Printer-friendly versionSend by emailPDF version

South African unions are large but fragmented, substantial but politically weak. They represent different political traditions and all are marked by serious organisational problems. They have little impact on the official public sphere. The unions need to work towards realizing a stateless, classless, self-managed society without hierarchy, based on political pluralism and freedom.

Among South Africa’s distinctive features is the largest union movement in Africa (as a proportion of the waged population), itself a reflection of another distinction, the country’s largely proletarian population and capitalist order. However, this movement is also deeply fractured, and at this stage, crippled.  It has contradictory tendencies, towards unity and division, towards revolution and reform.  

This paper provides a critical survey of South African labour, considering all the main forces. It starts from the premise that unions remain here, as elsewhere, indispensable forces for deep social transformation. There is no substitute for mobilisation on the labour front, given the structural power, numbers and resources that unions can bring to bear in workers’ struggles, and in the struggles of the oppressed more generally. Unions also play a key role as reservoirs of working class identity, and in keeping socialist ideas and spaces alive. With the potential to unite oppressed and exploited people across ethnic, gender, linguistic and other divisions, unions are also an essential foundation for internationalism and universalism. Unions have a demonstrated capacity to develop workers’ capacity to take control of production and the larger economy, as shown, for example, in the 1936 Spanish Revolution.

That said, unions in South Africa, as elsewhere, often fail to meet their potential. Unions are not inherently capable of making radical change. They have tendencies towards economism, conservatism, reformism, sectionalism and bureaucratisation, which have to be resisted. The routines of bargaining, and entanglement in state-run negotiations and processes, and in political party patronage systems, all dampen union power and workers’ control. Union leaderships often emerge that are positioned as a conservative layer of brokers between workers and bosses. Divisions in the working class and among the oppressed are often carried into the labour movement. This is certainly the case in South Africa, as discussed below.

Without a clear project for social change that rests upon the development of popular capacities and organisation, self-activity, ideological counter-hegemony and the construction of a class-based counter-power able to resist, then displace, the ruling class – including through workplace occupations by the unions – unions cannot fundamentally change society.

Size and scope

There were around 3.7 million union members in South Africa in 2015, out of a formal sector workforce of slightly over 10.8 million, and an informal sector workforce of slightly over 2.6 million.[1] Union membership is almost always voluntary (closed shops are extremely rare). This gives a union density of 27.6% (both sectors), or 34.2% (formal sector only).

Union size is remarkable, given the country’s frightening unemployment (5.2 million – or 7.6 million using an expanded definition), [2] massive processes of casualisation, outsourcing and mechanisation, low economic growth, aggressive employers and the impact of the world capitalist crisis of 2007 onwards.

It is facilitated by a Constitution (1996) entrenching union rights. The 1995 Labour Relations Act (LRA) provides union rights for all workers, and the 1999 General Regulations of the South African National Defence Force even permits military unions.

Ultimately, however, union strength rests on long-standing working class traditions, a politicised society, and material benefits. Unionists earn 6-7% more than non-union workers, rising to 16% in the private sector, and 22% in the state sector, when statutory bargaining councils are also involved.[3]

Left traditions

Unions have never been state-run, unlike in many neighbouring countries. There is, however, a long history of union links to political parties, notable examples being close links between white-only unions and the South African Labour Party (founded 1909), and the alliance between the South African Congress of Trade Unions (SACTU, formed 1955) and the African National Congress (ANC, founded 1912).

Also worth noting is the deep imprint of socialist politics on the two of the most important union formations, with few parallels in the region. The Congress of South African Trade Unions (COSATU, formed 1985, claiming around 1.9 million members in November 2015),[4] and the National Union of Metalworkers of South Africa (NUMSA, formed 1987, claiming 340,000 members in December 2016) [5] both describe themselves as “Marxist-Leninist,” i.e. orthodox Communist.

COSATU is formally allied to the now-ruling ANC, as well as the South African Communist Party (SACP, founded 1921). COSATU’s Marxist-Leninist claims and SACP ties help explain why it hosted the 2016 World Federation of Trade Unions (WFTU) congress in South Africa. Here, the president of COSATU’s 250,000-strong National Education Health and Allied Workers Union (NEHAWU) was elected WFTU president.  COSATU remains one of the most important working class formations in Africa, with a long history of struggle, including mass strikes post-1994.

The smaller National Council of Trade Unions (NACTU, formed 1986) has at times also declared adherence to socialism. NACTU’s numbers are disputed – the long-claimed 370,000 members is questionable – but were recently boosted by the growth of NACTU’s Association of Mineworkers and Construction Union (AMCU). Starting as a 1998 splinter from COSATU’s National Union of Mineworkers (NUM), AMCU now claims 200,000 members.[6] It played a key role in the 2012-2014 strike wave, indelibly associated with the 2012 massacre of strikers at Marikana.

Low-wage Black labour

However, these points all need to be qualified. Agriculture remains almost completely non-union. Most South African workers are not union members. Casual workers are rarely unionised. Many union members are inactive. Over 400,000 workers do not know whether they are union members. [7]

Many unions are small and weak. Out of 484 unions, only 184 were “registered” in 2016, enabling access to the statutory bargaining system.[8] Branches of larger unions often lack effective bargaining rights. Big unions have often failed casual workers, who have then relied on splinter unions or on ad-hoc strike committees. Only “23% of workers’ wages are determined through collective bargaining with only 9% determined through centralised bargaining.”[9]

Real wages stagnated between 1997 and 2013; some sectors like construction and mining have shown important gains, but median real wages have declined in agriculture and manufacturing.[10] Monthly median earnings in 2013 were low: R2, 200 in agriculture, R2, 700 in construction and business services, R3, 400 in manufacturing and R4, 650 in mining.[11] Most wage settlements are very modest in real terms, around 2%. [12] In contrast, CEOs’ average monthly remuneration was R1, 279, 976 in 2014 (excluding long-term incentives), and “executive directors took home more in a month than … a low wage worker would earn in 94 years.”[13]

These reproduce the racist cheap labour system that has long marked capitalism in South Africa.

Confining laws

The core of industrial relations law dates to 1924-1926, and is designed to limit strikes, contain unions and entangle their negotiators.[14] The law has been deracialised since 1979, and expanded to almost all sectors with the 1995 LRA – a major victory – but the basic system remains. Strikes (outside essential services) are only “protected”[15] after a lengthy procedure for conciliation has failed, and then only after a 30-day period plus 48 hours’ notice. Arbitration is prescribed in essential services (affecting life, health and safety). Solidarity and political strikes are only “protected” after a lengthy process, and 14 days’ notice.

Thus the system promotes a centralised union structure reliant on paid leaders tied up in negotiations. This is reinforced by the corporatism of the National Economic Development and Labour Council (NEDLAC), where unions have input into state policy. When these developments are coupled to union links to political parties engaged in the state, and the patronage and outside interference such parties bring, the scope for union autonomy and workers’ control is further reduced.

The law has scope for extending bargaining council agreements to workers outside of negotiations, and for sectoral minimum wage determinations by the Labour Minister. While these measures can benefit vulnerable workers, they often substitute for shop-floor organising. Minimum wages set by the Ministry are generally appalling.

Individuals can lodge cases fairly easily, but the system is slow and backlogged; employer compliance is difficult to enforce.[16] Until recently, individuals could only be represented by unions or lawyers – this disadvantaged the majority, with no unions and no money.[17]

Left or centre?

The union movement should not be reduced to COSATU, which only represents half the unionised workers, nor should union radicalism be exaggerated.

NACTU’s socialist commitment, for example, is vague, and the federation has serious problems. In the early 2010s, it closed its regional and provincial offices due to financial and leadership problems.[17]

Linked to Pan-Africanist / Black Consciousness racial nationalisms in the 1980s, NACTU has rarely been political since 1994.  From 2005, NACTU was involved in merger negotiations with the centre-right Federation of Unions of South Africa (FEDUSA, formed 1997, claiming 540,000 members) and the Confederation of South African Workers’ Unions (CONSAWU, formed 2003, claiming 290,000 members).[19] This led to NACTU and FEDUSA forming a loose South African Confederation of Trade Unions (SACOTU) in 2007. SACOTU still exists, but no merger has taken place. The sustained liaison with FEDUSA says something about NACTU’s politics.

In 2011, NACTU’s important Metal and Electrical Workers’ Union of South Africa (MEWUSA) purged members of the Democratic Socialist Movement (DSM). DSM militants had played a central role in revitalising the union, but challenged entrenched leaders. (The DSM later helped launch a Workers’ and Socialist Party in 2012, and a Socialist Trade Union Network).

NACTU’s AMCU is a fighting union, notable for demanding a R12, 500 minimum wage. However, AMCU’s focus is explicitly on bread-and-butter issues: its outlook is militant economism. It tends to dismiss socialism and politics because of their association with its hated rival, NUM.

AMCU represents a majority of black miners in platinum, but is a minority in coal and gold. No more than half of AMCU’s members are in mining, the only sector in which it has demonstrated an ability to win strikes. It has no strongholds elsewhere. In 2013 and 2016, AMCU has settled for modest wage increases in platinum. Although substantial and hard-won, these were well below the 50% increase required to reach R12, 500. [20]

The moderates, the right

FEDUSA rejects “ideological and party-political ties”: it takes as its “real mandate” “workplace matters and issues that affect its members.”[21] Its roots lie in the conservative tradition of the defunct Trade Union Council of South Africa (TUCSA, formed 1954). Centred on Coloured, Indian and white workers, TUCSA was side-lined by the new unionism that led to COSATU, NACTU and NUMSA. FEDUSA today is a majority-black federation, but it remains economistic, with a narrow focus on workplace issues. It projects itself as a “stable” federation that advances “the interests of employees and of the economy” in a “responsible manner,” [22] and avoids strikes.

Historically, FEDUSA represented more skilled, semi-professional and professional workers, some with roots in old craft and staff unions. FEDUSA’s South African Typographical Union dates to 1898. FEDUSA’s growth centres on attracting existing unions to affiliate, not mass organising. It has been hampered by disaffiliation of large unions like the 70,000-strong Independent Municipal and Allied Workers’ Union (IMATU) in 2012. 

CONSAWU has similar roots to FEDUSA, and also includes a range of union types: general, industrial, craft and professional – COSATU and NACTU instead stress industrial unionism.

However, almost half of CONSAWU’s members (130,000) come from one affiliate, Solidarity, itself a union centre. Solidarity’s roots lie in white, segregationist, pro-apartheid unions going back to the Transvaal Miners Association (1902) and its successor, the far-right SA Mineworkers’ Union.

Renamed Solidarity in 2002, it now identifies as an inclusive Christian-democratic union. It is, however, firmly located in a strand in Afrikaner nationalism that positions itself as a language and rights lobby in the new South Africa. Solidarity has a record of cooperation with unions like NUM and NUMSA; it even attended the 2003 SACP congress. However, its conservative economic views and opposition to affirmative action have helped entrench racial divisions in labour.[23]

CONSAWU, FEDUSA, NACTU and Solidarity are all reformist unions: their aims do not go beyond modest changes in the existing order. They simultaneously defend and divide the working class: products of class contradictions, they do not have a programme to remove class exploitation. 

Together, these reformist unions represent a third of unionised workers.

The Left unions: COSATU’S Marxism

Solidarity, unlike FEDUSA or the rest of CONSAWU, is a politicised union, like COSATU. But unlike COSATU, it is on the right.

Both COSATU and NUMSA define themselves as revolutionary and socialist.

COSATU’s links to the WFTU illustrates some of its politics’ limitations. The WFTU is committed to socialism and internationalism, but in the Marxist-Leninist tradition. Obviously, many WFTU members and affiliates are seriously committed to profound, revolutionary change. However, its political tradition identifies “socialism” with the largely-collapsed Marxist dictatorships of the old East bloc.

The WFTU was closely tied to the then-Soviet Union, a labour-repressive state that did not recognise basic human and union rights, and which operated an informal and formal empire. This link is why current WFTU affiliates include fake state-run “unions” in Cuba, North Korea and Vietnam, countries marked by bans on strikes and independent unions, by low wages, heavy repression and a complete absence of any working class or peasant power.

COSATU’S social democratic problem

COSATU is actually to the right of many WFTU affiliates. Its formal Marxism-Leninism co-exists with a strategy that can only be described as social-democracy with a large dash of nationalism. Since the 1990s, both COSATU and the SACP have sought change through expanding corporatism, co-determination, union investments, co-operatives, etc., and by intervening in state policy through NEDLAC and the ANC.[24] This is sometimes given a radical gloss with slogans like “Socialism is the future! Build it now!”

COSATU is completely correct that South Africa’s burning national and social questions cannot be resolved within a capitalist framework. COSATU is correct in identifying the ongoing power of “white monopoly capital” as a major obstacle to justice for the black working class majority (the term refers to the giant, white-run conglomerates that emerged under segregation and apartheid). COSATU is also correct to draw attention to the devastation that neo-liberal policies and restructuring have wrought on South Africa over the last decades.

But COSATU’s strategy is not a revolutionary one. It does not involve a decisive assault on the citadels of capitalism, smashing the state, or the mass working class mobilisation, not to march, but to take power and control production. Instead, it centres on the incremental accumulation of anti-capitalist gains through a series of reforms, supposedly shifting the balance of forces towards socialism.

In reality, none of the measures COSATU proposes abolishes capitalist relations or undermines the capitalist state. For example, union investments do not provide workers effective control over means of production: they fund capitalist companies based on exploiting wage labour and selling commodities.

Posing the problem as “monopoly capitalism” or “apartheid-capitalism,” rather than capitalism as such, suggests the answer is a better variant of capitalism – not socialism.

What COSATU proposes is, actually, a variant of the Keynesian welfare state. However, the Keynesian welfare state is in its death-throes everywhere, and, as I have argued elsewhere,[25] is an extraordinarily improbable future for South Africa. COSATU strategy and aspirations are trapped in a dead-end.

COSATU and its affiliates have achieved little through corporatism and efforts at shaping economic and industrial policy, besides bureaucratising the unions.[26]

Blind to the state

The focus on white monopoly capital also blinds COSATU to major changes, like the ongoing denationalisation of the economy, and the economic role of the state. The state is responsible for almost a quarter of the entire Gross Domestic Product (GDP), is the largest employer and landowner, and operates its own massive corporations.

The state is not an organisation that can be wielded by the working class. Hierarchical, centralised, an organ of minority rule that is used by political elites to accumulate power and wealth, it is structurally allied to private capitalists and landlords.

The state, itself, is never criticised by COSATU, however – only specific policies, politicians, personalities and parties. While COSATU condemns the way that Western imperialism imposes itself on South Africa, it is almost completely silent on the role that the South African state plays as mid-level imperialist power in Africa.

This sort of reasoning underlies COSATU’s alliance with the ANC, a multi-class nationalist party openly committed to neo-liberalism. COSATU doesn’t deny that corrupt politicians and powerful capitalists wield extensive influence in the ANC, but it treats this problem as primarily organisational and subjective i.e. it claims “working class bias” can be secured in the ANC through union and SACP interventions, more “working class” / “progressive” representation, and a renegotiation of the terms of the ANC/ COSATU / SACP alliance.

This does not recognise that the South African state is irrevocably a capitalist state, that the ANC is part of that capitalist state, and that the ANC-dominated political elite is allied to the private economic elite by common interests. Or, furthermore, that the South African capitalist state is deeply embedded in a global capitalist political economy, in which South African imperialism takes place largely through private and state capitalism, “soft power,” and multi-lateral structures. These objective conditions cannot be removed by fiddling around in the ANC.


Among the costs of COSATU’s statism is that it has no real project of preparing the workers to occupy, and self-manage, the workplaces. COSATU’s programme has the opposite: joint work with the bosses through co-determination and corporatism, and joint work with the politicians through elections and the ANC/SACP/COSATU alliance.

COSATU presents the ANC as a force of the left, and the heart of a “revolution” under fire – rather than a key enforcer of capitalist power. This logic led the NUM and SACP to denounce AMCU as a counter-revolutionary force, backed by “imperialism,” and to COSATU’s suspension and expulsion of NUMSA for rejecting the alliance with the ANC and SACP at a special congress in 2013.

This politics of labelling, with its potential for violence, is very dangerous; its flipside is intolerance to non-SACP leftists, a Stalinist-style politics.

“Revolution,” in COSATU and SACP terms, actually means “national-democratic” revolution (NDR) by a multi-class nationalist “people’s movement” against apartheid and its legacy – a “stage” towards socialism, by radically reforming capitalism.

But it is extremely unlikely that capitalism can be reformed to undertake “national” and “democratic” reforms on the scale needed to uproot the apartheid legacy. COSATU admits this, yet promotes a capitalist stage. Even if every single capitalist was black, this would remain true: the role of black capitalists like ANC leader Cyril Ramaphosa in the Marikana massacre shows they share their white counterparts’ reliance on cheap black labour. Blacks do not have common interests or experiences, as NDR theory suggests.

There is also no reason to suppose “national-democratic” reforms must lead to socialism: the ANC, supposed bearer of NDR, has proved a solid bulwark against further change. The ANC’s nationalist ideology – stressing national unity across class lines – is itself a major obstacle to class consciousness, tying labour to capital. The focus on using state power to achieve change – statism – helps entrench illusions in the capitalist state.

The left unions: where is NUMSA’s moment?

The radical NUMSA remains the wildcard in South African unionism; it is the key force, at this stage, which can shift unions onto a revolutionary track. At its 2013 congress, NUMSA called for a “movement for socialism” and a “united front against neo-liberalism,” and correctly noted that Marikana demonstrated that the ANC was irredeemable. (It also indicated its intention to expand beyond its base in metals / engineering).

To its great credit, NUMSA has taken bold steps. Of course, it takes a great deal of time for a union – especially one like NUMSA, which is quite democratic – to shift gears. There are no short-cuts to building an emancipatory movement.

NUMSA’s achievements to date are uneven. The United Front (UF) was launched in 2014: it has attracted a number of community-based and left formations, and significant goodwill. However, NUMSA’s relationship to the UF is not clearly defined, and it has not brought the union’s full power to bear: there are no strikes backing UF actions. Local NUMSA structures have played a key role in building the UF in several places, but the UF is under-resourced, weak and stagnating. UF candidates in the 2016 local elections fared very poorly.

NUMSA’s leaders clearly envisage the “movement for socialism” as a new Communist Party, but this has yet to be formed. But the space for a radical party outside the ANC has, meanwhile, been taken by the populist Economic Freedom Fighters, an ANC breakaway (formed 2013).  Elsewhere, Pan-Africanism / Black Consciousness have expanded as the frustrations of the black middle class and unemployed poor rise in a stagnant, unequal, racialised economy. These traditions reproduce the flaws of nationalism and statism, but NUMSA has not managed to contest their impact.

NUMSA struggles to project its voice beyond its traditional base. Initially, NUMSA planned to challenge its unconstitutional suspension from COSATU, insisting on the right to make its case at a special COSATU conference, where it would undoubtedly enjoy great sympathy. In 2015, this was replaced by a debatable dismissal of COSATU as “dead,” and plans for a new federation.

A successful summit with other unions in April-May 2016 was meant to lead to a new centre this year. However, the need to create a large centre is at odds with NUMSA’s aim of a Marxist-Leninist union movement, given that few of the unions engaging NUMSA are Marxist-Leninist. Key unions at the summit included AMCU and Solidarity. NUMSA has meanwhile cut itself off from the COSATU mass base.

NUMSA runs the risk of over-extending itself with such a wide range of projects, and these in addition to its ordinary union work.

Which movement, which socialism?

NUMSA has investigated options for socialism in a range of processes. Again, this is to be commended. However, there is little attention to issues like the self-management of production, or the insights that can be gleaned from experiences such as the Paris Commune (1871) or the revolution in Spain (1936), or from other left traditions, such as anarchism and syndicalism.

The possibilities for a shift from Marxism-Leninism, steeped in SACP traditions, are narrow. For many in NUMSA, the break with the SACP is organisational, not political i.e. the SACP is rejected for capitulating to the Zuma-led ANC; the problem is posed as betrayal or corruption - rather than as a failed model. So the solution becomes a return to classical SACP politics. Thus, NUMSA’s 2016 congress resolved that NDR remains “the most direct route to socialism,” and that the burning task is “to actively build a Marxist-Leninist Vanguard Revolutionary Workers’ Party.”[27]

Marxism-Leninism is only one strand in socialism, understood as an expression of working class interests and commitment to common ownership – and it is a strand that has serious flaws. But in NUMSA, it is often presented as the only true socialism.

The political party mirage

The long history of political parties betraying unions – from the ANC in South Africa, to the Democrats in the USA, to the Communists in the Soviet Union – needs to be engaged more seriously.

The problem with COSATU’s alliance with the ANC is not just that the ANC is a bad party. It is the reality that alliances with parties involved in the state subordinate the unions, entangle them in the state apparatus, and corrode them with patronage and corruption. The NDR and social democratic approaches are, as argued above, inherently flawed. But not even a perfect programme can survive the embrace of the state. No state can act “for” the workers, since all states are, by their nature, anti-working class. And every union will be damaged by immersion in the state and its parties.

As COSATU’s experience showed, politicians buy-off leaders and entangle unions in factional battles. In the late 2000s, COSATU (including NUMSA) threw its weight behind the Jacob Zuma faction in ANC (purging unionists who disagreed). Today, COSATU (including NEHAWU) is knee-deep in the inner-ANC battles, completely useless to the working class, over Zuma’s successor.

Unions can be political without parties, and parties do immense harm to unions.

Towards a brighter day

In closing, South African unions are large but fragmented, substantial but politically weak. They represent different political traditions, and all are marked, to a greater or lesser degree, by serious organisational problems. They have little impact on the official public sphere, where the media is dominated by middle class and elite voices, black and white, and have made few strides to build a proletarian public sphere and mass media. This all helps to explain why there has not been a single general strike against job losses and austerity in the ten years since the onset of the world capitalist crisis, and why there is no union mass media.

FEDUSA and CONSAWU come from a bureaucratic tradition, based on a low level of mobilisation and a passive membership. COSATU unions were founded on the principle of workers’ control, with shop-stewards kept accountable to assemblies, but this has been significantly eroded since 1990. This has run alongside a decline in union life, debates and education. The bureaucratic and authoritarian methods used to drive out NUMSA (and in an earlier period, anti-Zuma dissidents) indicate the problems. NUM’s splits, and NACTU’s decline, are the result of a similar bureaucratisation. NUMSA has fared much better, but is not free of these problems. AMCU, meanwhile, has struggled to consolidate democratic structures.

Simply setting up new unions does not resolve these ideological and organisational problems, as the problems are easily reproduced in new unions. For unions to move forward, it is important to have a lively internal culture and political pluralism. Intolerance within unions, and between unions— a notorious example is violence between AMCU and NUM – is a serious problem. Solidarity is needed: sectarianism pitting ordinary union members against each other undermines the working class.

Instead of failed statist solutions – nationalism,  social democracy, Marxism-Leninism – unions need to work towards the democratic elaboration of a libertarian and left project centred on self-activity. A socialism that does not just oppose capitalism, but aims at a stateless, classless, self-managed society without hierarchy, based on political pluralism and freedom.

Seeds of the new      

Contrary to the pessimistic dismissal of unions as inherently reformist, or as captured by capitalism, they can provide a key space for the development and experience of bottom-up democracy, workers’ education and the “new faith” of libertarian socialism, and “solidarity and fraternal sentiment” between the “workers in all occupations in all lands.”[28] They can provide, through democratic structures and mass mobilisation, both a lever for large-scale workplace takeovers and a core part of the administrative infrastructure for a self-managed society, forming “the living seeds of the new society which is to replace the old world.”[29]

These organisational and political shifts require a reform of the existing unions, not abandoning these key organisations that the working class has already built. It requires building the unions, rather than narrow loyalty to any union brand. And it requires serious reflection on the deep flaws in models centred on building political parties, trying to capture the state: the ANC and SACP failed to free the working class, because their leaders were incorporated into the capitalist state, and their leaders were incorporated because they followed a statist politics that led, inevitably, to that result. That is the fate of party politics.

* Lucien van der Walt ([email protected]) works at Rhodes University, South Africa. He has published widely on labour and left history and theory, and political economy, and on anarchism and syndicalism. He is actively involved in union and working class education and movements. Notable works include "Negro Vermelho: Anarquismo, Sindicalismo Revolucionario e Pessoas de Cor na Africa Meridional nas Decadas de 1880 a 1920" (2014), and "Anarchism and Syndicalism in the Colonial and Postcolonial World, 1880-1940" (2010/2014, with Steve Hirsch). He was southern Africa editor for "The International Encyclopedia of Revolution and Protest" (2009). His Ph.D. on black and white radicals, "Anarchism and Syndicalism in South Africa, 1904-1921," won both the Labor History and the Council for the Development of Social Science Research in Africa (CODESRIA) theses prizes.

End notes

[1] Quarterly Labour Force Survey, Quarter 2, 29 July 2015, pp. iv, xvii, xi.

[2] As previous.

[3] C. van der Westhuizen, H. Bhorat & S. Goga, 28 May 2013, “How Much Do Unions and Bargaining Councils Elevate Wages?,” Econ3x3,

[4] G. Quintal, 17 November 2015, “COSATU Membership Down by Thousands,” News24,

[5] “NUMSA 10th National Congress, 12-15 December 2016, Cape Town: Congress Declaration,”

[6] AMCU, “About AMCU,”

[7] Quarterly Labour Force Survey, Quarter 2, 29 July 2015, p. 52.

[8] NUMSA, 28 April 2016, “The Workers Summit and May Day 2016,”

[9] NUMSA, 28 April 2016, “The Workers Summit and May Day 2016,”

[10] T. Elsley and G. Mthethwa, 2014, “Wage Determination in South Africa since 1994,” Bargaining Indicators 2014: Twenty Years - A Labour Perspective, volume 14, pp. 19-24.

[11] T. Elsley and G. Mthethwa, 2014, “Wage Determination in South Africa since 1994,” pp. 21-23.

[12] T. Elsley and G. Mthethwa, 2014, “Wage Determination in South Africa since 1994,” p. 10.

[13] M. Taal, 2015, “The Maximum Wage: Directors’ Fees Report 2015,” Bargaining Indicators 2015, volume 15, p. 57.

[14] R. Davies, 1978, “The Class Character of South Africa's Industrial Conciliation Legislation,” E. Webster (ed.), Essays in Southern African Labour History, Ravan, Braamfontein, pp. 75-78.

[15] Participants cannot be fired simply for striking, although non-work-no-pay applies.

[16] E.g. B. Terblanche, 24 February 2012, “Conflict Reigns in the Workplace,” Mail&Guardian,

[17] S.F. Karim, 11 October 2016, “CCMA Victory for Non-Unionised Workers,” GroundUp,

[18] Raymond Kgagudi in K. Sosibo, 25 June 2013, “Why the Left Has Failed to Capitalise on Post-Marikana Massacre Cracks in the ANC Hegemony,” The Con,

[19] CONSAWU, “Establishment,”

[20] For example, the three-year deal struck with AmPlats in October 2016 is for an annual average of R1000 a month or 7% (whichever is greater), for workers in the bargaining unit, plus increases in perks: see Anglo-American Press Release, 28 October 2016, “Anglo American Platinum and AMCU Sign Wage Agreement,”

[21] FEDUSA, “About Us,”

[22] FEDUSA, “About Us.”

[23] See e.g. G. Mantashe, 2009, “The Decline of the Mining Industry and the Response of the Mining Unions,” MA thesis, Wits University.

[24] E.g.  COSATU/ SACP, 1999, Building Socialism Now: Preparing for the New Millennium, Johannesburg;  SACP, 2012, The South African Road to Socialism,

[25] E.g. L. van der Walt, 2011, “COSATU’s Response to the Crisis: An Anarcho-Syndicalist Assessment and Alternative,” ASR, number 56, pp. 11-13.

[26] E.g. P. Dibben, G. Klerck, and G. Wood, 2014, “The Ending of Southern Africa's Tripartite Dream: The Cases of South Africa, Namibia and Mozambique,” Business History, volume 57, number 3, pp. 461-483.

[27] “NUMSA 10th National Congress, 12-15 December 2016, Cape Town: Congress Declaration,”, pts 43-45.

[28] M. Bakunin, [1871] 1971, “The Programme of the Alliance,” S. Dolgoff (ed.), Bakunin on Anarchy, London, George Allen and Unwin, pp. 249, 252.

[29] Bakunin, [1871] 1971, “The Programme of the Alliance,” p. 255.