ARTICLES

 

Growing Apart: The Historical Construction of Difference in Northern Cabo Delgado, Mozambique

 

 

Ana Margarida Sousa Santos

Department of Anthropology, Durham University and Institute of Social Sciences, University of Lisbon. https://orcid.org/0000-0002-3440-7250

 

 

---

ABSTRACT

How do past experiences of violence map onto present day narratives of insurgency in northern Mozambique? In the insurgency that began in northern Mozambique in 2017, narratives of past violent encounters and animosity between different ethnic groups sharing the coastal space emerge as a way of making sense of the conflict. This article explores the ways these narratives stem from existing historical tensions between Makonde and Mwani inhabiting the northern districts of Cabo Delgado. Drawing on ethnography and oral history to highlight the continuities in narrative, I will explore the ways in which current violence is interpreted through references to pivotal moments of violence in the past. I will address the historical construction of difference between Makonde and Mwani to consider what it may tell us about the fractures that underpin the insurgency.

Keywords: Makonde, Mwani, Cabo Delgado, violence, memory

---

 

 

Introduction

In early October 2017, a group of armed young men attacked the police post in the town of Mocímboa da Praia, killing two policemen, stealing arms and briefly occupying the town.¹ They were successfully rebuffed by government forces shortly after. This attack was the opening salvo of what was to become a difficult and increasingly violent period in the northern districts of Cabo Delgado. The violence that ensued drew attention to a largely forgotten region in Mozambique and shed a light into local politics, as well as the fraught relationships between the groups inhabiting the north.

The attacks grew more violent and extended to Mocímboa's neighbouring districts of Macomia, Nangade, Muidumbe, Palma and, later, Mueda. 2020 and 2021 saw an unprecedented escalation of what had been, until then, a low-level, local, insurgency. The state security forces, and mercenary groups (including Wagner and the Dyck Advisory Group) proved unable to withstand the insurgents at this point. Moreover, they compounded the violence with a series of brutal engagements and abuses of the local population. This exacerbated existing resentments and grievances at the local level. In response to the continuous attacks and insecurity, the inhabitants of these districts fled some of the worst hit areas and made their way to IDP camps near the provincial capital Pemba, or the neighbouring provinces of Niassa and Nampula in increasingly large numbers. The dual effect of the violence and the destruction caused by the cyclone Kenneth in 2019 led to what is now a severe humanitarian disaster.

The arrival of a contingent of Rwandan troops in late 2021 to aid the Mozambican forces proved more effective in changing the landscape of attacks and control by the insurgents and leading to a decrease in violent events thereafter. The changes that followed included the retaking of Mocímboa da Praia (held by the insurgents for almost a full year) and other small villages and towns in the northern districts of Cabo Delgado. During 2022, the pattern of attacks changed with news of violence extending to Niassa, and a return to small scale, random attacks to remote villages. The shift in strategy shows a return to earlier tactics and a retrenchment to safer positions by the insurgent groups. The government forces, alongside the Rwandan military, have been able to keep the more devastating attacks to larger centres at bay but have been unable to ensure that the population feels safe enough to return to their homes. Despite the emerging stability and government assurances, the local populations have been slow to return to the areas formerly under insurgent control. The IDP camps outside Pemba are still large and in need of support.

The analysis of the insurgency has from its inception focused on external factors, shifts in the security landscape, endeavouring to search for causes that explain the enduring violence. While a number of analyses offer a good assessment of the ongoing security situation, there is a strong emphasis on narratives that align the insurgents with broader jihadist movements that operate elsewhere in Africa.² This has led to a neglect of the longer histories of proximity and animosity between the different groups inhabiting coastal Cabo Delgado. It is to this longer history and its actors, the Makonde and Mwani people who make the coast their home, that I will turn below. I will address local narratives of difference, the evolving patterns of proximity and distance, the discourses of belonging and the ways in which they build distinction through varying interpretations of a shared past. In doing this, I build on recent literature on autochthony and identity to consider how these patterns play out in discussions of the current violence and how they build on past memories to interpret the current predicament. To do this, the article proceeds as follows: I begin by considering how the current literature on autochthony may help illuminate some of the less discussed fractures in the social relationships in the north. I will discuss the context in which the insurgency emerged, followed by a discussion of narratives that were frequently alluded to in order to explain the growing distance between Makonde and Mwani along the coast. I end the article with a discussion of the implications of the historical construction of difference for the insurgency.

To probe the context in which the insurgency emerged, I build on long-standing research in Cabo Delgado. My research has focused on memory and storytelling as well as issues of identity and belonging and it is to these that I now turn in the hope of providing some much-needed historical background on the relationships between the inhabitants of the northern districts. It is upon this history of ambiguity and the multiple layers of history that I think the current violence should be interpreted and made sense of. However, much of the current assessments fall prey to the lure of debates around jihadism, and the radicalisation of young men at the expense of a deeper analysis of the longer historical dynamics and its interactions with the new realities of vast natural resources in a province that has for long seen little investment in infrastructure and is one of the least developed provinces in Mozambique. While true, these assessments are nevertheless partial and flawed in their reach and ability to explain the current, deeper divides of northern Cabo Delgado. I will consider the fraught and changing relationships between Makonde and Mwani and the ways in which these play out in accounts of the insurgency. I will now briefly introduce the history of Mocímboa where much of the early violence took place, and where some of the fiercest battles were fought in 2020 and 2021.

 

Identity, belonging, and local narratives of legitimacy

The discussion of the insurgency often hinges upon issues of religious belief and practice as well as emphasising international connections to extremist networks in Africa and beyond. The clear divide in religious affiliation - the Makonde are vocal about their Christian beliefs and the Mwani are Muslim - mark the separation that builds into the narratives of the insurgency and its connections with other cases. The assessment falters when considering internal fractures and longstanding historical divides that play out in the on-going relationship of the different groups inhabiting the north. This includes a lack of analysis that considers the ways in which these groups relate to the state and the territory they inhabit, as well as local understandings and narratives of identity. I argue that the increasingly fixed and inflexible narrative divides of identity and belonging between Makonde and Mwani inhabiting northern Cabo Delgado is central to this assessment. This has been fostered by long term patterns of intensifying competition for territory, political influence and resources at the local level. Before I present some of the ways in which this is discussed in the northern Cabo Delgado by reference to a few past examples, I will add a brief note on anthropological understandings of ethnicity and difference, as well as recent explorations of the rise of narratives of autochthony that develop alongside processes of democratisation.

Anthropological analysis of ethnicity has rightly pointed out its relational nature³ and its historically contingent character, noting the change in ethnic categories.⁴ However, discourses that are not ethnographically informed often remark upon the links between ethnicity and political competition, the manipulation, and the fixed nature of ethnicity, making little allowance for processes of change and flexibility. These are frequently shallow historical interpretations. Looking at belonging and the ever-evolving nature of ethnicity from a long-term historical perspective, the longue durée, and addressing questions of contact and shared spaces highlights how this narrative is at odds with the historiography.

Recent literature on ethnicity and identity draws attention to the dynamic nature of ethnicity, emphasising the perpetual redefinition of ethnic belonging. This continuous redefinition has been discussed in the context of ethnicity along the Swahili coast⁵ and built on long regional analyses of the relationship between the coast and the hinterland in East Africa.⁶ The multi-ethnic character of the coastal context of northern Mozambique, the discursive emphasis on movement, contact and a population described as permanently in transit, further highlights the fluidity and changeable categorical distinctions at odds with fixed local narratives of difference, as well as the endlessly negotiated nature of these identities which will be the focus of the discussion below.

Fluidity and permeability have been part of discussions that centre on porous borders (the kinds marked on maps but seldom respected)⁷ and boundaries (categorical distinctions associated with identity formations), and have been part of recent anthropological and historical literature. It is precisely this fluidity, perceived in longer-term studies of ethnicity, that contrasts with locally expressed idioms of fixity and opposition when it comes to describing the oppositional 'other'. Associated with the fluidity that is perceived as characterising ethnicity, is its mirror image - the increasing rigidity in classification that emerges from competition for political spaces and resources and the increasingly strong discussion about indigeneity and autochthony that has been developed in the past decade. This includes an attention to the responses to how 'the globalising forces of free markets and easy mobility have created displacement and competition that often trigger movements of exclusion, compartmentalisation and ethno-territoriality'.⁸ This is in line with the consensus on ethnicity that emphasises the need 'to map the historical trajectories of contemporary identities'⁹ as I intend to do here. The fluidity is not absolute, however. Local narratives of autochthony and establishing the link between territory and identity describe the shifting relationship between owners and guests, first comers and late arrivals, and play into discussions of rights, and feed both accounts of the past and present tensions.¹⁰

Categorical definitions, and boundaries between groups change continuously in northern Mozambique, but reformed and reframed boundaries between groups persist in part because 'categorical ethnic distinctions do not depend on absence of mobility, contact and information but do entail social processes of exclusion and incorporation whereby discrete categories are maintained despite changing participation and membership in the case of individual life histories'.¹¹ It is this process of exclusion and incorporation, the continuous redefinition of belonging and the grievances that emerge from it, that is associated with violent outbursts in 2005 or the on-going insurgency. There are historical bases for these interpretations. Southall's seminal article¹² highlighted the influence of the colonial administrations in crystallising ethnic divisions that were, until then, fluid and unremarked upon. I argue below that the process of division and increasing separation between Mwani and Makonde was set apace during the colonial period and in connection with colonial policies,¹³ and continued after independence, despite early efforts from Frelimo to create a 'new man' free from the shackles of ethnic identity and custom. Although the influence of colonial administrations was often uneven, the impact of shifts in power, and the reliance on rule through local leaders, régulos, led to the increased separation of some of these groups' crystallising identities, as did the use of forced labour which will be discussed below. The increasingly sharp separation between Makonde and Mwani is in large part connected with the differing experiences and memories of colonial and post-colonial politics and violence. These experiences and accounts reflect competing political claims to territory, power, and resources. These are often included in accounts that emphasise autochthony. Recent explorations of autochthony¹⁴ account for the emergence of idioms of belonging that place origin and territory at the forefront of conflicts, subsuming existing processes of incorporation that formed part of past relationships between groups.¹⁵ Some of the narratives discussed below suggest that local relationships were messier than memory allows for, and I would caution against drawing too rigid a distinction between the expressions of identity and understandings of belonging but instead focus on the changing nature of relationships, the specific historical context in which they develop, and the way past engagements are understood at times of conflict. This is an important dimension in a region where the wealth of natural resources and its exploration emerges alongside an increase in violence. It is to these experiences and the ways in which they are understood that I turn below.

 

'We are all the same' - emerging narratives of difference in Mocímboa

The northernmost province of Mozambique is no stranger to violence, with a long history of conflict, wars and forms of structural violence that mark severe underdevelopment. Cabo Delgado is one of the least developed provinces of a deeply underdeveloped country. Its history is marked by a succession of conflicts, but for this article I will focus on a few moments to explore the ways in which the experience of violence plays into narratives of difference.

It is to some of these narratives that I now turn. When discussing the past in Mocímboa, there was frequently a reference to how distinctive the experiences of Makonde and Mwani were and how separate they were. This mutual understanding was upended one morning when speaking with Ansumane. Ansumane is an old Mwani man who at the time lived in Pamunda, a neighbourhood near the old administrative centre of Mocímboa that grew in size in the 1980s and housed Makonde and Mwani. He had lived in Mocímboa his whole life. We sat on his porch, and he slowly explained how he viewed the relationship between Makonde and Mwani. His was a radically different view of the connections and fractures between the groups and stood out against the prevailing narrative. His was an extreme perspective, denying difference and emphasising a common origin. This narrative is both consistent with the history of contact and migration that characterises the coastal north and irreconcilable with the historical record.

Mocímboa da Praia is all Makonde. The Mwani clans (Kabile) do not exist. The Kabile that existed in Mueda or here were all the same. When I travelled and went to Ibo or Nampula they would say that the Makonde from Mocímboa were arriving. This area all belongs to the Makonde. The Mwani are part of the Makonde. I am va-Shitunguli which also exists in Mueda and Ibo. Mwani is the one who lives along the beach, so even if they are Makonde living here, they are Mwani. The lineages that exist here are the same that exist in the plateau.

Ansumane perceived Makonde and Mwani to be a single group because he believed they shared the same origin and clans. His understanding supported the claim that any distinction was of recent origin, and that the distinction that appeared sharp was only superficially so. Although unusual, Ansumane's perspective presented a more nuanced and continuous understanding of belonging in the area. He stated that most people who claimed Mwani or Makonde identity were in fact related. Ansumane's view conforms to some of what we know from the historical record, but it is not a perfect match, and is lacking in detail. What it does is highlight the fluidity and proximity between Makonde and Mwani and the shifting understandings of identity that are seen elsewhere along the East African coast.¹⁶

Most of my interlocutors, however, had vastly different understandings of these distinctions, claiming Makonde and Mwani to be quite different peoples. Listing clear distinctions in religion, political allegiance, language, cultural mores and practices, my interlocutors would emphasise difference while the features that brought them closer were often not remarked upon. It was only when probed further that they would mention how close they were to neighbours of a different group, instances of intermarriage or mutual help. While it was frequently acknowledged that both groups share the space in the northern districts of Cabo Delgado, and that there are close connections between them (including familial relationships), the prevalent view was that of distinctive identities. By discussing the mutual conceptions of the other, uttered continuously by Makonde and Mwani, I hope to shed light on mutual and persistent, if often changing, representations of ethnicity.

Makonde and Mwani have distinctive histories and experiences of territorial occupation in northern Cabo Delgado. These histories were built on longer regional connections, broader political transformations in the region that impacted relationships and the hold on territory. These included the long-distance trade of ivory, the enslavement of people, and on-going conflict in the region¹⁷ leading to major demographic shifts¹⁸ and a population in flux. Group identities were being defined and redefined through warfare, movement and conquest¹⁹. It is to the ways in which these developed and shifted that I now turn.

The impact of the slave trade in the region led to a retreat of the Makonde to the relative safety provided by the Mueda Plateau. This territory offered protection against slave traders.²⁰ The hinterland's inaccessibility was welcome in times of war, and the villages in the plateau were heavily fortified and hidden. Makonde elusiveness and fierceness was considered a defining aspect of their identity for a long time.²¹ The enslavement of peoples and the growing hostility between raiders and raided added to the differentiation between the coastal and hinterland identities, those of Mwani and Makonde. Some of this hostility emerges in present day narratives. Fardon's account of the history of the political organisation of the Chamba describes a process in which constant raiding and refuge played a big part in the definition of identities.²² A similar pattern can be seen at play here, as the slave trade and its impact grew from the mid-1700s onwards.²³

Dias suggests a few possible waves of migration that brought the Makonde to the plateau.²⁴ Originating south of Lake Niassa, the group grew, with smaller waves of migration added later. These included peoples arriving from the Rovuma valley, south of the Messalo River, the area near Montepuez and elsewhere in the region in successive waves of migration. Some of the clans were said to have come from the coast, which is consistent with Ansumane's account. West was given different versions of the history of Makonde settlement:

'We came to the plateau from Mataka's country, fleeing the Angoni', an elder told me in Mwambula. 'We were once Yao'. Said another: 'We are really Makua. We took refuge here from the slave caravans' […] 'many of those who today call themselves Makonde were once Yao, and many who today call themselves Yao were once Makonde'. […] 'the same thing is true here between Makonde and Angoni.'²⁵

The Makonde share some clans with neighbouring peoples and incorporated rituals and practices from others as well. These different waves of migration led to alliances between different settlements, creating larger groups.²⁶ Makonde society was incorporative of others who wished to join. Israel points to it being a maroon society, where the adoption of language, and the skin markings that until the 1970s were visible in adult Makonde, as well as the marriage into a lineage, made one a Makonde.²⁷

While the Makonde coalesced into a largely acephalous group that resisted the slave trade, along the coast the Mwani had long established a foothold and were part of wider Swahili networks in the Eastern African coast. The origins of the Mwani as a group are as hard to trace as those of the Makonde. The earliest references to a coastal population come from medieval Arabic documents showing the establishment of trading posts along the coast.²⁸ The word Mwani translates as people of the sea, coast, or margin - 'those who live by the coast', and later became a linguistic community.²⁹ Their language, Kimwane, is an admixture of Shimakonde and Kiswahili. The Mwani lived in relatively dense communities under the leadership of sheikhs connected with Shirazi clans from Zanzibar.³⁰ Like the Makonde, the Mwani were formed of a mix of different populations who had moved and settled on the coast,³¹ but who maintained strong links with Zanzibar, Madagascar, the Comoros, and the wider region. The Mwani who lived in what is now Mocímboa were under the leadership of Sultan Muicumba in the mid-19^th century and are considered the autochthonous population, and the 'owners of the land'. Some of the villages along the coast still claim Shirazi origin.³² The coast became progressively separate from the hinterland, with its own political and socio-economic dynamic, keeping their independence from other Swahili cities but maintaining close relationships of trade.³³ More than confirming the historical record, the earlier account presented in this section highlights the proximity between the peoples inhabiting northern Cabo Delgado, and the long tradition of contact between the coast and the hinterland³⁴ that parallels some of the relationships along the Swahili coast.

The elites that ruled the coast during the 19^th and early 20^th centuries would be progressively removed from positions of power in the later half of the 20^th century, fostering growing resentment. Before this, however, the establishment of a colonial administration added to the separation between coastal peoples and those who inhabit the hinterland. It is to the changes introduced during the colonial period and the impact of forced labour in the growing separation between Makonde and Mwani that I now turn. I will focus specifically on a sisal plantation established in Mocímboa which came to be known as Mpanga and was frequently mentioned by my interlocutors in accounts of their lives and the history of the area.

 

Mpanga and the colonial experience

Mocímboa became increasingly important during the colonial period due to its good port, used to ship out some of the area's raw materials, and easy access to the northern hinterland and Tanganyika. A number of sisal and cashew plantations were set up in the outskirts of town. These plantations were regularly supplied with workers from the hinterland to fulfil their chibalo (forced labour) obligations. The growing, if uneven, hold of the Portuguese administration, and its unpopular policies in the first half of the 20^th century, gave rise to migration to Tanganyika³⁵ and growing resistance in Cabo Delgado. The plantation economy and especially the conscription for labour were responsible for large numbers of migrants from inland to the coastal areas (mostly forced migrants) and stimulated the settlement of some of these migrants in Mocímboa da Praia long after this. Long-standing relationships between the coastal and hinterland populations changed. The dynamics which had been the result, not of friendly relationships, but of commerce, slave trading and raiding, had established patterns between the populations became upset, and later turned on their heads by the colonial administration and the post-independence government. Along with the movement from the hinterland to the coast, another form of migration stemmed from the policies of forced labour, and that was migration in large numbers to Tanganyika.

I first heard about Mpanga from a Makonde man who had not been there but had heard many stories about the place as a child. Some of these stories has been so horrific that they had left strong memories. He was told that people had been forced to work there in dangerous conditions. One of the workers had lost an arm in one of the machines used to process the sisal. He described this in vivid detail, almost as if he had witnessed the event himself, and the story had clearly left an impression. However, when probing other interlocutors who had worked there in different capacities and for different lengths of time, I was unable to corroborate the story properly. It appeared to be told widely in the hinterland but had little connection with the realities of working at Mpanga, and remained in the imagination of those who had not worked there, being too young or, in some cases, born long after it closed. Though people remember Mpanga less now, and stories about forced labour seem not to be told as often, there are still many people alive who remember working there in different capacities. The physical remnants of the plantation are disappearing fast too, and what was left of the machinery and train tracks which connected it to the town have long been used for other purposes.

Mpanga, the Mocímboa Sysal Syndicate, located just outside Mocímboa da Praia, was one of the places where Makonde were brought to fulfil their chibalo (forced labour) obligations to make up for the annual taxes. The work was physically demanding, and the payment deemed insufficient. The workers complained often and found ways of escaping. One Mwani who had worked there described the situation thus:

People complained a lot and others escaped because the work was very hard. [...] It was difficult to work on the first few days and people had to learn. Most of those who escaped went to Tanzania. People could escape during the work and no one would notice. There were many people. Too many. The plantation was very big - it took about a day to go around it. [...] When people escaped they just wanted to find refuge.³⁶

Another Mwani described how people found ways of escaping, emphasising the difference between Makonde and Mwani responses to the work.

Many ran away, especially Mwani, because Makonde can work. When a Mwani was caught, they would run to Tanzania. When someone missed a day no one paid attention, but on the second day they would be searched for, and by then they would be far. The Makonde were many, and came from Muidumbe, Mueda and Nangade. They were the ones who worked the most, but for those living close to the border it was easy to escape.³⁷

Along with the workers there were Makonde there too. People who were not working forcibly, but who were often there to support their family members, helping sisters with children, or parents, or other relatives. Older children would often go to help take care of younger siblings or nieces and nephews while their parents worked on the plantation. When a married man was caught and made to work in the plantation, his wife would usually follow so that she could take care of him and, sometimes, work in the plantation for money as well. Sometimes other relatives, especially children, would also come. Some of my interviewees came to Mpanga to help a brother or sister when they were there. Because they were too young, they were not 'caught' themselves, but still lived in the plantation in order to help with raising the children and preparing food or fetching water. The plantation became more like a village (albeit one where most of the population lived temporarily). It had a village-like organisation and a hierarchical structure where the people who worked on a more permanent basis had a higher status than the temporary workers. They also had, as would be expected, a more stable relationship with the people in charge of the plantation. The experience of the voluntary workers always, unsurprisingly, stood in sharp contrast with the memories of the Makonde who were forced to work there.

The plantation was closed shortly before the start of the liberation struggle in 1964. The effects of the continuous movement of population from the hinterland to the coast, and the differing experiences of Mwani and Makonde when it came to forced labour, along with the memories of the period, however, have had a lasting influence in local level relationships in the area. The administration of the province was first passed onto the Nyassa company, a chartered company,³⁸ whose repressive and often violent actions were met with resistance from the local population. The charter for the company was revoked in 1929, at the same time as the new state regime was getting its footing in Lisbon. The administration of the province passed onto the Portuguese state, but there were strong continuities in brutal methods employed and the responses of the local population. Various forms of low-level resistance, what Scott³⁹ terms 'foot dragging', were employed, as well as escape to Tanganyika.

The longstanding migration to Tanzania created a background of support for Frelimo's war efforts. Life on the Tanzanian side during the struggle also gave Frelimo the first taste of administrative duties. Establishing and managing refugee camps with populations which kept increasing and, facing a number of crises, Frelimo had their first experience of policy-making which went far beyond the war effort. Life on the other side of the border was not easy. Crossing the border was difficult enough. Even though people were generally familiar with the paths and routes which would lead to safe areas on the river, it was still a dangerous enterprise and could entail meeting Portuguese soldiers or running into a skirmish between the Portuguese and Frelimo, which could have dire consequences. Once in Tanzania, people had two options, neither very attractive. They could go to one of the refugee centres which were created by Frelimo with the help of Tanzania and international organisations, or they could try and find a place to live outside these and stay in one of the Tanzanian towns.

The autochthonous Mwani worked for the Portuguese administration and the sisal plantations the Makonde were forced to work at. The separation between Mwani and Makonde became increasingly sharp, developing further during the anti-colonial war (1964-1974). It is to this experience that I now turn, describing the sharpening divide between Makonde and Mwani as the liberation struggle progressed and as the post-independence policies were implemented.

 

Liberation struggle

Led by Frelimo, the war produced dramatic demographic shifts in Cabo Delgado. Large numbers of Makonde moved to what became known as the liberated areas.⁴⁰ A sizeable number of Makonde escaped to Tanganyika and either remained there for the duration of the war, in refugee camps and towns, or returned to Mozambique to fight.⁴¹ Mocímboa housed a large Portuguese garrison during the war, was fenced with barbed wire, and its inhabitants, largely of Mwani origin, remained under Portuguese control.⁴² The differing experiences of the anti-colonial war are frequently alluded to as a means of establishing legitimacy in claims to the state, and accounts of the war are central in competition for local power and have become increasingly powerful since independence (1975), the civil war (1976-1992) and especially in the wake of democratisation and the first elections held in 1994.

The war began in September 1964 and in early 1965 entire settlements were moved to the areas of Cabo Delgado under the control of Frelimo,⁴³ and the first military school in a liberated area was created.⁴⁴ In some cases, the heads of settlement moved as well.⁴⁵ Many of those who moved to the liberated areas would remain there until the end of the war. When the new settlements grouped people from various smaller settlements, people would group together according to the areas they originated from. However. in these areas, even when the heads of settlement accompanied the people, they seldom continued to rule them. Instead, this would be done by the younger generation who had started work with Frelimo earlier, and many of the former vashilo (the young men who visited at night to effect political change, translated to night people) rose to the control and organisation of the liberated areas,⁴⁶ continuing the shift in power and authority that had been occurring in the plateau for decades. Those who moved to the liberated areas provided Frelimo with food, shelter and information. Life in these areas was extremely difficult. Some accounts tell of suffering and fear while living in the mato (bush):

In the bush there was a lot of suffering, people had lice, couldn't wash, or look for clothes, they would wear the same clothes for days. At night if there was an attack they would have to run [...] and could only return after learning that there were no soldiers. When they came back sometimes, they would start to cook and would have to run away again. It was very bad. There were many attacks. They [the Portuguese] would come by helicopter, by foot. There would be no food and no salt. They could not look for salt or for clothes.⁴⁷

The first years of the war entailed continuous struggle with people moving continually and staying attentive to the movements and actions of the Portuguese:

The Portuguese would come but the population had signals and would play the horn or shout and warn of the arrival of the tropa [tropa - troops - was the term used for the Portuguese; Frelimo fighters were called camaradas - comrades] and the population would run into the bush to different places. After the ambush the helicopter would come to collect the Portuguese troops and the people would return to their houses.⁴⁸

While talking about the lack of food and basic goods, my interlocutors would also remember the excitement of building a free area within Mozambique which was to be controlled and administered by Frelimo. Emerging scholarship on the liberation struggle presents a nuanced and diverse picture of the experience of the war and the memories that persist in different articulations at present.⁴⁹

Those who stayed in areas under the control of the Portuguese army remember the war as a difficult time. Fatima and her daughter Tonga, two Mwani women, described life in Mocímboa during the war.

Life here was also running. Frelimo would bomb here while the Portuguese bombed there [in the interior, where the Frelimo had their military bases] and we took refuge at the beach. It was all surrounded by barbed wire. We went to the fields with the soldiers. They shot in the air and that was the signal to return. We had cards with our name and picture which we left when we went to the fields. They were returned when we got back. If someone didn't return, it was because they had been captured by Frelimo. Some ran away, but many were caught.

Far from being the easy option later ascribed to them, the life of those who stayed in town was dangerous and full of fear. Fatima and Tonga described this as well.

When the war came, I was here. Frelimo came until Nkomangane [a village near the town] and started bombing. Those in Milamba could see the houses burning and heard the shots. A bomb hit a store and it burnt down. A rocket launched at the market hit a woman who was going to the bathroom and she was cut in two. In the garrison a soldier and his wife died. They [Frelimo] couldn't come in, but bombed from afar. There was war here and there. Nowhere was safe. We had PIDE [the Portuguese secret police]. Many people were denounced, and were arrested, beaten, killed.

A Mwani woman who had lived in Mocímboa for the duration of the struggle described the level of distrust the Mwani population endured from Portuguese and Frelimo alike.

We suffered with the Portuguese here, because they suspected we were supporting Frelimo, and Frelimo suspected that we supported the Portuguese. My family was kidnapped [by Frelimo]. Sometimes the Portuguese troops would beat us. We were in a very difficult situation.⁵⁰

The memories of the struggle and the growing influence of Makonde in the political sphere are frequently alluded to as reasons for the growing divide between the people inhabiting northern Cabo Delgado. Some of the tensions alluded to in the next section were discussed as having originated in the differing experiences of the war and the political shifts that developed during the political changes introduced during independence. The policies implemented since independence, which aimed at the construction of a strong and modern Mozambican state, and the transformation of the rural areas, introduced extensive changes in power relations,⁵¹ with the Frelimo-supporting Makonde dominating the state administration in the northern districts. This further increased cleavages between Makonde and Mwani.

During the civil war (1976-1992), Mocímboa was sought for refuge and safety, and the population of the town steadily increased. After the war ended, large numbers of refugees and migrants returned to Cabo Delgado from other Mozambican provinces or Tanzania and Mocímboa's Makonde population continued to rise. Old neighbourhoods grew bigger, and new ones were built, changing the urban landscape, and shifting population and power dynamics. The hold on power and local politics from the autochthonous Mwani grew weaker. Mwani support for Renamo⁵² put them at odds with local political structures affiliated with the ruling party Frelimo, whose political control endured. The association between Frelimo and Makonde remains largely uncontested. The spatial divisions alluded to earlier became increasingly blurred, and the distinction between Makonde and Mwani progressively sharper. The town changed further in the aftermath of the peace agreements, with a significant influx of refugees and migrants who had lived in Tanzania for the duration of the civil war (and some who had been living there since the liberation struggle) and who now moved to the coast.⁵³ This increase in population led to the construction of new neighbourhoods, as well as the expansion of villages and the development of new encampments and villages along the main road. It is to the fractures that emerged in this period and the violence that followed elections in 2005 that I now turn hoping to shed light on to some of the enduring local divides.

 

Rioting - foreshadowing fractures

Throughout the 20^th and 21^st century the relationship between Makonde and Mwani and the growing differentiation between then sharpened. For reasons associated with politics, socio-economic change and religion, expressions of identity have become increasingly distinctive and expressed in stronger terms. The politically motivated riots that shook the town of Mocímboa at the end of the summer of 2005 encapsulate the differences in dramatic fashion. I will briefly describe the events below reinforcing how the riots and their aftermath may shed light on existing fault lines that emerge at a later stage in the insurgency. The following description and analysis should be prefaced with the caveat: although there are some parallels between the riots of 2005 and the present violence, there are plenty of significant differences in scale, destructive reach, duration, people involved and affected, and geographical span. The riots were confined in time, lasting only a few days. The uncertainty that ensued, and the anticipation of more violence, lasted far longer.

The sudden death of the local council president in December 2004 was followed by a special local election in May 2005. The electoral campaign was fought fiercely by supporters of Frelimo and Renamo. The vote was difficult, and the results contested. It was only after 13 recounts that Frelimo's candidate was announced as the winner of the elections. Even after this, the results were strongly contested. For months afterwards there were daily demonstrations of large numbers of Renamo supporters.

The new council president, Amadeu Francisco Pedro, was sworn in at the beginning of September. At the same time, the local candidate for Renamo was also sworn in, in a mirror ceremony held in Renamo's headquarters. It was after these that the violence began. On 6 September 2005 at dawn, Renamo supporters attacked the buses leaving town. They then moved to the central rows of Nanduadua neighbourhood destroying houses in their wake. The confrontations that followed resulted in many dead and wounded. Renamo supporters, largely of Mwani origin, fought against Frelimo supporters, who were mostly Makonde. The destruction caused by the riots was largely circumscribed to a few areas in one of the town's neighbourhoods, Nanduadua. A few houses were also burned in Chimbanga, just outside Mocímboa, but no destruction beyond that.

It is in the ways that the riots were discussed that I find some parallels with the ongoing insurgency that affects northern Cabo Delgado since 2017. I will discuss these below and highlight the areas where they mirror current assessments. Identity and belonging, religion, socio-economic divides, political allegiance, power, and local and national politics, are all present in very striking form in the narratives of these riots and discussions of the insurgency. Less immediate, but no less powerful in how they underpin these narratives, are the accounts of historical relationships in this area, the continuities and changes through colonial occupation, two wars, and a recent peace process and the democratisation process that ensued.

The first accounts I heard of the riots, long before I even got to Mocímboa, presented the violence in terms of a clear-cut difference: one side was Islamic, the other Christian; one group farmed, the other fished; one was Frelimo, the other Renamo, and so on. Deeper issues are also expressed by these conflicts - issues of land use, access to money, and political influence: these are the root cause of the riots.

A report prepared locally to consider the causes and impact of the violence, remarked on one of their interviewee's assessment of the tensions: 'The Makonde are invaders, hostile to Islam'. This short sentence encapsulates some of the issues that were highlighted in conversation at the time. I will discuss two central ones: religious belief and the relationship of hosts and guests in Mocímboa.

The religious divide featured quite prominently in conversation. The Mwani have been Muslim for centuries. Their religious belief and practice made them suspicious to the Portuguese during colonial times. When speaking of the presence of the Portuguese along the coast and the possibility of travelling to the hinterland, my interlocutors would remember the fear of being forced to eat pork near Catholic missions as a deterrent to travel. The Makonde converted to Catholicism during the first half of the 20^th century and have since made religion an important part of their identity. In 2000 the Makonde living in 30 de Junho, a predominantly Makonde neighbourhood in the outskirts of Mocímboa, erected a cross to mark the Jubilee. The cross was placed along the main road at the entrance of the town. It was the source of much tension and conflict between Makonde and Mwani. One of the most tense moments of the riots, I was told, was when a group of Mwani supporters of Renamo set out along the road to bring down the cross, and were met by a group of Makonde 'prepared for war'. The confrontation did not escalate because both groups were told that the armed forces were on their way.

The issue of land ownership features prominently in local narrative as well. The opposition between donos (the owners) and guests is discussed often in scholarship.⁵⁴ In Mocímboa, the tensions that emerged from the slow loss of power from the Mwani throughout the 20^th and 21^st centuries, despite their belonging at the coast, matched by the Makonde dominance in state institutions and local political hierarchies, were a source of enduring conflict. Land use and occupation, though easily overlooked, are fundamental matters, nonetheless. This is not in any way a new phenomenon and has been noted in a variety of other places. The use of land is intrinsically linked with the movement and population dynamics which has been an important feature in this area of Mozambique for a very long time. The tensions that developed throughout different periods of the history of this area have a strong bearing in the present relationships between the groups living here, and the upending of hierarchical relationships at the local level as a result of political change features heavily in local narratives. There were plenty of stories of Makonde who went to the hinterland, back to their old villages, and Mwani who escaped to the islands nearby. Some of these people, I was told, never returned. The majority eventually made their way back to the town, when the situation was quiet. One of the fruit sellers in the market by the beach (Zalala) only came back after about 6 months. He was said to be a Renamo supporter, and because he was not seen for such a long time after the riots was thought to have died during the violence. Some people decided to move to their villages of origin and stay there because they were afraid it might happen again.

I have noted earlier that a great deal of cooperation and cohabitation emerges, despite the tensions. While this is generally the case, it is underpinned by a series of assessments made regularly that are derogatory and shed light on some of the lines of tension. Both Mwani and Makonde devise derogatory ways to talk about the other, and base this, mostly, on the perceived rules that guide either group. This stereotyped discussion of the other is not at all unusual, and neither is the negative aspect associated with it. Fardon discusses how 'cultural habits, sexual morals and so forth of exclusive categories have been found wanting under the lens of the inclusive category's perception of its own life habits. Ethnic stereotypes are usually based on the self-regard of the classifiers.'⁵⁵ In Mocímboa, I heard disparaging comments from Makonde and Mwani on the habits and practices of the other group, setting themselves as the inclusive category by exclusion of the other. Issues such as religious practice ('The Makonde are like monkeys, they worship their God under trees'); food taboos ('They [the Makonde] will eat anything', or '[The Mwani] pretend they don't eat during Ramadan, but that is not really true'); to observance of marriage rules ('You see all these Mwani men walking from house to house in the morning' [going from the house of one of their wives to the other in order to make sure they spent the same amount of time with all of them], 'The Mwani will not marry Makonde women because they think we are impure'); these were remarks I heard throughout my fieldwork. The comments were about education or lack thereof (which also appear in accounts of the insurgency when discussing grievances and the rejection of state schools and preference for Islamic schools,⁵⁶ trustworthiness ('What do you expect, the Mwani are prone to causing confusion'), hygiene, among others. They marked a less hostile relationship than the comments I heard in the aftermath of the riots, but they are still examples of a relationship which is not quite easy between the two major groups.

One final point of contention between the groups is the perceived privileges afforded to the Makonde through their participation in the liberation struggle. These take the form of pensions paid to ex-combatants of the struggle, or access to state institutions. These are not available to many Mwani living in the north who see themselves as the owners of the land, but lack the political power, or the access to that power, to exercise their rights. These differences emerge at the generational level as well, with the younger generation being unable to access pensions, or having the ability to secure a job, leading to ongoing resentment.

While not exhaustive, the fractures that emerge in the narrative point to important distinctions in political allegiance, religion and socio-economic power. These are some of the issues that connect past and present expressions of identity and linger at the margins of the assessment of the current violence. The present violence goes well beyond Mocímboa, affecting neighbouring districts and extending into Niassa province to the west. It has proven far longer in duration, going into 5 years and leading to close to 800,000 displaced people. In many ways, the levels of destruction are reminiscent of past violence, especially the anti-colonial war and the civil war.

 

Conclusion

I spoke with a young Makonde man, Daniel, about local and national history, and about Makonde identity. In his view the things that were important were the language and religion and being born in a Makonde family. Others would add to this the initiation rites, the way they build their houses, what they eat. I heard similar assessments of identity from Mwani men and women. However, these lists ignore historical context and the changing relationships between groups. Identity has long been understood as being linked to places, but it is accepted that time is important in the understanding of what constitutes identity. Both space and place have a bearing on the construction of Makonde and Mwani identity, and the changing contexts play a part in how these are defined and understood. Trying to understand local expressions of identity and how these impact relationships between groups at times of strife highlights some current fragilities in analysis. Local narratives of identity highlighted the clear divides and the impact of history in changing the relationship between Makonde and Mwani, placing them at odds at pivotal moments, but devoid of historical context and acknowledgement of change.

The external factors that made them different were a product of historical change. The religious divide, for instance - which was introduced by the missionaries to some extent - was there to begin with, since the Makonde and the Mwani did not practise the same religion. However, present claims of being either Muslim or Christian are stronger and have created further division, which for some people are unbridgeable. Political affiliation and livelihoods, said to create divides between Makonde and Mwani, were, similarly, the result of recent transformation and the result of historical change.

Far from being immutable, ethnic categories are ever changeable and assume diverse importance at different times. Recent theories on ethnicity point towards a need to avoid fixity and to an understanding of ethnic groups as more flexible units, integrated in both local and global worlds.⁵⁷ However fixed they may appear, expressions of identity and belonging respond to local and national politics and differing experiences of the past. This is often absent from current discussions of the insurgency. In a region such as this, where so many different groups share a geographical area, matters of boundaries between groups become increasingly complicated and fraught with confusion if looked at from a perspective of continuity and historical change. If looked at with a basis on local discourses of identity and belonging to a group, on the other hand, these divisions acquire much clearer, sharper boundaries. As is usually the case, Makonde identity, and similarly that of Mwani, is relational, dependent on context, created and maintained by contact. Turton⁵⁸ emphasises the fact that identity, although an 'imagined' category, is not an 'imaginary' one, since it is a factor of motivation and mobilisation, and is also the product of historical situations.

Considering the sharp fault lines and divides between Makonde and Mwani obscures the myriad instances of cooperation, peaceful coexistence, and forms of mutual assistance that were part and parcel of everyday life, my intention in this article was to highlight how the spaces of tension and separation may emerge as profound areas of division at times of conflict. Assessments of past entanglements, the frictions created by a differently experienced and remembered past, are perceived as insurmountable when discussing past experiences of conflict, and I suggest that a close attention to the ways in which these are currently articulated would be productive when trying to understand the longer engagements between those inhabiting the north and suffering the effects of the violence.

 

 

1 E. Morier-Genoud, 'Why Islamist attack demands a careful response from Mozambique', 2017 https://theconversation.com/why-islamist-attack-demands-a-careful-response-from-mozambique-85504.         [ Links ]
2 E. Morier-Genoud, 'The jihadi insurgency in Mozambique: origins, nature and beginning',Journal of Eastern African Studies, 14, 3, 2020, 396-412;         [ Links ] E. Morier-Genoud,Towards Jihad?: Muslims and Politics in Postcolonial Mozambique (Oxford University Press, 2024);         [ Links ] Crisis Group Africa Briefing N°178 Maputo/Nairobi/Brussels, 10 February 2022, https://www.crisisgroup.org/africa/east-and-southern-africa/mozambique;         [ Links ] IISS, 'The Islamist insurgency in Mozambique', Vol. 27, August 2021, https://www.iiss.org/publications/strategic-comments/2021/the-islamist-insurgency-in-mozambique.         [ Links ]
3 L. Malkki, 'Refugees and exile: From "refugee studies" to the national order of things', Annual Review of Anthropology, 24, 1, 1995, 495-523.         [ Links ]
4 J. D. Fearon, and D. D. Laitin, 'Ethnicity, insurgency, and civil war', American Political Science Review, 97, 1, 2003, 75-90;         [ Links ] B. R. Ferguson and N. L. Whitehead, War in the tribal zone (Santa Fe, NM: School of American Research Press, 1992).         [ Links ]
5 K. Askew, Performing the nation: Swahili music and cultural politics in Tanzania (Chicago: University of Chicago Press, 2002);         [ Links ] J. McIntosh, The Edge of Islam: Power, Personhood, and Ethnoreligious Boundaries on the Kenya Coast (Durham: Duke University Press, 2009).         [ Links ]
6 D. Parkin, Sacred void: spatial images of work and ritual among the Giriama of Kenya (Cambridge: Cambridge University Press, 1991);         [ Links ] McIntosh, The Edge of Islam
7 A large body of literature has investigated cross-border movement and dynamics in Mozambique: H. Englund, From War to Peace on the Mozambique-Malawi Borderland (Edinburgh: Edinburgh University Press for the International African Institute, London, 2002);         [ Links ] J. McGregor, 'People without Fathers: Mozambicans in Swaziland 1888-1993', Journal of Southern African Studies, 20, 4, 1994, 545-567;         [ Links ] J. McGregor, 'Violence and Social Change in a Border Economy: War in the Maputo Hinterland, 1984-1992', Journal of Southern African Studies, 24, 1, 1998, 37-60;         [ Links ] D. M. Hughes, 'Refugees and Squatters: Immigration and the Politics of Territory on the Zimbabwe-Mozambique Border', Journal of Southern African Studies, 25, 4, 1999, 533-552;         [ Links ] and elsewhere in Africa: W. James, War and Survival in Sudan's Frontierlands: Voices from the Blue Nile (Oxford: Oxford University Press, 2007);         [ Links ] M. Leopold, Inside West Nile: Violence, History & Representation on an African Frontier (Oxford: James Currey, 2005);         [ Links ] P. Nugent, Smugglers, Secessionists & Loyal Citizens on the Ghana-Togo Frontier: The Lie of the Borderlands Since 1914 (Oxford: James Currey, 2002);         [ Links ] P. Nugent and A. I. Asiwaju, 'African Boundaries: Barriers, Conduits and Opportunities' (London: Pinter, 1996);         [ Links ] C. Lentz, '''This is Ghanaian Territory!": Land Conflicts on a West African Border', American Ethnologist, 30, 2, 2003, 273-289,         [ Links ] providing insights as to the specific issues that arise while living at the border. There is a wide emphasis on the fluidity of borders, but also a caution not to see the borders as meaningless, and a reminder of the impact of administrative and state practices in the development of distinctive national and regional identities.
8 McIntosh, The Edge of Islam, 7.
9 P. Nugent, 'Putting the history back into ethnicity: enslavement, religion, and cultural brokerage in the construction of Mandinka/Jola and Ewe/Agotime identities in West Africa, c. 1650-1930', Comparative Studies in Society and History, 50, 4, 2008, 920-948.         [ Links ]
10 A. M. S. Santos, '"It's not my story to tell": ownership and the politics of history in Mocímboa da Praia, Mozambique', Journal of the Royal Anthropological Institute, 27, 3, 2021, 672-690.         [ Links ]
11 F. Barth, Ethnic groups and boundaries: The social organization of culture difference (Waveland Press, 1969), 9-10.         [ Links ]
12 A. Southall, 'The illusion of tribe' in Passing of Tribal Man in Africa (Brill, 1970), 28-50.         [ Links ]
13 E. Medeiros, História de Cabo Delgado e do Niassa (c. 1836-1929) (Maputo: Central Impressora, 1997);         [ Links ] A. R. Da Conceição, Entre o mar e a terra. Situações identitárias do norte de Moçambique (Promédia, 2006).         [ Links ]
14 P. Geschiere, The Perils of Belonging Autochthony: Citizenship, and Exclusion in Africa and Europe (Chicago: University of Chicago Press, 2009);         [ Links ] P. Geschiere and S. Jackson, 'Autochthony and the Crisis of Citizenship: Democratization, Decentralization, and the Politics of Belonging', African Studies Review, 49, 2, 2006, 1-7;         [ Links ] B. Ceuppens and P. Geschiere, 'Autochthony: Local or Global? New Modes in the Struggle over Citizenship and Belonging in Africa and Europe', Annual Review of Anthropology, 34, 1, 2005, 385-407.         [ Links ]
15 R. Sarro, 'Map and Territory: The Politics of Place and Autochthony among Baga Sitem (and their Neighbours)' in J. Knörr and W. Trajano Filho (eds), The Powerful Presence of the Past (United States: Brill, 2011), 229-252.         [ Links ]
16 Parkin, Sacred void; McIntosh, The Edge of Islam.
17 E. A. Alpers. Ivory and Slaves: changing pattern of international trade in East Central Africa to the later nineteenth century (University of California Press, 1975);         [ Links ] E. A. Alpers, 'Trade, state, and society among the Yao in the nineteenth century', The Journal of African History, 10, 3, 1969, 405-420.         [ Links ]
18 M. Newitt, A history of Mozambique (Indiana University Press, 1995);         [ Links ] Medeiros, História de Cabo Delgado.
19 Ferguson and Whitehead, War in a tribal zone.
20 J. Dias, 'Os macondes de moçambique. Vol. I: aspectos históricos e económicos' (Lisboa: CNCDP/IICT, 1998).         [ Links ]
21 H. O´Neill, 'A three months' journey in the Makua and Lomwe countries', Proceedings of the Royal Geographical Society and Monthly Record of Geography (1882);         [ Links ] H. O´Neill, 'Journey in the district west of Cape Delgado Bay, Sept.-Oct. 1882', Proceedings of the Royal Geographical Society and Monthly Record of Geography (1883);         [ Links ] H. O'Neill, The Mozambique and Nyassa slave trade (LSE Selected Pamphlets, 1885);         [ Links ] P. Israel, '"Kumungalela Guebuza": the Mozambican general election of 2004 in Muidumbe and the roots of loyalty of Makonde people to Frelimo', Lusotopie, 13, 2006, 103-25;         [ Links ] H. West, 'Villains, Victims, or Makonde in the making? Reading the explorer Henry O'Neill and listening to the headman Lishehe', Ethnohistory, 51, 2004, 1-43.         [ Links ]
22 R. Fardon, Raiders and Refugees: trends in Chamba political development, 1750-1950 (Smithsonian Institution Press, 1988).         [ Links ]
23 For a detailed discussion of Makonde ethnogenesis see P. Israel, The history of the Makonde of Mozambique, forthcoming.
24 Dias, Macondes.
25 West, Villains, 25.
26 West, Villains.
27 Israel, The history of the Makonde of Mozambique.
28 S. Von Sicard, 'Islam in Mozambique: Some Historical and Cultural Perspectives', Journal of Muslim Minority Affairs, 28, 3, 2008, 473-490.         [ Links ]
29 Conceição, Entre o mar; Medeiros, História de Cabo Delgado.
30 L. Bonate, 'Matriliny, Islam and gender in northern Mozambique', Journal of Religion in Africa, 36, 2006, 139-66.         [ Links ]
31 Conceição, Entre o mar, 215.
32 Conceição, Entre o mar.
33 Conceição, Entre o mar.
34 McIntosh, The Edge of Islam; Parkin, Sacred void.
35 Z. Kingdon, A Host of Devils: The history and context of the making of Makonde spirit sculpture (London: Routledge, 2013);         [ Links ] H. G. West, Kupilikula: Governance and the Invisible Realm in Mozambique (Chicago and London: University of Chicago Press, 2005).         [ Links ]
36 Ali Bacar Inguasse, Mocímboa da Praia, 2006.
37 Issa Momade, Mocímboa da Praia, 2006.
38 L. Vail, 'Mozambique's chartered companies: the rule of the feeble', Journal of African History, 17, 1976, 389-416.         [ Links ]
39 J. C. Scott, Weapons of the weak: Everyday forms of peasant resistance (Yale University Press, 1985).         [ Links ]
40 West, Kupilikula; P. Meyns, 'Liberation ideology and national development strategy in Mozambique', Review of African Political Economy, 22, 1981, 42-64;         [ Links ] T. H. Henriksen, Revolution and counterrevolution: Mozambique's war of independence, 1964-1974 (Westport, Conn.: Greenwood Press, 1983);         [ Links ] M. Hall and T. Young, Confronting Leviathan: Mozambique since independence (London: Hurst, 1997).         [ Links ]
41 Kingdon, A host of devils; West, Kupilikula; J. Tague, 'A war to build the nation: Mozambican refugees, rural development, and state sovereignty in Tanzania, 1964-1975' (PhD dissertation, University of California, Davis, 2012);         [ Links ] M. G. Panzer, A nation in name, a 'state' in exile: The FRELIMO proto-state, youth, gender, and the liberation of Mozambique 1962-1975 (State University of New York at Albany, 2013).         [ Links ]
42 A. M. S. Santos, 'Performing the past: celebrating women's day in northern Mozambique', Cahiers d'Études Africaines, 50, 197, 2010, 217-234;         [ Links ] A. M. S. Santos, 'Resistance and collaboration: conflicting memories of the liberation struggle (1964-1974) in northern Mozambique', Social Evolution and History, 3, 2014, 135-59.         [ Links ]
43 West, Kupilikula.
44 W. Opello Jr, Internal war in Mozambique: a social-psychological analysis of a nationalist revolution. (University of Colorado at Boulder, 1973).         [ Links ]
45 West, Kupilikula.
46 West, Kupilikula, 137.
47 Eugenia Bwanda, Mocímboa da Praia, 2006.
48 Albano Amissi, Mocímboa da Praia, 2006.
49 C. Darch and D. Hedges, 'Liberation and Biographical Narrative in Mozambican Historiography: The Struggle in Cabo Delgado, 1962-1974', Journal of Southern African Studies, 47, 4, 2021, 605-625;         [ Links ] J. Nkulunguila and J. N. Tembe, 'Frente de Cabo Delgado', História da luta de libertação nacional, 1, 2014, 264;         [ Links ] P. Israel, 'Mueda massacre: the musical archive', Journal of Southern African Studies, 43, 6, 2017, 1157-1179;         [ Links ] P. Israel, 'The Mueda Massacre Retold: The "Matter of Return" in Portuguese Colonial Intelligence', Journal of Southern African Studies, 46, 5, 2020, 1009-1036.         [ Links ]
50 Lukia Ali, Mocímboa da Praia, 2006.
51 J. P. Borges Coelho, 'State resettlement policies in post-colonial rural Mozambique: the impact of the Communal Village Programme on Tete province, 1977-1982', Journal of Southern African Studies, 24, 1998, 61-91;         [ Links ] J. Cravinho, 'Modernizing Mozambique: Frelimo ideology and the Frelimo state' (PhD thesis, University of Oxford, 1997).         [ Links ]
52 Renamo - Resitência Nacional Moçambicana - Mozambican National Resistance.
53 Similar patterns of tension and transformation during the civil war have been described by anthropologists who worked elsewhere in Mozambique. Notable among these are: S. Lubkemann, Culture in Chaos an Anthropology of the Social Condition in War (Chicago: University of Chicago Press, 2008);         [ Links ] N. Wiegink, Former Guerrillas in Mozambique (Philadelphia, Pennsylvania: University of Pennsylvania Press, 2020).         [ Links ]
54 Santos, It's not my story.
55 R. Fardon, 'African ethnogenesis: limits to the comparability of ethnic phenomena' in L. Holy (ed), Comparative Anthropology (Oxford: Basil Blackwell, 1987), 168-188.         [ Links ]
56 E. Morier-Genoud, 'The jihadi insurgency in Mozambique: origins, nature and beginning', Journal of Eastern African Studies, 14, 3, 2020, 396-412.         [ Links ]
57 McIntosh, The Edge of Islam.
58 D. Turton, 'War and ethnicity: global connections and local violence in North East Africa and Former Yugoslavia', Oxford Development Studies, 25, 1, 1997, 77-94.         [ Links ]