Introduction
The new democratic dispensation in South Africa has been characterised by proliferation of work fragmentation and precariousness. Labour has been severely fragmented while employer control and flexibility on how work is organised has been enhanced. Central to these changes is the sustenance of a cheap labour regime and flexibility. A cheap labour regime has been consistent from the onset of the mining industry in South Africa in the late nineteenth century.
This paper draws from the Marikana 2012 strike and gives a descriptive account of the patterns of fragmentation and precariousness and the connection to strike violence and worker mobilisation. It illustrates the ambiguous nature of strike violence by showing how the workers attained and sustained ‘solidarity and militancy’ in the face of fragmentation and precariousness.
The killing of 34 mineworkers at Marikana by the police service has brought to the fore the changing nature of employment in South Africa. Neoliberalism drives the global economy with the primary motive to maximise capital accumulation of surplus value for both domestic and foreign capital (Theron 2003, Harvey 2005). Although underpinning the driving forces of the exploitation regime in mining are global forces, the Marikana massacre has highlighted that the negative effects of neoliberalism are most felt locally.
The Marikana massacre unfolded almost 20 years after the democratic transition in South Africa and posed questions regarding the new dispensation. Important to note is that democracy and violence are interesting themes in South African history that manifest in numerous contexts. Democracy is associated with the broadening of political participation, as well as balancing political rights and promoting nonviolence in making claims (Tilly 2003). It has also brought about the establishment of third parties, for example, conciliation and arbitration panels to prevent violent resolution of disputes in labour relations. However, in South Africa democracy is significantly characterised by the persistence of violence in claim making. Violence has remained an important phenomenon in the polity since the democratic transition. Violent outbursts often bring simmering social tensions to the fore.
In industrial relations theory, conflict and strike violence may be managed through institutionalisation. This assigns a key role to the development of institutions with the capacity to regulate and manage conflict (Korpi and Shalev 1979). Strike violence during apartheid was initially explained by the lack of institutionalisation of industrial conflict for black workers. Although black workers were later partially incorporated in the 1980s following the Wiehahn Commission recommendations, the nexus between political and workplace struggles undermined institutionalisation of industrial conflict (Von Holdt 2010).
In the 1980s black people attained industrial citizenship but had no political citizenship. The South African constitution and labour relations regime adopted after the demise of apartheid guarantees both industrial and political citizenship. Nevertheless, violence in industrial conflict as at Marikana remains an important phenomenon. A 2012 survey by the Congress of South African Trade Unions (COSATU) shows that 60% of the workers interviewed believed the use of violence was a necessary strategy in strikes (NALEDI 2012). Von Holdt (2010) argues that institutionalisation of industrial conflict is tied to the broader political and socio-economic struggle which in South Africa remains highly contested and characterised by gross inequality.
The context
Since the onset of the mining revolution in South Africa in the 1860s, the organisation of work has gone through changes mainly designed to minimise risk and maximise profits. The underlying principle that has remained consistent since then is the use of cheap black labour and the migrant labour system (Wolpe 1972) designed to maximise surplus value.
The changes in how mining work is organised since the democratic transition are linked to fragmentation and precariousness both at work and home. Fragmentation refers to the breaking down of activities into smaller segments. It serves many purposes but primarily enhances the control of workers. Mining capital has thus drifted from ‘compounded to fragmented labour’ (Bezuidenhout and Buhlungu 2010). It has shifted from permanent to third party employment as an ongoing exercise following the emergence of union hegemony and the democratic transition. This became more pronounced in later years following the demise of apartheid and the assertion of neoliberalism. Lifetime employment has severely declined as more jobs are now temporary and insecure.
In platinum, for example, over one-third are employed by third parties (Bezuidenhout and Buhlungu 2010). This works as a mode of control. For example, precariousness enhances employer control of the pace of production and discipline. It pushes the worker harder and instils disciplines as there is always a threat of dismissal (Smith 1998). As a result, many of the new jobs reflect how work has been re-organised. However, in many cases, third party employees are closely controlled as direct employees and integrated into the principal mining operations. Although the South African legislation has been appraised as progressive and labour friendly, in practice employers are able to circumvent this and create a precarious context.
Fragmentation of work in the current neoliberal dispensation is always ongoing and shifting. This has implications for worker mobilisation and dispels the possibility of worker solidarity (Smith 1998). The third party workers are the least accessible to unions because of heightened job insecurity which undermines union mobilisation and organisation.
Platinum has the highest rate of third party employment in mining. Over one-third in the sector are employed through subcontractors and labour brokers (Bezuidenhout and Buhlungu 2010). This dramatically changed from the 1990s as neoliberalism asserted its hegemony. There are a number of reasons why platinum shifted to a more labour-flexible regime. The sector before the 1990s was small and the employers were not organised centrally. As a result, the use of cheap labour became an entrenched strategy enhancing competitive advantage.
The mining industry in South Africa has from the onset been dependent on migrant labour. The system developed initially as a response to the temporary nature of mining and lack of permanent settlement (Burawoy 1976). It was also designed to preserve an economic and political system based on exploitation and repression of black workers (Wolpe 1972). This has nevertheless persisted with some continuities and discontinuities.
The compound and later hostel system of accommodation developed along with the migrant labour system in mining. Before the attainment of democracy, the majority of the mine workers lived in hostels where as many as 25 workers shared a single room. This emerged as a solution to housing problems and to control mineworkers. After 1982, the National Union of Mineworkers (NUM) was thus able to ‘subvert employers’ logic of control by turning the hostels into sites of mobilisation’ (Bezuidenhout and Buhlungu 2008).
A significant number of low-skilled workers in platinum are migrants from Lesotho, Eastern Cape, Mozambique, Swaziland and other areas. At Lonmin, for example, 82% fall in this category (Lonmin 2012). Many have two families – one at their place of origin and another at their workplace, and are expected to support them from meagre salaries. Many workers have been forced to live precariously in informal settlements without running water, toilets, electricity and roads. Employers offer a living-out allowance (LOA) instead of providing accommodation. In the Rustenburg municipality area – the hub of the world platinum mining production – only 42% of the population live in formal dwellings (Rustenburg 2012).
The 2012 Lonmin strike
Lonmin, the third largest producer of platinum in the world, employs 28,000 workers directly and another 10,000 through third parties (Lonmin 2012). Its mining operations are centred in Marikana, which accounts for over 80% of its output.
The 2012 Lonmin strike was part of a wave that started at Impala Platinum in January 2012. The strike broke out after rock drill operators (RDOs1) embarked on a strike challenging management's decision to award an 18% retention allowance to mine blasters. This mobilisation spread across the platinum belt and beyond. The strikes had similarities in the claims, violence, repertoires and the actors. The RDOs, later joined by other unskilled workers, were uncompromising. They demanded a wage award of over 200% outside the collective bargaining system, and exhibited unprecedented levels of militancy.
The demands were presented by independent committees at the shaft level to the local management. From the onset, the RDOs rejected representation by the National Union of Mineworkers (NUM) which then was the majority and only recognised union at Lonmin.2 The RDOs claimed NUM was not accountable to its membership, corrupt, indifferent to their plight and compromised by management. They alleged corruption by shop stewards and union branch officials in issues such as company house allocation and recruitment, in which the union plays an important role (Chinguno 2013).
In most of the strikes there was an underground workers' committee responsible for meting out violence to dissenters. The committee would close off all the roads leading to the various work stations and assault anyone reporting for duty. Violence was thus a means of overcoming worker fragmentation and enforcing and maintaining ‘solidarity’.
Ironically, when the strike ended, third party workers paid the utmost price as Murray and Roberts, a leading Lonmin contractor, terminated 2000 contracts.
The platinum mining workforce is fragmented along lines of ethnicity, skill, settlement, gender and others. The lack of union cohesion, common interests and objectives made violence susceptible and undermined worker collective solidarity. The strikes were marked by violence and intimidation at different levels – i.e., intra- and inter-union, vertical and horizontal, and within the workforce and state (police). The violence often targeted scabs.
Fragmentation undermines worker mobilisation and collective solidarity in many dimensions. The changes in the mine settlement pattern play a role in worker fragmentation. Before the democratic transition, all mineworkers lived in company hostels and were not part of the nearby communities. This has since changed. The influx control legislation which restricted black workers' migration and mobility was disbanded in the 1980s and new settlement geography emerged. While a few of the mineworkers have remained in hostels, the majority live in local villages, own homes in high- and low-density suburbs, or stay in family units and in informal settlements. A significant number of workers moved into informal settlements that sprouted across the mining belts to supplement their salaries through saving from the LOA. This is all linked to the cheap labour regime.
The mining communities are broadly divided into the traditional villages, which preceded the platinum mining, and the new informal settlements that emerged following the democratic transition. Lonmin has three hostel complexes where the majority of its staff lived up till 1994. After democracy, the new government, the ANC and COSATU resolved to phase out hostels, and they advanced construction of family units to allow workers to live with their families. They argued the move was part of a strategy to destroy the apartheid legacy and restore the dignity of black workers. The Chamber of Mines and NUM brokered an agreement on an LOA for workers. This displaced the employer obligation to provide accommodation but was apparently adopted after limited consultation with the affected workers.
In the Rustenburg platinum belt there are over 38 informal settlements, and at least 37% of all dwellings are informal, compared to a national average of about 13%. Between 2001 and 2007, formal housing in Rustenburg decreased from 47 to 42%, reflecting growing fragmentation and precariousness of the workforce (Rustenburg 2012).
The hostels are no longer central in the control of workers. Many of the third party workers are not entitled to an LOA and cannot afford decent accommodation. This has forced many into the informal settlements. Many of Lonmin's low-skilled workers live in the neighbouring Nkaneng informal settlement. Lonmin has 1798 houses, 544 family and 95 bachelor units in addition to three hostel complexes which accommodate unskilled and semi-skilled workers. It can thus only provide recommended accommodation to 2342 workers – less than 10% of its 38,000 workforce (Lonmin 2010). At least 50% of the population living within a 15 km radius of its operations live in informal settlements with no basic services (Lonmin 2010). This new pattern reflects a new geography and new contours of class, citizenship, race, ethnicity and status that have emerged post-apartheid (Chinguno 2013).
This new geography has an important bearing on labour relations. An important feature of many of the informal settlements is the weakness or absence of local state structures. These have been in some cases substituted by informal structures that use vigilante-type violence to maintain order. This created a context for alternative systems to emerge. Many of them are ‘no-go areas’ for the South Africa Police Service. The absence of state structures or their retreat is not synonymous with the mining settlement but is quite widespread in South Africa (Monson 2011).
Although the demise of the hostels and the new spatial settlement geography may be a source of power and leverage for the workers as it reduces dependence on the employer and control, in reality this advantage has in most cases not been fully realised.
Fragmentation of work organisation
Neoliberal globalisation, according to its proponents, is linked to profit maximisation through minimisation of risk: for example, the substitution of direct by triangular employment. The employment contract is thus substituted by a business contract. Its foundation, according to Harvey (2005), is tied to classical liberalism which purports to be linked to human dignity and individual freedom, viewed as central to civilisation. It is based on the assumption that individual freedom is guaranteed by free market and trade (Harvey 2005). Theron (2003) explores the changing nature of employment in South Africa and argues that the changes are driven by global economic changes shaped by ideology. There has been proliferation of externalisation of work in the post-apartheid order as a response to enhance flexibility and changes in the global economy in line with the neoliberal doctrine.
Sandvik, an engineering company, has a typical example of flexible work arrangement. Its South African offices are in Jetpark, Johannesburg, but some of its workers are at Lonmin in Rustenburg. It also subcontracts other labour brokers such as SAN (see Figure 1). The workers are contractually employed by SAN but directly supervised by Sandvik. Figure 1 highlights the typical employment relations in platinum mining in South Africa. The subcontractors and the labour brokers in this case are closely integrated in the core operations. Lonmin is able to exercise direct control including over occupational health and safety standards. Thus fragmentation of work may coexist with operational integration (Williams, Davis, and Chinguno 2013).
However, this severely undermines worker mobilisation and solidarity. A Sandvik worker that was interviewed lamented how it is difficult to be active in the unions. Sandvik is an engineering company which in this case is contracted in mining. It is ordinarily organised by an engineering union such as NUMSA. However, in Rustenburg it is in a sector organised by NUM and the Association of Mineworkers and Construction Union (AMCU). This leaves workers in a quandary. As a result many have opted not to be active in unions.
During apartheid, most black workers were confined in the low-skilled job categories. The skills levels for black workers since apartheid have broadened. Black workers are now found in any level of the occupational hierarchy. In the past, they easily forged solidarity as low-skilled black workers. This formed a strong worker identity which has since dissipated and undermines worker mobilisation and solidarity.
Gender
Mining is dominated by men. This is explained by its demand for physical strength and endurance. In the past, legislation in South Africa barred women from underground mining. As part of the social transformation and promotion of gender equality, the government instituted legislation promoting women in mining. The Mining Charter (2002) instituted a minimum threshold for women in mining. It targeted 10% by 2009, which was then postponed to 2014. The introduction of women is a part of the internal transformation strategy. Lonmin attained 7.45% of women at a mine and 4.3% women in mining (Lonmin 2011). ‘Women at a mine’ are those working for the mining house in general, while ‘women in mining’ refers to those directly involved in the production process.
Over the past 10 years, more mining houses have integrated women in line with the legislation. Lonmin (2009) highlights its support for the integration of women. It argues that the integration of women is part of its corporate social responsibility and claims that women are better and more reliable employees.
This has brought a new gender dynamic. The majority of the women are usually recruited from local communities and are less likely to be migrants than the men. The introduction of women brings a new culture and struggles. Research by Benya (2013) argues that women bring new values, challenges and ‘class’ orientation. In many cases they have no role models to learn from. Men, for example, have the experience of the role that unions played in the struggle against apartheid. They have the experience of the value of the union in defending workers' rights. Men are thus attached to victories of the past.
The women may have a different conception. They have no experience or attachment with the unions' past successes. Therefore, women may question the significance of a union in cases where there may be no tangible results. Men often perceive the union as an instrument that liberated them from the past repression and will normally ask no questions about its value. Women do not share a similar attachment. They do not live in hostels and a few reside in the informal settlements. They are thus not fully assimilated into the dominant mining culture and come with their own struggles such as sexual harassment.
At Marikana, women were initially barred from the kopje and were not directly active in the strike because of cultural values and beliefs. The introduction of women in mining may be viewed as a form of worker fragmentation which has brought in new repertoires and unanticipated challenges to union organising and mobilisation.
Ethnicity, local community and worker fragmentation
The recruitment of mineworkers in the past was often tied to socio-economic and political factors and other prejudices (Moodie and Ndatshe 1994). Certain jobs were preserved for some ethnic groups. For example, underground work was preserved for Basotho, rock drilling for Xhosas and machine operation underground for Mozambicans. There are continuities and discontinuities in this practice. At Lonmin, for example, nearly all the RDOs and most underground workers are from the Eastern Cape and Lesotho, despite a new policy that gives preference to locals. There are almost no Batswana (the locals) working as RDOs. These divisions have an effect on worker solidarity and mobilisation.
In the hostels, workers were usually divided along ethnic lines and an induna 3 appointed to oversee each group. As a form of control, migrants were separated in hostels on the basis of ethnicity. Continuity of this segmentation is reflected in the new informal settlements emerging across Rustenburg platinum district. Sondela, Nkaneng, Ikemeleng and Number 8 and 9 informal settlements in Rustenburg are predominantly Xhosa, Sotho or Shangaan, while the adjoining communities are more mixed and dominated by Batswana. The workers are thus not only divided at the workplace but even where they live.
The Mining Charter obliges mines to uplift the living conditions in the neighbouring communities. However, the changes in the organisation of work from ‘compounded to fragmented labour’ have created two distinct communities, apparently with different privileges and claims. At Marikana the distinct communities are the traditional villages near the mines and the informal settlements which sprouted after the democratic transition. These communities sometimes have parallel local structures and organise and mobilise differently.
Association of Mineworkers and Construction Union (AMCU)
The NUM had a recognition agreement at Lonmin. This guaranteed absolute rights and shut off competition from rival unions. The dramatic loss of the NUM legitimacy which unfolded during the strike bankrolled the growth of AMCU, a breakaway union. The AMCU's membership escalated from 23,000 in January 2012 to 150,000 by November 2012 (Chinguno 2013). In all the strikes, the dispute mutated into union rivalry and AMCU emerged as the new dominant union. It is important to note that when the dispute emerged initially it was linked not to union rivalry, but rather to what workers claimed were genuine shop-floor problems, yet later developed into this as a manifestation.
AMCU rose from the ashes of NUM, which lost members from defections. It adopted a militant stance and its agenda was alleged to be driven from below. However, AMCU lacks a clear ideological perspective, which poses a challenge for the sustainability of its strategy and hegemony.
AMCU spread to all the major platinum mines and into gold and iron ore, and threatened the dominance of the NUM. In the past, the NUM emerged and spread its wings in a similar way. This suggests that union membership in these mines is somewhat precarious.
Conclusion
The Marikana strike and massacre has brought to the fore the impact of the changing nature of employment in South Africa characterised by fragmentation and precariousness. It has shown how this undermines institutionalisation of industrial conflict and worker organisation, and that it is connected to strike violence. The paper has shown the different forms and dimensions of fragmentation, i.e. work organisation, ethnicity, settlement patterns and gender. This separates workers, undermines union mobilisation and solidarity, and increases the propensity of violence.
Marikana highlights the ambiguous nature of strike violence. Strike violence may mean lack of solidarity on one hand, but conversely may be a means of forging workers' collective ‘solidarity’. It highlights worker fragmentation but at the same time militancy and solidarity. The violence at Marikana underpinned rather than undermined worker collective solidarity.