Negotiating Territoriality in North-Western Zimbabwe: Locating The Multiple-Identities of BaTonga, Shangwe, and Karanga in History

Multiple-identities are not an event, neither are they overnight occurrences. They undergo constructions and reconstructions over time. The BaTonga, Shangwe and Karanga speaking people in the Musampakaruma Chiefdom of north-western Zimbabwe are not an exception. Forced colonial displacements and post-independence involuntary (and/or voluntary) migrations resulted in their settling in the Musampakaruma Chiefdom from which they have now come to negotiate for space, and ultimately their identities too, in the Zimbabwean mainstream nation-state making process. For years, these three ethnic groups have had a primodalist alliance to identity wherein their identification with ancestral places of origin appeared to have been common. This, however, has changed as the new terrain has offered them new options prompting rethinking of identity and ethnicity concepts. Using qualitative and historical ethnographic data obtained in Musampakaruma from April to September 2017, this paper reports the historical and contemporary socio-political experiences of the people in the area advancing the multiple-identity phenomena. Taking Musampakaruma as a case, the broad nation-state identity is re-engaged in the paper from the perspective of so-called marginalised groups showing that while landscape and socio-ethno-identities are determinants of ‘multi-personalities’, deep theorisation of identity and ethnicity is required in nation-state development because ethnicities are based on interactions resulting in negotiated identities.


Introduction
Issues around a nation-state, democracy, sovereignty, and development project major challenges within the realm of nation-state formation and development. In Zimbabwe, Ndlovu-Gatsheni (2011) believes that the aforementioned is complex and against the notions of ethnicity and ethnic identities. Woldeselassie (2017) also notes that ethnic transformation in terms of on ethnicity, belonging, and identity are very intricate, while Jinadu (2004) is quick to say democratic principles based on individual rights are forced to face ethnic realities in Africa. In Zimbabwe, ethnic groups such as the Shona, Ndebele, and Whites live together; Tendi (2020) saw these ethnicities as visible game changers. Despite the existence of other groups, such as BaTonga, Shangwe, Nambya, Sotho, Venda, Coloureds, and Chewa, to mention but a few, scholarship has deliberately projected them as subgroups to the Shona or Ndebele or placed them side by side with the Whites. Tendi (2020) further states that during the country's struggle for independence from colonial rule, identity facets were used to locate and demarcate groups involved in the fight for political control as well as in the remaking of the nation-state.
This means that conflicts of the Zimbabwe African National Liberation Army (ZANLA) and Zimbabwe People's Revolutionary Army (ZIPRA) or Rhodesian Front (RF) military wings, Tendi (2020) believes, in a sense were to address representative ethnic groups as well as prominent ideologies which sought control of Zimbabwe. Post-independent Zimbabwe witnessed aspects of identity within the making of an inclusive nation-state. It was comprised of once rival dominant groups, now active players, who found some degree of common ground to rebuild the nation, this time as 'Zimbabwe' instead of 'Rhodesia'. In the 1980s Gukurahundi (Ndlovu-Gatsheni, 2012) witnessed dissident identity constructions and reconstructions leading to violence and atrocities exerted against inhabitants mostly of the Midlands and Matabeleland provinces as the independent government, using the Korean Fifth Brigade (Ndlovu-Gatsheni, 2012), sought to 'fish out' so called 'dissidents'.
In the late 1990s, right up to the greater part of the new millennium, identity reconstructions were alluded to again, this time around the land. Two hostile camps politically divided, the Zimbabwe African National Union Patriotic Front (ZANU PF) and the Movement for Democratic Change (MDC), became dominant political groups. In terms of ethnic composition, both parties had a cocktail of ethnic groups in the country. They only differed on ideological orientations and political subscriptions. In their differences around the land, identity constructions and deconstructions were also witnessed as the two tried to undo each other. ZANU PF accused Whites of being unrepentant and natives who supported them were regarded as sell outs and stooges of the Whites. In response, MDC characterised ZANU PF as demonic, corrupt, authoritarian, cruel, chaotic, and unfit to rule Zimbabwe. The 'our land' claim and consideration of it as part of an attempt to gain independence (Tendi, 2020) clearly demarcated identity boundaries defining who should own what, how, and under what circumstances. As aptly concluded by Nmoma, (2008), the then Robert Mugabe government had taken a bold decision to redistribute land but was failing to make the land policy mutually beneficial among the races in Zimbabwe.
Within the larger nation-state realm, Brubaker (2014:805) treats "ethnicity, race, and nationhood as a single integrated family of forms of cultural understanding, social organisation, and political contestation." Thus, reference to race reflects this basic identity tug-of-war between indigenous inhabitants and former colonialists over resources. Further, from 1999 to present indigenous inhabitants who questioned Mugabe's governance style amid the land distribution, politically or otherwise, were labelled sell-outs, stooges of the whites, puppets, spineless, detractors meant to reverse the gains of independence, and so on, with Whites reimagined once again as unrepentant colonialists. Those labelled as such in response reconstructed their counter labels. These ranged from naming the then Mugabe regime oppressors, corrupt, authoritarian, thieves, abductors, and in some cases killers. Such labels gained much of their prominence during the 2000s as the country continued searching for a formula to the nation-state question.
Against the above, the BaTonga, Shangwe, and Karanga in Zimbabwe are part of marginalised groups which, within the context of the nation-state, exist as if they are an extension of either the Shona or Ndebele. Their cry for visibility as well as autonomy for years has triggered robust demands for space in Zimbabwe's development of the nation-state. Historical realities tied to identity deserve scholarly attention to locate existing multiple-identities but are ignored in nation-state discourses. To do so, defining, and theorising identity against the ethnicity concept are vital before engaging with the aligned issues of the BaTonga, Shangwe and Karanga in Musampakaruma.

Defining and Theorising Identity
Despite an array of definitions of identity, there is no consensus as to what it should mean. Matanzima and Saidi (2020) say identity are narratives or stories people tell themselves or others about who they are (and who they are not). Gobulovic (2010) says identities are usually understood through self-definition as well as differentiation of oneself/a group from a perceived or existing other. Most views on identity, therefore, involve ideas of self-definitions. These self-definitions can be broken down into different qualifiers. Manyena (2013) identified four of these, which are national, ethnic, religious, and cultural. This paper addresses the 'ethnic' qualifier to locate BaTonga, Shangwe and Karanga multiple-identities in history within the Musampakaruma Chiefdom of north-western Zimbabwe. Ethnicity has come to be appreciated beyond mere cultural symbols and is now considered based on both local and universal social relations of production (Jinadu, 2004). Antweiler, (2015:26) using a cultural anthropological perspective, insisted that ethnicity "is a socially grown collective identity, which assumes a common history and origin as well as shared traditions, and claims to define a culture as different from (all) others." Tying ethnicity to primordialist (Berman, 1998) concepts has become a diluted endeavour (Schraml, 2014). In other words, ethnicity has now become a function of various turfs and systems of social relations that confront ethnic individuals on a day-to-day basis. Based on the day-to-day experiences of the social relations of production and related aspects, individuals appear to be forced, ultimately, to assume multiple-identities to belong.
Identities are linked to ethnicity (Jinadu, 2004) in that prevailing circumstances anywhere in the world regard that individuals define themselves and are defined by others in terms of the ethnic group to which they belong. Hence, to which ethnic group one belongs matters within the universal continuum of identity. People deliberately choose to have multiple identities against the notion that identity is more relational wherein ethnicity comes in as a set of such relations (Woldeselassie, 2017). While the relationship between ethnicity and identity appears blurred, scholars have attempted to mark out the line of these differences, however, noting aspects of convergence. For instance, Antweiler (2015) believes identity, whether collective or individual, is partially static in the context of others. In other words, it entails difference and demarcation; and its dynamics regard inclusion or exclusion -in some cases discrimination, because ethnicity is "an aspect or variant of collective identity" (Antweiler, 2015:26). Despite intellectual attempts to theorise and define the above concepts, it is generally agreed that identities, regardless of their qualifiers, must be negotiated and the people within geopolitical spaces continue to negotiate for identities.
The above, therefore, reveals that efforts should not be to explain what constitutes identity or ethnicity, but to theorise on the two concepts, as this has been the missing link within ethnicity and identity debates (Woldeselassie, 2017). Defining identity, ethnicity, or both to study a community is usually devoid of underlying ideological principles which are the bedrock of the concepts. Theorisation, is instead required because cultural frameworks create boundaries that guide and constrain the choices available to individual people (Schwartz and Petrova, 2018). It is important, therefore, to locate identity and problematise it within the larger geopolitical space. Ross (2007:289-287) rightly observes implications identities have on civil rights, especially what she calls "civic rights to minorities…in particular focus on the possible enhancement of rights (which are] a powerful vehicle for learning…deliberative democracy." As such, singular identity in a modern world appears non-existent against what is observable in contemporary societies, which is the multi-identity conception. The claim and reality are now that individuals or groups no longer exist or do not exist in a homogenous society with a single identity. In fact a repertoire of different identities (Ross, 2007) are observable, forcing individuals and groups to negotiate for identities to belong and partake in various engagements. This means ethnic groupings from which individuals come from or belong to have a repertoire of identities, making individuals reflect the respective groups they belong to.
Identity and multiple-identities at the level of concept are used here interchangeably because they assume and are based on seemingly similar ideological principles of identity and ethnicity. Group identities are manifested in and by individuals as they partake in different societal engagements. As Ross (2007:287) noted, "the individual will use each of these [identities], or a combination of them, contingently on where they are, whom they are with, and the particular social setting in which they find themselves." Thus, the microcosm of identity or multi-identity is the individual. This is because an individual reflects what he/she may have been socialised to with respect to identity or ethnicity. It is the individual who ultimately must relate to others for survival.
A fundamental and inclusive national identity ethos is overly complex in Zimbabwe. This is because, as alluded to above, there are various reasons ethnic identities are subscribed to, imposed, or otherwise are within the larger realm of state formation and development. Historical realities and marginalisation, especially in political and economic spheres, appear to make some groups have little conception of themselves as Zimbabweans because of their seeming neglect. It is not correct to claim that so called marginalisation problems are, as it were, a result of the fact that dominant ethnic groups have grip and control over the nation-state because one finds various individuals from most ethnic groups in the country occupying even influential positions in government, politics, industry, sport, and other areas key in the nation-state's functions. However, spatial identity is part of a larger niche of efforts that have become prominent lately involving the academic theorisation of the construction of identity in its broadest sense. These endeavours attempt to account for spatial elements or placerelevant aspects beyond language practices to consider "the function of other practices and semiotic domains, such as symbols, embodied movement and gesture, in the production of place/space and identity" (Benwell and Stokoe, 2013:204).
In this study, bias is not to consider place/space as "produced in and as a topic of discourse, [or] place/space as the location for [multiple-identities] discourse[s]" (Benwell and Stokoe, 2013:204), but consider places as determinants of those discourse as well as identities. As such, geo-semiotics, "the social meanings of the material placement of signs" (Benwell and Stokoe, 2013:209), is reclaimed to consider historical, socio-cultural, and religious material placements such as ancestral graves, sacred mountains, hot springs, hunting grounds, fishing, and crossing spots on the Zambezi river, as they all come in handy in defining geo-semiotic resources that give impute to identity claims as manifested in naming as well as related claims. In other words, 'Basilwizi' or 'Goba' in the case of BaTonga and Shangwe respectively, would be meaningless if the above instances of geo-semiotic resources are not located in a particular place or space. In the Musampakaruma Chiefdom, historical outreach to 'Basilwizi' or 'Goba' let alone 'MaVhitori' (Victorians) may have meaning by virtue of seemingly distant places and times, hence referential power may be lacking, especially if approached at the very surface. On a deeper level though, indexical connections are what allow specific ethnic group positioning against others within and outside the Chiefdom. This, hence, permits problematising between space and time to develop a full picture of identity constructions.
Objective identity refers to some characteristics one has, but not necessarily identifies with. Thus, subjective identity, Bilgrami (2018:160) believes, alludes to subjective belonging to a group of people who "undergo some sort of dislocation from their deep and longstanding roots". Thus, this paper is based on an ethnographic study carried out in the Musampakaruma Chiefdom of northwest Zimbabwe from April to September 2017. Observations of people's daily lives and how they engaged in multiple practices were noted. Contexts involved daily chores such as fetching water, fishing, firewood gathering, gardening, and farming activities as well as conduction of various economic activities around economic centres in the area. Religious activities such as church attendances and social contexts of sporting and/or drinking were observed, including household daily practices. Informants randomly sampled were also interviewed during the study, notably village heads by virtue of them being custodians of their communities. Informants were from the Karanga, BaTonga, and Shangwe speaking people, of various age groups, and were both male and female inhabitants in the area. Language use, dress, and other cultural practices were used to identify and group informants. Interviewing people across generations was important as it helped appreciate identity issues over time and across generations.

Of Identities in Musampakaruma
BaTonga, Shangwe, and Karanga allude to multiple-identities in Musampakaruma. Kang and Bodenhausen (2014:548) define multiple-identities as a collection of identities available for individuals to identify or be categorised with. Space, culture, history, religion, tradition, social behaviour, and language have influenced the formation of complex and multiple-identities (Raheem and Akande, 2019) in Zimbabwe. While CiTonga speaking people identify themselves as the BaTonga, they also have other identities they subscribe to such as the Basilwizi River People (Alexander and McGregor, 1997). This same situation is also prevalent among the Shangwe and the Karanga speaking people, where 'Goba' and 'MaVhitori' are used as identity labels, respectively.
Multiple-identities sometimes intersect to create more particular identities. Undoubtedly, when identities are constructed and reconstructed over time, intersections occur with phenomenal results. These often follow multifarious elements that determine identity formation, such as language and landscapes. Though the Karanga and Shangwe have individual group-identities, they are both identified as Shona speaking people in the broader national identity categorisation attributed largely to aspects of identity intersections. In the past, studies tended to focus more on the difficulties and challenges associated with multiple-identities. Attention is now being drawn towards opportunities that emerge from possessing distinct multiple-identities or having overlapping group-identities (Kang and Bodenhausen, 2015) to which this study is a part. Using Musampakaruma as a case, it emerged that people with multiple-identities draw numerous social as well as political benefits from this status quo. For instance, BaTonga people refer to themselves as 'Tonga' on one occasion and in the next subscribe to 'Basilwizi' and this was seen as well calculated. When seeking strong ties to landscape and territorial historical indigeneity or origins, the 'Basilwizi' identity was used; whereas 'Tonga' tended to be connected more to language and culture as well as used to mark group distinctive place within the nation-state matrix.
Colonialism had the negative effect of not inventing identities from scratch, but it also reinvented, rigidified, and politicised existing ones in several ways (Ndlovu, 2011). 'Shona' is a product of missionaries, colonial administrators, and scholars' efforts reinforcing the notion that identities are a coinable phenomenon (Sithole, 2018:414). Before colonialism in 1890, the term Shona was never used to refer to local people as the absence of a central authority in the Mashonaland region made local people define themselves politically as subjects of a particular chief, but this was not done linguistically or culturally (Ranger, 2010). This created multiple-identities for the Shangwe at the time as they began to be identified or identify themselves as either Shona people or Shangwe speaking people at the advent of colonial administration. The Shangwe further assumed many more identity categories during this time and even after colonialism showing that imperialism was instrumental in the creation of multiple-identities for many Zimbabwean cultural groups. This making and unmaking of identities in Zimbabwe did not die with colonialism as the same processes continued after independence.

Migration of the BaTonga and Shangwe into the Zambezi Valley
The BaTonga and Shangwe settlement in the valley occurred at different times (Matanzima and Saidi, 2020). Early inhabitants of the area who came to be known as Musampakaruma were Tongaspeaking people. Archaeological evidence locates the BaTonga in Southern Zambia and Northern Zimbabwe around the start of the present millennium (Matthews, 1976). Saidi (2019) says the BaTonga might have interacted with people of the late Stone Age. Their adaptation to the extremely scotching climatic and environmental conditions of the Zambezi Valley landscape indicate that they had occupied the valley for a long time (Colson, 1960;Sitambuli, 2016). Their stronger cultural attachment to the valley-landscape demonstrates that they inhabited the basin area for centuries (Matanzima and Saidi, 2020). Undoubtedly, they had stronger attachments to the Zambezi than the latter minority groups of the valley such as the Shangwe immigrants.
Further, the BaTonga are reported as having had no ethnic label yet in the precolonial era, they identified themselves as Basilwizi, or People of the Great River/River People (McGregor, 2009:22), a fact buttressed by some BaTonga elders in Musampakaruma. Scholarship has onomastically as well as linguistically traced their identity and the ethnic label, Tonga, as having been imposed on them by the Shona, specifically the Shangwe-speaking people they interacted with. The term Tonga means to rule or judge. They were regarded as Tonga because they had no paramount Chief. In other words, they ruled or judged themselves, meaning their unique system of governance was different from that of the Shona or Ndebele, hence they were regarded as Chiefless people (Ncube, 2014). Pre-colonial societies in Zimbabwe were fluid enough to influence each other when it came to cultural and identity issues. Furthermore, it is important to note that their status as a 'River People' shows the longstanding attachment to Kasamba besi, the Great River (Zambezi), which was practically apparent in their exceptional skills and knowledge of fishing and even crossing it despite it being infested with riverine dangers. The river at the gorge flowed very rapidly and its course was very rough to an extent that crossing it was "a highly specialised skill" (McGregor, 2009:25), hence the Kasamba besi axiom. Outsiders who needed to cross the river sought help from local BaTonga ferrymen. Even this landscape knowledge was a determinant identity factor to claim belonging.
Though, the ethnic label Tonga became popular in the Valley, people retained their 'River People' status right up to the contemporary era. This identity marker has been maintained because the BaTonga regard Kasamba besi as their ancestral landscape. Colson (1960) confirms that very few tell stories of having originating outside the Valley and Musampakaruma BaTonga elders maintain the same. It can be assumed that entitlement to the Zambezi region could now be deliberately done to claim and negotiate for political as well as landscape space, control, and access to resources therein. BaTonga's identification with the great river indicates centrality of the canal in all areas of their lives (McGregor, 2003). It also shows interconnections between landscape and identity, especially affordances of the landscape when it comes to identity-making in north-west of Zimbabwe. The status provided by the 'River People' also distinguishes the BaTonga of the plateau from those of the Zambezi Valley, "though they shared [similar] language and cultural traits" (McGregor, 2009:25).
Similarly, Shangwe speaking people also connect themselves with the Zambezi landscape in relation to identity. Shangwe speaking people are a sub-Shona group. Origins of the term Shangwe are not clear, but there were several versions given in different oral accounts. Nyambara (2002) gives the impression that the term Shangwe began to be used by immigrants from the south in the 1960s to refer to indigenous people they found in the north of Zimbabwe. He also gives evidence showing that these people began to be referred to as Shangwe people in pre-colonial times. Citing Marr (in Nyambara, 2002:292) Nyambara says, "Abashankwe was used by the Ndebele to refer to the peoples of Gokwe (northern Zimbabwe)." To indigenous people of Gokwe, the term Shangwe describes frequent droughts and famines that often devastated that area in the pre-colonial period. Gokwe is generally a very dry area and prone to periodic drought and famine (Nyambara, 2002). Therefore, landscape qualities and features were used to cast derogatory identities by dominant ethnic groups upon seemingly weak ones. There is always a motive behind derogatory uses of terms by dominant ethnic groups to identify others as primitive. The Ndebele regarded themselves as superior to the other ethnic groups they dominated and found weak or inferior. Hence, they looked down upon Shangwe-speaking people of the north. Like other smaller groups, the Shangwe were deliberately supressed and as such got assimilated into the Shona broader category during the colonial period (Ndlovu, 2011) despite having never identified themselves as such; they were known as the Shangwe or Goba (Lancaster, 1974).
Shangwe immigrants are believed to have migrated into the Zambezi Valley at various time intervals from the beginning of the 15 th century up to the 19 th century (Lancaster, 1974). But, Colson (1960) says some immigrants may have arrived in the early 20 th century, and these people, perhaps, were running away from the expansion of the colonial settler regime and established themselves in the northern Zimbabwean plateau. Scudder (1962:25) notes that More recent immigrants…entered the Valley as refugees from the Ngoni and Ndebele raids during the 19 th century. Where they penetrated the Valley, they have been absorbed by the Valley Tonga except in the Lower river region (i.e., Musampakaruma). Here they retained more of their linguistic and ethnic heritage.
This refugee status emerged from the verity that the Shangwe are known to have arrived running away from different forms of violence in other parts of the Zimbabwean plateau. It was relayed that Shangwe speaking people were late comers in the Musampakaruma Chiefdom and that they identified themselves as the Goba (Valley) people because of their occupancy of the valley which they came to regard now as home (Matanzima and Saidi, 2020). This shows that they held multipleidentities in the Zambezi Valley before their subsequent displacement in 1958 to pave way for the construction of the dam. Various socio-political and environmental processes were to later entangle identities among the Shangwe. The BaTonga and Shangwe cases demonstrate how notions of identity were enabled and structured through the Zambezi landscape. Though both the BaTonga and the Shangwe identified themselves with the Zambezi landscape, they did so in very different ways. As shown, the BaTonga identified themselves with the River (Lwizi) whereas the Shangwe identified themselves with the valley (Goba). Evidence from Musampakaruma indicates that some of the Shangwe immigrants were absorbed or assimilated into the BaTonga culture, though some retained their linguistic traits.
This process of assimilation facilitated the formation of other different categories of identities among the Shangwe and BaTonga alike. McGregor (2003) says that the riverine society for the BaTonga was not tightly bound because they shared broader identities with other minority groups of the Zambezi, such as the Shona. Lancaster (1974) observed that some Goba people ended up identifying themselves as Tonga, especially when dealing with the outsiders. In some cases, therefore, minority groups borrow identities of the majority to avoid marginalisation. According to Scudder (1962) and Colson (1960), the BaTonga were the majority in the Valley whereas other groups such as the Shangwe were a minority group. Oral evidence attests to interconnections between the Shangwe and BaTonga in the Zambezi Valley, hence one Tonga Chief said, "BaTonga and the Shangwe people of Musampakaruma are very close relatives. Our forefathers lived side-by-side along the Zambezi before they were removed from the river to where we are today" (Interview with Chief Mola, 12 June 2017).

From the South to the 'North -The Karanga Displacement
The Karanga were forcibly removed from the southern parts of Zimbabwe (Masvingo Province) by the colonial government which intended to decongest Tribal Trust Lands in that province, subsequently creating colonial medical frontiers against tsetse flies in Gokwe in the 1960s. Demarcation was also meant to protect white-owned cattle ranches in Kwekwe (Mudzimu, 2016:52) from tsetse. Land pressure in the southern parts of Zimbabwe had become untenable and the sparsely populated northwest of Zimbabwe became a dumping ground for these displacees (Nyambara, 2002).
These Karanga immigrants from the southern parts of Zimbabwe and those from other parts of the country were labelled Madheruka by the Shangwe. Shangwe speaking peoples of Gokwe and Musampakaruma hosted Karanga immigrants. The Madheruka in turn regarded the Shangwe and Tonga they found in north-western Zimbabwe as primitive, backward, and resistant to change since they were stereotyped as not wanting to resort to newly introduced cotton production in their region. The Madheruka regarded themselves as modernised because they had been exposed to the modernisation introduced by Europeans in areas of their origin. For example, the Karanga (or Madheruka) had mastered cotton growing, which was introduced in Gokwe during the time they immigrated into this region (Nyambara, 2002). The Karanga came to be part of the Shangwe and Tonga people of the Gokwe district. Gokwe shares a boundary with Musampakaruma. The border is very porous and people in both areas constantly cross the boundary for everyday socio-economic activities. It was observed that people from Musampakaruma herd their cattle in Gokwe and even take their children to Gokwe schools. It was also noted by the researchers that people from Gokwe bring their cattle to dip tanks in Musampakaruma. Inhabitants residing in the immediate vicinity of the boundary did this as economic or social services would be far from their villages within their Chiefdoms.
Over time, some Karanga who had occupied Gokwe became part of Musampakaruma. Several factors led to this situation. Firstly, many reported that it was a result of voluntary migration from Gokwe to Musampakaruma in search of land for agriculture. The other factor was the political redrawing of the boundary of the two areas in the 1980s, especially in the north-western parts of the Musampakaruma Chiefdom. Explaining the boundary issues, a female inhabitant explained that: "When we came from Masvingo we were part of the Gokwe District. At that time, the demarcation of Gokwe and Musampakaruma was Hwadze River. Over time, the boundary was changed from being Hwadze River, to that mountain that you see. That is how we became part of Musampakaruma Chiefdom" (Interviewed 5 April 2017).
Some Karanga voluntarily moved to Gokwe and Musampakaruma from the southern parts of Zimbabwe after independence was gained. One of the reasons why they moved, as one female immigrant explained, was: "For us we had problems in the family. Family members would get sick, others even died under circumstances we could not understand. When we consulted traditional healers, it came out that we were being targeted by witches so had to move from the south to here. There could be some who share the same history with us, but others may have their own reasons." Thus, fear of witchcraft in the south, though each person or family could have had their own reasons for migrating, stood out as topical in explaining reasons for migrating. Since then, some Karanga of Zimbabwe became part of the Musampakaruma Chiefdom and as such commenced social interactions with the Shangwe and BaTonga people. Hence, these three ethnic groups came to occupy this Chiefdom because of displacements and/or voluntary migration. They all came from somewhere and this history now shapes the way identity issues are negotiated as well as shared in the Chiefdom in contemporary times.

Displacement of the BaTonga and Shangwe from the Valley
The BaTonga identity as a River People and the Shangwe identity as Valley People were disrupted. Today, they continue to say they are river and valley people respectively even after their lives ceased to be river/valley focused. These groups have maintained landscape-focused identities for political, socio-economic, and cultural reasons related to accessing Lake Kariba's resources. The dam led to the forced displacement of 57 000 Tonga (Zambian and Zimbabwean) (McGregor, 2009), Shangwe and the Kore Kore (another sub-Shona group in northwest Zimbabwe) from the river, which had been the backbone of their livelihoods, and moved to the dryland tsetse infested plateaus with poor soil without agricultural potential. The river had necessitated activities such as fishing and riverine cultivation. Since the 1950s those displaced are still economically impoverished as evidenced by their outlook in Musampakaruma. Affected groups were also uprooted and disconnected from their ancestral graves, sacred sites, as well as places of religious worship traditionally central to their socio-cultural and religious beliefs as well as practices.
People of Musampakaruma were the first to be displaced as they were in the immediate vicinity of the Kariba gorge. Hence, they are the ones who were the most negatively affected by the swiftness of the move. The Musampakaruma people were moved to an area approximately 200 kilometres from their ancestral lands. Both the BaTonga and Shangwe in the Chiefdom had their identities disrupted (or later enhanced) by the move. Ironically, Europeans quickly forged their own identity and belongingness to the Zambezi Valley in the process of creating Lake Kariba. This process of forging artificial identities and belonging to the lake was meant to delete and replace the BaTonga/Shangwe in order to benefit directly from the newly created lake. From 1958 to 1980 Europeans involved in the dam construction engaged in socio-economic activities that served to cement their own belongingness to the Zambezi landscape. Hughes (2010:1) declares: Imperial colonisers do not seize land with guns and plows alone. To keep it, especially after imperial dissolution, settlers must establish a credible sense of entitlement. They must propagate the conviction that they belong on the land they have just settled. At the very leastand this may be difficult enough-settlers must convince themselves of their fit with the landscape of settlement. In other words, while excluding natives from power, from wealth, and from territory, overseas pioneers must find a way to include themselves in new.
The same was also the case at Lake Kariba and to symbolically appropriate the landscape, Hughes, (2010:59) further argues that "one had to do more than evacuate blacks from it." Mashingaidze, (2020:7) also says, Government trucks had moved the Tonga physically. Spatial and temporal scales associated with exploration pushed them metaphorically to the margins (and the result of this process was the creation of 'white hydro-social identities' in the waterscape. The landscape in and around the Zambezi (Lake Kariba) was named after European men/women, mainly in respect to those involved in the dam construction, animal rescue, and tranquilisation exercises, known as Operation Noah. With varying degrees of directness, place names coupled the lake to white economic developments, history, and culture. A clear example is the Fothergill Island which came to recognise and honour Rupert Fothergill, who was considered a hero of Operation Noah (Robins and Legge, 1959). The post-independent nation-state has not promoted meaningful BaTonga, Shangwe, and Karanga access to the Lake (Mashingaidze, 2013:9-10). McGregor (2008) argues that although the law was deracialised after Zimbabwean gained independence, there were notable continuities in the lake's regime that further barred the BaTonga, Shangwe, and Karanga from accessing the Lake.
The BaTonga, Shangwe, and Karanga have become inhabitants in the Musampakaruma Chiefdom, and they speak CiTonga, and CiShona (Karanga) languages. One of the researchers was born in northwest Zimbabwe and is well versed in CiTonga, Shangwe, and Karanga languages and as such his multilingual abilities came in handy during fieldwork. Some of the BaTonga people in the Musampakaruma Chiefdom claimed to be related to those of Mola. Mola and Musampakaruma Chiefdoms were neighbours in the Zambezi Valley, and according to tradition there are some Shangwe speaking people who were also part of the Mola Chiefdom. These Tonga speaking people also claimed to be related to the Shangwe. According to oral reports, these two groups lived together peacefully. As one local elder explained: "The Tonga and the Shangwe people are inseparable. They are closely interrelated. Some Shangwe people are able to speak CiTonga language and the Tonga people are able to speak CiShangwe language. Though I identify myself as Shangwe. My grandmother was a Tonga speaker" (interviewed 6 April 2017).
Tonga speaking people continue to identify themselves with the Zambezi Valley landscape as they did before the resettlement. They continue to narrate stories that connect them to the Zambezi Valley, and call themselves River People. Their stories of attachment to the Zambezi are often nostalgically narrated in ways and for many elders bring back memories of the resettlement of the late 1950s. Tonga-speaking people who claim attachment to the Valley are often motivated by political ambitions to access the lake's resources (Matanzima and Saidi 2020;McGregor, 2009). One Tonga politician in Siachazangwa Village indicated that: "We the Tonga people came from the Zambezi Valley. The Lake you see today is ours. We used to go and appease our ancestors in the Lake as the Tonga and the Shangwe people. Even the resources in the Lake are ours, the water is ours" (Interviewed 1 April 2017).
One is expected to have a specific identity as nobody is commonly identified and expected to be coming from the Zambezi today. The contemporary practice is that people come from different Chiefdoms established after resettlement in north-western Zimbabwe. Therefore, identity is variedly and told in different social and environmental contexts. Stereotyping still prevails especially as perpetrated by outsiders, as explained by a female participant in the study who said: "We are looked down own upon by many people. People say we are backward. We are not clever…" (Interviewed 1 April 2017).
Such negative narratives directed towards Tonga-speaking people are widespread in Zimbabwe. Due to the negative connotations created by these stereotypes, some Tonga speaking people residing in urban areas now avoid identifying themselves as such. This is done as a defensive strategy to avoid being looked down upon. However, sometimes it is difficult to distinguish BaTonga from the Shangwe as they claim to be related and share various cultural and religious activities within the Chiefdom. Some inhabitants identify themselves as both BaTonga and Shangwe. Shangwe people are also able to speak CiTonga and are related to those in Gokwe. Though the Shangwe in Gokwe and those in Musampakaruma might have different histories shaped by their different encounters with colonialism, their language is similar to that of the Kore Kore language spoken in the Hurungwe District, which is also part of north-west Zimbabwe, but differences do occur.
Despite having been displaced from the Zambezi Valley, the Shangwe still claim attachment to it. When Shangwe elders were asked where their origins are, they consistently mentioned that 'Takavha kuGoba' (we came from the Goba Valley). They do not identify Musampakaruma as their place of origin. In contrast, local youths are the ones who claim origins in Musampakaruma because they were born there. Furthermore, the Shangwe, like the Karanga, are regarded as Shona people mainly by other groups from other parts of Zimbabwe. This is largely because of their language which draws links with Karanga, one of the five dialects categorised as CiShona. Thus, nationally they are identified as the Shona people. The Karanga form a minority group among the three in the Chiefdom.
The Karanga people of Musampakaruma also identify themselves as MaVhitori. When Zimbabwe was colonised, European colonialists went on to name various places in the country including the Zambezi Valley during the Kariba dam construction. This reflected a colonial tendency almost everywhere on the continent to name much of the terrain they had colonised. Masvingo was no exception as it was once known as Fort Victoria, named after the Queen Victoria of England. It was renamed when the country gained independence to Masvingo. The post-independent landscape identities refer to native Africans from Masvingo, and much of the province, as the Victorians. This was localised, used by other ethnic groups elsewhere, and by themselves, as MaVhirori. As such, the Karanga linguistic identity, in most cases its unique accent of their dialect brought forth many links to the Vhitori (Victoria) identity. The Madheruka identity has ceased to be a signifier, presumably because the colonial identity died with the end of colonialism.

Conclusion
The formation of contemporary multiple-identities found in Musampakaruma Chiefdom was not an overnight process but evolved over time. Historical processes, from pre-colonial to the postindependent era, involved the making and unmaking of identities. Identities of these groups were constructed and re-constructed over space and time where several factors necessitated these identity (re)constructions. For the BaTonga, landscape influenced their identification with the river; their interactions with the other local ethnic groups resulted in other groups naming them the Tongas, meaning a Chiefless people. Europeans constructed the primitive identity for the BaTonga. As for the Shangwe, their pristine and infertile soils identified as Shangwe necessitated their identification as Shangwe. They also connected themselves to the valley and identified themselves as Goba and, during colonialism, they were also regarded as primitive. The Karanga were once identified as Madheruka, and this was connected to their economic activities and were associated with the widely held perception of them as being modern. Though they are called Karanga, in contemporary times, they have also come to call themselves and be called MaVhitori, stemming largely from the European naming of the Masvingo as Fort Victoria at the turn of the 19 th century.
Across time, each of the three groups has had their identities imposed by the more powerful others. This in a way shows the fluidity of pre-colonial African communities when it comes to identity issues. Society was also fluid to the extent that members from one ethnic group could even identify themselves with identity markers of others. For example, the Shangwe historically identified themselves as Tonga. They took advantage of having lived among the BaTonga for a long time and being able to speak the CiTonga language. Because of that, they identified themselves as Tonga when dealing with the outsiders. There is no doubt that territory has played a major role in the construction of identity in Musampakaruma. The Shangwe and the BaTonga subscribe to Goba and Basilwizi primordialist identities constructed based on the landscapes they were historically displaced from. They have retained these identities despite the impact of colonialist and globalisation forces. These landscape-related identities are mentioned in the politics of the landscape at Kariba in which these groups contest for space with various regimes and powerful others who have historically excluded them from accessing Lake Kariba.
Further studies need to focus on how these three groups interact politically, socially, economically, and religiously. Results from such studies can be used to unravel the grand making of the nation-state as well as bring judgment to the celebrated land reform programme to assess how the three groups reacted or benefited from the Fast-Track Land Reform Programme, dubbed the 3rd Chimurenga. What this article has done is give a background to the emergence of the entangled and multipleidentities existing in northwest Zimbabwe.