This article examines the social construction of difference as part of a complex cluster of historical factors that have shaped Khwe identity and livelihoods in west Caprivi, Namibia. It develops recent discussions of the mobilisation of ethnic identity by the Khwe since Independence and by other San groups elsewhere in Namibia. Lastly, it investigates the contributions of the colonial and apartheid administrations towards shaping Khwe and Mbukushu identities and ethnicities and the relationships between them.Footnote 1 In a context where historical and contemporary evidence suggests that Khwe and Mbukushu have long had relationships of economic exchange, interdependence, and social intercourse, I ask why west Caprivi's residents so often represented their respective identities in a polarised fashion in the period 2004–06 and how the themes of land, ethnicity, and political authority became the fulcrum of Khwe–Mbukushu politics.
A well-established historical literature on ‘tribes’, ‘tradition’, and ethnicity in Africa informs my understanding of identity and agency and the ways in which African ‘customs’, institutions, and identities have been intimately shaped and transformed by interventions that in part served empire and apartheid.Footnote 2 In particular, I am mindful of how historians at first accredited colonial actors with the principal role in ‘inventing’ tribe and tradition but later highlighted the contributions of Africans themselves to processes of identity and institution-building.Footnote 3 The literature on ethnicity and rural politics also informs my attempts to account for the wide range of institutions that interact at any one time to foster local identity-building.Footnote 4 I am particularly guided by recent work that calls for historical analyses that recognise the multi-layered, multi-authored, process-driven and political nature of identities that are often popularly conceived as primordial and timeless.
Taking the dialectical social construction of identity and ethnicity as its departure, this article shows that, from the early twentieth century, west Caprivi and its surrounding regions were a site of shifting power relationships, and struggles, between pre-colonial and colonial states and their subjects. It was also a site for numerous migrations and displacements. The article pays close attention to the ways in which colonial and apartheid administrations shaped relationships and identities through institutions of political authority. In particular, it shows how, through veterinary interventions, ‘Bushmen’ policies, and the relationship of the Khwe to the South African military from 1940, these administrations made active contributions to cementing Khwe and Mbukushu identities.Footnote 5 As Alexander puts it, ‘Colonial “inventions” did not so much determine the social order as provoke debates which themselves shaped struggles over authority and resources.’Footnote 6 Through interventions that fostered the differential treatment of Khwe and Mbukushu and created opportunities for Khwe alliances with the state, the South West Africa (SWA) administration altered the balance of sociopolitical power between Khwe and Mbukushu and their respective access to land and natural resources. The administration also increasingly recognised west Caprivi as a ‘Bushman’ area. These factors had significant implications for identity politics and associated claims to territory and authority in west Caprivi, and continued to influence these politics after 1990.
Scholars have pointed out the inadequacies of the terms used to describe the approximately 100,000 people who were historically hunter-gatherers and who speak a Khoisan rather than Bantu language. The most common terms are ‘Bushman’, ‘San’, and ‘Basarwa’, all of which are externally ascribed and have derogatory origins, but which are employed to some extent by the groups to whom they refer. I use the term ‘San’ here to refer to this collective. The San umbrella incorporates five linguistic families and many more distinct languages. I use the self-ascribed term Khwe to refer to the majority section of the San group living in west Caprivi.Footnote 7
ETHNIC RELATIONS AND SOCIOPOLITICAL CHANGE IN THE PRE-COLONIAL AND EARLY COLONIAL PERIODS
In the past … the Mbukushu would just come and attack your family and steal your daughter. Or the family would give their daughter [away], out of fear. I heard that story from my grandmother, [about] how we were suffering [because of] the Mbukushu. So myself, I hate Mbukushu even today, because I remember the bad things.
(Elderly Khwe woman)Footnote 8If Mbukushu found Khwe footprints in the bush they'd follow them and steal their [hunted] meat. Our history with them goes long back … They used to kill our old people if they found them [out in] the bush [alone] … Today after Independence we are still thinking about the past. We can discuss and we can even laugh about it, but we won't forget.
(Khwe man)Footnote 9In the past, the Khwe and Mbukushu were together, with no problems, living under one chief. Many tribes respected the Mbukushu chief because of his rain-making powers.
(Elderly Mbukushu man)Footnote 10We were staying together with the Khwe … Khwe would bring fruit and meat to [their] Mbukushu friends … The Khwe weren't ploughing [in the past], only gathering. [We were] exchanging … meat and mangetti [Ricinodendron rautenii nuts] with mahangu [millet]. Relations were good … It was a real friendship.
(Mbukushu woman)Footnote 11As elsewhere in southern Africa, the early twentieth century in west Caprivi and its surrounding areas was a time of considerable flux and change.Footnote 12 However, narratives about the past, and in particular those about the nature of the relationship between Khwe and Mbukushu collected between 2004 and 2006, are often static and polarised. In these accounts, the contrast between Khwe and Mbukushu representations of the past is striking. While Khwe accounts often referred to abduction, serfdom, and even murder, Mbukushu narratives referred to amiable relationships of exchange and reciprocity. Archival and secondary sources reveal that there is evidence for both sides of the story in different ways and at different times. Khwe and Mbukushu representations were informed by shifting processes of identity formation and power struggles between different polities. Historically, Khwe–Mbukushu relations were undoubtedly much more complex than their present representations. Historical explorations and explanations are crucial for disentangling the politics of memory in the present, explaining patterns and shifts in narratives about the past, and understanding contemporary concerns and relationships.
While early explorers' descriptions and accounts based on archival documents are rich in detail, rigorous historical analysis of the later nineteenth century in this area is sparse.Footnote 13 Most of this work, such as that of Larson and Fisch, focuses on the Mbukushu.Footnote 14 In these accounts, the Khwe are marginal and the nature of the relationship between the two groups is obscure. Like other San groups, as Wilmsen observes, the Khwe were often historiographically lost from sight.
Mbukushu living in west Caprivi since Independence in 1990 claim to have originated in south-west Zambia. However, during the twentieth century, driven by a series of divisions and migrations, they lived in Botswana, Namibia, and Angola, as did the Khwe. Fisch suggests that, in the second half of the nineteenth century, Khwe were incorporated as subjects under the Mbukushu chief Dimbu I (1865–96), whose rainmaking abilities were the basis of Mbukushu political power and for which Mbukushu authorities received tribute from the Khwe and others.Footnote 15 Fisch claims that the latter part of the nineteenth century was a period of close association and interdependence between the two groups.Footnote 16 While the lack of historical analysis in English makes a cross-examination of Fisch's claim difficult, arguments describing the integrative tendencies of pre-colonial southern Africa, including the uneven incorporation of San and other groups into expanding polities, support its plausibility.Footnote 17
The late nineteenth century saw the Tawana polity strengthening and the Lozi and Batawana chiefs vying for control over Mbukushu.Footnote 18 However, the extent of their interference in Mbukushu political structures and economy is debated. Batawana influence on Khwe–Mbukushu relations is also unclear, but relations appear to have become strained and more unequal over time. Larson argues that the Mbukushu in Ngamiland enjoyed relative independence under Tawana authority because of their remoteness, the weaknesses of Tawana rule, and the political power that Mbukushu chiefs derived from rainmaking.Footnote 19 In contrast, Fisch claims that diminishing tributes for rainmaking, due to the broader effects of Tawana systems of taxation and also to conflicts over rainmaking powers, led to the impoverishment of Mbukushu along the Kavango.Footnote 20 These two scenarios are not mutually exclusive and both may have fed into processes of incorporation and subordination of the Khwe by Mbukushu. This probably bore similarities to the ways in which Ngamiland Khwe were incorporated under the Batawana kgamelo system of political authority. In this system, access to political power was reflected in hierarchical ethnic and social labelling and in the organisation of space.Footnote 21
Complex relationships developed in the context of incorporation and subordination. The notion of ‘slavery’ is used in both archival sources and by Khwe informants in ways that oversimplify the multifaceted nature of Khwe–Mbukushu relationships. While there are references in the archival and anthropological literature to Khwe ‘enslavement’ by Mbukushu as late as 1953, these references need to be interrogated, not least because of the ambiguity over ‘slavery’ and serfdom in colonial discourse.Footnote 22 Also, evidence suggests that it was not only Khwe who were subjugated in this way and that some Khwe may have handed their own children into serfdom.Footnote 23 Moreover, as Tlou points out, careful differentiation is needed between voluntary clientship and hereditary serfdom in nineteenth-century Ngamiland. Tlou argues that European witnesses at the time often failed to distinguish these systems because they evolved over time and did not follow strict ethnic lines. As he explains, in some cases, ‘what began as a voluntary service [carried out by people of a lower socioeconomic status] turned into a hereditary one’.Footnote 24
This multifaceted scenario is backed by strong evidence in the San literature, especially from the revisionists and their followers in the epic ‘Kalahari debate’.Footnote 25 This work shows how San in most parts of Namibia and Botswana were incorporated into the social formations of other groups, becoming and remaining an underclass as the result of broader social, economic, and political processes in the nineteenth and twentieth centuries. These processes included African state expansion as well as the rise and fall of colonial and merchant capital in the Kalahari. By the late nineteenth century, Wilmsen argues, these processes contributed to the relegation of San speakers to peripheral political and economic positions.Footnote 26
Under the Mbukushu chiefs Diyeve (1895–1915) and Disho I (1915–29), Khwe were reportedly often treated violently, as they were by Batawana.Footnote 27 According to 1940 accounts of the missionary Wüst and those of linguist-ethnographer Köhler in 1989, punishments for relatively minor crimes against Mbukushu included drowning or being burned alive.Footnote 28 Crimes of rebellion were also dealt with severely, perhaps reflecting the diminishing power of Mbukushu leaders. Diyeve, for example, punished a Khwe settlement that tried to avoid paying tribute by poisoning their springs, causing numerous deaths.Footnote 29 Settlements of both Khwe and Mbukushu that had not paid tributes were raided for serfs. While violence certainly marked San subjugation to powerful others, processes of domination and the creation of dependency and serfdom were not necessarily characterised only, or at all, by violence, as Widlok argues in the case of the relationship between Hai//om and Owambo.Footnote 30 Such processes may have included a significant degree of acquiescence from San, similar to that suggested by Tlou for the evolution of serfdom in Ngamiland.Footnote 31
Despite the inadequacy of historical sources on the exact nature of Khwe–Mbukushu relationships, the broader literature on the inequality that marks San relationships with Bantu-speaking groups suggests it is likely that Mbukushu subordination of Khwe by the early twentieth century was successful.Footnote 32 It is also likely, based on oral accounts collected in the last decade, that practices such as interpenetrating land rights, intermarriage, concubinage, and exchange were, and still are, integral to this relationship and, concomitantly, that identities were more fluid and less distinctive than current representations would indicate.Footnote 33 The period 1900–30, in particular, brought considerable displacement, migration, and resettlement of both Khwe and Mbukushu in what are now Angola, Namibia, Botswana, and Zambia. While the paramount chief of the Mbukushu remained in Namibia on the Kavango river at Andara, the complexities of migration make any contemporary claims to singular identities and ‘traditional’ territory more complicated than they first appear. Significant migrations resulted from fear of slave-trading and compulsory taxation and forced labour, to which Angolan Mbukushu were subjected under the Portuguese. Political divisions among Mbukushu, including bitter conflicts over rainmaking and colonial veterinary interventions, also led to migration.Footnote 34 The implications of migrations for claims to land and land use between 1900 and 1930 were compounded by the effects of veterinary policies in the 1930s and 1940s and by war in Angola and Namibia after 1960.
STATE CONTRIBUTIONS TO THE CONSTRUCTION OF ETHNICITY, 1930–89
In this section, I explore the implications of the SWA administration's veterinary interventions in the 1930s and 1940s and its post-1947 ‘Bushmen’ policies for Khwe political authority. I also examine the impact on identity and ethnic relations of the South African Defence Force (SADF)'s occupation of west Caprivi in the 1970s and 1980s.
Like other colonised countries, Namibia was subjected to extensive social, economic, and political engineering, the rearrangement of space, and the establishment of institutions that best served the interests of white settlers.Footnote 35 San groups were repeatedly a target of extreme violence under German rule, the subjects of what Gordon has recently termed a series of ‘forgotten genocides’.Footnote 36 Inhabiting areas beyond the settler belt, however, Khwe appear to have escaped this brutality. In addition, their interaction with the South African colonial administration appears to have been different from that of other San groups whose territory increasingly fell under the incursions of the colonial and apartheid administrations.
This interaction contributed to the cementing of both Khwe and Mbukushu ethnic identities over time and resonates with examples from elsewhere in colonial southern Africa.Footnote 37 Admittedly, the available archival sources tell us much about how colonising parties saw Khwe and Mbukushu, particularly before the 1980s, but much less about how these groups saw themselves. Nevertheless, enough evidence is available to argue that Khwe engagement and alliance-making with the colonial and apartheid states had profound implications for local political authority and access to land and natural resources.
In 1929, the administration of west Caprivi was handed to the SWA administration by the Bechuanaland Protectorate. Alongside efforts to control newly defined boundaries and to inhibit local trade in northern Namibia, cattle movement and animal epidemics became a central administrative concern from the early 1930s.Footnote 38 Most of these veterinary interventions were directed at protecting white-owned livestock and the white settler economy in the police zone to the south. They had complex and far-reaching consequences in west Caprivi, especially when institutions created to deal with livestock control began to interact with policies towards the San. Veterinary control measures in the 1930s and 1940s are one factor in how west Caprivi came to be seen over time as the ‘traditional’ area of Khwe, to the exclusion of the Mbukushu, even though Mbukushu continued to visit the area. This perception complicates contemporary claims to, and narratives about, land occupation and use, even if Khwe had exclusively occupied west Caprivi prior to Mbukushu immigration in the nineteenth century, which is likely but for which evidence is sparse.Footnote 39 Further, despite differential treatment by the administration, which cultivated Khwe–Mbukushu ethnic boundaries, archival records and oral sources indicate that relationships of exchange and/or interdependence between Khwe and Mbukushu continued.Footnote 40
As early as 1929, correspondence refers to the prohibition of ‘nomadic movements of Mampukushu [sic] from Angola’ alongside the movement of cattle through west Caprivi.Footnote 41 Following an increase in cattle epidemics in the Kavango area in the 1930s, the administration ordered that west Caprivi be cleared of ‘natives’ and stock. This intervention supports Bollig's account of the creation of buffer zones to the west in Kaokoland in the same period.Footnote 42 The population of west Caprivi and its distribution at this time is not clear from the records but residents were relocated either to the western bank of the Kavango or to the eastern bank of the Kwando in East Caprivi.Footnote 43 Fisch, for example, reports that eight small Mbukushu villages were resettled from the western to the eastern bank of the Mashi/Kwando River in 1939–40.Footnote 44
Bollig notes that attempts to control disease and protect white-owned livestock were also a move to control the mobility of pastoralists.Footnote 45 Pastoralists posed a particular threat for the establishment of administrative boundaries, so central to the ideology of the colonial state. It would seem that, despite their mobility, Khwe escaped displacement partly because they did not own cattle. It is also possible that they were less visible to the colonial administration, thanks to their more mobile livelihoods. Most importantly, they were not subject to the same restrictions as Mbukushu. These factors perhaps explain the administration's laissez-faire approach towards San groups in northern Namibia until after the Second World War, along with a lack of state capacity. But official attitudes suggest, even then, that ‘Bushmen’ occupied a different space in the colonial imagination from ‘Bantu’. Footnote 46 This differentiation also led to the dry land between the Kavango and Kwando rivers becoming increasingly recognised as a ‘Bushmen’ area.
In the 1940s, colonial officials considered west Caprivi to be inhospitable, waterless, and difficult to cross. Given that men in the colonial service were stationed at least 250 kilometres away, their visits to the area were few, and their knowledge limited and inconsistent despite official concern about livestock movements through the Caprivi strip.Footnote 47 In 1941, the SWA administration employed two ‘native guards’ to patrol the strip and its borders, keeping it clear of ‘natives’ and livestock.Footnote 48 Muyatwa and Mukoya were stationed at Bwabwata and Mahango, respectively. Officials described Muyatwa's Bwabwata outpost on the Angolan border as ‘one of the most lonely imaginable’.Footnote 49 Considering that the area of the strip is some 6,000 square kilometres, and the borderline some 200 kilometres long, the efficacy of the two guards' surveillance is questionable. Nevertheless, Muyatwa proved useful to the administration in other ways. Most importantly for understanding contemporary local politics, Muyatwa, who was Lozi, became the first intermediator between Khwe and the SWA Administration. He was also employed to gather military intelligence during the Second World War.Footnote 50
In 1947, the Administration commenced a scheme to bring the ‘Bushpeople’ under greater ‘protection’. As Gordon has demonstrated, the reversal of the government's earlier laissez-faire policy towards Bushmen beyond the northern police zone was tied to an attempt to counter stock-raiding on white farms, veld fires, and the killing of game.Footnote 51 Within the police zone, the administration's attempts to deal with the ‘Bushmen problem’ through violence and coercion had failed, and they proposed new measures through which Native Commissioners should ‘befriend’ these groups and supply them with tobacco, salt, and maize meal.Footnote 52 These measures were also an attempt to appease international opinion about apartheid policies and were cast as a positive contribution to scientific research. From 1947, ‘Bushman guards’ were installed at San settlements in the Kavango region to maintain law and order, protect game, protect San from other blacks, and encourage them to become sedentary, adopt agriculture, and practise animal husbandry. The concentration of such guards reached its peak in 1962, when they numbered 16.Footnote 53 In 1949, the administration established a Commission for the Preservation of the Bushmen. Chaired by Professor P. J. Schoeman, a Stellenbosch University architect of apartheid, this commission had far-reaching effects on all Namibian San groups in terms of identity politics and land appropriation.
The ways in which Khwe were affected by the emerging ‘Bushmen’ policies of the late 1940s appear to have been contradictory and uneven. The Schoeman commission, focussing on the Khwe situated close to the Kavango River, designated them ‘hybrids’ or ‘black Bushmen’ who were not worth ‘preserving’. It recommended that, like other blacks – but unlike other San, for example !Kung (Ju/'hoansi), who had been observed to ‘often fret and pine to death’ in situations of ‘captivity’ – the Khwe should become a new pool of migrant labour.Footnote 54 As a result, many Khwe men were recruited and tied to sixteen- or eighteen-month contracts on the Witwatersrand mines in South Africa. The change was dramatic for, as Orth claims, prior to the 1950s only a few Khwe had engaged in migrant farm labour, owing to poor working conditions.Footnote 55 Contradictions abounded. While the ‘hybrid’ Khwe were incorporated into the regional labour economy, the Rundu Native Commissioner (NC) appears to have applied ‘pure Bushmen’ policies in west Caprivi. Official policy, however, was gradually to ‘tame’ ‘wild’ Bushmen by bringing them under administrative control and incorporating them into the settler economy.Footnote 56
The objective of ‘Bushmen protection’ and ‘taming’ policies seem to have been directed at steering San towards labour in the service of the colonial administration and economy. The vast distance between Bwabwata and white farms to the west meant that stock theft by Khwe was an unlikely explanation for the enforcement of protection policies. It is therefore more probable that the purpose of sedentarisation was to draft Khwe into the regional labour market, specifically mine work. In 1948, the Rundu NC wrote that a key objective of the guard's duties was ‘to encourage [bush people] to do light labour in return for tobacco and salt’.Footnote 57 It is not clear what this ‘work’ included, or whether it was, at Bwabwata, for Muyatwa's private purposes or for the administration. The idea of labour was in line with the commission's recommendation that ‘Right from the start the Bushmen must be taught they must work for their livelihoods and the dangers of demoralisation inherent in any free feeding scheme must be carefully avoided’.Footnote 58 These policies had interesting implications for Khwe–Mbukushu relationships, given that economic exchanges revolving around food, goods, and labour were already well established. The colonial administration intervened in these relationships such that they assumed new (and sometimes monetarised) forms. This may have reinforced the historical serf-like status of Khwe, but it also took power away from Mbukushu. As the NC reported in 1951,
The [Mbukushu] Chiefs say that the bushmen do useful work for them. They till and reap, weed and herd stock. In return the bushmen [must] receive cash or produce. The relationships between the Bushmen and Bantu [are] cordial … In two cases the bushmen were badly treated … The wrongdoers were punished. It is pointed out that the Bushmen are always to be regarded as falling under the protection of the Administration.Footnote 59
New ‘Bushmen’ policies thus appear to have brought change to Khwe–Mbukushu relations. However, only when we consider the implications of these policies in the realm of political authority can an argument be made for a new ‘alliance’ between the Khwe and the apartheid state. Key to the NC's approach was his view that the San he interacted with – both Ju/'hoansi and Khwe – were leaderless:
I could not, much to my own amazement, find any bushmen leaders worthy of the name ‘Kaptein’ … the bushman is not yet nearly ready to take a great interest in his own government and is quite content to leave it to the ‘Mahona’ [NC].Footnote 60
The bushmen, themselves, prefer to have Native Guard as no bushman is strong enough to govern others.Footnote 61
The significance of the native guards as leaders is highlighted by the account of a native commissioner who, when he did not find a suitable ‘tribal head’ among the Khwe, instructed Muyatwa to be their representative.
In order to fill the gap caused by the Barakwengu [sic] having no ‘Kaptein’ or other leaders I have instructed Muyatwa to take a fatherly interest in the Bushmen of his area and have issued similar instructions to [Border Guard] Mukoya … at Mahango … Until Head Office decrees otherwise the Bushmen will regard these two persons as people to whom they can appeal when in trouble [with Mbukushu].Footnote 62
Muyatwa's role appears to have been extended from veterinary surveillance to political authority after the Bushmen Commission of 1950. Although the exact nature of his duties as an ‘interim’ Khwe representative are not clear, Muyatwa probably came to play an influential role in the relationship between Khwe and Mbukushu, and in their respective relationships with the administration.Footnote 63 Muyatwa's presence also affected Mbukushu access to west Caprivi land and resources, as described in a 1948 report:
The trouble between Muyatwa and the Mambukushu of the Eastern Caprivi Zipfel appears to have its origin in the fact that Muyatwa placed a beacon on the eastern side of the Malombe pan thereby indicating that the Natives from the Kwando could not drink there when they were out hunting.Footnote 64
As oral histories collected since the mid 1990s indicate, Khwe accounts of their relationship with Mbukushu in respect of land and authority commonly emphasise their historical subjugation by Mbukushu.Footnote 65 Against this backdrop, Muyatwa is cast as a liberating leader-figure who protected Khwe from Mbukushu harassment, enslavement, and land appropriation:
SB: Before Muyatwa came, the Mbukushu came in a respectful way [from Angola] asking for a place to live. We also showed them where to find water. Then Muyatwa arrived. After the Khwe gave them the place, they started to trouble the Khwe. Then Muyatwa was the one chasing them out, [to] send them back to where they came from.
JT: Why did Muyatwa chase only Mbukushu?
SB: Because the place was for the Khwe. After chasing the Mbukushu, he was friends with the Khwe, just helped them. He employed the young ones as cattle herders and married two Khwe women. We were just getting free at the time of Muyatwa.Footnote 66
Whilst Muyatwa is today represented as a protector and liberator by Khwe, European observers held a different view. In their interpretation, the native guards had a tendency to abuse their positions of power.Footnote 67 Köhler reported in 1975 that the Khwe at Bwabwata were ‘subjects’ of Muyatwa and carried out menial labour for him.Footnote 68 Similarly, the NC reported that Muyatwa
is by nature a Dictator and inclined to conscript bushman labour. I explained to the bushmen that they were free to work or not for Muyatwa in exchange for the protection he afforded them from Angolan raiders who apparently still enslave bushmen children … Muyatwa has been instructed not to enforce labour but rather to encourage the bushpeople to work for grain and tobacco.Footnote 69
Concurrent with the view that Muyatwa was dictatorial, is the NC's enquiry into an allegation that ‘the Mambukushu had incited the bushpeople to kill Muyatwa’.Footnote 70 While the NC was unable to discern further information about this allegation, if it held any truth, it suggests that relations between Muyatwa and Khwe were not nearly as harmonious as portrayed from the late 1990s. Rather, it suggests that Khwe and Mbukushu had ongoing interactions and alliances outside those apparent to the administration.
That contemporary Khwe and historical European interpretations of Muyatwa are so inconsistent provides insight into the broader question of the nature of relationships between San and Bantu-speaking groups. Scholars have documented the power dynamics over land, labour, and natural resources that have fostered contemporary San economic and political marginalisation in Namibia and Botswana. Relations between Ju/'hoansi and Herero, Hai//om and Owambo, and Khwe and Batawana have received considerable attention.Footnote 71 In these accounts, discourses of San subjugation and the erosion of independent hunting-gathering livelihoods and access to land are located within wider discourses of the benign paternalism of more powerful, livestock-owning groups. For example, Herero in Omaheke often relied on the incorporation and cooperation of low-status San labour as an economic strategy to counteract the draining of Herero male labour into the white economy.Footnote 72 San resistance to these processes, including simply moving away in the case of Basarwa in Ngamiland, has been mixed with a significant degree of acquiescence to socioeconomic integration, dependency, and land appropriation.Footnote 73 Acquiescence brought access to economic and social networks that sustained San in the face of the transformation of their livelihoods, or provided the opportunity to retain at least a degree of autonomy.Footnote 74 Through this lens, Muyatwa's different guises as protector, liberator, and dictator are reconcilable.
Details of the emergence of government-recognised political authority among the Khwe are inconsistent. Contrary to the NC's assumption that the Khwe were leaderless prior to administrative intervention, Boden argues that, according to oral historical evidence collected in the late 1990s, the Khwe did have established institutions of leadership, although these may not have been as hierarchical as those of other groups and were organised on a more local level:
Most prominent were so-called dixa-//'ae-//ua … best translated as ‘responsible owners of settlements’. They granted and controlled access to land, water and other resources, settled conflicts within the community and represented the local community in conflicts with outsiders. Personal qualities and a good reputation were as important for legitimacy as a certain genealogical relationship to a predecessor.Footnote 75
This view complicates both administrative and Mbukushu narratives about (the lack of) Khwe political authority. While Mbukushu leaders in Namibia had long had direct contact with the administration through a recognised chieftainship, Muyatwa's era appears to have been a turning point for the degree of Khwe exposure to, and contact with, the state apparatus. In the late 1950s, a Khwe man called Martin Ndumba became a recognised voorman (headman) in Mutc'iku. There are discrepancies among oral and archival accounts as to whether it was the administration or the Mbukushu chief who installed him, and when this installation was effected. To the east, in Bwabwata, Muyatwa made Khwe residents elect a leader of their own before his death in the early 1960s. They chose Kandunda Kaseta, the eldest man of the family group generally said to be the owners of Bwabwata. Boden refers to two additional voormanne in smaller settlements who were also recognised by the administration.Footnote 76 Their duties are not specified in the existing literature or records, but these representatives most likely embodied significant change in Khwe sociopolitical structures, which, according to the NC's observations, had been more egalitarian in the past.Footnote 77 There is no evidence in archival material for these Khwe representatives being referred to as ‘chiefs’.
These developments reflect some of the contradictory processes at play during the 1940s and 1950s. While colonial ‘civilising’ policies may have reinforced certain historical inequalities between Khwe and Mbukushu, they undid those inequalities in other respects. Despite the ambiguity surrounding the nature of his relationship with the Khwe, Muyatwa symbolised the forging of a new alliance between Khwe and the white administration that gave them a substantial, and possibly unprecedented, degree of power in their relationship with the Mbukushu, at least in the eastern part of west Caprivi. As one of Boden's informants stated,
The bad treatment stopped because the Whites accepted Khwe leaders. The guavas [Mbukushu] were afraid of the Whites. The Khwe leaders told the guavas they should stay away. The Whites brought tobacco, clothes and food. We only depended on the Whites. They told us that we should report bad treatment by the guavas. We told the guavas, ‘Now we have got the Whites’.Footnote 78
Foreigners and whites thus became significant actors and symbols in Khwe–Mbukushu relationships, especially in their role as ‘protectors’ of Khwe and as promoters of differential treatment of the two groups as early as the 1940s.
The administration repeatedly intervened in and undermined Mbukushu political authority. Both Chief Ndara (1929–39) and Chief Disho II (1939–47) were deposed by the SWA administration on account of various factors and practices that made them ‘unfit’ for leadership.Footnote 79 Chief Disho II was subsequently exiled to Botswana. This outcome, and the fact that he took the chiefly rainmaking powers with him, significantly affected the political and economic status of the incumbent Chief Makushe (1947–69) and Mbukushu perceptions of the administration.Footnote 80 Van Tonder's doctoral thesis on the Kavango region testifies to this. He reports that Mbukushu were
unfriendly and suspicious, a disposition which at times bordered on extremity. The reason for this was mainly due to some unfortunate clashes with the SWA Administration, before the Department of Bantu Affairs … took over … Even now, on account of the remote control of the central authorities and the lack of knowledge reflected in the local approach, the relationship between the Hambukushu and their white neighbours is under a pressing … strain.Footnote 81
An example of the administration's intervention in Mbukushu political authority is found in a dispute between Chief Makushe and 16 ‘Barakwengu’ in 1953, following the chief's attempt to exact tribute from them. In line with policies regarding San exemption from tax-paying, the NC stated at the hearings that ‘no Bushman has ever been told to pay tax or tribute to anyone’. He also reported that the Mbukushu chief ‘has never been asked to exercise any jurisdiction over the Barakwengu who are a bushmen clan having a language of their own, customs of their own and their own mode of livelihood’, indicating that the administration recognised the Khwe as a distinctive, autonomous group. Footnote 82 The NC and his colleagues intervened and thus rejected the Mbukushu chief's power over the Khwe, which was already diminished due to the loss of rainmaking gifts. The Khwe were once again ‘protected’ against Mbukushu demands for land, labour, and tribute.
Thus, archival evidence shows that, between the early 1940s and the early 1960s, SWA's veterinary and Bushmen policies, despite their uneven nature, significantly shaped both Khwe and Mbukushu access to land and the extent of their respective political authority. As such, these policies were key contributors to the emergence of more distinct Khwe and Mbukushu identities, and to the ways in which these groups perceived the apartheid government.
THE KHWE MILITARY IDENTITY
From the late 1960s, the most crucial factor in the ongoing formation of contemporary identities, political authority, and claims to resources in west Caprivi was Namibia's independence struggle, which began in 1964. West Caprivi's two-decade-long occupation by the South African military forces, when the area was used as a springboard for operations into Angola against South West Africa People's Organisation (SWAPO) liberation forces, deeply influenced Khwe relationships with both Mbukushu and the post-apartheid state. West Caprivi was declared a Nature Park in 1963, despite recommendations from a key apartheid land-planning body, the Odendaal Commission of Enquiry into South West Africa Affairs, that it be converted into a ‘homeland’ for Bushmen. This decision took place in the context of bureaucratic rivalries between the Departments of Bantu Administration and Nature Conservation over the ‘scraps’ of land unwanted by white farmers.Footnote 83 The later decision to upgrade west Caprivi's status to that of a game park in 1968 was linked to its strategic geopolitical importance, given the emergence of independence struggles in neighbouring Angola and Zambia, and subsequently in Namibia. In about 1970, west Caprivi was declared a military zone and conservation officials were denied access to the area, with the SADF appointing their own ‘nature conservation’ authorities. Both the park's declaration and its subsequent occupation substantially altered the livelihood strategies that the Khwe employed at the time and contributed to a process of land and resource dispossession.Footnote 84
In 1974, the SADF began recruiting Khwe men and !Xun from southern Angola, many of whom were refugees. Given the sensitivities of conducting research on SADF collaborators in post-independent Caprivi, there is little information available on how and why these groups made the political choices that they did or the extent to which they were coerced into doing so.Footnote 85 In response to a question about this period, an elderly !Xun woman in Omega said, ‘In Angola we were suffering; there was [armed conflict]. P.W. Botha brought us [here] from Angola … the Boers told us we'd die [there] if we didn't leave.’Footnote 86
In the wake of their recruitment, anthropologists, UN officials, and SWAPO representatives alike represented !Xun as ill-equipped to make informed decisions:
[The anthropologist] Lee is appalled by this recruitment of Bushmen. He says that the politically naïve Bushmen do not understand the realities of the situation … [and] the SWAPO representative to the United Nations says, ‘It is really tragic that South Africa has to go so low as to use these people, who are not conscious of what they are doing’.Footnote 87
As the anthropologist Edwin Wilmsen reported, however, !Xun were clearly aware of more than the material benefits of military conscription: ‘They get a steady diet, prestige, and are treated, in their view, more like other Africans and less like lower class citizens’.Footnote 88
Within four years, the military base at Omega housed 3,000 people and the SADF began to promote a romanticised and essentialist discourse about their ‘Bushmen soldiers’.Footnote 89 SADF Colonel Breytenbach, for example, described Bushmen recruits as ‘stone age hunter-gatherers’ who lived an ‘innocent and idyllic life’ and who ‘could have shown us the way back to living in harmony with nature … the secrets of the bush … [and] how to feel at one with the spirit of ancient Africa’.Footnote 90 These same qualities were claimed to enhance their capacity for tracking and soldiering since ‘the Bushmen were intrepid ambushers and skilled stalkers by nature. They love to stalk an unwary enemy, to shoot the hell out of him, and then disappear like a shadow into the bush.’Footnote 91 While some SADF representatives acknowledged that ‘Bushman’ society was threatened by their military involvement, others denied this. One lieutenant claimed, ‘Our aim is not to try and Westernize them … but to make them better Bushmen.’Footnote 92
The identity politics for which the SADF served as catalyst were, of course, much more complex than Breytenbach's excerpts suggest; so, too, was the SADF's view of San much more ambiguous, contrary to Gordon's argument.Footnote 93 Khwe and !Xun, with their respective identities, relationships, and allegiances, had already been influenced by experiences of the Angolan colonial war, which ended in 1974, before they went on to provide a pool of recruits for the SADF. In 1975, in the course of the SADF's Operation Savannah into Angola, the Angolan Khwe who had expert knowledge of the area came to be seen as ‘crack soldiers’. It was after this, helped by the media and visiting politicians in the late 1970s, that the ‘Bushman soldier’ myth became fully developed. As Sharp and Douglas explain, the myth also created a problem for the SADF, because it was the smaller, lighter-skinned !Xun who fitted the public stereotype of ‘bushmen’, and not the taller, ‘blacker’, Bantu-like Khwe. From the SADF's perspective, however, the !Xun did not show the military prowess of the Khwe. Hence, to maintain the image of the SADF's ‘crack ethnic unit’, !Xun and Khwe were kept together.Footnote 94
In line with apartheid ideology, processes of racial and ethnic ‘othering’ were central to the SADF's strategy in west Caprivi.Footnote 95 This ‘othering’ reinforced Khwe fears of and antagonism towards Mbukushu and other Bantu groups. Khwe narratives during my fieldwork reflected this tension but they may also have projected present concerns onto the past, as seen in the following reference to HIV/AIDS, a disease that was only identified in the mid 1980s:Footnote 96
All [the Khwe in] West Caprivi were in SWATIF … [The army said] Khwe should stay [segregated] both for training and accommodation. [They said] if you mix with blacks you can get HIV/AIDS. [They said] if you mix, you'll get disease. If you mix, then you'll dislike white people also.Footnote 97
As Rousset documents, and as my own interviews revealed, Khwe were continually categorised by the SADF as ‘different’ from other Africans and encouraged to exercise this power.Footnote 98 Boden reports that
Mbukushu and other blacks daring to come into the military camps were declared fair game for the Bushmen soldiers, [who were] encouraged to do as they pleased with blacks … [P]eople who were children at the time remembered [being] encouraged to throw stones at black beggars in the camps.Footnote 99
Meanwhile, some SADF representatives cast ethnic and racial tensions between Khwe and other blacks as primordial: ‘A Bushman's hate for SWAPO will give you the shivers … they hate SWAPO because [SWAPO] enslaved them and took their daughters for prostitutes’.Footnote 100
The SADF restructured pre-existing ethnic differences and divisions along political lines while simultaneously underpinning the idea that west Caprivi could serve first and foremost a ‘Bushman’ domain. This, of course, was a concept that the earlier Khwe alliance with Muyatwa had already catalysed. As one Mbukushu headman told me,
The Boers said that Mbukushu are SWAPO; that's why they chased them out. But they were not all SWAPO, [the Boers] were just chasing them. When there were no whites, we were living nicely with Khwe. When the Boers came, that's when this separation [happened].Footnote 101
Sharp and Douglas caution against overstating the SADF's agency and success, and challenge Gordon's interpretation that San were simply victims of the ‘Bushman myth’. They argue that the San are not simply independent or dependent, nor merely prisoners or opportunists, but that they have adopted each of these ‘identities’ at different times, depending on the context.Footnote 102 It is likely that Khwe manoeuvred within and contributed to SADF discourses about ethnic difference but they did not entirely buy into such discourses either. For example, Sharp and Douglas report that, in 1994, some relocated Khwe in Schmidtsdrift, South Africa, believed that ‘they had never really been regarded as “proper Bushmen” in relation to the “authentic” !Xun’.Footnote 103 The SADF, however, had provided Khwe with a variety of opportunities over time, including an exit strategy from Angola's long civil war for those Khwe of Angolan origin. With this in mind, Khwe may have contributed to SADF ethnic discourses in so far as those discourses bolstered economic privilege and employment in the SADF and possibly to the extent that they reinforced their claims to territory. Detailed research into how Khwe engaged with or resisted apartheid and nationalist politics and discourses has yet to be conducted.
In the longer term, the SADF's occupation had deeply negative impacts on Khwe livelihoods.Footnote 104 Although little research has been carried out on the monetisation and sedentarisation of the Khwe economy through men's incorporation into the SADF, its effects – including a legacy of alcoholism – appear to have radically influenced their long-term situation. The SADF's provision of a host of free services such as running water, pharmacies, schools, clinics, a bank, and self-service stores also influenced the Khwe economy.Footnote 105 SADF food provision transformed livelihoods previously sustained through hunting, gathering, limited agriculture, and exchange. As one informant explained,
There was no need to go to the bush [for gathering] because we were employed. Everything was free, everything was plenty. I [even] bought a car during the time of the army. When the army left us we got a lot of pain and suffering.Footnote 106
The SADF was not only seen in a positive light, of course. Talking about the Namibian Defence Force (NDF) employment of Khwe after Independence, one of Boden's informants reported,
They put the Khwe only where death is. The SADF has put us in it. If you have a dog, you can take your best dog who runs fast and is a killer. The Boers did like that with us Khwe. Maybe because of that the Namibians say that we know how to kill. Now they turn to us and want [us to] join the NDF. They do the same like the SADF. They give us the dog's work.Footnote 107
Such statements are part of a consistent pattern in which Khwe have constructed themselves as victims of history, at least since 1990.Footnote 108 This pattern contrasts with the historical record, which suggests that Khwe often benefited more than other groups from their interactions with the state. This situation changed at Independence. The SADF services, alongside army salaries, came to an abrupt halt when the SADF withdrew in early 1990, taking with them four thousand Khwe and !Xun who chose to emigrate to South Africa, and leaving adrift some 5,000 people, the vast majority being Khwe who had become dependent on the SADF for their day-to-day survival.Footnote 109 These people were, uncharacteristically for southern Africa, allowed to remain inside the Caprivi Game Park, whose status was now restored. Much like the policies of the colonial administration, the SADF's occupation contributed to the forging of west Caprivi as an ‘ethnicised’ territory. It also contributed to the legacy of a stigmatised ethnic identity and the association of the Khwe with political subversion after Independence.
CONCLUSION
I began this article with the problems presented by polarised Khwe and Mbukushu narratives and representations of history collected between 2004 and 2006. These often portrayed a static past, particularly with regard to Khwe–Mbukushu relations and territory. The twentieth century in west Caprivi, however, was a time of continuous flux, change, and migration, of alliances and divisions, and of colonial and apartheid interventions. Khwe–Mbukushu relationships appear to have been interdependent, if unequal, during the early part of this period.
Successive administrations contributed to the shaping of an ethnic identity and an ethnic ‘territory’ for the Khwe, first through veterinary policies in the early 1940s and later through the policies recommended by the 1950 Commission for the Preservation of Bushmen. Linked by the influential figure of Muyatwa, these two sets of policies, which consistently treated Khwe and Mbukushu differently, set in motion significant changes in the realm of political authority for both groups. If Khwe had previously fallen under the control of paternalistic Mbukushu masters, they now fell under the ‘protection’ of the apartheid administration. While the contributions of the colonial and apartheid administrations to the construction of ethnicity in west Caprivi should not be overstated, as Khwe and Mbukushu identities may have remained fluid, this article has argued that state interventions hardened these identities in significant ways and buttressed Khwe authority.
This article has built on recent revisionist studies that attempt to achieve a more historically conscious, nuanced understanding of how San identities are fashioned. It has demonstrated that, for the most part, Khwe were not ‘isolates’ but engaged with external interventions, with other groups, and with identity-building in very similar ways to non-San groups all over the region. At the same time, however, they often became the subject of state interest and interventions precisely because of their categorisation as ‘Bushmen’ by officials and other outsiders. Consequently, they had greater opportunities for engaging with external parties, as well as with discourses and historical processes that extended far beyond west Caprivi's borders.
