Speaker 1: The fact that we are Africans doesn't automatically make us the subjects of chiefs. We can also be independent families and clans. We are landowners too, and to be a landowner you do not need a chief.
Speaker 2: When you call a meeting, you can only call the beneficiaries. You cannot call all the people in the area; the people who only arrived in the 1990s and so on. It can only be the descendants of the families who originally purchased the land. And their names are in the book. (Contributions from the floor at the ‘North West Mining-affected Communities’ Workshop’, Tlhabane, Rustenburg, 18 May 2015)
Introduction
Following the great strike waves of 2012 and 2014, and their brutal yet failed suppression at Marikana, academic work on the platinum industry has largely focused on the deepening crises of social reproduction faced by migrant mineworkers, their titanic battles for a ‘living wage’, and the wider ramifications of their historic break with the African National Congress (ANC)-aligned National Union of Mineworkers. Yet, for all its national significance, the class struggles of mine labour form only one dimension of a much greater and more complex field of social conflict and political possibility on the South African platinum belt. For one of the most distinctive, yet least remarked, features of the platinum industry is that its mineral reserves are overwhelmingly located within the former homeland areas of the North West and Limpopo provinces and, consequently, the expanded reproduction of platinum capital is conditional upon the expanding alienation of communal land under the local administrative control of traditional authorities (Capps 2012a). Platinum, then, also presents a particularly fascinating example of the wider relationship between accumulation and dispossession, which has manifested so sharply of late across the continent in the context of large-scale land acquisitions by foreign investors and domestic states (Hall 2011; Peters 2013).
This article explores how intensifying struggles over communal land in the platinum zones around Rustenburg are being mediated through conflicts over group boundaries and identities, and how this in turn is articulating a potentially new yet contradictory rural class politics. In a context where chiefly authorities are themselves becoming major shareholders in local mining operations, the burning issues are whether the ‘tribe’ should be treated as the only legitimate African land-holding unit or whether the collective ownership of mineralised land should reside in smaller socio-political groups, and who should represent their proprietorial interests vis-à-vis platinum capital. Drawing on a detailed study of three village-level disputes in the Bakgatla-ba-Kgafela traditional authority area (Mnwana and Capps 2015), we argue that first, as elsewhere on the continent, contested constructions of ‘community’ are emerging as a significant means of defending or advancing claims on rural resources, but in ways that are also rooted in the specific histories of the Rustenburg region, and in particular its distinctive legacies of group land-buying and tribal registration. Yet, second, that these corporate forms of organisation are at the same time riven by gender, generational and other social divisions, and are prone to replicating the tribalist logics they seek to challenge. We thus conclude that while the proliferation of small-group struggles against the emergent mine–chief nexus is the source of a potentially progressive rural politics, the attempt to establish private property rights through more exclusionary group definitions may itself express a new dynamic of class formation ‘from below’ – and one that could equally act as a divisive force against those labelled ‘outsiders’, not least migrant mineworkers.
The Rustenburg pattern
In a magisterial survey, the anthropologist Pauline Peters has shown that rising competition over land across sub-Saharan Africa is in many cases leading to its de facto privatisation through ‘new and/or intensified efforts to demarcate [group] boundaries and to delimit the definition of those to be included as “owners”’ (Peters 2004, 304). These processes of ‘African enclosure’, as Woodhouse (2003) terms them elsewhere, are taking ‘the form of stricter definitions of those who have legitimate claims to resources’, while excluding others from ‘rightful claims’ through a ‘narrowing in the definition of belonging’ (Peters 2004, 302). Moreover, argues Peters (2004, 291), there is ‘not only intensifying competition over land but deepening social differentiation, and though this differentiation takes many forms – including youth against elders, man against women, ethnic and religious confrontations – these also reveal new social divisions that, in sum, can be seen as class formation’.
We argue that the continental trends identified by Peters are assuming a particularly exaggerated form in the traditional authority areas around Rustenburg. This is due to the legacies of a regionally distinct yet broadly common African property regime, which historically combined the group-purchase of land with its compulsory registration to large state-defined tribes. This section will first briefly trace the evolution of this ‘Rustenburg pattern’ of land and political relations under conditions of colonial and apartheid rule. It will then consider how its intrinsic tensions have combined with recent legislative developments to forge new alliances between local chieftaincies and platinum corporations, while shaping the terms of popular struggle against them. With this historico-regional context established, the next section will initiate a detailed exploration of three such ‘claims from below’ in the Bakgatla-ba-Kgafela traditional authority area, with particular reference to the ways in which these also generating contradictory ‘dynamics of division and exclusion, alliance and inclusion’ that may themselves ‘constitute class formation’ (Peters 2004, 305).
The tribal-trust regime
The major traditional authority areas of the Rustenburg region have two defining features. The first is that they straddle the greater part of the Western limb of the Bushveld Complex, and hence encompass a substantial swathe of the world's largest platinum reserve. The second is that they comprise large numbers of farms that were historically purchased by African groups and registered to the state ‘in trust’ for a recognised local ‘chief and his tribe’. The antecedents of this ‘tribal-title-trust’ property regime (hereafter, tribal-trust regime) can be traced to the colonial heartlands of mid nineteenth century Transvaal.1
Following Afrikaner conquest in 1837, the entire land area of its Rustenburg region was rapidly surveyed and parcelled out among the white settlers, and the Tswana communities occupying these farms converted into a semi-servile labour tenantry. However, although blacks were legally barred from the emergent title market, by the mid 1860s Africans had begun to form syndicates to ‘buy back’ land through the informal intermediation of white nominees – most often missionaries. These land-buying groups were typically bound together by political identities that not only facilitated the collective mobilisation of resources, but also demarcated the social boundaries of who would benefit. But there was also considerable variation in their scales and motivations. At one end of the spectrum, chiefs mobilised their followerships in a collective endeavour to regain control of erstwhile territories, while at the other – and perhaps more frequently – clan-based groups of diverse origin acquired their own farms, not only free of the oppressive demands of white landlords, but also the coercive jurisdictions of chiefs. There was thus from the beginning a vital interconnection between the assertion and/or formation of discrete group identities and the collective acquisition of land across the Rustenburg region.
Following a brief imperial occupation of the Transvaal by the British, the bar on black land ownership was partially raised in the early 1880s. However, the independent entry of Africans into the title market would be increasingly subject to two distinct yet complementary forms of trusteeship. First, an appropriate state authority would have to sanction the transaction and then assume trusteeship of that property on its purchasers’ behalf. Second, it would only be possible for a land-buying group to acquire title via the recognised chief of a designated tribe, acting in his presumed role of ‘traditional custodian’ of its communal property. This was further buttressed through the definition of the ‘tribe’ as a ‘legal personality’ with the capacity to enter into contracts with third parties through its chiefly authority. These parallel institutions of ‘state’ and ‘tribal’ trusteeship would progressively, though unevenly, evolve over the remainder of the colonial period, and then be fully incorporated into the segregationist and apartheid states. Yet, the effect was to produce a quasi-private tribal-property form around Rustenburg that embodied significant tensions over the question of ownership at the respective levels of ‘tribal’ and ‘state’ trusteeship.
The first tension was within the tribe between the landed clans and the chiefly trustee over the control of group-purchased farms. Under the new property regime, the title previously acquired by different corporate groups through the aegis of missionaries would now be simultaneously transferred to the formal trusteeship of the state, and re-registered in the name of the recognised tribe within or near whose jurisdiction it fell. Moreover, because the colonial authorities worked within an ideological framework that assumed all Africans to be members of tribes (Delius 2008; Mamdani 1996), they would only sanction new group land-purchases if they were tribally based. This in turn encouraged such groups to seek out the nearest chief, or simply reconstitute themselves as a ‘tribe’, in order to enter the land market (Claassens and Gilfillan 2008, 298–299; Keegan 1988, 109–138, 146–148). Tribal landownership would thus be privileged over all other forms of corporate possession, while the legal identities of the original purchasers would be denied by the tribal registration. Yet, the facts remained that the political identities of these groups were inextricably tied to their collective acquisition of this land, and that the fiduciary powers of their chiefly trustees were conceived in terms of customary rather than statutory law. As will shortly be seen, there was thus considerable scope for contestation over the customary rights and powers of each party, and whether the parameters of corporate possession should be defined in terms of the landed clan or the tribal authority.
The second tension was between the state trustee and the chiefly authority in respect of the representation of the tribe's proprietorial interest vis-à-vis external parties. This was most vividly expressed through the development of the local platinum industry during the apartheid and homeland eras. Since tribal-trust land was often purchased with mineral rights attached, the Rustenburg chieftaincies were endowed with the capacity to appropriate substantial royalty revenues from the platinum corporations that mined it. They were thus constituted as a distinctive form of modern landed property vis-à-vis mining capital, casting the former into a contradictory relation with the latter, in which each agent struggled to capture the greatest share of the mining surplus in the respective forms of profit and (royalty) ground-rent (Capps 2012b, 317–318). Yet, while this opened the way for potentially lucrative rentier forms of chiefly accumulation, the ‘state trusteeship’ element of the Rustenburg property regime would also ensure that tribal minerals ownership posed minimal obstacles to the profitability and free development of the platinum industry (Capps 2012a, 71–72).
As the ultimate trustee of tribal-trust land, the apartheid state mediated every aspect of the relationship between the mineralled chieftaincies and the mining corporations, and could unilaterally consummate or modify the requisite lease agreements. This not only raised questions of which entity – the tribal authority or the trustee state – formally represented the tribe's corporate interests at the ownership level, but also endowed the latter with the capacity to fundamentally weaken, if not entirely abrogate, the property rights of the former in favour of mining capital.2 When Bophuthatswana gained its fictive independence from South Africa in 1977, the powers (and ambiguities) of state trusteeship over Rustenburg's tribal-trust land were transferred wholesale to the new homeland regime, and from there to different departments and tiers of the post-apartheid state. More recently, however, the radical reform of mineral property relations has combined with other legislative developments to launch the Rustenburg chieftaincies onto an entirely new trajectory of ‘accumulation from above’. This has effectively resolved the historic tension around the ownership of tribal-trust land at the level of state trusteeship, while intensifying that between the landed clans and the chiefly trustee within the tribe itself.
Chiefly accumulation from above
The Minerals and Petroleum Resources Development Act (MPRDA) of 2002 is the critical legislation here. Its goals have been to simultaneously accelerate competitive accumulation in the national mining industry and transform its racial composition, alongside other social objectives.3 This has entailed: eliminating the barriers to investment posed by private mineral rights, through their effective nationalisation; proscribing black shareholding levels, as a condition of the new mine licensing system; and encouraging those tribal authorities that were receiving mining royalties to convert them into direct equity stakes in local mining operations, in the names of Black Economic Empowerment (BEE) and ‘community participation’. The effects around Rustenburg have been fourfold.
First, all the major chieftaincies in the region have either converted their existing royalty streams into mine shareholdings, or are in the process of doing so. These and other ‘tribal assets’ have typically been vested in dedicated holding companies and trusts, which act as a springboard for a growing range of secondary investments in sectors as diverse as construction, telecommunications and banking. In shifting from the passive receipt of royalty-rent into these highly financialised forms of accumulation, the Rustenburg chiefs have thus been reconstituted as a distinctive fraction of the new BEE elite that has formed at interface between the ANC party-state and the mining multinationals.
Second, these royalty-to-equity conversions have not only helped the major platinum producers meet their proscribed ‘empowerment’ targets, and hence retain their mining licences, but have also incorporated the chiefs as junior partners in their local operations. Now sharing a common interest in the profitable development of these mines, the Rustenburg chiefs have increasingly acted as ‘land brokers’ for the platinum corporations, unilaterally signing over the surface rights to tribal farms in their capacity as the representatives of the ‘community interest’. The way is thus being cleared for unfettered mine expansion in through the aegis of these new configurations of tribal authority and extractive capital.
Third, the chief's ‘enclosing’ role is being juridically strengthened through a parallel set of legislative developments. After a brief experiment with democratising rural relations in its first term, the ANC government has increasingly (re-)turned to the chieftaincy as the primary means of governing the black countryside, reproducing its apartheid-era jurisdictions and entrenching colonial versions of its ‘customary’ powers (Claassens and Cousins 2008). In particular, the Communal Land Rights Act (CLARA) of 2004 attempted to transfer virtually all state land in the former bantustans to traditional authorities, and while the Act was subsequently overturned in the Constitutional Court on procedural grounds, this has clearly remained the intent of the ANC's ‘tenure reform’ policy (Weinberg 2015). Meanwhile, Rustenburg chiefs have taken matters into their own hands, seeking the exclusive registration of tribal-trust land to themselves in the lower courts. The overarching tendency is thus towards chiefly privatisation, dissolving the old, regulatory bonds of state trusteeship, while consolidating the tribal ownership of group-purchased land at the local level.
Finally, the cumulative effect of this new dialectic of chiefly accumulation and popular dispossession has been to bring the intrinsic tensions of the tribal-trust regime to the fore. As Andrew Manson (2013) has shown in an instructive survey of the Rustenburg chieftaincies, the platinum boom has been accompanied by growing conflicts over chiefly corruption, legitimate incumbency, tribal service-delivery and mine-related jobs.4 These in turn are being fought out through the courts, within and between different government departments and, most dramatically, through local protests and demonstrations, which are invariably met with state repression. Running through these myriad forms of social conflict and the diverse actors they involve is the common thread of land, and in particular the critical question of whether its ownership should exclusively vest in tribal authorities, or in those clan-based groups claiming descent from its original buyers. In order to explore the political dynamics of these ‘claims from below’, we now turn to the case of the Bakgatla-ba-Kgafela traditional authority.
The Bakgatla case
Encompassing some 40 farms, the Bakgatla-ba-Kgafela (hereafter, Bakgatla) territory spreads over the northern foothills of the Pilanesberg Mountains, some 60 km north of Rustenburg (see Figure 1). With an estimated population of 160,000 residents living in 32 villages (Hamilton 2012, 35), it is one of the largest traditional authority areas in the North West Province. Bakgatla has also been implicated in the regional platinum economy since prospecting began in the late 1920s, but it is in recent years that mining has really taken off with the broader expansion of the industry.
Mining proper began in Bakgatla in 1968. That year, Chief Tidimane Pilane signed a mineral lease with Rustenburg Platinum Mines – the forerunner of Anglo American Platinum (Amplats) – to extend its Union operation5 over the tribally registered farm, Spitskop. Royalty payments began in 1982 and were worth R6.2 million per annum by the mid 1990s (Manson and Mbenga 2014, 155). Then, in 2006, an entirely new agreement was reached with Amplats in line with the BEE imperatives of the MPRDA. The projected value of the tribe's future royalties was converted into a 15% equity stake in the Union mine, while Amplats ‘released’ the mining rights to two tribal farms for development with a new partner. This opened the door to a complex series of transactions that resulted in, first, the development of the Pilanesberg Platinum Mine (majority owned by Platmin, a subsidiary of the Jersey-registered consortium Pallinghurst Resources), and then, in December 2012, the formation of a major new producer – Sedibelo Platinum Mines Ltd – through a joint venture between Pallinghurst (42%), the Bakgatla traditional authority (27%) and the Industrial Development Corporation (IDC) (16.2%).6 Dubbed the African Queen Project, and with a mining right area of 150.19 km² that encompasses nine tribal farms, this is now set to become one of largest platinum operations in South Africa.
The current chief, Nyalala Pilane, is credited with masterminding these mining deals and transforming the Bakgatla tribal authority into a major BEE concern. Comprised of a spiralling web of mine ventures, holding companies and diversified investments, the Bakgatla business empire now has an estimated worth of R15 billion (Khanyile 2012). Yet the contradictions of this tribal accumulation project are manifold. A former businessman and General Secretary of the Rustenburg United Taxi Association, it is widely claimed that Nyalala ‘treats the tribe's assets as his own property, registering companies and bank accounts in his own name as opposed to the tribe's, and unilaterally deciding on mining and investment deals without consulting the wider community’ (Claassens 2012). In 2006, the chief, his cousin and two associates were found guilty of 42 counts of theft, corruption and fraud (Mnwana 2014, 24). Although their conviction was later overturned on appeal, an internal audit of the tribe's accounts in 2012 revealed that only between one and five per cent of its expenditure was on community upliftment projects, while financial mismanagement was legion (Claassens and Matlala 2013, 125). The chief has however demonstrated a penchant for prestige developments, erecting new tribal authority offices, a R174-million football stadium and a gleaming shopping mall in Moruleng, his principal village.
There have been numerous attempts to depose Nyalala or at least hold the tribal authority to account (Claassens and Matlala 2013; Matlala 2014; Mnwana 2014). These have involved complex and shifting combinations of rival royal factions and more popular initiatives, like the Concerned Bakgatla Anti-Corruption Organisation (COBACO). Yet, the Provincial courts have invariably sided with Nyalala against his opponents, while appeals to the Provincial government have largely fallen on deaf ears, fuelling perceptions of high-level collusion between the tribal authority, mining companies and senior ANC politicians. Land is a particular flashpoint. The chief has not only negotiated numerous lease deals with existing mining partners, but has also ceded the surface rights to other farms to a dedicated entity aimed at attracting further investors. At the time of writing, a major legal struggle is also raging over the question of whether thousands of hectares of high-value land that were previously restituted to the tribe through a successful land claim should be administered by an elected Communal Property Association (as was first decided), or vested in a ‘trust’ under the exclusive control of the tribal authority (Evans 2015).
This section focuses on the village-level struggles that have unfolded against this backdrop of deepening antagonisms around the chief's ‘class project’. Intensive fieldwork was undertaken in three settlements – Lesetlheng, Motlhabe and Sefikile – between July and November 2013, supplemented by follow-up visits over 2014.7 Marked on a map of the Bakgatla traditional authority area (Figure 1), all are sites of popular contestation over farms historically acquired through the tribal-trust regime, but there are also significant variations in the relationships between property, community and authority in each case. We shall now see how the specific histories of the three villages have shaped the distinctive ways in which these small-group claims have been articulated in relation to the wider tribal polity, but how this has also generated other lines of lines of exclusion and differentiation that may themselves be indicative of ‘class formation’ (Peters 2004, 305).
Lesetlheng
The first claim to be considered here is by residents of Lesetlheng village over the farm Wilgespruit. One of the oldest settlements in the area, Lesetlheng is largely comprised of people who trace their origins back to a series of fissures in the Bakgatla ruling lineage during the mid nineteenth century (Morton 1998, 87). A group broke away from under the leadership of a brother of the then chief and established an independent settlement close to what is now the Sun City resort. They subsequently purchased the farm Kruidfontein though the intermediation of a local missionary, and decamped there in 1888 to found Lesetlheng. By the turn of the century, however, the community had been reincorporated into the ‘official’ Bakgatla tribe, and its leadership demoted to the status of ‘headmen’ by the colonial authorities. Nevertheless, the community retained a sense of its former political autonomy, which was further strengthened through the collective acquisition of Wilgespruit between in 1916.
There is strong written and oral evidence that the purchase price for Wilgespruit was raised by an independent syndicate of 13 clans from Lesetlheng (Mnwana and Capps 2015, 17–18). The village headman oversaw the collection of cattle and recorded the contribution of each family. However, by this point in the evolution of the tribal-trust regime, such groups were compelled to register their purchases through the aegis of a recognised chief. It was consequently the name of the Bakgatla chieftaincy that would be inscribed on the title deed.
Yet, in practice, the land not only remained under the immediate control of the Lesetlheng syndicate, but was further subdivided among the 13 clans that comprised it. Each was allotted a large ploughing area in line with the size of its contribution, which was in turn distributed among its constituent families. Moreover, it was a clan's prerogative to grant usufruct rights to groups or individuals who affiliated after the farm was purchased, as this informant explains:
Kgoro refers to a clan which is a group of households that mainly share the same surname and same identity. However, some clans expanded by allowing other families who either married or sought refuge to become members of their kgoro … These families were allocated pieces of land within the portion of the adoptive clan to cultivate and feed the children. (Interview: Pretoria, 10.11.2013)
Wilgespruit was one of the most productive farms in the Bakgatla area (Breutz 1953, 282), and a significant source of livelihood for the Lesetlheng villagers. However, in 2007, they were informed that the Pilanesberg Platinum Mine (PPM) had acquired the mining rights to the farm via the chief and it soon began to be fenced off. There was no compensation for the loss of this land, and when the mining revenues began flowing to the tribal centre in Moruleng nothing came back in return. The villagers responded by forming the Lesetlheng Land Committee (LLC) to contest the tribal registration of Wilgespruit. Their aim was not to secede from the chiefdom, but rather to have their property rights formally recognised within it. This would then enable them to negotiate directly with the mine and enjoy a fair share of the benefits:
[Chief] Nyalala has been calling meetings and telling people that we, the people of Lesetlheng, are claiming land which belongs to the 32 villages of Bakgatla. … [But] they have never used that land. Our farm is divided according to the 13 clans – those who bought it. These clans must get their share of mining revenues first and then a certain percentage will go to the Bakgatla tribal office. That percentage will be shared by the whole community. (Interview with LLC leader: Lesetlheng, 07.08.2013)
According to informants, the families that bought Wilgespruit are called dibeso (sing.: sebeso), while incomers were commonly known as bagotsi-ba-mollo – ‘those who make fire for others’. These were usually landless immigrants of different ethnic origin who had affiliated to the buying clans, and been granted ploughing rights on Wilgespruit, and sometimes loaned cattle and equipment. In return, the bagotsi were required to wake up early every morning and make fires for the main families in the clan, and also assist with ploughing, herding cattle and other tasks. They were, in short, incorporated into these clans as labour tenants. Such differences were now being revived as the basis of more exclusive identities in the Wilgespruit claim. But in some cases they may also have been invented. A number of informants reported that the notion of sebeso was of recent origin, and had emerged as a means of distinguishing the descendants of the buyers from the non-buyers. ‘We grew up using the word kgoro,’ said one, and it was only ‘in 2008 when the word sebeso started to be used more and more’ (Informal conversation: Lesetlheng, 12.08.2013). In a manner not dissimilar from more conventional restitution claims, the very process of claiming Wilgespruit has thus revealed ‘a series of sharply differentiated historical experiences’ and ‘widely divergent interests’ within the corporate community that are simultaneously obscured by its political demarcation as an historically independent entity (James 2008, 146).
Motlhabe
If the prevailing logic of the Lesetlheng claim is to secure the community's private rights within the framework of the tribe, then in Motlhabe it is to realise an historic claim to land through the recognition of the community as a tribe in its own right. The drive to secede from the established Bakgatla-ba-Kgafela chieftaincy is being led by a royalist faction in the village, which argues that the community was established on the farm Witkleifontein during the nineteenth century.8 Its founder, Kautlwale Pilane, was the first son from the 11th House of the great chief, Pilane, and at this point ‘we were all equal as Bakgatla and the name baKgafela was not used to indicate a separate tribe.’ During the 1920s, the baKautlwale raised the purchase price of a neighbouring farm, Welgewaagd, but were forced to formalise the transaction through the recognised chief in Moruleng, under the auspices of the tribal-trust regime. Welgewaagd was consequently registered in 1926 to ‘the Minister of Native Affairs in trust for the Bakgatla tribe’. Soon after, in 1932, the baKautlwale were evicted from Witkleifontein by its white owner, and resettled on Welgewaagd, the site of the current Motlhabe village. Nevertheless they continued to graze cattle on Witkleifontein, which they consider their ancestral land, and exercised independent control over their local affairs on Welgewaagd.
In 1953, however, the political circumstances of the BaKautlwale took a turn for the worst with the formal constitution of the Bakgatla-ba-Kgafela Tribal Authority:
The government redefined the Bakgatla as the Bakgatla-ba-Kgafela … at the same time, the other royal houses of the Bakgatla were converted into being ‘headmen’ under the Bakgatla-ba-Kgafela instead of separate and independent Bakgatla royal houses as they had been previously … and we have never been able to escape this historical mistake. (Breutz 1989, 339)
In 1998, a restitution claim was lodged over Witkleifontein by the descendants of the baKautlwale who had been evicted from the farm in the early 1930s. In the intervening period, Witkleifontein had been acquired by the South African Development Trust and integrated into the Bakgatla ‘location’, and has more recently has been fenced off for the northerly expansion of the PPM mine. Yet, the restitution claim was recorded ‘under the name of the Bakgatla-ba-Kgafela and [Chief] Nyalala is now claiming that the land, which only we as the baKautlwale ever occupied, rightfully belongs to him’. Confronted by the alienation of their historic grazing land, but thwarted in their pursuit of an independent claim over it, the baKautlwale leadership again began to agitate for secession. In 2009, notice was given that Motlhabe and its ancestral territory would no longer fall under the jurisdiction of the tribal administration in Moruleng, and that the baKautlwale would henceforth negotiate directly with PPM over its use and benefits. As they subsequently explained, ‘Nyalala is putting forward versions of custom that prop up his power to steal our land and our mineral resources’: we want to be ‘recognised as equal and separate from the BaKgafela’, as do ‘other royal houses in surrounding villages’.
Matters came to a head in January 2010 when Mainole's successor, Mmuthi Pilane, attempted to convene a village meeting to discuss secession in the name of ‘Motlhabe Tribal Authority’. This nomenclature provided Nyalala with the pretext to interdict the meeting on the grounds that the baKautlwale leadership had created the false impression of having an official status in terms of the 2003 Traditional Leadership and Governance Framework Act (TLGFA). The following year, the Bakgatla chief was unequivocally backed by a North West High Court ruling that ‘any action by a parallel but unsanctioned structure that is neither recognised by law or custom, seeking to perform or assume functions that are clearly the exclusive preserve of recognised authorities, ought to incur the wrath of law.’9 However, with the support of the Legal Resources Centre, the judicial tables were turned on Nyalala in 2013. In a landmark judgment, the Constitutional Court found that the ‘interdicts adversely impact on the applicants’ rights to freedom of expression, association and assembly’,10 and that village meetings could be organised in tribal areas without a chief's permission. Buoyed by this legal victory, Mmuthi Pilane soon after applied for his familial claim to the headmanship of Motlhabe to be formally recognised, again in terms of the TLGFA. Yet, like his father before him, this was not merely to undo the injustice visited on the baKautlwale lineage by the baKgafela chieftaincy at the level of village politics. It was also to provide a platform for Motlhabe to demand complete independence of the Moruleng autocracy, and thereby secure the clan's sovereign rights over its mineralised land.
A Commission on Traditional Leadership Disputes and Claims was duly appointed to investigate the matter by the North West Provincial Government. This was still in progress at the time of fieldwork. However, by this point, it had also become apparent that the secessionist strategy had generated significant tensions in the village. These were of two orders. First, although the court battle against Nyalala had provided a focal point for more generalised discontent with his rule, the fact remained that a substantial section of the village wished to remain under the baKgafela chieftaincy, even if they were critical of the current incumbent. In particular, there was doubt about the legitimacy of the baKautlwale's claims to independent chiefly stature, as well as suspicion that it was using wider opposition to Nyalala and the mine to further its own ambitions. The effect was to politically cleave the village into supporters of the baKautlwale and of the baKgafela.
Second, Motlhabe's youth were largely alienated from the gerontocratic politics of secession, and more concerned with immediate issues like the dearth of village infrastructure and, especially, mine employment. As one youth leader explained:
As the youth, we see that the mine close to this village, but it is doing nothing for us. Unemployment is too high. Development is very slow. We are oppressed by the chief. All the levies [revenues] from this mine [PPM] go to the Bakgatla-ba-Kgafela in Moruleng. The descendants of Kautlwale are also fighting for chieftaincy and they claim that they bought this land. But the title deed does not confirm their claim. We are confused. Whenever the baKautlwale call a meeting they talk about chieftaincy – not about our needs as the community. If they want chieftaincy they must leave us alone. The youth want jobs. They want development! (Interview: Motlhabe, 05.04.2014)
Sefikile
Clustered around a small rocky outcrop on the northern perimeter of the Bakgatla territory, the village of Sefikile has been historically characterised by a uniquely inclusive political culture. Whereas the leading clans in Lesetlheng and Motlhabe trace their origins to fissures within the Bakgatla ruling lineage, the multi-ethnic roots of Sefikile are indicated precisely by its name: meaning ‘we have arrived’ in isiNdebele, it was an offshoot of Mzilikazi's raiding state that founded the settlement during the turbulent years of the early nineteenth century. This slowly grew through the incorporation of diverse groups and individuals to number some 50 to 59 households by the early twentieth century (Mnwana and Capps 2015, 29–30). But it was the collective purchase of the farm Spitskop – on which the settlement was located – that would really define the baSefikile political identity.
According to our informants, Spitskop was acquired in two lots between 1910 and 1918 by a syndicate of 52 families – virtually the entire population of the village. The purchase price was raised independently of the Bakgatla chieftaincy (which by this point had administrative jurisdiction over Sefikile) under the leadership of a local African minister of the Dutch Reformed Church, who recorded the names of the buyers and their contributions. However, it was inevitably the name of the Bakgatla chief and tribe that was inscribed on the title deeds under the tribal-trust regime:
People bought land [here] before and after 1913. But the apartheid laws deprived black people of the right to own land, even if they bought it … If a group of people bought land, they had to give away that land into the custodianship of the chief. No chief ever bought a piece of land! Chiefs owned land by virtue of them being guardians. (Interview: Sefikile, 08.11.2013)
The unique combination of political autonomy and ethnic inclusivity that defined the corporate community was further institutionalised and reproduced by its distinctive system of local governance. Despite its formal integration into the Bakgatla Tribal Authority, Sefikile's headmen were neither derived from the core Bakgatla lineages, nor did they inherit their position from their fathers. Rather, the tasks of village leadership – adjudicating disputes, allocating residential or ploughing land, representing the community in the tribal administration in Moruleng – fell to influential men who assumed the role voluntarily. This broadly democratic form of self-government (women and youth were nevertheless excluded), lasted until 2009 when residents removed the incumbent headman on the alleged grounds of his corruption. A group of activists who identified themselves as the Bakgatla-ba-Sefikile Traditional Community Association (BBSTCA) took over the leadership of the village and elected a committee, which administered community finances and the allocation of stands. This would in turn become bound up with a major dispute over the ownership and control of Spitskop that, as elsewhere in Bakgatla, was sparked by the development of mining, but with a form and timing quite specific to the locality.
As was seen above, a significant portion of Spitskop has been leased to Amplats’ (formerly Rustenburg Platinum Mine's) Union operation in terms of an agreement signed with the then Bakgatla chief in the late 1960s. However, local residents claim that they were neither consulted about the deal, nor were they compensated for the loss of their ploughing and grazing lands. The mine consequently became known as ‘Sinkgalaleng’ – ‘don't undermine us.’ In the late 1990s, village youth took up the struggle against Union, arguing that the village had never benefited from mine jobs or infrastructural development. Tensions surfaced again in 2006 when rumours began to circulate that Chief Nyalala was negotiating a direct equity stake in Union, through the conversion of the tribe's royalty stream. Villagers claim that up to this point they were unaware that Union had been paying royalties for its Spitskop operation, and tempers were further inflamed when it emerged that the mine had been illegally encroaching on land outside of the lease area. What had initially begun as a youth revolt against the mine now rapidly spread through Sefikile to include elder residents, who in turn educated the youngsters about the farm's purchase history. Tracing a radically different path to both Lesetlheng and Motlhabe, and drawing on the villages’ distinctive history, land thus emerged as the unifying issue in a common struggle that subsumed potential differences of ethnic origin and generation within the claimant community.
Representing the descendants of the 52 families who had acquired the farm between 1910 and 1912, and supported by a prominent human rights lawyer, the BBSTCA filed a court application to challenge the tribal registration of Spitskop in 2010. However, when the matter was finally heard in the North West High Court in 2011, their case was dismissed with costs. The judgment not only relied on colonial conceptions of the chiefly custodianship of communal land, but held that there was a lack of sufficient documentary evidence to support the claim that the baSefikile syndicate had raised the purchase price. Financially depleted, the BBSTCA was unable to appeal the judgment, but by now the committees’ popular support had waned as allegations surfaced about its own misuse of community funds. Moreover, parallel to this, other kinds of land disputes had been unfolding, which point to the potentially exclusionary nature of the prevailing conception of corporate community in Sefikile.
The impact of mining is far more visible on Spitskop than any other farm in the Bakgatla area. This is not simply because of the alienation of land for mining purposes, but also as a result of the exponential growth of informal settlements that are now housing thousands of migrant mine workers and job seekers. In contrast to the newer PPM mine, which is a capital-intensive open-cast operation, Union is a deep-level mine that relies on a much larger workforce, recruited from the traditional labour-sending areas of South and Southern Africa. After 1994, when the mines started to pay living-out allowances, these migrants began to move out of the hostels provided by Union and establish informal settlements on surrounding land. Thus, in Sefikile, the incursion of the mine has increasingly been experienced as an incursion of ‘foreigners’, who are not only ‘stealing’ the village lands, but also the jobs that should be reserved for locals in ‘compensation’ for the loss of rural livelihoods. As one village activist explains, these tensions have been further inflamed by the actions of the mine itself:
Foreigners are the mine workers from the Free State and the Eastern Cape who are occupying our land by force. This all started when these guys began moving out of the hostel and occupied [mine] land behind [it] … [But] unfortunately, the management of this mine removed them. Police came and demolished their shacks. That was when they began to occupy our ploughing and grazing fields at Sefikile. We tried to stop them. A conflict started. Two of our youth leaders were killed during that conflict. (Interview: Union Mine, 07.09.2013)
However, in the 1990s, as land pressures mounted and the broader political climate changed, male villagers began to sell portions of their ploughing fields as residential plots to single women – a phenomenon noted elsewhere in post-apartheid South Africa (Claassens 2013). As more unmarried women entered the informal land market, this area soon developed into a new section of the village known as ‘Komasingling’ – ‘the place of the singles’. This was further transformed when the advent of the living-out allowance created a new market opportunity. A new breed of ‘shacklord’ – often isiXhosa-speaking mineworkers themselves – started buying up more plots from locals and erected rows of shacks to rent to other migrants. This has proved a lucrative business. Between 8 and 20 shacks can be found in one plot, without any form of public infrastructure or services, and the settlement has since been renamed ‘Machela Pata’ – ‘across the road’ – to reflect its changing social composition. While new group boundaries have been erected against the ‘incursion’ of migrant workers at one level, they have been undermined at another through the effective commoditisation of village land initially allocated to residents through its corporate structures.
Conclusion
Intensifying competition over land across sub-Saharan Africa is manifesting in struggles to establish more exclusive boundaries and identities within and between corporate groups of different scales and compositions (Amanor 2008; Bernstein 2007; Peters 2004; Woodhouse 2003). We have argued that these continent-wide processes are assuming a particularly sharp form around Rustenburg, due to the legacies of a distinctive property regime that historically combined the group-purchase of land with its registration to state-defined tribes. The expansion of the platinum industry under changing legislative conditions has bought the historic tensions of this tribal-trust regime to the fore. On the one hand, mineralised tribal authorities are converting previous royalty streams into direct equity stakes in local mining operations, thus simultaneously reconstituting themselves as a fraction of the new BEE elite while clearing the way for further mine expansion. The recent history of the Bakgatla chieftaincy exemplifies this new trajectory of tribal ‘accumulation from above’. However, as the material stakes multiply and dispossession proceeds apace, a new wave of popular struggle has erupted around the question of whether the ownership of this land should exclusively vest in tribal authorities, or in those clan-based groups claiming descent from its original buyers. We conclude with three comparative observations from our investigation of these ‘claims from below’ in Bakgatla.
First, a remarkably similar history of group land-acquisition has emerged from the three study villages. In each case, farms are being contested that were purchased by African land syndicates in the first quarter of the twentieth century. It is claimed that these were acquired independently of the Bakgatla chieftaincy for the exclusive benefit of the groups that raised the purchase. The collective acquisition of this land thus also entailed the assertion and/or formation of discrete political identities. In the cases of Lesetlheng and Motlhabe, the land buyers distinguished themselves from the wider Bakgatla polity on the basis of their descent from royal lineages that had split off from the ruling core at different points during the nineteenth century. In Sefikile, an entirely new corporate identity was actively constructed to facilitate the acquisition of Spitskop, which unified and defined its diverse political community. Nevertheless, all three land-buying groups were forced to register these properties through the aegis of the Bakgatla chief in order to meet the administrative requirement of a tribal purchase. This resulted in formal registrations that bore no trace of the actual buyers and legally compressed these divergent configurations of group land-ownership into a singular tribal property.
Second, there is a reassertion – if not recreation – of these discrete political identities by today's land claimants as a means of establishing their private-group rights. Yet this process of political demarcation is happening in different ways, and with different degrees of legitimacy. In both Lesetlheng and Motlhabe arguments are being rehearsed about past genealogical seniority and independence from the ‘recognised’ chiefly lineage in Moruleng. However, whereas in Lesetlheng this is to facilitate a claim to independent land title within the framework of the tribe, in Motlhabe it is to secede entirely. This not only reproduces a much older strategy of securing claims to land by seeking official recognition as an independent tribe, but is suggestive of the extent to which the ‘rules of the game’ that were established in the early twentieth century have themselves been reproduced in the post-apartheid period (Keegan 1988, 147–149). Yet now, as then, it is a game in which the dice remain loaded against those small groups articulating independent claims to resources on the basis of their historic independence from the chiefly lineages inscribed in the ethnographic handbooks of the state (Comaroff 1974, 45–46). In both Motlhabe and Sefikile, the North West courts sided with received notions of homogenous tribes and chiefly trustees, while the broad thrust of ANC land policy has increasingly been to privilege tribal ownership over all other forms of corporate possession.
Whether by accident or design, this in turn is crafting a form of large-scale landed property that facilitates the access of mining capital to platinum and other high-value mineral resources in the former homeland areas, by eliminating the barriers to investment that would arise from the legal recognition of a plethora of small-group rights. At the same time, chiefs are being locked into these arrangements as BEE partners, with a shared interest in unfettered mine expansion. Seen from this perspective, then, the small-group claims erupting around Rustenburg are not merely a matter of confronting the historic injustice of tribal land registration, and realising property rights that were racially denied. They also thus far represent by far the most significant popular challenge to what the South African commentator Nomboniso Gasa (2014) has aptly termed the new ‘triumvirate’ of chiefly elites, ANC politicians and mining corporations, and hence are on the frontline of the struggle against South Africa's new political economy of extraction.
Finally, as the village cases have also shown, the attempt to establish private property rights through more exclusive group definitions is itself generating new lines of social demarcation and division that have the potential to undermine and fragment this nascent counter-movement. In Lesetlheng, the very process of making the claim has revived and perhaps even created older differences between buyers and non-buyers, politically expressed in the distinction between ‘bona fide’ clan members and affiliates. In Motlhabe, the secessionist strategy of the ba-Kautlwale group has seemingly been rejected by the youth and other villagers who fear it will simply replicate the logic of chiefly power at a more local level. And in Sefikile, significant tensions have emerged between the claimant community and mine migrants and single women struggling for independent access to residential land. All of these different social cleavages and divisions point to the dangers, if not the limits, of seeking social justice through narrowly defined claims to private-group ownership, in which ‘notions of entitlement’ are ethnically based on ‘descent from particular ancestors’ (James 2008, 154).
Moreover, to recall Peters (2004, 305), it may well also be the case that what we are witnessing in these village-level struggles are the incipient ‘dynamics of division and exclusion, alliance and inclusion’ that are constitutive of rural ‘class formation’. Thus the claimant's drive to assert their bourgeois rights vis-à-vis both the mining corporations and the chieftaincy may not only be a defensive response to their dispossession and a historic response to their oppression as political minorities within the tribe. A successful claim also holds the promise of an entitlement to a direct share of mining revenue, and hence itself may represent a project of ‘accumulation from below’. At the same time, the realisation of an exclusive right to land could result not merely in the exclusion of those social categories within the political community who do not qualify as ‘beneficiaries’, but also the expulsion of mine migrants and all others defined as ‘outsiders’ from the land itself. None of this is inevitable, multiple social forces are involved and many other scenarios may unfold. Nevertheless they point to the challenges that will need to be politically confronted by a new South African left that seeks to link workplace and community struggles and forge them into a united movement on the platinum belt.