New Agenda 92: Special Issue – Economics & Labour in the Global South

40 years of shop floor resistance

Racial ordering and history of militant culture

By Sithembiso Bhengu

Dr Bhengu is the Director of the Chris Hani Institute and Senior Research Associate with the Sociology Department, University of Johannesburg. He holds a PhD in Industrial Sociology from the University of KwaZulu-Natal.

Based on four decades of ethnographic research into growing workplace resistance and militancy in the 1980s and 1990s at a Durban rubber factory, SITHEMBISO BHENGU adds his own study of the factory workers – their struggles, identities and everyday lives – more than a decade into democracy. In piecing together their narratives over a span of four decades, he identifies the clear continuities in the objective conditions workers live and work in, as well as in their subjective consciousness, and also the significant discontinuities between the two critical eras of worker formations and struggle in South Africa.

A comment on method

Sitas’ longitudinal ethnography of factory workers in Durban (1984, 1986, 1989, 1996, 1996a, 1997, 2002 and 2004) is an attempt to move beyond the limitations of labour process theories in studying and theorising the resurgence of militant worker formations from the late 1970s in Natal.1 Sitas’ cultural formations approach extended burgeoning labour studies and highlighted new terrains of struggle in South Africa during the 1980s and 1990s. His research was not only concerned with a cultural formations approach, but also with coalescing cultural formations into the broader struggle for freedom and linking workplace and community struggles at the height of the anti-apartheid struggle.

The ethnographic approach also became an active research approach through which cultural associations and social formations used by African workers in everyday life for leisure and de-stressing became avenues for social mobilisation and struggle. Sitas researched and engaged with factory workers around brewing and cultural activities such as indlamu, isicathamiya, plays and poetry as a means for mobilising, political education and struggle.

The majority of militant and vibrant shop stewards in the 1980s and early 1990s took part in plays, sometimes travelling across the country and even abroad to perform theatre depicting workers’ lives under the apartheid work regime.

This article draws on 16 months of critical ethnography in which the author revisited factory workers in the second decade after democracy to examine the social reproduction of the African working class in post-apartheid South Africa.2 It argues first that social reproduction, lives and livelihoods of working-class South Africans are organised, reorganised beyond the rural-urban divide. Wage income remains the most important resource in the production and reproduction of the African working class in post- apartheid South Africa. Second, institutional production regimes continue to be organised through a racialised ordering, a lack of substantive transformation on the shop floors (Bhengu, 2014). These precipitate antagonistic relations between workers and management as well as a militant workforce.

Persisting migrant labour identity

Migrant labour identities constitute a continuing set of relations and meanings in both objective conditions and subjective consciousness of factory workers. The everyday life of workers and their households continue to reach beyond the rural-urban divide. By the 1980s most factory workers were second-generation urban residents, but many had active links with the Natal (Kwa-Zulu) countryside. Revisiting shop floors between 2007 and 2010, the research show that some workers are third and fourth generation urban residents, some with fully functional households in urban townships, yet many still maintain active links with homesteads in the countryside.

While factories no longer use apartheid recruitment patterns in KwaZulu-Natal, the migrant identity amongst workers is still prevalent. By the 1980s Sitas (1989) observed that migrant identities on the shop floor represented both the objective reality as well as the subjective identities workers invoke in making meaning of everyday work life. The migrant identity itself became a constellation of workers’ experiences and the invoking of popular memories and popular histories, which have become a fabric of worker culture and worker consciousness on the factory floor. By invoking migrant discourses, workers mould and reproduce a workplace culture of resistance, militancy, unity and solidarity on the shop floor.

Rural-urban linkages also endure because the countryside continues to be a place of meaning and a place of refuge for workers. While family networks in the countryside seek work in the city and send children to the city for school, etc, workers continue to return to the countryside to perform rituals which constitute social capital, as well as retreat to the countryside when they retire or are unable to work (either because of old age or frailty). Almost all workers and worker leaders interviewed indicated a wish to retreat to the countryside after they retire as a place to finally rest with their ancestors. Sitas (1989) argued that workers invoked notions of inextricable connections to the countryside, even among the second-generation migrants in the 1980s and 1990s. A popular saying in isiZulu is ‘inkaba yami igqitshwe le emakhaya, (“my umbilical cord is deeply rooted in my rural homestead”).

Urban-rural linkages are not one-dimensional; a common assumption of the unilinear direction of urban-rural connections neglects the reciprocal nature of urban-rural networks. While workers send remittances, household networks in the countryside fulfil other functions of social reproduction in African working-class households.

Racialised institutional apparatus of production

Multiple generations of workers from the 1980s to 2010 expressed similar experiences and invoked similar stories of shop floor struggles and intransigence of factory bosses towards workers. The 2007 rubber strike almost resembles in every form the historical 1984 Dunlop strike. In both strikes workers expressed frustration at what they regarded as the intransigence and unwillingness of factory management to engage with workers reasonably. The friction, the sparks, the explosion and the discipline of the strikes were the consequences of a collision between a strong, confident shop steward leadership and the traditional managerialism of a tough, anti-union company.

Associated with these tensions was also a strong feeling by workers of continued racial ordering of industrial relations at the factory. While in 1984 the political apparatus of apartheid was intact and manifest in the structural setup of industries in South Africa, between 2007 and 2010 workers lamented the lack of transformation in managerial layers of the company.

During the 2007 rubber industry strike (and its aftermath) workers still used racial categorisation to differentiate between shop floor and management. During the strike the content of most revolutionary songs and slogans highlighted the workers’ views of the employers and factory bosses.

Many of these songs date back to the 1980s and 1990s during times of resistance to apartheid and white capitalist oppression. In these songs and slogans words like ‘capitalists’, ‘whites’ and ‘employers’ are used interchangeably to refer to factory bosses. The songs declare “Ongxiwa basaba umbimbi, ngoba befuna ukuphatha umhlaba ngabodwa,” (“Capitalists are afraid of the collective alliance because they want to own the land by themselves”); “abelungu bayasicindezela,” (“These whites are oppressing [exploiting] us”); and “umqashi naye kumele abuzwe ubuhlungu balesisiteleka” (“the employer must also feel the discomfort of the strike”) (Bhengu, 2010:208). Furthermore, workers highlighted the lack of transformation at the factory and shop stewards showed company organograms that demonstrated that management was still white and male.3

In one of the meetings during the strike some workers put forward a motion demanding factory bosses address transformation in line with national priorities (employment equity), especially at management levels. In the same discussion, workers adopted a demand for the factory to transform, noting however that employment equity and Black Economic Empowerment are not a panacea for all workers’ problems. One striking worker cautioned that they (the workers and shop stewards) have lobbied for Black (African, Coloured and Indian) workers to be promoted to foremen, supervisors and even middle management. He went on to say: Umuntu sifike simulwele, simxhase, uma elwa nabaqashi ukuthi aphakamiswe. Umaesebekiwe phezulu bese yena futhi asijikijele ngamatshe. Uyababona nje, nalamaqabane esawayisa epalamente, asesijikijela ngamatshe manje sikhuluma nje! (“We as workers fight for people to be promoted, all of a sudden once they have been promoted people throw stones at us, even comrades that we have deployed in parliament, are now throwing stones at us!”)

A racialised workplace ordering results in worker valorisation, precipitating shop floor militancy, animosity towards factory bosses and tenuous industrial relations between workers and management. As indicated above, worker militancy is accentuated by the perceived lack of transformation in management in line with national equity priorities.

The actual construction of shop floor militancy and how it plays out in everyday life on the shop floor is another interesting observation that emerged from engaging with Dunlop workers. Workers construct and mobilise shop floor militancy by invoking what Sitas (1997) calls the popular memory (popular history) of trade unions and through the shop floor socialisation of new and younger workers by their older familial networks.

Generational history of shop floor militancy

Worker militancy is also explained by the experience of persistent super exploitation. In the 1970s and 1980s the factory was regarded as a better paying institution, but workers lamented that the factory still paid the workers too little given the profits and productivity the workers consistently produced. When older workers speak about the span of time from the 1940s to the present, the themes of “thankless toil” – sacrificing years for nothing in return – and managerial harshness predominate. One worker said, “For over thirty years, I have been working for this company … and the only thing I managed to buy over and above the bare necessities was a bicycle” (Sitas, 1984: 32).

However, this too proved to be useless after his move to KwaMashu where the bicycle was damaged by rust and the worker exchanged it for part of the price of a head of cattle. Enlarging on the theme, a worker said: “Dunlop has been a very, very hard firm. The saying goes that if you work there, stay unmarried, for you can’t afford a family … wages, night shift, tiredness … if you are married already, don’t have any more than two children or you will be crying … They feel no pain there … they make more and more profit but don’t pay the worker anything to match it. It annoys us a lot to read about their profits in the Dunlop Gazette …” (Sitas, 1984: 32).

The relationship between profit extraction and income has been a source of great dissatisfaction. As one of the elders commented: “I used to make 120 tyres for them a day, that was about R6,000-R7,000 … You want to know what my wage saw of all that?” (Sitas, 1984: 33) The messages transmitted down the generations to younger workers were simple: they too should expect nothing from Dunlop. It was not the benevolent employer it claimed to be.

Worker militancy represents a constant tussle, what Sitas calls trials of strength between traditionalist, anti-union employers and strongly organised workers. “It is like a Mlabalaba game,” explained a worker, “where the winner takes all. You throw the stone. He blocks it. You throw another one. He blocks, but at the same time he is on the offensive … You block … The difference is that there is no time for the struggle to end. In each minute that goes by, you both lose something. In the beginning you lose a finger, by the second week you are a cripple. The same with him. He loses money, profits. So, one of you will have to stop and say okay” (Sitas, 1984: 34). Trials of strength are costly for all concerned: “They are in many respects,” asserted Richard Hyman, “the industrial equivalent of war between nations” (Sitas, 1984:34).

During the 2007 strikes, older workers narrated how strikes constituted trials of strength and how strike victories were uncertain, short lived and always contested. They related that in the aftermath of the groundbreaking 1984 strike, the victory of trade union recognition and wage agreement was followed by a plethora of disciplinary actions and dismissals of workers by the factory.4

After the workers’ victory in the 2007 strike, factory bosses began to contest the resolutions on labour brokers as well as attempt to force a new shift pattern, which workers had successfully rejected and resisted at the end of 2003. Worker leaders also suspect that management attempted to manipulate the 2007 November shop steward elections to remove shop stewards that management viewed as a problem and replace them with new management-friendly shop stewards. The plot backfired and workers ordered the regional executive of the National Union of Metalworkers of South Africa (Numsa) to administer another election, which resulted in the return of all the previous militant shop stewards and the replacement of two of them with older militant workers who had been part of the shop stewards committee during the 1980s.

Older workers also spoke of the contradictory possibilities and simultaneous challenges of their struggles. In discussions with shop stewards in 2008, they acknowledged the victory of democracy – the gains for workers because of legislative instruments such as the Labour Relations Act of 1995, the Basic Conditions of Employment Act of 1997, the Employment Equity Act of 1988 and many others – but they also mentioned that at some point between 1984 and 1992 many victories won on the factory floor through struggle were regulated and, in some cases, reversed by the new legislation.

Worker retention at the rubber factory also contributed to the strength of workers and trade union on the shop floor. Figure 4 show that the majority of workers had been working in the factory for more than 10 years. Almost half of the sampled workers had been working at the factory for more than 20 years and more than 10% worked there for more than 30 years. Two of the inaugural Metal and Allied Workers Union (Mawu) shop steward committee members were still working at the factory during the course of this research, one of whom was re-elected onto the shop steward committee in 2008. The vast majority of the members of the shop stewards committee of 2007/2008 had been working at Dunlop since the late 1970s or early 1980s. Many of them had been shop stewards before.

In conversation, Dunlop shop stewards highlighted the crisis in many trade unions in the region. They raised concerns about the lack of training of shop stewards on basic shop steward functions like representing a worker in a disciplinary hearing or declaring a dispute with management. They highlighted that Dunlop workers are lucky because shop stewards and workers train each other on these matters. The chairperson even acknowledged that he was only in his second stint as a shop steward, but during his first stint (2003 to 2006) older and more experienced shop stewards helped him and other younger shop stewards to understand what it meant to be a shop steward and how to deal with management.

Sitas argues that the continuity of service amongst older workers at Dunlop ensured the survival of a mercurial factory tradition: partly a mechanism of adjustment to mass production, partly a lever of discontent. This factory culture, handed down over the years from older to younger recruits, influenced both behaviour and expectations from management. One such lineage of factory socialisation runs as follows: Mr Melebane, now retired, was employed in 1944 in the fabrics department and became the reference point and adviser of Mr Kambule who was employed in 1946. In 1954, Mr Kambule socialised Mr Banda in the “Dunlopillo” department; he, in turn did the same to Mr Bhaca in the light-truck department, and so on (Sitas, 1984: 28).

In the factory there was not so much an informal organisation of workers but a crisscrossing of reference groups, leaderships and brotherhoods made up of a multiplicity of defensive combinations of workers and a repository of useful public knowledge (Sitas, 1984: 29). It demarcated an ‘us’, which belonged together, and a ‘them’. Furthermore, within the Black labour force itself, it also demarcated an “us” against ‘them’, the “impimpis” (police informers/ traitors) (Sitas, 1984: 30) This contest was violent; over the years many such disputes were settled in the factory, in the locations or in the bus queues.

In 2007, workers still used generational and familial networks to socialise younger workers into shop floor militancy. In both the 2003-2007 as well as the 2007-2011 shop steward committees, the majority of the 12-member committee had family members of different generations working with them at the factory. Bomshel Khumalo came to the factory through an elder relative in 1976 and his son joined the factory in 2004. Buthelezi heard from the hostel that the factory was recruiting and was employed in 1982 and in 2006 his son joined through labour broker workers but became a full-time employee after the strike. Mthethwa joined the factory in 1988 after being told about the job by his uncle (who still works there). In 2007 his nephew joined the factory a month before the strike. Sompisi joined the factory in 1975 because he knew workers from his hometown of Flagstaff working there. He had a brother who worked there, although he was retrenched in 1993 and has a nephew working at the factory.

The workers utilise these familial and generational networks and connections to build a militant work culture on the shop floor. In the 2007 rubber strike, almost all labour broker workers (new recruits) joined the strike and attended strike meetings consistently. Permanent workers devised a strategy to include new recruits in the strike by holding a regular open gathering session in the canteen at 8:30am. Workers would open the meeting with a prayer, followed by a revolutionary song and marching outside the factory gates, which new recruits joined, before marching back inside the factory floor to discuss the issues on the agenda.

Workers attribute the high continuity of employment at the factory to the fear of the alternatives: “You stay at the factory because it is slightly better with the wage than most of the others. This doesn’t mean that it pays well” (Sitas, 1984: 27). Workers stated unequivocally that given the profits they made for the company, their wages were minimal.

What was remarkable in the interviews with workers was the low level of trust between workers and management. Given the lengthy service records of many of the employees this “low trust” has a historical and contemporary dimension that has cemented over time into a tradition of mistrust and a grumbling acquiescence to managerial authority. Both memory and contemporary experience are important in explaining why a degree of polarisation and trial strength of such proportion could develop on the shop floor.

Concluding remarks

Workers at Dunlop have reproduced their shop floor militancy by invoking histories of workers’ struggles and through the socialisation of younger workers into the militant shop floor culture by older workers. Workers have appropriated generational familial recruitment to reproduce shop floor militancy. Older working men socialise their sons and nephews drawing on the memory of their struggles and victories.

The research also showed that persistent racialisation on the shop floor at Dunlop contributes to the anti-employer, anti-management and anti-discrimination attitude among workers. Persisting racial ordering and the experiences of untransformed management reproduces antagonistic industrial relations between management and workers on the shop floor.

Finally, African and migrant workers make meaning of their lives on the shop floor though developing solidarity in response to oppressive work regimes and culture, dance, songs, sports, etc. became everyday expressions through which workers constructed their lives on the shop floor (Bonnin, 1987; Moodie, 1994; Sitas, 1996a, 1997; Von Holdt, 2003).


Bhengu, S. 2010. Workplace regimes, identity, and everyday life of African working men in post- apartheid South Africa – the case study of Dunlop, Durban in L. Heinecken & H. Prozesky (eds.), Society in Focus – Change, Challenge and Resistance: reflections from South Africa and beyond. Cambridge: Cambridge Scholars Publishing.

Bhengu, S. 2014. Workers, Wages and Livelihoods – a case of Dunlop factory, PhD Thesis, Durban: University of KwaZulu-Natal.

Bonnin, D. 1987. Class Consciousness and Conflict in the Natal Midlands: the Case Study of BTR Sarmcol Workers, MA Thesis, Durban: University of Natal.

Burawoy, M. 1979. Manufacturing Consent: Changes in the Labor Process Under Monopoly Capitalism. Chicago: University of Chicago Press.

Burawoy, M. 1991. Ethnography Unbound: Power and Resistance in Modern Metropolis. Berkeley CA: University of California Press.

Burawoy, M. 2000. Global Ethnography: Forces, Connections and Imaginations in a Post-Modern World. Berkeley CA: University of California Press.

Hart, G. 2006. Denaturalizing Dispossession: critical ethnography in the age of resurgent imperialism, Antipode, 38 (5), pp.977-1004.

Hyman, R. (ed.), 1989. Strikes. (4th ed.), Basingstoke: MacMillan.

Moodie, D. 1994. Going for Gold: Men, Mines and Migration. Johannesburg, Berkeley CA: Wits University Press, UCLA Press.

Nzimande, B. 1991. The Corporate Guerrillas – class formation and the corporate petty bourgeoisie in South Africa, PhD Thesis, Durban: University of Natal.

Sitas, A. 1984. Accommodation and Resistance in a Large Mass-Producing Factory, 1940s – 1980s, Durban: University of Natal.

Sitas, A. 1985. Where Wealth and Power and Blood Reign worshipped Gods, South African Labour Bulletin, 11 (4), pp.85-121.

Sitas, A. 1989. Class, Nation and Ethnicity in Natal’s Labour Movement, Institute of Commonwealth Studies: The Societies of Southern Africa, 15 (38), pp.267-78.

Sitas, A. 1996. The Sweat was Black: working for Dunlop in P. Maylam & I. Edwards (eds.), The People’s City: African Life in the Twentieth Century Durban, Pietermaritzburg: University of Natal Press, pp.222–42.

Sitas, A. 1996a. The New Tribalism: hostels and violence, Journal of Southern African Studies, 22 (2), pp.235-48.

Sitas, A. 1997. Neither Gold nor Bile: industrial and labour studies of socio-economic transformation and cultural formations in KwaZulu-Natal, South Africa’, African Sociological Review, 1 (1), pp.99- 111.

Sitas, A. 2002. Autobiography of a Movement: Trade Unions in KwaZulu-Natal 1970–1990 in R. Kriger & A. Zegeye (eds.), Culture in a new South Africa, Cape Town: Kwela Books, pp.237–59.

Sitas, A. 2004. Thirty Years Since the Durban Strikes: Black Working-class Leadership and the South African Transition, Current Sociology, 52 (5), pp.830-49.

Sitas, A. 2004a. Voices That Reason: theoretical parables. Pretoria: University of South Africa Press.

Turner, R. 1972. The eye of the needle: towards participatory democracy in South Africa. Johannesburg: Ravan Press.

Von Holdt, K. 2003. Transition from Below: forging trade unionism and workplace change in South Africa. Scottsville: University of Natal Press.

Webster, E. & Von Holdt, K. 2005. Work Restructuring and the Crisis of Social Reproduction: a southern perspective, in E. Webster & K. Von Holdt (eds.), Beyond the Apartheid Workplace: studies in transition. Scottsville: UKZN Press.


  1. Until the ground-breaking writing of Turner (1972), literature in labour history, leftwing historians and Marxist sociologists did not believe that African workers could become a militant social formation in South Africa because of the nature of the repressive state apparatus of apartheid and the banning of progressive organisations in the 1960s. In the aftermath of the 1973 Durban strikes, Marxist-leaning industrial sociologists attempted to explain the explosion of militant African trade unions using the labour process theory. Sitas (1984, 1989, 2004), as well as Bonnin (1987), Moodie (1994) and Nzimande (1991) made a significant contribution to labour studies by introducing a cultural formations approach.
  2. Bhengu’s PhD thesis incorporates Burawoy’s (1991, 2000) extended case study method as well as Hart’s (2006) critical ethnography, what she calls relational comparison. Bhengu visited the factory floor for over 16 months, having conversations, administering questionnaires and conducting interviews with workers and shop stewards, as well as visiting several rural households of the workers (and worker leaders). In his research, workplace and factory regimes represent sets of relations that reach beyond the boundaries of the shop floor. Bhengu critiques a conception of workers’ lives in South African macro-economic literature that presents rural- urban, workplace-household, employed-unemployed and first-second economy as binaries. Instead he argues that the everyday life of African workers represents a complex set of relations and workers construct complex identities and meanings that play out on the shop floor and in their everyday life experiences as workers, comrades, migrants, husbands and men.
  3. The international holding company that owns the factory established a new board in 2011 for its African operations and appointed to the board two local African women as non-executive members, a South African Indian man as the general manager of the Durban factory and a Brazilian CEO (and President of African Operations).
  4. Strikes as trials of strength are depicted in worker struggles in the rubber factory throughout the 40-year continuum. While workers would celebrate strike victories, management would respond by instituting disciplinary proceedings against militant workers. Workers though intimated that these reactive measures by employers, especially between 1985 and 1992, were always challenged by the strength of the workers and the trade union militancy and unity on shop floor.

How to Cite

Bhengu, S. (2024). 40 years of shop floor resistance: Racial ordering and history of militant culture. New Agenda: South African Journal of Social and Economic Policy92(SI).

Leave a Reply

Your email address will not be published. Required fields are marked *