of the area of each district for most of the war. Frelimo’s radius of action was reduced to five miles in the towns, and Renamo units frequently attacked the towns and surrounding areas. In both districts, the army contingent was small and the existing troops faced a lack of military supplies. Necessary reinforcements from the military headquarters in Mocuba – more than forty miles away from each district town – were slow and often arrived too late to effectively respond to Renamo attacks. The population frequently fled to the bush and, particularly in Lugela, suffered from severe hunger. The rebels’ strong presence during the war is reflected in the electoral victory of the rebel-turned-opposition-party Renamo in the first multiparty elections in Mozambique in 1994. In Namarrói, Renamo gained 53 percent of the vote for parliament (compared to 30 percent for Frelimo), and in Lugela 56 percent (compared to 33 percent for Frelimo). In all of Zambézia province, Renamo gained 52.5 percent and Frelimo 31.5 percent (Mazula 1995).

Although both districts faced similar challenges, Naparama forces only spread sustainably to Lugela. In Lugela, Naparama constituted a large and influential force, while in Namarrói its size and activities were limited and only survived for a short time in one village. How can this difference in Naparama’s formation in the two districts be explained? What accounts for the variation in diffusion processes of community-initiated militias across districts in Zambézia province?

6.2 Sustained Diffusion of Naparama to Lugela

6.2.1 The Military Situation in Lugela

Even before Renamo reached Lugela district in 1982, the district had been an area of opposition activity. Starting in the late 1970s, PRM/Africa Livre units created insecurity in the rural areas. Local government reports from Lugela in 1979 link this “enemy activity” to infiltration from across the border in Malawi to the neighboring district of Milange.⁶ Milange borders on the administrative post of Muabanama in Lugela, about forty-five miles from the district town of Lugela. Muabanama was the location of the only communal village in the district, and thus a target of frequent attacks by those who opposed Frelimo’s policies after independence.⁷

Much of this early violence perpetrated by PRM/Africa Livre targeted the Frelimo state and focused on securing access to resources. The local administration accused “counter-revolutionaries” of mobilizing monetary support from local residents, stealing household goods, plundering health posts and stores, and burning administrative buildings and residences of Frelimo secretaries.⁸ On April 20, 1981, armed men murdered Muabanama’s administrative head and pillaged administrative buildings, the health post, and the main store.⁹

Renamo attacks from 1982 onward had similar targets as PRM/Africa Livre’s – symbols of Frelimo’s economic and social policy in the east and north of the district. In the fall of 1982, rebel units attacked and destroyed the Emochá and Madal tea factories and destroyed the bridge over the Munhamade River in the administrative post of Tacuane, about twenty miles to the east of the district town.¹⁰ Renamo occupied the tea factories and surrounding areas shortly thereafter and maintained control of these areas until the end of the war.¹¹

Renamo pursued both military and political objectives in Lugela. The rebels mobilized the people for their cause and – especially in the beginning – received much support. When the rebels convened community meetings, they declared that Frelimo did not treat residents well and offered them their protection. They also mobilized collaborators, who would join mujeeba, Renamo’s main informants and local police.¹² Initially, community residents welcomed the opposition movement and provided the soldiers with food.¹³ One of the strategies to gain popular support was to trick people into thinking that Renamo already controlled major towns – for example, by providing rare goods such as salt. In this way, the population felt it did not have a choice but to support the new ruling movement.¹⁴ As in other regions, local chiefs overwhelmingly supported Renamo because it promised that they would regain the power that Frelimo had taken from them at independence.¹⁵

However, Renamo’s military objectives guided their behavior toward civilians more than their political ambitions. Popular support decreased when Renamo started targeting civilians indiscriminately. While the rebels’ initial operations focused on the intimidation of Frelimo party and state representatives, indiscriminate violence increased after Renamo abducted and trained youths from the areas under occupation and expanded its military presence in the rural areas.¹⁶ Many of Renamo’s abductees were forced to carry pillaged metal to Malawi, where it was sold. Whoever refused was killed.¹⁷

During Renamo’s second military offensive across Zambézia in 1986, the military situation in Lugela worsened significantly and Renamo occupied the district for several months. Rebel units attacked and occupied the district town in March 1986, occupying it four more times until 1988. Whenever Frelimo retook the district town, its radius of action remained limited to about five miles. Access roads to surrounding villages were impassable, and travel to Mocuba, the next large town and location of Frelimo’s provincial military headquarters, was only possible on bush paths. Even when Frelimo had regained control of the district town, rebels constantly attacked, and control of the town and the surrounding areas changed frequently.¹⁸ As soon as the army was able to organize convoys with supplies from Mocuba, Renamo returned and assaulted the convoys and stole the supplies.¹⁹

The frequent attacks and change in territorial control meant community residents and government representatives perceived the situation as chaotic. Since they constantly had to flee their homes, the people lived a “nomadic existence.”²⁰ Community residents searched for security as well as food, as several years of low crop yield had led to severe hunger. After the March 1986 rebel attack and occupation, members of the local administration and most of the population fled into the mountains and then to the locality of Namagoa close to Mocuba, where the government set up a camp for the displaced on the grounds of a former Sisal fiber factory.²¹ Frelimo actively mobilized people to leave Renamo-held areas and assisted them in resettling to the camp in Namagoa throughout the town’s occupation.²² Namagoa was the only place where people could find some form of relief; the remaining areas of the district became ungoverned and anyone who tried to cross them risked being killed or arrested by Renamo or Frelimo armed forces:

From then on, the war spread, any person could be killed and people were confused; there was no government or anything. Many of us went to Mocuba, following Frelimo. Frelimo was over there, and here was Renamo. When you crossed the Lugela River to Namagoa, you were killed, and from here to there you were arrested. When someone left here to go to Mocuba, he was welcomed and we were given maize, sardines, beans, oil – many things. We were given clothes; men received pants and a coat. But when coming from there to here – [when] I forgot something, my own brother would inform [Renamo] and they would catch you and kill you out of fear that you came to collect intelligence.²³

The district administration was unable to cope with the situation. The district administrator repeatedly complained in his reports to the provincial government about the lack of military and logistical support from the province. Only one battalion – half of which consisted of newly trained local forces – protected the district.²⁴ A local government official estimated that 40 percent of the force was either malnourished or wounded and unable to participate in combat.²⁵ Thus, each time Renamo units approached the town, the district commander had to call for military reinforcement from Mocuba. Moreover, since the military did not have any supplies, it had to live off the population and often did so by force.

Frelimo’s grip on power over the district began to slip. Lugela’s administrator made it clear in reports that without provincial logistical and military support Frelimo control could no longer be upheld in the district: “If we delay any longer, the population will drink the enemy’s politics and their recuperation will be very difficult.”²⁶ As a solution, the district administrator proposed to train local forces as part of the country-wide “territorial security and defense system.” The situation only improved from 1990 onward, partly due to a new military commander who committed himself to not leaving the district until the situation had improved.²⁷

6.2.2 Initial Diffusion of Naparama to Lugela

As the district administrator recognized, only the local population’s involvement in the defense of the town held any promise of improving the security situation. Naparama units formed in the camp for the displaced in Namagoa, presumably in late 1990. By that time, António had moved from Alto Molócuè, a district close to Nampula province, to Mocuba, where he was working closely with the provincial command of the military, headquartered there.²⁸

Although the military and local administration later supported mobilization for Naparama forces, they were not involved in the initial diffusion of Naparama to Lugela. António arrived in Namagoa on his own initiative, then introduced himself to the local authorities and requested authorization to mobilize youths to be treated with the Parama vaccine. Joining was voluntary and mobilization occurred in cooperation with lower-level administrative elites such as neighborhood secretaries.²⁹ Naparama’s main task became patrolling the camp at night.³⁰ Later, the designated Naparama leader in Namagoa brought his group to Lugela to improve the security situation in the district town once it was retaken by Frelimo.

Community residents welcomed Naparama activities as they were convinced that António would be able to liberate them from the hardships of the war and help them return to their homes.³¹ Community residents’ narratives reveal the fascination with the Naparama forces:

[Manuel António] Naparama is someone to admire – [he is able to] fight with machete [and] spear. God helped him to help the population.³²

Naparama! They put in the work, they captured the enemy alive! They didn’t retreat. Renamo attempted to shoot them, but they didn’t die. Because this was a magical war! That’s why they did not have anything to fear, they were not afraid, they went there, captured the enemy, and brought him back together with [his war] material. In those days, Naparama worked [hard]. They [really] worked [hard].³³

The way community residents were intrigued by the power of Naparama and the group’s leader demonstrates that cultural bonds existed between Lugela and the communities in which Naparama previously formed. Community residents believed in António’s power to defend the people with nothing but spears and machetes, and they were even convinced that he had divine powers. Individuals in the community understood the cultural references Naparama made, and they believed that Naparama units were able to prevail in a “magical war” with Renamo. Thus, cultural bonds between communities facilitated Naparama’s initial diffusion.³⁴

Naparama’s diffusion was closely linked to stories about their success in one of Renamo’s main bases in the Mocuba district. Naparama units formed in Namagoa after António supposedly killed the famous Renamo regional commander Calisto Meque in the Muaquiua rebel base.³⁵ After Naparama’s successful attack on Muaquiua, António asked a Lugela community leader named Namudavale to take a group of Naparama combatants to Lugela.³⁶ As Namudavale narrated,

So it was at that moment [after the battle of Muaquiua] when Manuel António demanded, “You will take your combatants to your district of Lugela and help the [military] commander who is there.” So I brought my 128 men here to the district. When I arrived here, I increased the number of combatants. [António] gave me some of the medicine to vaccinate these men. When I arrived here, I asked for more men. Volunteers showed up, I gave them the vaccine; we went to fight in Mdedereia [and] we entered Mdedereia.³⁷

A resident of Namagoa explained how closely he and his community monitored events in the neighboring district of Mocuba, comparing the Mocuba residents’ situation to their own:

In the area where [Naparama] emerged, miracles occurred and the population came to live well, while we were in Namagoa and watched the good situation in which the population of Mocuba was in. Since [Mocuba] was a district that was very close [to Namagoa], [we were] talking about Naparama, what they were like, what their work was like.³⁸

In sum, the spread of information about Naparama enabled the diffusion of Naparama to Namagoa and subsequently to the town of Lugela. This was made possible by two causal mechanisms. The first was the (temporary) migration of diffusion agents – António and Namudavale – who linked communities with strong ethnic and cultural bonds. Due to these ethnic and cultural bonds, the social conventions underpinning Naparama’s power resonated in these communities. In addition, through the demonstration effect of successful missions in other areas – such as the battle of Muaquiua – people learned that they would be better off supporting a new Naparama force in their district.

6.2.3 Sustained Diffusion of Naparama to Lugela

The Naparama militia became an important force in Lugela, meaning that the local administration integrated the militia into the local security apparatus. Naparama collaborated closely with the local administration and the armed forces stationed in the district. The local administration supported the growth of the militia by aiding the Naparama members’ mobilization process, and when Naparama moved to the town of Lugela the force reached a size of about 240 men.³⁹ The recruits stayed in huts at the militia’s headquarters and went to their homes during the day when they were not sent on missions.⁴⁰

Naparama’s sustained diffusion was facilitated by the militia leader’s standing in the community and the faith that people put into the group as a result. Naparama leader Namudavale was a trusted community leader respected by local authorities, and thus elite conflicts did not arise. As a former combatant during the liberation struggle, Namudavale became secretary of the Frelimo party’s dynamizing groups after independence. When the district population relocated to the displacement camp in Namagoa, Namudavale organized and distributed food to the people – an important task in times of war when such resources are scarce. Namudavale remained a loyal Frelimo party member throughout the war and beyond, and at the time of our conversation he served as a member of the district party committee.⁴¹

In order to gain and retain the trust of community residents and local authorities, militias need to prove their service to the community. Namudavale’s Naparama forces proved their loyalty by retrieving the population that had been abducted by Renamo. The district administration was worried at first that Naparama would formulate political and/or monetary demands after the war. However, when militia members proved that they were willing to take risks to go to Renamo camps and managed to retrieve people who had been abducted by the rebels, the district administrator decided to tolerate the group’s activities:

The government did not support Naparama. However, as the people were seeing good results from Naparama’s actions, the government did not organize a force to eradicate Naparama. Because Naparama said over there is an advanced military position or camp – there is a Renamo base. Naparama was prepared to go to these bases. And in some way they brought with them some uncles from these areas. They brought some ladies from these areas back here. Taking them from [Renamo’s] base! The disposition of such an individual [António] is welcome. “Ah, that’s how [he does things]!” An individual who reached a dangerous place in order to retrieve his uncle. [They] retrieve your wife who had been with the bandit, with Renamo. Even retrieve your wife! From then on, there was no way for the government to prohibit the work of the Naparama.⁴²

The militia’s success in bringing back abducted family members generated extensive popular support for Naparama. The administration could therefore not prohibit Naparama’s operations without alienating community residents. In fact, the local government came to consider the militia as a last resort to fulfill the people’s desire for peace.⁴³ The situation was thus ripe for the community-initiated militias to form. As a government official pointed out, “[e]ven if Naparama had not appeared, the people would have asked the military for firearms [to defend themselves].”⁴⁴ The Naparama leader in Lugela explained to me that people could no longer rely on the army for help; it was a time in which everyone had to defend themselves at their own risk.⁴⁵ The militia demonstrated that it was able to take over defending the people and not threaten Frelimo control, and this is what allowed the local administration to make the militia part of the local security apparatus.

A relationship based on mutual trust also evolved between Naparama and the armed forces. When asked about allegations that the military stole relief aid during the war, Namudavale demonstrated great loyalty by strongly denying any such corruption existed within the armed forces. He also emphasized that Naparama and the Frelimo army always conducted joint operations, during which each group remained independent and respected each other:

Were the military headquarters always aware of Naparama operations?

They knew about them, we worked together. The army commander with whom I had spoken even told [the soldiers], “Work with [the Naparama] together as if you were brothers. Wherever you have to go, go there together, work, and return. You just can’t command [the Naparama]. Your commander can only command your own men. Though whoever disobeys his own commander, will die.” He did not tell us what to do – no, we went together.⁴⁶

Although Naparama had a degree of autonomy, the army monitored their activities closely, and often joined them on missions to Renamo bases or to accompany military convoys to Mocuba.⁴⁷ Lugela’s Naparama leader lived in the very center of town, close to the army commander, and his assistant controlled activities in the displacement camp by the river, about a mile from the town center.⁴⁸ Naparama also collaborated with the popular militias that possessed firearms. Ten militiamen and twenty Naparama combatants were assigned to each administrative unit of the district town, and in the late afternoons all of them were expected at the control post for their nightly patrol.⁴⁹

Former Frelimo combatants confirm the positive relationship between Naparama and the army stationed in Lugela:

We [Naparama and the military] considered ourselves friends, because [Naparama] were also fighting for the same purpose, and that is why there was no difference. In battle we were all military men, no one could say “I’m Naparama, I’m from the Frelimo army,” no, we were all soldiers, we were fighting and returned victorious.⁵⁰

This affirmation by Frelimo soldiers of the close collaboration between Frelimo and Naparama is remarkable, as in other districts and at the provincial level, Frelimo representatives usually downplayed their reliance on Naparama forces. Officially, Frelimo could not concede that the party compromised its commitment to a socialist ideology that is hostile to traditional forms of power (Dinerman 2006).⁵¹

The trust between the militia, the army, and the local government further found expression in the fact that when António died and Naparama’s strength seemed to wane, the local administration supplied Naparama with firearms:

So when the government saw that their leader had already died – [that Naparama] were dying – that is when it gave them firearms, saying that [Naparama] could not lose, they were already losing the drugs’ force, thus they gave them firearms, saying – epah! – you can now carry firearms. [Naparama] were fighting [Renamo]. It was already the end of that drug. [They were] already dying.⁵²

The local government needed Naparama’s support and trusted them sufficiently to hand them weapons when they realized that the powers of the Parama vaccine were not as reliable as Naparama claimed them to be.

In sum, the Naparama militia diffused sustainably to Lugela. The community-initiated militia grew quickly in size, was integrated into the local security apparatus, and played a vital part in the protection of the displacement camp and later the district town. Diffusion was made possible by the cultural bonds that existed between communities and the success of the movement in other parts of the province. Migrating diffusion agents and local residents’ acquired beliefs that Naparama would help them curb the violence convinced community residents to support forming Naparama units in their district. These mechanisms only led to sustained diffusion, however, because the Naparama leader was a trusted community leader who would not challenge Frelimo dominance. As I will explain below, Naparama did not enjoy such support among the people and the administration in the Namarrói district due to ruptures in the relationship among political and military elites.

6.3 Limited Diffusion of Naparama to Namarrói

6.3.1 The Military Situation in Namarrói

Renamo’s war reached Namarrói from the neighboring Lugela and Milange districts in September 1982. As in Lugela, respondents and local government reports point to frequent movements of “bandits” – presumably linked to Africa Livre – in the district before September 1982.⁵³ When Renamo first entered the district in 1982, it soon controlled the district’s north, where it established its major regional base (see map in Figure 6.2).⁵⁴ The base was located in Nantutu but was then transferred to Maquiringa, an area deep in the forest close to the Regone locality, which remained occupied from 1983–91.⁵⁵ The mountain range that crosses the district from east to west became a frontier between Frelimo-held and Renamo-held areas. Frelimo controlled five localities; two remained occupied by Renamo until the end of the war. Renamo attacked the localities under Frelimo control from the northern part of the district and from Lugela district to the south.⁵⁶

As in Lugela, opposition movements initially enjoyed popular support. During the early years of the war, the local population collaborated with Africa Livre and Renamo by identifying representatives of the Frelimo state. This way, opposition forces were able to target state and party officials, professors, and health workers, and destroy critical economic infrastructure.⁵⁷ A teacher who was among the targeted explained:

During this time, the population distanced itself from the political cadre, teachers, and nurses. For example, teachers [who were] seeking shelter among the population – they turned [them] away, and they even reported them to Renamo, claiming that they were living [too] well.⁵⁸

Similar to the situation in Lugela, support for Renamo waned when the rebels no longer limited their violence to representatives of the Frelimo state. When Renamo began targeting the population indiscriminately, “a feeling of solidarity among the population and the officials [emerged]. In the beginning [of the war], people thought that after the punishment of the officials, the war would end.”⁵⁹ Instead, Renamo kept a tight grip on the population living in areas under their control. For example, the rebels received the support of chiefs by promising them benefits and status after the war.⁶⁰ But when the chiefs did not comply with the rebels’ demands, they were killed, often along with their families and sympathizers, as happened in a neighborhood in the district town in 1986.⁶¹ Mambos and mujeeba collected food from peasants and closely controlled people’s movements, and they often did so by force.⁶² Renamo also committed several massacres. On February 22, 1983, for instance, between fifty and ninety people were killed in Mualiua, a neighborhood in the district town, after residents had been called from their fields to a meeting.⁶³ Thus, support for Renamo diminished among the population in Frelimo-held areas.

In its response to Renamo’s violence, Namarrói’s district administration faced two major dilemmas. First, Renamo benefited from a wide network of local informers. Many of Renamo’s combatants stationed in the main base in Maquiringa were from Namarrói and their family members served as collaborators at all levels of the Frelimo state and party apparatus. Political sympathy for Renamo was strong across the district. Thus, similarly to Lugela’s district administrator who was worried that the population would “drink the enemy’s politics,” Namarrói’s district administrator maintained in a local government report from 1985 that Namarrói needed serious “political-military work and not just simple speeches.”⁶⁴ Second, as in Lugela town, the armed forces stationed in Namarrói town were weak, as they only consisted of poorly trained and poorly equipped popular militias. Local government reports of the time frequently express concern that the district’s small military force lacked weapons and supplies.⁶⁵ The armed forces were entirely made up of locals who did not stay at the military barracks due to logistical problems. Most of them were trained as popular militias in 1979 and then retrained to join the military when the war reached Namarrói.

Like Lugela, Namarrói suffered from long-term rebel occupation, during which the population fled to a neighboring district. Frelimo’s inability to defend meant that the district town fell to a group of Renamo combatants trained in Kenya on December 29, 1986.⁶⁶ Coming from Malawi, Renamo had started a major offensive that would lead to the occupation of many district towns across Zambézia. Most of Namarrói’s population fled to Ile, the neighboring district about forty-five miles from the town. The district administration moved its offices to Ile’s district town. The armed forces remained stationed in the Muémue locality to the south of the town on the road linking Namarrói to Ile, but they struggled to defend the remaining three localities under their control.⁶⁷

The security situation in Namarrói remained difficult even after the army had succeeded in retaking the district town after almost a year of occupation. Frelimo forces tried twice, unsuccessfully, to retake the town.⁶⁸ Only on December 16, 1987, did the army reoccupy Namarrói with the support of a major reinforcement, helicopters, and MIGs, and a new military commander called Jacinto Rudes, commander of battalion 054. The new commander and the reinforced troop contingent managed to defend the town in the following years. On January 2 and December 21, 1988, Renamo attempted to occupy the town with a large contingent, but both times the rebels were unsuccessful. Subsequently changing their strategy, Renamo units divided into small groups of four to six people and infiltrated areas around the town.⁶⁹

Namarrói experienced some relief after Frelimo attacked and destroyed the main base in Maquiringa (also known as Nantutu) in September 1988: “[The base] was believed by local people to be impregnable until special units of the Mozambican armed forces stormed it in September. The destruction of Nantutu consolidated the Mozambican army’s hold on the district.”⁷⁰ The destruction of the base facilitated the population’s return from Ile starting in 1989. The administration settled people into communal wards so they could be supplied with relief aid and their movement controlled. Renamo only staged new attacks when parts of the military contingent left the district, for instance, on December 18, 1991.

While the larger Frelimo troop contingent stationed in the district after retaking the town in 1987 had reduced the rebel threat, Frelimo soldiers increasingly became a threat to the population. Local government reports sent to the provincial government at the time frequently spoke of the bad relationship between the soldiers and the population, and the soldiers and the local government. Soldiers were often drunk, killed civilians arbitrarily, and threatened to kill party and state representatives who criticized their behavior.⁷¹ The state-initiated popular militias were disorganized, if they existed at all. Efficient and capable command was lacking, weapons were in short supply, and desertion was widespread.⁷² The few militia units that existed mistreated the population on their patrols in residential areas.⁷³ The local administration therefore trained locals for new (state-initiated) militias. By the end of 1988, Frelimo had mobilized 1815 men, organized into 137 control posts, who were tasked with regular patrols and the collection of intelligence from strategic areas under Renamo control.⁷⁴

Overall, the military situation in Namarrói resembled the situation in Lugela. Namarrói faced a similar threat as Renamo perpetrated indiscriminate violence against civilians, and Frelimo forces were too weak to adequately respond to the threat in either district. As in Lugela, the local administration in Namarrói organized community residents into (state-initiated) militias, and these became increasingly important to the defense of the town because of the army’s unreliability. However, community initiatives to form militias did not have the same level of success in Namarrói as in Lugela.

6.3.2 The Rise and Fall of Naparama in Namarrói

Naparama leader António made a lasting impression on the people when he visited Namarrói and demonstrated the power of his medicine. However, the group he left behind quickly fell apart. Cultural bonds between neighboring communities and Namarrói and migrating diffusion agents facilitated the spread of Naparama to Namarrói. However, the case of Namarrói shows that the lack of opportunity to learn about Naparama’s successes and the existence of elite conflicts prevented local officials from institutionalizing the community-initiated militia.

Namarrói’s residents still vividly remember Naparama leader António’s initial visit because the cultural references he used during a public ceremony resonated with the community. When the Naparama leader came to the town in late 1990, he introduced himself to the administration, which then organized a meeting with community residents. The district administrator made it clear to António that his endorsement of António’s work did not imply a promise to pay salaries, and he sought to limit Naparama’s activities to acts of self-defense.⁷⁵ António performed a ceremony in order to demonstrate his powers, during which he re-enacted Christ’s resurrection.⁷⁶ Many respondents had a distinct memory of this ceremony:

On the day [António] arrived, it was at the [student housing building], here at the secondary school, when he arrived he said, “here I am, I am Naparama.” He called together all the people, us as well, in order to meet Naparama and he said, “I am Naparama, whoever doesn’t know me, here I am. Today I want to go to [the area] where Renamo is, I will collect everything and I will kill all the Renamo combatants and bring weapons.” All the people were there. He dug [a hole] into the ground, went into [the grave] and [others] put sand [on it] and he said, “I died.” After some time, he came out and said, “I will go and capture them all.” So he vaccinated some youths who were there and volunteered. The vaccination occurred with a machete. He took some medicine and [used] the machete to make cuts [into the person’s body], but the machete did not [hurt the person], this was the vaccine. When they were finished, he went at night and captured all of [the rebels], killed them and took their weapons.⁷⁷

Another respondent remembered the performance like this:

The first day I saw Naparama was on a Sunday. The Administrator Mucutueliua had invited many people in order to see how [Naparama] did things, there by the church. We didn’t wait long and went to watch. They dug a grave as if it was a cemetery and they started singing and carried [António] over to [the area] where the student housing is today and buried him there. They put a mat [in the grave] and a big rectangular mirror and he started talking. That there are this many troops and that many weapons in the base of Sahia and in the base of Mussisse that many. Everything that was at the base. Then they covered the grave. When he left [the grave], they started to sing and started a fire with big pieces of firewood. He went [to the fire], sat down and started to take piece after piece [out of the fire] and put them on his chest. They brought some leaves and a mortar …. The youth who wanted to join the group came and took a leaf from the mortar and when they took a machete and struck [his] body, they put the leaf [on the wound] and the wound healed right away. The next day, they left and went to test whether the medicine worked. They advised the new Naparama combatants that whoever ducked down or retreated [would die]. Even though the enemy had a weapon, you should not withdraw – they even captured [the enemy] and brought the captured men here with their weapons.⁷⁸

The awe with which respondents told these stories and the detailed recollection of the events demonstrate that the cultural references – both to traditional medicine and Christian religion – invoked during the ceremony resonated with the community. As in the Lugela district, the cultural bonds among the communities helped spread the violent collective action repertoire. António served as a diffusion agent; he traveled to Namarrói and formed new militia units.

However, cultural bonds and migrating diffusion agents only led to Naparama’s initial diffusion. Respondents pointed out that Naparama activity in Namarrói was short-lived:

When [António] left, the day he went to [the district of Morrumbala], [Naparama] did not continue going [on missions]. Also, when [António] died [in December 1991], no one continued with this work.

Did he leave a commander behind?

… Here he had left [someone]. But when [Naparama in Namarrói] heard that [António] had died, they stopped doing their work, and the war ended. It didn’t take long [for the war] to end, [during] the time of Naparama.⁷⁹

The strong impressions of António’s ceremony contrast with respondents’ narratives of Naparama after António had left Namarrói. Several respondents emphasized that António died soon after he had left Namarrói, after which the group fell apart. Even though António only died in December 1991, respondents do not have any recollection of major Naparama activity in Namarrói after António’s initial visit in late 1990. The little they remember was that the Naparama combatants who were left behind did not form a strong group and disregarded the rules that ensured the medicine’s effectiveness:

Did Manuel António leave someone behind who could vaccinate people?

No, when he returned [to where he came from], he just left behind his troops and a commander. He didn’t leave behind instructions to continue to vaccinate. Only those who had been treated [with the vaccine] stayed. But when they started to contradict the orders received, they began to suffer, because one time when Renamo men came and attacked, they fled. They were not allowed to eat feijão jogo [a certain type of beans], sweet potato leaves, but they started eating it. Then, the [effect of the] drug started to disappear.⁸⁰

This statement implies that the Naparama militia was a weak group of combatants with little discipline, and that the militia soon fell apart. Thus Naparama’s diffusion was not sustained.

6.3.3 Explaining the Limits of Naparama Diffusion to Namarrói

What explains the failure of sustained diffusion of Naparama to Namarrói? Comparing Naparama’s manifestation in Namarrói to that in Lugela, I identify two factors that explain Naparama’s different trajectories in the two districts: First, Naparama in Namarrói lacked integration into the local security apparatus because people did not have an opportunity to learn about the positive effects of Naparama’s work on security in and around Namarrói. From the perspective of community residents, the security situation was not sufficiently similar to the one people had heard about in other districts where the Naparama militia was active. Second, the district’s elites were in conflict with each other, so the Naparama leader came to pose a threat to local authorities rather than providing welcome support. This situation resulted in, at best, lukewarm support for Naparama among the population.

Regarding the first explanatory factor, in contrast to Lugela’s residents, Namarrói’s population was not convinced that Naparama forces could have a lasting effect on security in Namarrói. Respondents reported that when António left, Naparama combatants died on missions, fled in the face of an attack, or no longer went on missions at all.⁸¹ Residents also pointed to the fact that the war was already at its end when Naparama emerged, and thus, in their memory, Naparama did not operate for a long time.⁸² Moreover, the district administrator had made sure that Naparama’s activities were limited to patrolling and accompanying military convoys. They reportedly never had any direct encounters with Renamo troops, which further diminished their pertinence to the improvement of the district’s security situation.⁸³ Thus, although people were fascinated with António’s personality, they did not experience that his work was crucial in improving their situation.

The perceived limited impact of Naparama on the security situation in Namarrói can be attributed to the timing of the Naparama leader’s arrival. At the time, troops in the district had strengthened and reorganized and the district had seen fewer attacks. The former district administrator explained that the security situation did not allow António to demonstrate his powers in battle:

We organized a meeting with the population – I was there, I wanted to know what he will say to the people. He vaccinated some people during this meeting. But he didn’t have much impact in Namarrói – he couldn’t show what he was able to do since during the three days he was there, there was no attack, and so he didn’t have the opportunity to show off. In the other districts he had more impact since he could show his work.⁸⁴

Residents remembered the time after the end of the town’s occupation as relatively peaceful. In people’s memories, the improvement of the security situation was not due to the Naparama’s emergence, but to the military commander Rudes who was in charge of the operation to liberate the town from Renamo occupation. Even if people praised what Naparama had done in the region generally, they did not consider the militia’s activities to have brought about a significant change to their own district’s situation. In contrast, commander Rudes was remembered as a powerful man, as he succeeded to recapture the town in 1987. To underline Rudes’ effect on security, Namarrói residents pointed out that when the military commander left the district in 1991, Renamo restarted their attacks, forcing people to flee to the bush.⁸⁵ In a way then, the Naparama leader’s services were rivaled by those of the military commander, who had demonstrated his effectiveness in a powerful way that could not be repeated by the Naparama leader.

Regarding the second explanatory factor, there was a lack of trust between Naparama and Frelimo elites. Rivalry and mutual distrust prevented the armed forces from cooperating with the militia. The district administrator pointed out that “the military thought [António] was crazy and laughed at him.”⁸⁶ The Naparama district commander, Makosso, was a Frelimo combatant who had fought during the liberation war in Lugela and Namarrói, but had left the army in disgrace.⁸⁷ According to a local government official, before becoming a Naparama commander, Makosso had been discharged from the military due to bad conduct – allegedly killing civilians arbitrarily.⁸⁸ Others maintained that he had been a member of a Frelimo death squad.⁸⁹ Before António came to Namarrói, Makosso lived without work in his home village Muémue and then made use of the opportunity to become involved in the war as a Naparama commander. He nominated himself for the militia leadership, and António delegated the task of recruiting more Naparama combatants to Makosso after his departure.⁹⁰

Naparama combatants did not think highly of the army either. From the militia combatants’ perspective, cooperation with the army was undesirable. The way Naparama combatants fought – standing upright in plain sight – was incompatible with the way the army fought: Soldiers “[took cover and] crawled and we did not crawl, [so] we were afraid of them because they could crawl and we would go in front of them, [and then] they [would] shoot [and] hit us.”⁹¹ Some Naparama combatants adopted a rather arrogant attitude toward the armed forces:

Although all fought for the same purpose, there was some alienation … between Naparama combatants and soldiers. They disrespected the soldiers, telling them that they were nothing, that, although they used weapons, they did not succeed in destroying bases, and [Naparama] with spears and machetes was able to do so.⁹²

By pointing to their superior power, Naparama combatants claimed to be entitled to the food aid meant for the army.⁹³ Militia members even held the army responsible for their losses during battle. The rivalry between the army and Naparama culminated in a clash between the two forces in Inlugo in the locality of Mudine after António’s death in late 1991, resulting in the death of several Naparama combatants.⁹⁴

The distance between the military and Naparama could also be felt geographically. Presumably due to the strong presence of the armed forces in Namarrói district town and their attitude toward Naparama, the Naparama district commander created the militia’s headquarters in the Muémue locality on the road toward Ile.⁹⁵ Most of the Naparama combatants – about eight – were stationed in Muémue.⁹⁶ Naparama’s headquarters were at the opposite end of the locality from those of the small armed force in the same village. The Naparama force in Muémue overpowered the military contingent, which was only one section of twelve men.⁹⁷

Conflict and rivalry were also present between Naparama and Frelimo’s administrative apparatus, for example in Muémue. Not long after the formation of the Naparama unit in Muémue, the then-head of the local administration told me that Makosso got into conflict with him because the Naparama commander sought to take control of the locality:

[Naparama] went to Lugela and successfully brought back some women who they captured over there and took them to me. I placed them into families who could provide them with food. So these people stayed and lived [with these families]. While they were living [with these families], the time came when the amnesty law was passed. With the amnesty law, [violence] decreased a little. So when the women saw that the situation had [improved], they wanted to return to Lugela. When these three women with their children were on their way to Lugela, militiamen captured them. They asked, “Where are you going?” – “We are returning home to Lugela.” [The militiamen] sent [the women] back [to Muémue]. When they sent them back, they took them to the Naparama. At that time, I was at a meeting in the town. So Naparama started to investigate: “Where did you go?” – “We were returning home.” Commander Makosso said, “You wanted to betray us, go and provide [Renamo] with information so that they could attack us.” So instead of waiting to inform me, [Makosso] wanted to assassinate these people. [Naparama] had already distributed the children among them. So when I returned from the town and arrived at home, my secretary informed me that these women who had lived in the house of that person no longer lived.⁹⁸

In the absence of the local administration head, Makosso assumed responsibility, ignored the stipulation of the new amnesty law, charged three women with treason and ordered their assassination. When the head of the locality learned that one of the women had survived and returned to Muémue, he sent a nurse for her and brought her clothes. This led the Naparama commander to threaten him:

So when this commander [Makosso] heard about this, he gathered his group to assassinate me, as I had ordered to rescue someone who was betraying him. So this group came, encircled my office, someone came and told me that a group was outside that wanted me to come outside and talk to them. I went outside. I started asking, “What is the problem?” – “Ah, you are a mujeeba because you ordered to rescue a person that we ordered to [kill].”⁹⁹

Confronted with this threat, the head of the local administration called the police. The policemen took Makosso with his three men to the district town, where he was arrested and sent to Ile for imprisonment. After this event, Naparama stopped operating in Muémue. This story demonstrates the deep mistrust that existed between the army, the local government, and the Naparama militia, as the local government perceived the militia to rival and threaten Frelimo’s control over the district and to help fight elite-level power struggles.

In sum, Naparama activity in Namarrói was erratic and riddled with challenges. Community residents remembered Naparama’s activities as short-lived. Two factors prevented Naparama from sustainably diffusing to Namarrói. First, community residents were more impressed by the Frelimo military commander who had liberated the town than by the Naparama leader who did not get the opportunity to demonstrate his powers, so they perceived Naparama’s impact on the improvement of security as limited. Second, elite rivalry prevented the militia’s institutionalization and its integration into the local security apparatus. The army was relatively well organized and suspicious of Naparama, and the militia leadership entered into conflict with the army and the local administration. Instead of supplementing the army, Naparama sought to rival the army. Instead of expressing its full loyalty to Frelimo, the group sought to replace the Frelimo administration. Instead of remaining a people’s movement, it evolved into a personal army of a single commander.

6.3.4 Alternative Explanations

Two alternative explanations for the observed differences of militia diffusion to the two districts in Zambézia province deserve consideration. First, the type of agent might matter for sustained diffusion. Naparama’s leader, António, arrived in the Namarrói district on his own accord, and not as a consequence of the dislocation of local residents who brought him to Namarrói. This could have led to a lower level of community support for Naparama and thus to limited diffusion. However, the sustained diffusion of Naparama to Lugela was also not based on the residents’ dislocation and demand of a Naparama force and yet militia diffusion was sustained.

A second alternative explanation is that Namarrói’s residents may not have believed that cultural practices would aid them in their quest for security. Cultural (and ethnic) bonds were not strong enough to convince Namarrói’s residents to include a traditional militia in its collective action repertoire. However, Namarrói is known for its people’s spiritual powers, and during the war several traditional healers are said to have contributed to the protection of the town and its inhabitants.¹⁰⁰ In fact, some respondents attributed Frelimo’s success in retaking the town to the help of a traditional healer who had treated soldiers before their mission and advanced in front of them when recapturing the town.¹⁰¹ Thus, cultural bonds were actually very strong, which was necessary for initial diffusion, but on their own they were not sufficient to enable sustained militia diffusion.

6.4 Exploring the Implications: Naparama Diffusion to Mecubúri

If the argument outlined above holds, we should observe similar dynamics in the diffusion processes of the Naparama militia to other districts. I explore whether the arguments developed through the comparison of the two districts in Zambézia province hold in a different case in a different province, that of the diffusion of Naparama from Murrupula to Mecubúri district in Nampula province (see map in Figure 6.3). Naparama in Nampula evolved independently of Naparama in Zambézia province. One of António’s confidants, Ambrósio Albino, formed his own militia in the Murrupula district, and Murrupula’s traditional healer, Nampila, became an important Naparama leader. Mecubúri faced similar wartime challenges as the two districts discussed above in terms of its remoteness, proximity to Renamo’s major bases, and levels of violence. The independent evolution of Naparama in Nampula and the similar wartime conditions thus make Mecubúri an excellent case to explore whether the arguments developed in the comparison of Lugela and Namarrói hold in a different context.

Figure 6.3. Map of Mecubúri and Murrupula in Nampula province.

Note: Cartography by Sofia Jorges

The conditions and mechanisms that account for the divergent trajectories of Naparama in the cases of Lugela and Namarrói districts lead to the following expectations for Mecubúri (see a summary of the argument in Table 6.1): (1) Ethnic and cultural bonds facilitate the spread of collective action repertoires through the mechanism of migration. Diffusion agents strengthen these bonds and community residents react positively to cultural tropes used by these agents; (2) successes in neighboring regions facilitate the diffusion of collective action repertoires through the mechanism of learning. The receiving community considers its security situation as sufficiently similar to that of neighboring communities and expects the community-initiated militia to have the same impact in their community; and (3) the integration of Naparama forces into the local security apparatus is facilitated by the absence of elite rivalry, which increases trust among local elites and the community. A high level of cooperation and coordination between Naparama, the local administration, and the armed forces stationed in the district exists.

Table 6.1. Overview of factors, mechanisms, and indicators explaining sustained diffusion

Factor	Mechanism	Indicator
Ethnic and cultural bonds	Migration	Dislocation of diffusion agents and community residents’ positive reaction to cultural tropes
Success of militia in neighboring regions	Learning	Community identifies with neighboring communities and considers its security situation as sufficiently similar to that of neighboring communities
Absence of elite rivalry	Trust	High level of cooperation and coordination between Naparama, the local administration, and the armed forces stationed in the district

The following sections first discuss the military situation in Mecubúri, which provide the context for the rise of the Naparama militia, and then analyze the conditions for sustained diffusion of Naparama to Mecubúri.

6.4.1 The Military Situation in Mecubúri

When Renamo entered Nampula province in May 1983, it advanced along the train line from Cuamba to Nampula, which crosses the south of the Mecubúri district. Mecubúri lies fifty miles to the east of the provincial capital Nampula. With major Renamo bases in Murrupula (Namilasse) to the south and Muecate (Namahia) to the north, Mecubúri became a transit area for Renamo units moving between these bases.

As in Lugela and Namarrói districts, Renamo attacks first focused on economically viable areas, the destruction of (state) infrastructure, the kidnapping of people, and the intimidation of party and state representatives. The war reached the district in June 1983, and the district town of Mecubúri in June 1986. Coming from Ribáuè district to the east, Renamo units attacked Tocolo twenty miles to the northwest of the district town in 1983, kidnapped several people, cut Frelimo officials’ ears off, and pillaged people’s belongings.¹⁰² On June 24, 1983, Renamo units attacked the communal village of Nametil to the northeast of the district town, pillaged and burned houses, and told people to return to their ancestral lands.¹⁰³ On June 24, 1986, Renamo units coming from Murrupula district attacked the district town of Mecubúri for the first time, pillaging the warehouse and people’s belongings.¹⁰⁴

The district saw some of the worst violence between 1988 and 1990, shortly before Naparama emerged in the area. During this time, the district administrator was killed during an attack on the communal village, Mucheluia.¹⁰⁵ After the military battalion, which had been stationed in the district after Renamo’s initial attacks in 1984, had left in July 1989, violence increased significantly.¹⁰⁶ Within a year, the number of displaced people in Mecubúri rose from 2,000 in 6 accommodation centers in December 1988 to almost 9,000 in 12 accommodation centers by the end of 1989.¹⁰⁷ Two more attacks on the district town occurred in 1989 and 1990.¹⁰⁸

In its response to the violence, the local administration faced similar challenges to those in Lugela and Namarrói. When the war began, most of the population was settled in communal wards and villages, which allowed Frelimo to surveil the population. However, these wards and villages were also among the major targets of Renamo attacks to pillage relief goods.¹⁰⁹ Moreover, the district had an insufficient number of armed forces to counter the Renamo menace. Youths refused recruitment into compulsory military service by moving to Nampula city, mostly due to the military’s problems with food and supplies for the armed forces.¹¹⁰ After the military battalion stationed in the district had moved to Pemba in Cabo Delgado province, villages remained without any army contingent. Due to limited supplies, the few military forces present preferred working their plots to carrying arms, or even stealing food aid meant for the population.¹¹¹

6.4.2 The Emergence of Naparama in Mecubúri

In the context of this difficult military situation, Naparama emerged in mid-1990, spreading from Murrupula district on the initiative of local residents.¹¹² The militia was strong in Mecubúri, counting about 200 men in the district town alone.¹¹³ Before Mecubúri residents traveled to Murrupula to meet the Naparama leader Nampila, António, who later became Zambézia’s Naparama leader, had visited the district and initiated a few people. According to Mecubúri’s residents, however, António’s group did not curb the violence, and so António left the district shortly after his arrival.¹¹⁴

Naparama units formed in Mecubúri when residents of Nahipa, a village close to the district town, heard of Naparama’s successes in Murrupula district. They decided to go and meet Nampila and bring him to Nahipa to form a militia unit. The young men who met the Naparama leader in Murrupula belonged to a community-initiated militia that was formed before Naparama’s emergence. Youths in Nahipa sought to counter the activities of “criminals” who stole people’s belongings. The group was called the grupo decidido (the committed group), and was presumably linked to the local office of the Mozambican intelligence service (Serviço Nacional de Segurança Popular, SNASP).¹¹⁵ When the group heard about Naparama, they sent a small group to Murrupula to meet the traditional healer Nampila. Their main motivation was to benefit from the additional protection Naparama promised, as one of the first members narrated:

We heard that there was a traditional healer in Murrupula who vaccinates people and [these people] don’t die from weapons. So we said we are the grupo decidido, but we die from weapons, so twelve of us left …. We went to Murrupula, looked for the traditional healer, met him and got vaccinated, returned to Nahipa, introduced ourselves [to the administration] and began to work. From then onward, we took five people from every village and vaccinated them. [Naparama’s] center was Nahipa, and when the traditional healer came here from Murrupula, he stayed in Nahipa.

Thus, migration was the mechanism that led to the initial diffusion of Naparama between communities with strong cultural and ethnic bonds. People across Nampula province belong to the Makua-Lomwe linguistic group, and Naparama made use of many of the traditional rituals known to community residents.¹¹⁶ Naparama’s rituals resonated with Mecubúri residents, and the members of the grupo decidido initiated the spread of Naparama to their district. In contrast to Lugela and Namarrói, however, the diffusion agents in Mecubúri were local residents in search of new ways to curb the violence, and not the provincial Naparama leaders in search of new members.

As in Lugela, Naparama spread to Mecubúri and across the district because the population learned of Naparama’s success in other areas, became convinced that the situation in Mecubúri was sufficiently similar to that of other areas, and that Naparama could be the one to respond to it most effectively. At the time of the community residents’ decision to become Naparama followers, Nahipa was a destroyed village and most of the population lived in the bush. When the initial twelve Naparama members returned from their trip to Murrupula, they successfully mobilized displaced people to return to the village, join the new militia, and defend the village.¹¹⁷ Moreover, after Naparama forces from Nahipa had helped to retake the district town from Renamo occupation, the town’s population sought to form its own Naparama force, as a former Naparama combatant from Mecubúri town explained:

At that time, Renamo had come and attacked the district town and the first Naparama combatants here expelled them. We [from Mecubúri] came to the conclusion that they did a good job. We asked these Naparama where they came from, they said they were from Nahipa. Then we decided to go there.¹¹⁸

It was only after the Naparama forces from Nahipa had demonstrated their strength that people from Mecubúri realized the value of the new force, as this former Naparama combatant’s remarks show:

There was war here and we fled to Nahipa. When we arrived there, we met members of the Naparama militia. They [wanted to] mobilize us, but we didn’t agree [to join] and returned here [to Mecubúri town]. Here we met a group of Naparama who had come and rescued the town, and its leaders mobilized us. Then we decided to return to Nahipa to get vaccinated. We came and stayed in the district town to defend [it] in case of any eventualities.¹¹⁹

Thus, after learning about the success of Naparama in Mecubúri town, youths joined Naparama on their own initiative.¹²⁰ A former Naparama combatant from Nahipa confirms this story:

One of the times [when Renamo attacked], Renamo stayed for one week in Mecubúri. The military force had fled. Reinforcements from Nampula came twice, but without success. So we, twelve men, left [Nahipa for Mecubúri], succeeded in recuperating [the town]. So the population accepted and confided in us and followed us here [to Nahipa], they came to receive the vaccine.¹²¹

The mechanisms of migration and learning led to Naparama’s initial diffusion because local residents reacted positively to the cultural tropes used by Naparama and became convinced that the community-initiated militia enhanced the security situation in their area.

6.4.3 Sustained Diffusion of Naparama to Mecubúri

Diffusion was sustained in Mecubúri when elite conflicts could be averted so that the Naparama militia was integrated into the local security apparatus. The militia cooperated closely with the local administration. The local administration not only tolerated, but even promoted the diffusion of Naparama forces across the district, as this former Naparama combatant explains:

What did the local government think of Naparama?

The government felt relieved, that is why it accepted the emergence of other groups. The secretary of [the administrative post] Muite came and requested that we go there and vaccinate the population. From then on, the group emerged in Muite. After here, Nahipa, people were vaccinated in the district town and in Muite.¹²²

It was on the initiative of Muite’s party secretary that Naparama spread to the village to the north of the district town. In Mecubúri town, a local government official confirmed that there had been a close relationship between the population and the local government in deciding to form a Naparama force:

Because of the fatigue with the burning of houses and random killing of people, when the traditional healer arrived, he contacted the local authorities and since these were tired … – just imagine an administrator who never slept in his residence! [Naparama emerged] in agreement with the local government to resolve the problem – and we [the people] agreed that this was real because the enemy did not reach areas where Naparama units were [present]. As soon as they founded Naparama, the enemy no longer came here. When the youths went into battle, the enemy died and it was in battle that [Naparama] convinced the people and the local government [to support Naparama].¹²³

Traditional leaders and local party secretaries actively mobilized youths to join Naparama, even forcing some to be initiated into the group, as this former Naparama combatant testified:

You mentioned that some youths were forced to receive the vaccine – by whom?

By the leaders, because Naparama worked together with them and the secretaries. There was a meeting, where some volunteered [to receive the vaccine], and since their number was small, they had to force some other youths.¹²⁴

Although there was initial mistrust, elite conflicts did not arise. The reasons for this seem slightly different in Mecubúri than in Lugela. Instead of evolving as equal to the Frelimo armed forces, the Naparama militia simply overpowered the army contingent stationed in Mecubúri and came to replace them with the army’s approval. As a former Frelimo combatant stated, Naparama dismissed collaboration with the army contingent, but the army did not challenge Naparama:

Did Naparama coordinate with the armed forces stationed in the district?

No, they did not because Naparama said to the armed forces that they did not do anything. … They went to battle jointly, but in some cases, Naparama did not accept [joint operations]. Even I told them that I was a soldier and could do anything [to support them], but they refused, saying “You won’t succeed.” … [The armed forces] wanted to give [Naparama] [advice on] tactics in case the enemy came, but they didn’t want it.¹²⁵

The Naparama militia did not want to collaborate with the armed forces, claiming that the soldiers would die in battle.¹²⁶ A former Naparama combatant explained that from Naparama’s perspective, militia tactics were incompatible with those of the armed forces. He accused the military of being afraid of Naparama, which is why they did not cooperate:

[The armed forces] were afraid to go with us [into battle]. The manner in which we fought was different. The armed forces fought from a distance and Naparama always wanted to catch the men, this is what made the armed forces afraid [of us].¹²⁷

There was some mistrust between Naparama’s and Frelimo’s armed forces, but mutual mistrust did not undermine the basis of cooperation or provoke lasting conflict. Several respondents confirmed that Naparama militiamen received weapons from the armed forces – in particular those who had been members of the state-initiated popular militias before receiving the Parama vaccine and were trained in using a rifle. Those Naparama combatants with weapons went on missions while the armed forces stayed behind.¹²⁸ Rather than rivaling the army, the Naparama militia came to replace the military forces with the latter’s approval, as this former Naparama combatant explained:

One time, [Renamo combatants] came from Lalaua and passed by Metuba, and we went to meet them. We confronted each other and we recuperated a mortar. When we went to present it to the army headquarters, [the army] decided to provide all of us with weapons and the troops suspended going to battle. Until the war ended.¹²⁹

In sum, in the case of Mecubúri district, the mechanisms of migration and learning operated in a similar way as they did in Lugela and (in a limited manner) in Namarrói districts. However, the initiative for Naparama’s formation in Mecubúri came from the local population instead of the provincial Naparama leaders. This created a basis for strong popular support for the new Naparama force, which in turn translated into a slightly different model of sustained diffusion of Naparama in Mecubúri. Instead of being integrated into the local security apparatus and acting in parallel with local armed forces, Naparama replaced the existing local security apparatus. The few armed forces stationed in the district ceased all their operations when it became clear that Naparama was stronger than the armed forces could ever be.

6.5 Conclusion

What explains the differences in the diffusion of Naparama to two adjacent districts, Lugela and Namarrói, in Zambézia province? The two districts faced a similar set of challenges in terms of rebel presence, occupation, and violence, but they followed different trajectories regarding the mobilization of community-initiated militias. Incorporating mechanisms of diffusion and the conditions under which they operate into existing theoretical frameworks helps us to understand this variation.

As discussed in Chapter 5, Naparama spread across communities because of its innovative character, providing a competitive supplement and even an alternative to prevailing local security arrangements at the time. Naparama spread to Lugela and Namarrói because diffusion agents arrived in the district and the population learned about Naparama’s success in neighboring communities. Naparama was able to become a lasting force in Lugela because its integration into the local security apparatus was not hampered by elite conflicts. A highly regarded community leader took on the role of Naparama leader and was able to mobilize loyal members. Moreover, Naparama enjoyed the trust of the local administration and acted in close coordination with the armed forces. In the adjacent district of Namarrói, people did not perceive their situation as sufficiently similar to that of neighboring communities, so they did not consider Naparama an important force to improve their security situation. The militia also suffered from weak leadership that was unable to overcome the army’s suspicions and, in fact, sundered relations with the military and administrative elite altogether.

The case of Mecubúri in Nampula province helps to develop these arguments further. The main difference between sustained diffusion in Mecubúri and Lugela was that in Mecubúri, elite conflicts did not arise because Naparama replaced the weak armed forces with their approval. In Namarrói, the armed forces were strong, and thus they did not allow for a forceful and potentially rival Naparama militia to evolve. The only – limited – chance of survival for Naparama in Namarrói was to evolve into a private army that represented a third party to the conflict and threatened Frelimo hegemony in government-controlled areas.

In sum, even when a “trigger” moment occurs and newly invented collective action repertoires spread from one community to the other, the community’s elite constellation influences whether the new militia is maintained. Whether a community-initiated militia is institutionalized does not say much, however, about how the militia can attract more members. The armed organization’s growth once it is diffused from one region to another is the subject of Chapter 7.