PRECARIOUS LIBERATION Workers, the State, and Contested Social Citizenship in Postapartheid South Africa

Franco Barchiesi

To my mother, Cesira, the memory of my father, Ugo, and to Miranda, with gratitude

If you are caught up in someone elses dream, you are lost (Gilles Deleuze)1

Table of Contents List of Maps List of Tables Preface and Acknowledgements 7 7 10 19 20 20

Note on South Africas Racial Terminology Introduction -

The Promise of Wage Labor in South Africas Democratization

The Nexus of Work and Social Citizenship as a Contested Field of Signification 28 Work and Citizenship in Postcolonial and Postapartheid Modernity Conclusion and Summary of Chapters 53 57 37

1. Redeeming Labor. From the Racial State to National Liberation Introduction 57

Schooling Bodies to Hard Work: Labor, Modernity, and the Policy Discourse of the Racial State 58 68

The Hopes and Disappointments of an Inclusive South Africanism

Apartheid Social Engineering and the Coercive Enforcement of Wage Labor Discipline76 Black Workers Struggles and the Redemption of Wage Labor, 1973-1994 Conclusion 105 108 85

2. The Work-Citizenship Nexus of Postapartheid South Africa -

Resistance is Futile. The Governance Project of the ANC in the New South Africa 108 The Changing Face of Precariousness 124

Building the Patriotic Worker: The Democratic Constitutionalization of Wage Labor 133 Conclusion. Disciplining Citizenship 4 152

3. Contesting Commodification. Social Policy Debates in the Crisis of Waged Employment Introduction. Governing in the Shadow of Precariousness Social Policy as a Technology of Self-responsibility 159 157

Laudable Citizens and Silly Fools. Work, Families, and the Developmental Social Welfare Idea 173

The Wage-Income Relationship is Breaking Down: Basic Income and Contested Decommodification 193 210

Conclusion. Precarious Employment as the New Peoples Contract? 217 217

4. The Changing World of Work in Gauteng -

Introduction. Dreaming of Modernity in the Place of Gold City of Industry. The East Rand/Ekurhuleni and the Promise of Work

219

Economic Restructuring and Employment Decline: The East Rand in Transition 227 Johannesburg Municipal Workers and the Corporatization of Local Service Delivery 234 Conclusion. Invisible Workers and the Discursive Production of Postapartheid Spaces251 254

5. Translation Troubles. Signifying Precarious Work on the Shopfloor Introduction 254

Coping with Something Strange: The Disappointments of Workplace Transformation in East Rand Factories 258

New Canaan, New Egypt. Workplace, Community, and Identity among Johannesburg Municipal Workers 281 290

We Feel Sort of Redundant. Surviving the Flexible Workplace

Entrepreneurs of the Self. Individual Strategies and Life after Waged Employment 301 Conclusion 311 5

6. Like a Branch on a Rotten Tree. Recovering Agency after Wage Labor Introduction 316 319

316

Commodification and the Reconfiguration of Workers Lives

A Future Unlike It Used to Be: Visions of the Apocalypse and Labors Politics of Melancholia 332

The Fog of Activism. Working Class Agency and the Uncertain Quest for Citizenship Alternatives 345 363 368 380

Conclusion

Conclusion

Appendix on Methodology Notes Bibliography 384 411

Maps 1. Gauteng Metropolitan Area 2. Research Sites for the Book Tables 5.1 Seniority and Job Changes of Interviewed Workers 6.1 Remittances and Family Support Networks of Respondents 6.2 Impact of Social Provisions Expenditures on Respondents Wages 6.3 Respondents Housing Conditions and Expenditures 286 321 321 325 8 9

[MAP 1 HERE: Gauteng Metropolitan Area]

[MAP 2 HERE: Research Sites for the Book]

Preface and Acknowledgements

I would not have written this book without a ruthless reassessment of the reasons and the myths that brought me to South Africa. Over the past twenty years I have studied the contribution of workers struggles to antiapartheid popular opposition and the transition to representative democracy. Eventually I realized that my early work, mostly focused on labor organizations and workplace identities, did not adequately problematize the relationships between employment and emancipative politics. In conceptualizing this project I chose, therefore, to look at work and production as objects of contrasting images with which the postapartheid state, labor ideologies, and ordinary workers discourse represent citizenship and articulate social claims. Soon after the first democratic elections of April 1994 I moved to Johannesburg, where for six years I have taught in the Department of Sociology at the University of the Witwatersrand. At the time I was mostly attracted to the triumphs of black workers struggles, which I had already taken as the subject of undergraduate research in my native Italy. I found the decisive impact of radical class-based mobilization on South Africas liberation enormously inspiring as an alternative both to labors slow demise in liberalizing Europe and to its subordination to nationalist regimes in postcolonial Africa. The South African events of the 1980s and early 1990s titillated my early activist and theoretical practices within operaista (or autonomist in a broad English-speaking usage) Marxism, which posited the capacity of workers struggles to shape social relations independently from the dialectics of production and the requirements of organization. The South African labor movement shone as an example of militancy that, originating from the wage relation, subverted and transcended it to ignite a broader revolutionary transformation. 10

Once in South Africa, therefore, I started investigating black working-class politics by situating unionization within complex and multilayered material, symbolic, and discursive practices. By then most trade unions were clearly aligned with the African National Congress (ANC,) the leading force of national liberation, which rose to power in 1994 and has held it firmly ever since. A new entity, postapartheid South Africa, emerged from majority rule, a liberal democratic constitution reputed to be the most advanced in the world, and the prospect to heal past traumas through reconciliation and human development. As its antithesis the government placed the legacy of apartheid and its challenges, which media and scholarly parlance embodied in specific problem populations grappling with poverty and vulnerability: the rural landless, the urban unemployed, the shack dwellers, the maladjusted youth. Liberal democracy considered these subjects as potentially unmanageable for two reasons: their fiscally unsustainable expectations and their alleged disruptive propensities rooted in a culture of entitlement. In official discourse, therefore, work, production, and a morality of personal responsibility heralded the political incorporation of democracys unruly antithesis despite enormous social disparities. Becoming a worker was supposed to be the purest, most virtuous expression of citizenship. Liberal commentators, social scientists, and left critics agreed that jobs were the unquestionable solution to the countrys social ills. They therefore vilified demands not subordinated to labor market participation and reified the poors agency into abstract statistical categories active and disillusioned jobseekers, structurally unemployed, informal microentrepreneurs, non-working populations predominantly defined by occupational status. In 1996 the government adopted the Growth, Employment and Redistribution (GEAR) strategy, which turned the ANCs developmental agenda towards market-driven globalization. 11

Now in power, the liberation movement shifted its focus from resistance to economic profitability, political stability, and the disciplining of social radicalism. Whereas in the past wage labor had laid foundations for solidarity against exploitation, post-1994 policies reemphasized its status as a commodity and a factor of production responsive to the requirements of national competitiveness. Organized labor continued, nonetheless, to glorify employment and advocate job creation as an avenue to socialism, even if the political hegemony of the ANC translated liberation into a non-class nationalist narrative of progress and modernization. Meanwhile, destitute, unemployed, and working poor multitudes were left outside decent jobs and the social dividends of democracy and took to articulate their grievances into mounting collective dissatisfaction with the ANC government. I was then part of the editorial collective of a small magazine, Debate. Voices from the South African Left, busy discussing how, under the imperatives of economic liberalization, a democratic transition can simultaneously enable political participation and constrain the range of claims brought before its institutions. As crucial allies of the new ruling party, trade unions had gained significant political and policy influence precisely when labor and production, which historically underpinned their redemptive discourse, constantly deteriorated and failed to materialize their promises. Black workers continued to question the democratic experiment with insubordinate desires for social justice, but the relationships between production-based identities and political liberation appeared increasingly ambivalent and fragile. In 1998, with such questions in mind, I started researching this book. Most left and labor supporters of the ANC were then confused by the partys embrace of macroeconomic orthodoxy and recurrent accusations of neoliberalism flew. Amidst their acrimonious exchanges, the government and organized labor continued to share, nonetheless, the view that joblessness is the 12

main cause of poverty and that waged employment is the cornerstone of progressive policy responses. My initial hypothesis, on the contrary, somehow reflected the minority view in South African social sciences that employment relations indeed decisively contribute to marginalization and vulnerability. I consequently chose to investigate black unionized formal employees in mostly male occupations, a group that mainstream analyses tended to assume as relatively betteroff. A vast sociological literature has strictly identified the truly disadvantaged in the new South Africa with subjects outside the economically active population. My research started to show, instead, that even many wage earners suffered from employment situations that did not translate into once expected social advancement. Factory workers biographies foreshadowed the democratic transition as an uncertain and open-ended predicament, torn between labors past expectations, its current disappointments, and the states rigidly employment-centered normativity. Even without much of the pathos scholars recursively associate with other personifications of African suffering and victimhood (or perhaps because of that,) the precariousness of employment destabilizes progressive postcolonial narratives, be they of nationalist, socialist, or liberal inspiration. Deep-seated practices of domination that once marred South Africas racial capitalism have blurred into modern imperatives of productivity and flexibility, problematizing the idea that postapartheid is a resolute discontinuity in a cumulative trajectory of improvement. The centrality of work, and its contradictions, for a disparate set of narratives reinforces instead deep connections between profoundly different historical contexts, regimes, and ideological forces. I started interviewing Johannesburgs heavily unionized municipal workforce and industrial workers in the adjacent East Rand expecting to encounter traces of the confident, 13

determined working-class collectivities that during the 1980s were central to the regions social conflicts. What I found was rather disorientation, apprehension, and disillusionment towards changes that did not resemble erstwhile imageries of work as a bulwark of social stability and human dignity in a democratic society. A self-styled revolutionary state was now in power, but revolution did not come. Old working-class radicalism morphed into pleas for the government symbolized by the saintly figure of Nelson Mandela to deliver, and yet the countrys new rulers were calling for sacrifices, layoffs, and public spending stringencies in the name of globalization. The concept of precariousness allowed me to capture black workers feelings not just in terms of perceived insecurity capitalist employment has hardly been, after all, stable and dignified for the majority of South African workers but as manifestations of a discursive and significational gap. On one side stood a mythology of labor woven in decades of proletarian resistance and in the official scripts of trade unions and community activism. On the other, such rhetoric sounded hollow and muffled for workers experiencing labor market participation as fundamentally inimical to basic everyday survival, let alone human dignity. Employment precariousness has also, however, increasingly denoted in critical social analysis for example in the European usage of precarity or precarit not only a reality of domination, but the capacity workers have to signify labors declining existential centrality and stability. It allows, in other words, understanding insecurity not just as a paralyzing blow from inscrutable economic forces or the result of individual failures to produce, but as a condition for strategizing alternative political possibilities. Responses to the inquietudes of work are not necessarily progressive and can indeed be easily articulated into exclusivist and chauvinist collective appeals, as the ascendance of the 14

extreme right in the West and some sobering remarks in this books conclusion on recent South African developments remind. A focus on the discursive processes that signify employment invites, nonetheless, to transcend conventional sociological definitions of precarious work as the mere result of market discipline and approach it, instead, as an unsettled terrain where intersections of production, citizenship, policies, and subjectivities are continuously contested and negotiated. As a category problematizing market-driven democratization precarious employment vindicates the place of work at the core of social research, even when workers organizational and ideological expressions become fragmented and contradictory. Postapartheid South Africa hardly reflects a functionalist paradigm where the state and the market harness social processes towards inclusion and citizenship. Work is, instead, central to governmental and social practices that are mutually and unpredictably constituted. To include social subjects the state has first to imagine them as targets of governmental intervention and market discipline. Therefore the ideal virtuous worker as citizen opposes the imagined social pathologies of disruptive radicalism and entitlement claims. But actual workers retain, nonetheless, the ability to autonomously contrast work as an icon of state normativity with their troubled material experiences of employment. As a normative universal the employment imperative can only incompletely signify the subjectivities it presupposes, thereby laying the foundations for new claims and antagonisms.

In the ten years spent on this project, colleagues and friends provided invaluable human and intellectual support. I am indebted to many for my exciting years in Johannesburg from 1994 to 2002, and for enriching my firsthand experience of South Africas transition. My teaching and research took place in a world-class scholarly observatory, the University of the Witwatersrand, 15

specifically its Department of Sociology and the Sociology of Work Unit (SWOP,) where I hugely benefitted from the guidance and mentorship of Eddie Webster and the intellectual stimulation of colleagues like Glenn Adler, Shireen Ally, Andries Bezuidenhout, Belinda Bozzoli, Jackie Cock, Louise Hagemeier, Jon Hyslop, Yoon Jung Park, Paul Stewart, and Lucien van der Walt. The encouragement and inspiration I gained from Bridget Kenny and Elsa van Huyssteen cannot be adequately synthesized in such a limited space. In South Africa I have also learnt a great deal from many people with whom I shared activist practice, extra-academic discussions, and critical engagement; this book owes much to exhilarating, exhausting conversations with Heinrich Bohmke, Ashwin Desai, Rehad Desai, Stephen Greenberg, Fred Hendricks, Ulrike Kistner, Hein Marais, Dale McKinley, Darlene Miller, Andile Mngxitama, Prishani Naidoo, Andrew Nash, Noor Nieftagodien, Rebecca Pointer, Greg Ruiters, Melanie Samson, Ari Sitas, Peter van Heusden, Ahmed Veriava, and Frank Wilderson. Many others provided unfailing personal encouragement, and I wish here to thank Alice Brown, Kim Christodoulou, Jenny Haggard, Polly Horton-Dewhirst, Debbie Hunt, Khaled Mansour, and Viky Perschler. Michelle Pickover and the staff at the Department of Historical Papers at the University of the Witwatersrand were immensely helpful and reliable in allowing me access to vast archival collections of the South African trade unions and antiapartheid movements. I gained access to workplaces and respondents and I was able to negotiate the often complex related interactions thanks to decisive help from various trade union officials and organizers. They enabled my research in a spirit of comradely cooperation and selflessly took time away from pressing commitments to give me advice, introductions, and logistical support. This book would not exist without them, and I would like to express my particular gratitude to 16

John Appolis, Hlubi Biyana, Jenny Grice, the late Zweli Livi, George Magaseng, the late Victor Mhlongo, Shirley Miller, Jeffrey Ndumo, Andile Nyambezi, Rob Rees, Meshack Robertson, Dinga Sikwebu, and Andrew Zulu. The workers who accepted to be anonymously interviewed were more than respondents; in a sense, they are the actual authors of this book. Various officials and experts assisted me with access to government contacts and documentation, and I wish to thank Cynthia Alvilar, Lucky Mphafudi, Sonya Myburgh, Lydia Nthenga, and Alex van den Heever. Several friends outside South Africa have provided ideas, commented preparatory drafts, raised valuable questions, and sharpened my ability to problematize received assumptions. I thank in particular Amanda Alexander, John Beck, Tom Bramble, Michael Burawoy, George Caffentzis, Marlea Clarke, Silvia Federici, James Ferguson, Nigel Gibson, Michael Hardt, Gillian Hart, Mark Hunter, Peter Limb, and Antina von Schnitzler. At the Ohio State University I met a warm, vibrant, and stimulating professional and intellectual environment, where I greatly benefitted from the friendship of, among others, Mathew Coleman, Ken Goings, Christine (Cricket) Keating, Ike Newsum, Walter Rucker, and Rebecca Wanzo. Jim Degrand graced this book with superb, powerfully explanatory maps. My editor at SUNY Press, Larin McLaughlin, has believed in this publication from the start and followed it through with patience, commitment, and impeccable direction. In 2003, while working on early drafts, I conducted research in US institutions thanks to the hospitality of Mahmood Mamdani at Columbia Universitys Institute of African Studies and Stanley Aronowitz at the Graduate Center of the City University of New York. From 2002 to 2005 I have taught in the Department of Politics at the University of Bologna (Italy,) where I gained valuable insights from Anna Maria Gentili to which also goes my everlasting 17

appreciation for getting me interested in the study of Africa as an undergraduate student Sandro Mezzadra, Arrigo Pallotti, and Mario Zamponi. Funding for my research was generously provided by the National Research Foundation (South Africa,) the Consiglio Nazionale delle Ricerche (Italy,) the University of the Witwatersrand Research Council, the College of Humanities at the Ohio State University, and a Marco Polo grant of the University of Bologna. In parts of Chapter 6 I revised content from Franco Barchiesi (2010,)Wage Labor, Precarious Employment, and Social Inclusion in the Making of South Africas Postapartheid Transition, African Studies Review 51 (2): 127-36, which is here reproduced with permission of the African Studies Association. At this point, many give a closing space to someone particularly important in making a work like this possible. In my case, that person is my wife, Miranda Martinez, whom I met when my resolve was wavering and prospects of completing this book were almost as precarious as its subject matter. Mirandas love, sharp intellect, and passion accompanied the completion of my endeavor and infused it with a new sense of purpose.

18

Note on South Africas Racial Terminology

The apartheid regime divided by law the countrys population into four distinct groups for the purpose of racial discrimination and segregation, and such definitions also entered the language of policymaking and socio-scientific knowledge. Based on that classification, African (in different times the terms Native, Bantu, or Black were also used) generally refers to descendants of peoples inhabiting todays South Africa at the beginning of European colonization. White denotes the population of European descent, Indian designates the descendants of populations imported from India (the word Asian included other groups of Asiatic origin,) and Colored is applied to groups descending from mixed unions, usually between individuals of European or Asian and African origins. Scholars opposed to apartheid have generally rejected this terminology as racist, and they have rather used the word blacks to define the totality of the population (African, Indian, and Colored) oppressed and segregated under apartheid. The post-1994 democratic government has, however, retained the words White, African, Indian/Asian, and Colored in official statistics and policies of socioeconomic redress, as in the case of employment equity. I must also, therefore, commonly use such terms for the purpose of clarity and descriptive precision, even if I prefer to use whenever possible black in relation to socioeconomic conditions and political claims determined by the apartheid system and its present legacies.

19

Introduction

Had Cetswayos 30,000 warriors been in time changed to labourers working for wages, Zululand would have been a prosperous, peaceful country instead of what it now is, a source of perpetual danger to itself and its neighbours. (Theophilus Shepstone, former Natal secretary for native affairs, 1879)1

The Promise of Wage Labor in South Africas Democratization

Recalling the birth, in the mid-1970s, of the Commercial, Catering and Allied Workers Union of South Africa (CCAWUSA,) Emma Mashinini, the unions first general secretary, wrote: It is hard for black workers in South Africa to identify with other workers problems. Other workers are seen as human beings, and the black workers are seen as underdogs. . . . It is all the menial jobs, all the lowest jobs in the workplaces, that are the jobs of the black workers. And as a black worker, if I speak about a transport problem I am speaking about a different transport problem from anything the white worker will have to suffer. We have these very long distances to travel, and we have the poorest possible transport facilities, and our problems concern the pass laws and schooling, and hospitals, exhaustion, and poor diet. And while white mothers have problems of their own, such as having to see one of their boys leave to fight on the border, we can understand them, because we also must lose our children to the security forces, or to fight against 20

apartheid. But white mothers in this country do not have to suffer anxiety over what we call breadline problems. . . . Breadline problems are questions of who will care for the children when their mother goes to work? Who will pay the bills when the grandmother or friend cannot come one day and the mother must stay at home, even though she is not paid enough to be able to afford to lose that one days money? Who will pay when she has to spend a day at the hospital waiting for an appointment? No. Our problems are not the same. We had to fight for our identity as a black union. (Mashinini 1991: 39-40) Mashininis memories of militancy in South Africas independent trade unions the mostly black workers organizations, like CCAWUSA, challenging apartheid state repression and the power of established white unions echo two important themes. First, black workers could not experience waged employment as part of decent human life. Second, the conjunction of capitalist domination and racial oppression separated black and white workers and, by admitting only the latter to the full rights of South African citizenship, disavowed the idea that work could be a foundation of equality. As discriminatory legislation underpinned and enabled racist managerial despotism in the workplace, black employees experienced wage labor as a reality of humiliation, violence, poverty, uncertainty, and denial of basic social provisions. Historians and sociologists have long debated how South Africas racial state and capitalist development intersected to create cheap black labor (Levy 1982; Crush, Jeeves and Yudelman 1992) as a socially vulnerable and politically excluded condition for the majority of the population. In parallel to the expropriation of African land, labor exploitation underpinned a hierarchical ordering of citizenship (James 2007.)

21

The connection of oppression inside the workplace and political exclusion outside defined the precariousness of non-white labor. Mashininis words expressed the independent unions yearning for equal rights, meaning not only political liberation but also social citizenship, or the benefits and income security supposed to be labors fair reward. The plight of African women, at the bottom of the apartheid racial and gender hierarchies, most visibly embodied the problematic entanglements of work, citizenship, and redress. Unionization intervened as a force capable to transcend the narrow boundaries of bargaining over wages and working conditions. For many workers like Mashinini trade unions promised to redeem wage labor, turning it from a condition of oppression, degradation, and precariousness into a prospect of inclusion and human dignity. Not merely the expression of claims for economic and productive improvements, the black labor movement heralded fundamental social change as integral to the imagination of a future democratic South Africa. The redemption of wage labor was not only crucial for independent trade unionism, but also permeated the imagination of the national liberation movement. By the 1980s the African National Congress (ANC) still outlawed and exiled alongside its ally, the South African Communist Party (SACP) had become the main voice of opposition to apartheid. In an articles section titled Work. The Source of Security, Mosiuoa Terror Lekota, leader of the ANC-aligned United Democratic Front (UDF) and later minister in the post-1994 democratic government, reminisced how precarious life was in his segregated black township where, without access to land, for their survival our people have to rely entirely on gainful employment (Lekota 1986: 201.) In a socially oppressive reality, waged work bore the mark of insecurity and indignity: I recall distinctly those suppers of porridge with black coffee when the old man was out of work. Sometimes it was just dry porridge (Lekota 1986: 202.) The liberation struggle, however, 22

promised to restore work to a dimension of solidarity and self-realization. The ANCs historic program, the 1955 Freedom Charter, declared that there shall be work and security, which for Lekota meant that work, no longer under the yoke of capitalism and racism, would become an inviolable right of the individual in a new social order (Lekota 1986: 203.) In the last two decades of apartheid, black trade unions were decisively involved in political struggle. A formal alliance, lasting to this day, bound the ANC, the SACP, and the countrys main labor federation and flag carrier of the independent unions tradition, the Congress of South African Trade Unions (COSATU.) The majority of the black labor movement thus adopted the national democratic revolution agenda, initially elaborated by the SACP and then endorsed by the ANC in exile. In a distinctly modernist imagery, the national democratic revolution rejected the pseudo-traditional ethnic and racial divisions imposed by apartheid and brought to the forefront of the struggle industrial individuals (modern humankind) as the authentic bearers of the nation (Chipkin 2007: 106.) The approach ideally regarded industrialization and proletarianization both as midwives of apartheid and the causes of its ultimate downfall. Black waged workers embodied such a dialectical movement towards democratization and socioeconomic modernization. Modalities of struggle not derived from wage earning lost, conversely, importance. National liberation papered over historically complex dynamics of independent subsistence production and recalcitrance towards capitalist discipline, with which Africans resisted, cushioned, and negotiated the necessity to work for wages (Hart 2002.) 2 The connections of citizenship and labor are vital to understand the redemptive message of South Africas democratization. The rise of the ANC to power following the first democratic elections of April 1994 announced that work, emancipated from the shame and violations of the 23

past, could now contribute to democratic nation-building as the formerly oppressed rightfully reclaimed their land and an equitable share of the wealth they produce. The creation of free and dignified jobs became an imperative for ANC lawmakers. The partys powerful unionized working class constituency demanded it, but the new rulers as well saw jobs as the solution to abysmal poverty and inequality with their related, persistent risks of social explosion. Work, in brief, promised to infuse democratic citizenship in the new South Africa with unprecedented social and ethical qualities. With the advent of democracy, the glorification of productive employment outlined what Ivor Chipkin (2007) termed a normative public space, or the marriage of abstractly equal citizenship rights with the idea that virtuous citizenship resides in hardworking individuals, loath to complain and ready to sacrifice for the wellbeing of the nation. As imaginative projects have material consequences (Cooper 1996: 457,) a work-centered meaning of citizenship characterized both disciplined membership of the postapartheid political community and the procedures through which the state could selectively incorporate, co-opt, marginalize, or disqualify specific social claims. The first democratically elected president, Nelson Mandela, used all his iconic authoritativeness and moral stewardship to request patriotic citizens To exercise moderation in wage and salary increments, . . . to invest in greater capital injections and thus create possibilities for hundreds of thousands of people to be absorbed into the mainstream of the economy. We should all frankly acknowledge that there will be sacrifice.3 The vast symbolic and political investment on work, turned into a critical element for the survival and stability of the new democracy, under ANC rule motivates this books basic questions: Has employment fulfilled its promise of emancipation and dignity in democratic 24

South Africa? How did work relate to diverse visions of citizenship in the first postapartheid decades? How did government agencies, trade unions, and rank-and-file workers imagine such relationships? In what ways does the persistent precariousness of employment impact on workers identities, discourse, and collective solidarity? Is it still possible to think of labor as a progressive subject of social transformation? Many scholars debating South Africas post-1994 challenges have linked nation-building to political inclusion and racial reconciliation as ways to overcome past injustices and antagonisms. A common concern is how to gear formal democracy to the material needs of the formerly oppressed (Marx 1998; Alexander 2002; Chipkin 2007.) At the core of analysis have often been the normative foundations of the postapartheid transition, its constitutional elaborations, institutional projects, and the ideological pronouncements of its main players. Liberal commentators have focused on the compatibility of the ANCs transformative nationalism with constitutional provisions of procedural democracy and individual rights. Left and labor observers have, conversely, wondered how a negotiated, nonrevolutionary political transition could safeguard and revive the radical changes announced by the national democratic revolution. More needs, however, to be written on how official images of the transition related to the material practices and mundane meanings of ordinary South Africans. The ideas that shaped the postapartheid political space did not merely configure a new national collectivity, but also foreshadowed the citizen as an agent with claims and responsibilities, virtues and vices. The democratic constitution and public policies attached citizenship rights to certain explicit injunctions, allusive encouragements, and normative assumptions showing to the formerly oppressed the conducts befitting true citizens worthy of their newly found democratic freedoms. 25

The official discourse of the South African transition praised patience over radical demands, reconciliation over conflict, and personal initiative over reliance on public programs. Work and wage labor test the boundaries of postapartheid citizenship because they are quintessentially liminal categories encompassing, in the everyday lives of black South Africans, material precariousness, memories of past expropriations, potentials for social solidarity, and expectations of a better life under nonracial democracy. At the same time, the democratic promises of citizenship, social rights, and workers protections opened new spaces for claim-making and contestation similarly to what Frederick Coopers (1996) research showed on labor policies in decolonizing Africa. The states imagination of rights as counterparts to responsibilities did not necessarily match workers views of rights as tools to remedy structural social injustice. The intersections of production and inclusion remained sites of competing strategies underlying the quandaries of representing the African worker as citizen. Clashing with the abstract normativity of employment in the ANCruled state were the declining conditions of black workers, whose urgent concerns questioned earlier promises of emancipation. In 2009 the rate of unemployment stood at approximately 23 percent of the economically active population, rising to 27 percent among Africans, but could be as high as 30 percent if discouraged work seekers are counted (Statistics South Africa 2008a.)4 In the mid-2000s, the large majority of unemployed youth had never had a remunerated occupation (Bhorat and Oosthuizen 2005.) Such figures seemed largely unaffected by years of rapid economic growth, and could indeed deteriorate as in 2008 the country joined a deep global recession.5 Narratives of industriousness and work ethic in the national-democratic revolutionary ideology did not necessarily provide solace to jobless multitudes entering the new democracy with no meaningful prospect of decent employment. Even many union members worked for 26

poverty wages and scant protections. For them, the glorification of work was a poor substitute for the pressing necessities of daily survival. Employment problems are in fact not confined to joblessness but have also to do with the type of available jobs. In a globalizing South African economy, the jobs that liberalization and informalisation have created are often precarious, lack benefits and have low wages (Webster 2006: 21.) Informal and contingent occupations have absorbed many of the unemployed created by the restructuring of formal enterprises, and may now involve as many as four in ten workers. The South African Reserve Bank calculates that, as a share of national income, workers compensations have declined from 55 percent in 1970 to 42 percent in 2007, the lowest level in forty years. The share of company profits, conversely, climbed from 28 to 32 percent over the same period, approaching the 33 percent peak touched in 1981, in the heyday of apartheid labor repression (Mohammed 2008.)6 As democratization largely benefitted business, the deep crisis of waged employment amplified social marginality. How the South African government deals with its labor question will carry important lessons for other postauthoritarian transitions in highly unequal societies confronting the ideological hegemony of neoliberalism. The persistent centrality of work as a solution to poverty and social exclusion has accompanied in the governments discourse the condemnation of dependency on state social programs (Meth 2004,) fueled by the alarm over the continuous increase in the number of indigent South Africans claiming noncontributory state grants. Whereas antiapartheid labor and social movements saw job creation and redistributive policies as mutually reinforcing and complementary, the ANC as the ruling party has increasingly understood them as substitutes, with the emphasis clearly placed on the former. For president Thabo Mbeki expanding noncontributory provisions for economically active citizens unable to find work would amount to abandoning them.7 Minister of Social Development Zola 27

Skweyiya, in charge of social assistance programs, argued that social grants could promote dependency instead of boosting self-reliance,8 and pushed for measures to put recipients to work. Minister of Labor, Membathisi Mdladlana, saw as the aim of political liberation not to create a welfare state, but rather a developmental state where communities do not live off state hand-outs.9 The problematic relationships between the adverse social predicament of work and its persistent centrality in institutional imagination opens spaces where competing discourses of citizenship can be enunciated and enacted.

The Nexus of Work and Social Citizenship as a Contested Field of Signification

In most studies of the socioeconomic challenges of the South African transition, poverty and inequality identify specifically vulnerable groups like the rural poor, the urban unemployed, the residents of informal settlements, and some low-wage non-agricultural occupations like domestic and construction workers which seem to be left behind in the new democracy. Africans, women, and youth with poor skills and inadequate schooling are usually considered to get the roughest deal. The lack of a stable waged job is allegedly at the roots of their adverse predicament (May 2000; Bhorat et al. 2001.) Conversely, most urban waged workers, especially males with regular jobs and formal contracts of employment, hardly register in scholarly and policy representations of disadvantage and exclusion. A recent, influential analysis of inequality in the new South Africa by Jeremy Seekings and Nicoli Nattrass places indeed the waged working class among the winners of the transition. For them those with jobs have come to be a relatively privileged group (Seekings and Nattrass 2005: 46,) which prospered (Seekings and Nattrass 2005: 129) thanks to the 28

unions, often connected to the ruling ANC, and access to collective bargaining, company-based social provisions, and political clout. Earning a wage decides here the boundary between inclusion and exclusion, privilege and marginality, prosperity and poverty. Social citizenship appears as a clear-cut line separating the regularly employed from the underemployed and the jobless. After 1994, South African workers and their dreams of substantive citizenship and social rights confronted, nonetheless, the harshness of economic liberalization and the counterutopian (Comaroff and Comaroff 2000) imagery of market-driven globalization. Robust doses of fiscal restraint and public spending thrift mitigated redistributive ambitions. In 1996, the ANC-led government adopted the Growth, Employment and Redistribution (GEAR) strategy, turning democratic governance in the direction of free markets, fiscal discipline, and private investors confidence, domestic and international. Its alliance with the ANC notwithstanding, COSATU immediately denounced GEAR as neoliberal. The union federation saw the new policy framework as an abdication from the socialist aspirations and the redistributive demands that cemented the anti-apartheid front of the working class and the national liberation movement. In the end, however, GEARs constraints and the ANCs unassailable parliamentary majority consigned much policymaking to processes insulated from political contestation and dominated by technocrats, experts, and consultants. Responding to expectations for radical social change, public policies extolled solid data and cutting edge research as objective, technically superior measures of the possible and feasible. The rulers patience with competing ideological orientations ran rapidly short, reflecting minister Manuels conviction that its the governments job to govern (cit. in Gevisser 2007: 664.) A new jargon praising pragmatic solutions and realistic objectives circumscribed aspirations to social justice still evoked in official rhetoric. The 29

1996 Constitution had formally stated that resources made available by prevailing economic policies could condition the scope of redistributive programs. As the states deployment of expert knowledge depoliticized and moderated social claims, analysts invited labor leaders to move from the old politics of resistance to a politics of reconstruction (von Holdt 2000.) In the end, the new democracy calibrated its social citizenship imagination according to the imperative of minimizing disruptions to an economic order that continued nonetheless to exclude the majority. I started visiting factories in the East Rand three years after GEAR was passed. Economic liberalization had strengthened a highly globalized financial sector uncommonly powerful in the African context, but manufacturing suffered. Amidst rising competition and restructuring, the East Rand was losing jobs by the tens of thousands, while fixed-term and casual workers were replacing unionized employees. Among workers there was a widespread perception that the crisis had something to do with what a shop steward called the monster of globalization. Many were at a loss, however, to explain the role of the supposedly friendly ANC government. Fears of impending layoffs replaced workers optimism and solidarity with what Ari Sitas (2004) described as feeling stuck or deteriorated. Imagined connections between waged employment, emancipation, and social advancement were no longer obvious. During one of my factory visits, I witnessed a shop steward being verbally abused for a minor negligence by an African human resource official who used to be a union comrade. The union representative drew from the episode the bitter lesson that the ANCs policy of corporate affirmative action benefitted only few black workers hired as low-level managers and the racial window dressing of white-controlled companies, while driving new divides across labors rank and file.

30

Most workers I interviewed complained that their wages were hopelessly inadequate even for basic necessities. To corroborate their point, many could estimate household budgets and expenditures in minutely detailed calculations, immediately assessing down to decimal figures how much they needed weekly for food, their childrens schooling, remittances for some unemployed relative, or to borrow from friends and colleagues to repay loan sharks. In parallel to the new democracys celebration of personal responsibility, respondents saw themselves as participants in collectivist webs of solidarity as much as calculating, adaptive individuals. Echoes of past struggles incongruously resonated in a pragmatics of life where poverty wages have to be sagaciously administered to ensure daily survival and turn imponderable uncertainties into measurable risks. Heroic memories of unionization could hardly map clear routes from past triumphs to a grim present. Workers rather looked back at their lives in the union to find the specific turns where things started to go wrong, power relations shifted, concessions overrode contestation. My workplace interviews powerfully drove home what Dipesh Chakrabarty (2000: 76) discussed as the gap between labor as a universal one of the key categories in the imagination of capitalism itself and labor as an irreducibly singular, constitutively fragile human experience elaborated through life strategies and cultural practices. The universals claims to transparency, its ambition to flawlessly translate material workers lives into normative images of solidarity, consciousness, and progress, Chakrabarty (2000: 89) continues, are only possible by silencing daily distress and indignities that are too scandalous and shocking to be translated. In South Africa, the abstract universality of the employment contract, to the extent it endorses and reproduces labor market hierarchies and the growing casualization of jobs, has replaced overtly racial juridical categories and blunt political coercion as a determinant of insecurity for 31

the majority of the working population. But to understand why the postapartheid liberation of labor turned out to be precarious and hollow a focus on employment conditions is of limited use. Rather, the precariousness of black workers lives needs to be analyzed as a social and existential reality akin to what Claus Offe (1997: 82) termed shakiness and harmful unpredictability. For him, precariousness entails the contrast between the declining centrality of the labour contract in a social order where jobs are insecure (precariousness of work) and the norms that keep work central for individuals and households affected by the retrenchment of public programs and the official praise of work over welfare (precariousness of subsistence.) One needs to look, at this point, at how changes in the world of work diverge from and question governmental representations of economic activity. The normative use of citizenship to enforce work discipline was foundational for the imagination of the modern nation-state as it incorporated ideas of social rights and welfare. T.H. Marshalls classic definition of social citizenship encompassed the right to a modicum of economic welfare . . . and to live the life of a civilised being according to standards prevailing in society (Marshall 1964: 11.) At the core of such civilized standards he posited the generalization of male waged employment and the associated gendered division of household labor. The universalist framing of social citizenship, in spite of its material inequalities, aimed to incorporate the working class in the capitalist economy and defuse its potentially anti-systemic desires and demands (Fox-Piven and Cloward 1993.) Social citizenship was the reward for hard work and productivity and a bulwark against idleness, unruliness, and unrest. The nexus of social citizenship and work operates as a technology (Cruikshank 1999: 67-85) that produces the very subjects it takes care of through regimes and programs of social security. In what Michel Foucault (2001: 221) called the art of government the ability to structure the possible field 32

of action of other people social citizenship constitutes a specific biopolitical domain, or the endeavor . . . to rationalize the problems presented to governmental practice by the phenomena characteristics of a group of living human beings constituted as a population: health, sanitation, birthrate, longevity, race (Foucault 1997: 73.) A rich literature inspired by Foucaults concept of governmentality has investigated how modern social policies shape orderly, predictable, and responsible laboring subjects (Rabinow 1989; Procacci 1993.) Socio-scientific studies on living conditions, demographic assessments, and expert policy reviews, help naturalize the labor market as the realm where sovereign individuals as economically active citizens freely choose to empower themselves and participate in the production of the social (Hindess 2004.) The nexus of work and citizenship claims veracity impervious to empirical challenge because it blurs any internal distinction between normative judgment and scientific argument, views of human nature and prescriptions of rational behavior (Somers and Block 2005: 265.) Neoliberal policies of welfare retrenchment abandoned Marshalls redistribution but retained, nonetheless, his emphasis on the discipline of work. Both the welfare state and the antiwelfare notion of workfare (Handler 2004) enforce the centrality of wage labor in the social order, albeit by radically different means: fiscally funded social rights in the former and the compulsion to work as a condition for public social provisions in the latter. The neoliberal shift to work-based social inclusion implied a gradual erosion of decommodification, defined as the provision of social benefits (such as state-funded social assistance or employer-funded social insurance) as a matter of rights, independent from employment status (Esping-Andersen 1990: 22.) The ideology of workfare deprecates indeed the persistence of decommodification as an object of workers claims and on a line powerfully reflected in South Africas policy discourse reads it as a pathological manifestation of dependency habits. Policymakers during the 33

Mbeki administration (1999-2008) recursively evoked images somehow reminiscent of the third way of Britains New Labour or president Clintons welfare reform: self-reliance as a simultaneous indicator of rationality and character, hard work as the abstract condition of human dignity, the idea that welfare benefits would abandon to recall Mbekis expression citizens to a life without responsibility, while welfare cuts would compassionately empower them. The states official narrative of work-based citizenship claimed autonomy from evidence showing that low-wage jobs and meager welfare benefits are crucial factors in the devastation of working class lives. Social policy representations indeed powerfully shaped how labor markets work; in Somers and Blocks (2005) formulation, they acted as ideational causal mechanisms to cognitively embed the assumption that work, and work only, allows individuals to be fully functional. The postapartheid work-citizenship nexus tackles South Africas legacy of poverty and inequality by shifting the driving force of emancipation from structural social change to the abstract deracialized individual citizen it conjures up. Jacques Lacan (1977: 123; 1998) used the notion of chains of signification to define symbolic processes that produce subjects by anticipating their meanings while constantly deferring their fulfillment. Signification does not aim to attain something real but is a fluctuating, unstable system constantly producing new significations (Stavrakakis 1999: 22.) In a similar way, the virtuous and rational worker-ascitizen in postapartheid official imagination represents a subject for another signifier (Lacan 1998: 157,) or the moral binary opposition between discipline, initiative, and respectability on one side, and sloth, laziness, and dependency on the other. The work-citizenship nexus operates as a mode of biopolitical governmentality through the continuous signification of its subjects. Signifying citizenship through the work imperative reinforces the acceptance of the labor market, 34

its hierarchies, and inequalities, as the source of legitimate conducts while keeping more unruly desires at bay. Slavoj Zizek suggested that the states institutional significations tend to constitute society as a totality that normalizes all possible antagonisms so that social claims deflect the possibility of fundamental change. For him, desiring citizenship and empowerment within such confines is largely illusory and ultimately amounts to a hysterical demand for a new master (Zizek 1993: 211.) Signification, however, cuts both ways. Its continuously deferring the satisfaction of desire cannot in the end, as Deleuze and Guattari (1984: 1-44) argued, ensure stable government. Social claims are not merely the inert object of the states psychic life of welfare (Schram 2000: 160); they also interrogate the gaps and incongruities in governmental imagination and reveal a leftover (Zizek 2000: 171-243) of unsignified, unregulated desire that originates new aspirations and political possibilities. The work-citizenship nexus of postapartheid South Africa, therefore, is not merely a product of institutional rationality, but constitutes a contested field of signification. The governments ambition to shape disciplined laboring subjects is only part of the story because workers too signify their labor and autonomously imagine citizenship. Some elaborate ideas of dignified jobs to criticize their occupations, seen as exploitative, unrewarding, and meaningless. Others turn the discrepancy between work as it is and how it is supposed to be into an overt political challenge to the ruling party. Others, finally, refuse to place degraded employment conditions at the center of citizenship claims and explore alternative meanings. The contested nature of the work-citizenship nexus is not a new theme in South African studies. Between the 1980s and the 1990s, as apartheid declined and the possibility of a democratic government emerged, social scientists departed from early structuralist analyses that looked at state ideologies and institutions mostly as the expression of socioeconomic dynamics 35

and class interests. The apartheid state appeared then as a realm of conflicted adjustments and adaptations to the mundane practices of the disenfranchised, rather than a consistent and flawlessly rational planner (Posel 1991.) After 1994, however, the study of relations between official discourse and ordinary subjectivities declined and institutional dynamics took center stage. Many scholars wondered how the ANC in power would reconcile the national democratic revolution with macroeconomic moderation and the need to contain and control popular expectations. Of related interest were the operations of the elites, particularly the personal connections and patronage networks with which the ANC nurtured the loyalty of once fiercely independent labor cadres. Critics argued that, in their slow drift towards political cooption, trade unions had relinquished grassroots mobilization and social contestation, relying instead for their influence on direct access to powerful politicians (Calland 2006: 133-161.) Despite their useful insights, such debates tended to restrict the space of ordinary workers voices. Critical social scientists recognized that employment remained precarious for many, and could well be increasingly so, but they mostly framed this reality in terms of domination and disempowerment, whereby helpless workers confronted uncontrollable economic forces and relied deferentially on the state to deliver rights and protections. Workers were rarely seen as capable of signifying their own precariousness, entwine it to life strategies within and without the workplace, or use it to question the abstract centrality of work in the governments imagination of citizenship. Moreover, the socialist celebration powerful in COSATU, the SACP, and the left of the ANC of full-time wage labor as the normal condition of emancipative consciousness did not allow precariousness to foreground a factor Paul Gilroy (1991: 233) detected across the black postcolonial and diasporic experience: a political and philosophical critique of work and productivism. A look at South African workers struggles reveal, nonetheless, the salience, 36

shared with workers in the rest of the continent, of precarious work as a way to recast social relations and imagine political alternatives.

Work and Citizenship in Postcolonial and Postapartheid Modernity

It is still common among historians and social scientists to consider South Africa as an exceptional reality, set apart from the rest of Africa by its advanced, diversified economy, the extent of formal employment, and the strength of the unionized working class. For Seekings and Nattrass (2005: 122) South African exceptionalism rests on the unstoppable march of proletarianization and deagrarianisation, which, aided by the racial state, precociously led to the steady elimination of the African peasantry. Allen (2006: 3-4) comments that todays South Africa is unequalled in Africa due to, among other things, its developed infrastructures, financial services, and capital productivity. Thus the countrys poor and unemployed can allegedly enjoy opportunities without comparisons in the rest of the continent to enter a modern workforce. Critics of South African exceptionalism have rejected its intellectual and political parochialism (Webster 1997: 280) or its elitist pretense that the country is a Western society that just happened, accidentally and inconsequentially, if irritatingly, to be situated at the foot of the dark continent (Lazarus 2004: 610.) Exceptionalist narratives continue, however, to shape the view of postapartheid politicians that by successfully competing in global markets South Africa will escape the grim fate of other nationalisms in power, with their excesses of authoritarianism, dirigisme, and corruption. Those who deprecate exceptionalism express the need to locate the South African transition within Africas trajectory towards postcoloniality, but 37

often neglect labors role in such continuities. Mahmood Mamdani underlined that the South African racial state and most African colonial systems dealt with the native question in essentially similar terms, including the undemocratic government of rural areas through African chiefs. But when it comes to what he calls the labor question he also accepts the view that South Africa is unique. For him, It is only from a perspective that focuses single -mindedly on the labor question that the South African experience appears exceptional (Mamdani 1996: 28.) While it would surely be unwise to overlook how colonialism defined and ruled the native, the South African states responses to the labor question are not, however, fundamentally peculiar. Rather, complex commonalities and differences, involving similar processes occurring at different speeds and non-coincidental temporalities, connect labor regimes in twentieth-century South Africa with Africas transition from European rule to independence. In both cases, as Gillian Hart (2002) indicated, institutional and ideological dynamics are not easily separable from processes of accumulation and class formation. A focus on South Africas evolving work-citizenship nexus provides a powerful case for seeing the shift to majority rule much more as part of Africas postcolonial condition than as the premise of an exceptional story. Or else, maybe the continental trends to which South Africa is supposed to be an exception are not that uniform and homogenous after all. James Ferguson (1999) offered a trenchant criticism of the teleological linearity, espoused by colonial and postcolonial governments alike, of what he calls izations: Proletarianization, urbanization, industrialization, modernization. Underpinning them all is a narrative of modernity and progress marked by the inevitable advance of wage labor, commercial agriculture, and Western-type urban lifestyles. The narrative is indeed deeply tributary to a late colonial developmental imagery conveyed to African postcolonial leaders and abetted by international organizations and aid 38

agencies (Cooper and Packard 1997.) One of its aims was to incorporate African workers into the fabric of the state as a predictable and productive collectivity (Cooper 1996: 14.) The use of waged employment, no matter how confined it was to small sectors of the population, to propagate Western ideas of development and citizenship profoundly characterized Africas postcolonial condition, but also contributed to its fragility and contradictions. Integral to the late colonial experience were social policies designed to stabilize urban non-European workers. The dignity of work was one of the main rhetorical devices colonial administrators used to subjugate African labor and initially it was usually translated into coercive and repressive practices (Penvenne 1995; Isaacman 1996.) Where colonialism confronted a rapidly growing and potentially militant African working class, however, it tried to graft onto the discipline of work a modernist and progressive discourse intended to replace coercion with persuasion. In the words of an official report on Northern Rhodesian (Zambian) mineworkers (Davis 1933: 126,) Africans needed to be instructed into the true inwardness of the White mans moral system whereby labour in industry, on the mines and more especially on the farms of local White settlers could overcome the former magical and superstitious practices of village life. Early colonial att empts to stifle and control potentially troublesome working classes had, in the final decades of European rule, to give way to the colonizers grudging recognition of their growing dependence on African labor. The late colonial state tried consequently to gain African workers acquiescence by legalizing trade unions, regulating employment conditions, and introducing new social provisions. The goal was also to endow African working-class families with the respectability predicated upon male breadwinning (Lindsay 2003; Wilder 2005.) In the imagination of the time, the Western ideology of work operated as a heterarchy (Grosfoguel 2002,) or a combination of institutional practices, modes of knowledge, and normative ideas. 39

Social sciences, family planning, welfare state notions, and productivity pacts promised to reconcile the African nature to the immutable laws of development and, as Lucy Mair (1944: 111-112) optimistically concluded, harness [African] sense of community, and of duty to the tribe, to some modern movement. The fusion of socioeconomic imperatives of control and stability with the ideological aim of translating waged work into the normative vehicle of noncoercive governance pervaded the official discourse of late colonialism. The ambition of a French official in the 1940s, for example, was to fill Africans with the mystique of work and progress so that the indigenous wage worker will no longer be a deracinated person (cit. in Cooper 1996: 140.) Social and labor reforms were, however, confined to the small minority of regular wage earners and excluded women, rural workers, casual employees, and what was on its way to being designated as the informal economy. The poor without a stable monetary income were supposed to rely on self-help aided by residual policies of community development (Eckert 2004.) Despite the reforms pretensions, finally, even the waged workers they benefitted were generally deemed unsuitable for the remunerations and benefits of their European counterparts. The process towards decolonization sanctioned, therefore, the contradictory status of work in state imagination and practice: despite its universal civilizational message and the promise of modern rights, wage labor structured a highly hierarchical, profoundly unequal social order (Barchiesi 2007a.) Its universal values became hardly commensurable in semi-proletarian lives which, although inserted in capitalist production, were not intended to fully partake of the fruits of citizenship. It was, nonetheless, in the passage from late colonialism to independence that Western policymakers and local elites shared an imagination of Africa as a space made governable by the responsabilization within juridically free contracts of employment of its 40

modern, male waged working class. International organizations loaned their expertise in framing Africas labor question as the challenge of turning segments of local populations into orderly and productive industrial men. The late colonial experiment with governing Africa through a modernist labor regime failed because African workers appropriated it to make claims to power and social equality that authoritarian institutions were reluctant to accommodate (Cooper 1996; Brown 2003.) African workers also resisted working for wages, which, rather than reflecting redemptive prospects of progress, remained usually synonymous with unfair treatment, managerial abuse, and inadequate earnings. Casual employment was often a preferred option, despite its insecurity, as it cushioned the impact of capitalist work discipline, allowed to safeguard cultural norms, and preserved multiple networks of production and exchange across urban and rural spaces (Cooper 1987; Atkins 1994: 78-95; Ferguson 1999.) Workers claims were then expressed through the refusal of wage labor as a structuring principle of life as much as via workplace bargaining or productivity increases. In the hierarchical world of African waged employment, according to Cooper (1993: 134,) capital and the state had not created a reserve army of the unemployed but a guerrilla army of the underemployed who fought workplace discipline with desertion, slowdowns, and efforts to shape their own work rhythms. If African workers primarily struggled, as Cooper put it, against becoming working classes, precarious employment was not a condition of disadvantage but enabled opposition to the labor-centered citizenship of Western modernity. As the colonized refused to identify themselves with the laboring subjects the colonizers fantasized about, it eventually became convenient for European powers to have independent African leaders deal with workers demands and the social consequences of economic processes (Cooper 1989: 758.) 41

In the postcolonial nation-state, the precarious reality of labor within vulnerable, dependent economies underpinned challenges to authoritarian attempts to co-opt workers organizations (Freund 1988: 81-109) and rally them around the imperative of disciplined nation building as a new deal of the emotions (Balandier 1967.) African workers lives unfolded across multiple cultural formations, economic activities, and collective identities rather than focus on the point of production as the repository of consciousness and solidarity. What James Ferguson (1999: 96) calls workers signifying practices or a capability to deploy signs in a way that positions the actor in relation to social categories were weapons as powerful as formal organizations. Signifying practices do not express a material social condition, a transparent form of membership, or a self-evident cultural affiliation, but rather autonomously elaborate a multilayered system of meanings responding to duress, necessity, and socioeconomic crisis. But signification also reclaims a political space out of what would otherwise be mere survival: it expresses the subversive claim that the work-citizenship nexus of official discourse is incommensurable with, and untranslatable into, workers quotidian experiences. Combined to deep misgivings about capitalist work as simultaneously embodying a promise of emancipation and its betrayal, workers significations are liminal, shifting, indefinite grounds for potential insubordination (Das 1989; Bhabha 1995.) The unpredictability of workers signifying practices was an important constraint for the nationalist idea of political power as an omnipotent tool to reshape nature and social relations. Multiple livelihoods and the impermanence of employment undermined the states stabilizing fantasy of a cohesive working class as a subject of popular sovereignty within the idealized people on whose behalf the new nationalist elites claimed to speak. As the colonial governments before it, the African nation-state confronted the threatening indeterminacy of the 42

borders between predictable labor demands and more elusive social claims, and between respectable workers and dangerous classes (Burton 2005.) Following in the footsteps of the late colonial project of labor modernization, postcolonial governments deprecated casual labor as backwardness and a temporary phenomenon only (Freund 1988: 75.) In many countries an arsenal of regulations, legacies of colonial governmentality, tried to discourage the refusal of work as an avenue to immorality, criminalize idleness and vagrancy, and force unemplo yed youth to work in government projects (Lubeck 1985; Momoh 2000.) As only a tiny minority could enter decent jobs with benefits, and the blessings of capitalist employment were not obvious to most, repression joined moral exhortation in teaching citizens the virtues of work. Yet, as under its predecessor, in the postcolonial state the power to inflict violence did not match the power to force people to work (Bayart 1989: 23.) Africans persistent recalcitrance to wage labor discipline and the ability to resignify work according to local meanings are subjects for a rich historical literature. Congolese mineworkers saw the wage as barren wealth (Higginson 1989,) while in East Africa rumors represented Western production technologies as vampires feeding on the lifeblood, actual and symbolic, of communities (White 2000.) Among Tswana mineworkers in South Africa, wage labor violates the very definition of doing (Alverson 1978: 137.) Doing (go dira or tiro) as purposeful action orientated to meaningful social relations, opposes in fact wage labor (go bereka or mmereko, a foreign word from the Afrikaans werk) as alienated, senseless, externally imposed performance (Alverson 1978: 118-19; Comaroff and Comaroff 1987.) From the early 1980s, African governments applied structural adjustment programs to control a largely unwaged or intermittently waged working population. Structural adjustment decimated the ranks of the working class and organized labor through economic liberalization 43

and the downsizing of government employment. The evisceration of public social provisions and the destruction of formal jobs marked the end of a postcolonial social contract (Carmody 2002: 53) centered on the promise of employment creation. National elites redefined themselves as privatized intermediaries selling local resources under the supervision of international financial institutions. As healthcare, education, and municipal services largely became unaffordable commodities, expert discourse celebrated informal social security, or the poors resort to their initiative and social networks (van Ginneken 2003.) Mirroring trends in Western welfare reform, technocratic policymaking invested the disadvantaged with ethical attributes of individual responsibility as an alternative to decommodified public services. Academics recommended social policies that help poor people to help themselves (Wood 2004: 70,) for example by replacing social protection with social investment and relief with preparation.10 Ideas of self-help, community development, and decentralized capacity building, which once used to target the unwaged periphery of the labor market, occupy a central position in neoliberal governmentality (World Bank 2001.) Casual labor and the urban lumpenproletariat are now regarded as productive subjects and not just as anomic and unpredictable social aberrations. The idea of the informal economy evokes active, industrious, and entrepreneurial poor to replace the undisciplined and troublemaking casual or intermittent workers of colonial and early postcolonial memory. Rather than equalizing economic opportunities, however, the combination of informalized work and the growing commodification of life strengthened households dependence on the labor market as monetary income became increasingly inadequate. Declining social safety nets, rising prices of basic necessities, and the crisis of subsistence economies reinforced the compulsion to work for wages precisely when stable waged occupations dwindled. 44

The precariousness of African labor under neoliberalism is, in the end, simultaneously an enabling and an undermining condition (Weiss 2004): informality and undocumented work disclose new avenues for accumulation, but by favoring circulation and trade they are to a large extent disentangled from production. Access to goods and services continues, conversely, to depend on the very social institutions family, education, employment, political patronage undermined by liberalization. As the postcolonial promise of decent, stable work fades, ordinary people respond to social precariousness echoing Deleuze and Guattaris (1987) notion of deterritorialization, or the loosening of ties with established principles of social order by decoupling economic activities and life strategies from the sites of production. The meaning of employment in much of neoliberal Africa has transcended the realm of workplace interactions or national regulations, coming to rely on broad, interlocking, mutable social networks encompassing a panoply of local intermediaries, including political entrepreneurs and public officials, who provide often off the books and in the crevices between legality and illegality economic opportunities, social services, and a decentered sense of community (Mbembe 2001: 66-101.) In Mbembes (2001: 84) words, now citizens are those who can have access to the networks of the parallel economy as formal businesses and regular jobs are confined to narrowing pockets and enclaves. Yet, the borders of formal and informal are porous and permeable, as unrecorded and unregulated precarious entrepreneurs produce for formal companies, while undocumented trade, counterfeiting, smuggling, and migration circulate finished products locally and globally. In its attempt to incorporate precarious labor neoliberalism has paradoxically contributed to the implosion of work as a domain of disciplined actors, orderly conducts, and citizenship universals. Crucial to defining work in contemporary Africa are rather singular, messy, and 45

volatile transactions, as most laboring subjects devote their efforts not to the predictable production of commodities, but to the highly situational activation of social connections and the improvisation of doing things (Simone 2004a.) The contradictory colonial and postcolonial predicament of work, caught between the universalism of its values and the specificity of its injuries, reveals the persistence of what Anibal Quijano (2000) calls the coloniality of power: labor hierarchies inherited from past modes of domination continue to reflect the uneven penetration of a global capital hungry for Africas resources but with little to offer to any but a minority of its working people. Yet, even while suffering the damages of economic liberalization African workers remain capable, not without irony and cynicism, to autonomously strategize around precarious work beyond the conventional images of hopelessness and victimhood from the left or initiative and opportunity from the right. Underlying the postcolonial condition, in Africa as elsewhere, are bricolages (Cooper 1996) of discursive practices steeped in religion, ethnicity, and local moral economies immeasurable as abstract labor and unmanageable as activities reproducing capital (Chakrabarty 1988.) Ordinary people can demand job creation to gain political visibility and define themselves according to categories of governmentality the poor, the unemployed in order to give their claims the moral attributes of a community (Chatterjee 1994.) Grassroots tactical appropriation of official discourse, however, is not the same as buying into the progressive narrative, the collective identities, the sense of responsibility, and the idea of citizenship the governmental ideology of work underwrites. The cunning of survival informed by historical experience, pressing needs, and scant wages hardly warrants beliefs that decent work is an ordinary African condition and warns against bequeathing hopes of social inclusion to waged employment. 46

The contested vicissitudes of work and its ambiguous relations with citizenship place the comparison between South Africa and the rest of the continent into a sharper focus than exceptionalist views probably allow. South African developments reflect, despite local social and economic specificities, certain broad continental trends but with remarkably different timing and tempo. When French and British colonialism were navigating their way through social reforms, the apartheid regime in power since 1948 opted to eschew the stabilization of African labor through urban residential rights and trade union recognition. It rather reinforced a coercive system of circular labor migration that confined African citizenship to racially and ethnically separate, allegedly independent homelands. African workers, especially those classified as migrants, experienced work and citizenship as directly antagonistic terms. Forced to leave their families in the appalling poverty of the homelands and legally deprived of permanent urban residence, African migrant workers were easily employable into low-wage and insecure jobs, which regular urban dwellers could refuse. In its repressive and discriminatory conditions waged work was hardly intended to accommodate African social, let alone political, claims. Only in its last two decades, amidst a massive socioeconomic and political crisis, did the apartheid government embark in reforms somehow similar to the late colonial labor stabilization thirty years before. With the reforms of the 1980s, the white minority regime tried to gain the loyalty of black South Africans towards capitalism and govern insubordinate subjects by grafting elements of citizenship onto waged work. New legislation partially deracialized labor relations while maintaining the political disenfranchisement of the majority. Black trade unions were legalized and barriers to urban residence for African workers were relaxed. In its quest for legitimacy the state tried to recast employment as a free and uncoerced contractual relationship regulated 47

uniquely by the impersonal forces of the market. Similarly to late colonialism, the experiment failed as black workers rejected the agenda of reforms accelerating the downfall of white rule. After apartheid, as after colonialism, however, nationalist elites ended up responding to popular expectations by keeping the imperative to work as a disciplinary icon regardless to the injuries of degraded employment conditions and economic liberalization. The imagery of the South African racial state had opposed the productive African waged worker to the unruly native recalcitrant to capitalist production. The post-1948 National Party government regarded Africans as permanently stuck in the tribalized universe of the homelands rather than moving towards European lifestyles. It nonetheless deployed for black and white workers alike a narrative where, to paraphrase Adam Ashforth (1990: 208,) progress was the guiding star to which hard work was the compass. The idea that working for wages is the condition of full human life while redistributive programs are carriers of social disease gained its epistemic privilege (Somers and Block 2005: 265) under white rule and inward-looking growth, and survived the shift to representative democracy and globalized capital. When apartheid ended, central to the ANCs view of a new South Africa was the establishment, for the first time in the countrys history, of a fully-fledged, juridically free capitalist labor market where equal rewards for individual worth and effort replaced racial privilege. Cleansed of its legacy of unfreedom, wage labor reclaimed its role of promoting nonracial citizenship and social inclusion. Even if the work-citizenship nexus acquired the flavor of democratic legitimacy, however, it remained hard to translate into the persistently precarious lives of black workers. Past struggles against class exploitation and racial domination had enabled powerful counternarratives drawing, as in much of Africa, from rich traditions of resistance to proletarianization. A substantial scholarship discussed the importance of refusal of wage labor for black opposition to 48

white rule and its racial and gendered hierarchies. The unauthorized movement of African women to the urban areas and their preference there for casual jobs from domestic services to prohibited activities like sex work or beer brewing exemplified life strategies that undermined both the patriarchal order of the homelands and capitalist work in the city. The subversion of the physical and gendered geograph y of apartheid was a way to exist at a distance from the disciplines of time, productivity, and monotony that the more deeply proletarianized sections of the work force were experiencing (Bozzoli 1991: 145.) But informality, self-employment, multiple intermittent jobs offered to a broad range of working people, including unskilled Afrikaner proletarians (Van Onselen 2001,) alternatives to the relentless and ferocious regularity of the mines and the factories. As South Africa industrialized, young residents of African urban townships refused factory work to the point of forcing the state to expand the recruitment of migrant workers (Posel 1993,) whose unionization eventually gave decisive impetus to labor conflicts. During the resurgence of labor activism in the 1970s and 1980s, analyses of the new independent trade unions were not confined to formal organizations and ideologies, but looked at the complex prism of cultural practices, discursive formations, and collective identities through which black workers signified union membership. Ari Sitas (1983) and Eddie Webster (1985) showed that black unionization was not unproblematically premised on the linear development of class consciousness, but emerged out of negotiations among workers identities and meani ngs enmeshed in gender, ethnicity, religion, and migration. As the unions entrenched themselves and gained important victories, however, South African labor scholars started to abandon the study of waged work as an object of complex signifying practices. They took, instead, to celebrate employment as the linear destination of socioeconomic emancipation, the true foundation of 49

proletarian consciousness, and the vehicle of workers political demands within national liberation politics. Discussing Alversons opposition of mmereko (alienated wage labor) and tiro (autonomous, meaningful, and purposeful doing,) Dunbar Moodies study of African mineworkers identities emphatically elevated the former in a teleological modernist narrative where now the men of the union are replacing the men of migrant cultures as contemporary workers who are completely dependent upon wage labor . . . have no home in which to practice tiro, no umzi [homestead] to build . . . Wage labor is their life . . . the trade union is a home of a sort (Moodie 1994: 306. Own emphasis.) Precariousness, being outside waged work, or living in-between waged and unwaged activities ceased therefore to be seen as conduits for meanings and strategies. They rather became conditions of uprootedness and speechlessness. In the repressive 1980s, organized labor was the most powerfully organized domestic opponent of apartheid. Concomitantly, labor scholars overwhelming interest in organizations and ideologies consigned to irrelevance and invisibility workers expressions not derived from occupational or political dynamics. The periods classic analyses of labor struggles were either productivist, workplace-orientated, and based on the primacy of class struggle (Friedman 1987,) or nationalist, ANC-orientated, and focused on the centrality of national liberation (Baskin 1991.) They converged, nonetheless, in glorifying wage labor and depicting it no longer as a target of resistance but an immanent force of liberation and social empowerment. Others explicitly dismissed contestation and refusal of waged work as aberrations to be liquidated as topics of inquiry (Seidman 1994.)11 The agency of casual, informal, and precarious workers was pushed to the sidelines as labor-friendly intellectuals mostly regarded them as unorganized masses, waiting to be rescued by the unions. The public transcripts (Scott 1990) of organized labor, or its explicitly formulated acts of resistance, supplanted workers hidden transcripts as 50

inconspicuous, unseen daily acts subverting production discipline. The late-1980s alliance between COSATU and the ANC, with which national liberation became labors principal battleground, symbolically sealed the process. With democratization, South African labor studies joined the emerging political consensus hailing the supposedly accomplished redemption of work. Policymakers and scholars alike deemed class antagonisms a thing of the past and urged labor and business to cooperate in advancing productivity and industrial peace for the greater good of national reconstruction. Karl von Holdts (2003) ethnography of workplace politics during the transition saw the unions as bulwarks of disciplined behavior, responsible demands, and institutionalized interactions against disruptive conducts, grassroots extremism, and similar temptations to ungovernability. Dissonant perspectives came, however, from researches on the material transformations of work. Many studies pointed out that economic liberalization did not make wage labor less precarious, but further fragmented employment conditions through restructuring, outsourcing, and casualization (Theron 2004; Kenny 2005; Clarke 2006.) The persistent marginalization of workers also questioned deep-seated union identities as their adversary was no longer a visibly oppressive regime blatantly intruding in employees lives but a hazy, complicated combination of market dynamics and technically defined macroeconomic priorities. In the new intellectual mainstream, waged work became nonetheless coterminous with hope and fulfillment. A government-commissioned report on the poors motivation to productive activity concluded that interviewees demonstrated a strong commitment to work; indeed the plea for jobs was the most consistent and fervent message to emerge from th e interviews (Surender et al. 2007: 54.)12 Some went further to argue that to gain employment the poor are ready to make significant sacrifices, defer socioeconomic expectations, and accept jobs at almost 51

any condition (Charney 1995.) In their very important work, Jeremy Seekings and Nicoli Nattrass (1997a; 2005) claim that class is replacing race as the main determinant of social inequality, whereas they essentially define class in terms of employment status. Therefore, class struggle in South Africa no longer opposes capital and labor, but the employed and the unemployed: The excluded are no longer defined in racial terms, although almost all are black. Now, the excluded are the unemployed. The new faulty line in the labour markets and in the provision of welfare is between those citizens with employment and the many without (Nattrass and Seekings 1997a: 464.) As waged workers ascend in this account to the ranks of the privileged, the poors hopes for social advancement cannot rest on redistribution and decommodification, but must depend on looking for jobs that, due to limited skills, would only pay very low wages: The best that a labour surplus economy such as S[outh] A[frica] can aspire to is an American-style welfare state regime with a very inegalitarian labour market, where the state provides minimal and stringently means-tested public welfare (Nattrass and Seekings 1997a: 476.) The main obstacles to the emancipation of the poor are then, for Seekings and Nattrass (2005: 230-31, 282,) the selfish preference of unionized workers for higher wages, and collective bargaining provisions that discourage private business. Few critics have directly taken aim at the normativity of work and contrasted it with the duress of actual waged employment. Journalist and commentator Jonny Steinberg argued that state social grants should be not just a residual intervention for the indigent outside the labor market, but a shelter against the compulsion to work for poverty wages: 52

It is taboo among the grant lobby to say these things, as if even whispering them might cause the sky to fall on our heads. Yet the proper answer is: so what? To my mind, it will be a good day when chronically ill young men do not need to carry beer on their backs in order to put food in their stomachs. It will be a good day when the hungry no longer need to queue outside an industrial furnace at 6am.13 The postapartheid official nexus of work and citizenship, however, forecloses the analysis of how the social precariousness of work problematizes the states normative imagination. It is nonetheless at that level that contrasts between promise and delivery, official commitments and mundane expectations, discipline and desire destabilize the marriage of democracy and liberalization. The ANCs continuous, massive electoral support amidst such contradictions and persistent labor radicalism contributes to its dilemmas. Perhaps underneath South Africas precarious liberation lies, after all, an exceptional feature, or the fact that, as Mark Gevisser (2007: 692) writes, in few other places bottom lines are so contested by people claiming the same political roots. The legacies of national liberation politics shared by the ruling party and its labor critics do not necessarily attenuate, however, present social antagonisms or reduce the importance of what is ultimately at stake: the social quality of political emancipation. Central to this issue will remain questions of what work means, what it entitles to, what are its rewards, and which desires it enables.

Conclusion and Summary of Chapters

53

The project on which the racist regime collapsed and the ANC is testing its governing abilities linking citizenship to work in a politically legitimate narrative of social change remains unaccomplished. This introduction provided three conceptual tools to assess that project, its limits, and alternative possibilities. First, the work-citizenship nexus is a technique of rule to produce governable social subjects by normatively categorizing the attitudes, behaviors, and proclivities individuals have towards employment. Second, the work-citizenship nexus is also a contested field of signification involving official discourse, organized labor ideologies, and workers meanings of work. Third, the social precariousness of employm ent is not just a matter of occupational insecurity and labor market uncertainty, but is shaped by the mismatch between the official imagination of work and significations derived from its ordinary material experiences. As a corollary, precariousness does not disable, but indeed encourages workers production of sense, strategies, and political possibilities. The first three chapters of this book discuss the official discourse of social citizenship and inclusion in South Africa from white minority rule to its demise. Chapter 1 provides a historical background to the work-citizenship nexus from the early racial state to the democratic transition. It focuses on the continuities and changes through which distinct social policies used waged employment to normatively signify the condition of citizen. Chapter 2 looks at the imagination of wage labor in the social policy discourse of the postapartheid state. It examines in particular the ANC governments refashioning of a work-centered public morality in conjunction with its opposition to welfare dependency. Chapter 3 zooms in on a particularly important moment of discursive contestation in postapartheid governance, the controversy surrounding the proposed introduction of a universal basic income grant as a measure not dependent on employment status. The ANCs opposition ultimately killed the proposal and reasserted the centrality of work. 54

However, left supporters of the grant also framed their argument in productivist terms, rather than as a critique of work-centered citizenship. The last three chapters examine how workers in my case studies signify their changing employment universe, and their specific predicament in it, during and after the political transition. The aim is to find out what happened, in their views, to the promise of work after the end of racial domination. Chapter 4 is an overview of production changes and employment instability in the two areas I researched, the East Rand manufacturing concentration and the greater Johannesburg municipality, as they dealt with economic liberalization, organizational change, and industrial restructuring. Chapter 5 addresses workers narratives of change as marked by perceptions disappointment at unfulfilled expectations of work with dignity, the specter of insecurity, the powerlessness of union organizing that underlie the incommensurability between the official glorification of work and its experienced realities. Workers did not perceive waged employment as a foundation of decent life, but rather endured it with a view to escape, which was often articulated through ideals of self-employment sometimes with religious overtones. Chapter 6 deals with how workers articulate politically their desires to transcend a grim, precarious workplace life. Some try to grapple with change through an updated activist imagination appealing, beyond the walls of the shopfloor, to community mobilization and demands for decommodified social services. More widespread is, however, the continuous reliance on the ANC for policies of job creation and protection. Seemingly in contradiction with the low esteem workers have for their own jobs, such claims reveal, in what I term an emerging politics of labor melancholia, aspirations for an idealized social order where work guarantees authority relations based on gender, age, and nationality. Such developments raise the

55

disquieting possibility that, by maintaining work at the core of its imagination of citizenship, emancipative discourse can easily and inadvertently feed chauvinist and authoritarian fantasies.

56

1 Redeeming Labor. From the Racial State to National Liberation

I worked my way up from nothing to a state of extreme poverty. (Groucho Marx)

Introduction

Until the last decade of the twentieth century, racial domination has shaped the lives and employment experiences of the vast majority of South African workers. The realm of production was indeed crucial in determining the status of blacks as second-class citizens. The National Party government, rising to power in 1948 with its program of apartheid, subjected non-whites to a particularly harsh and pervasive system of labor control at a time when most African colonial regimes opted for labor reforms and stabilization. Within four decades, black workers would become protagonists of popular resistance. As a response, the racial state mimicked late colonial experiments and tried to tame labor militancy by making waged work an avenue to limited social entitlements for the disenfranchised majority. The failure of that project was decisive in the collapse of apartheid and left the ANC, triumphant in the first universal suffrage elections of 1994, with the task of responding to black workers expectations of social redress. This chapter discusses the contentious ways in which waged employment was woven in the normative imagination of the racial state, national liberation politics, and black workers unions. For all of them productive economic activity was a normative universal, albeit with 57

radically contrasting implications. Even when it was most committed to enforce rigid racial hierarchies and exclusionary citizenship, the pre-1994 state preached wage earning as a tool of social elevation for blacks and whites alike. The ANCs discourse drew from a complex mix of themes, but as the country industrialized it ended up placing the formally waged working class in a central symbolic position, which displaced earlier emphases on resistance to proletarianization. For the independent black labor movement resurrected in the 1970s, finally, waged work dialectically embodied a grim, devalued social existence and a necessary condition of solidarity that could ferry the oppressed to the shores of true freedom and working-class power. The fact that actors so distinct, when not overtly antagonistic, shared a vision of social redemption based on employment indicates the deep roots of imaginations that continue to bear fruits in the postapartheid states perceptions of, and responses to, South Africas social questions. It also highlights the contradiction between labors abstractly universal values and the material hierarchies, social conflicts, and practical claims that shape actual experiences of work.

Schooling Bodies to Hard Work: Labor, Modernity, and the Policy Discourse of the Racial State

As South Africa industrialized, historian Stanley Trapido (1971: 313) noticed, it has not incorporated the major part of its working class into its social and political institutions. The response of local capital to the mobilization of black workers has largely resisted the allure of welfarist ideas, which influenced limited experiments with inclusive social policies in late British-ruled Africa. For the South African non-white proletariat, work and social citizenship were destined to march on entirely separate tracks. 58

In the period often referred to as segregation (1910-1948) the power of mining capital and South Africas political unification within the British empire defined citizenship as an instrument of white supremacy. The then Union of South Africa designated, in particular, African societies as culturally distinct and unsuitable for European modernity. Imperial policies spatially confined African political belonging to rural native reserves designed through land expropriation. Segregation combined, nonetheless, political exclusion with the subaltern economic incorporation of blacks into waged employment. Cecil Rhodes argued in 1894 that monetary taxation would act for Africans as a gentle stimulant to remove them from that life of sloth and laziness, . . . teach them the dignity of labour and make them contribute to the prosperity of the state (cit. in van der Horst 1942: 149.) Echoing colonial ideologies of the time, the virtues of work would ideally nurture conducts fit for civilization, even when denying the enjoyment of its fruits. An 1898 editorial in the Natal Mercury summarized the aims of colonial governance as to overcome an inbred disposition for idleness and irresponsibility on the part of these Natives (cit. in Dhupelia 1982: 37.) Official discourse made progress rest on individual attitudes to the labor market. As Bozzoli (1981: 57) argued: Work was to the mineowner, therefore, what education and Christianity had been to the missionary not merely a fact of self-justification, but the symbol and apex of a social order. . . . Work was depicted . . . as having purifying and dignifying aspects to it. Africans resisted wage labor by defending independent agriculture as an alternative to capitalist production relations (Bundy 1988.) In response, the state deployed a vast array of rules limitations to blacks land claims, pass laws constraining labor mobility, and industrial legislation preventing black unionization which channeled Africans into low-wage jobs. 59

Landlessness and segregation in destitute, overcrowded reserves turned most African workers into domestic migrants while eroding independent production. As local agriculture declined, families in the reserves became increasingly dependent on the wages earned by relatives working in white South Africa. The migrant labor system also sanctioned gender hierarchies by identifying breadwinning with masculinity and confining African women to the unpaid work of household reproduction in the reserves. Yet, African proletarianization was, far from a linear and cumulative process, discursively and culturally mediated, and as such it was neither fully uniform nor irreversible (Bonner, Delius, and Posel 1993.) Migrancy led African workers to a highly precarious existence, but for the youth it also meant an alternative way to earn ilobolo (bridewealth,) which enabled claims to independence from household hierarchies and obligations. As migrant workers managed to keep rural support networks alive, they could use them to escape particularly oppressive jobs and defend noncapitalist work rhythms (Harries 1994: 41.) Migrant labor even revealed unexpected opportunities to elude waged work, depending on workers ability to enter urban self-employment rather than regimented mining jobs. African women looked for unauthorized urban occupations as a way to challenge their confinement in the patriarchal universe of the reserves. In the unwelcoming milieu of the city, they sought part-time domestic services and illegal liquor production or sex work as valuable alternatives to working for wages (Koch 1983; Bonner 1990.) As proletarianization gathered steam, idioms of escape from wage labor articulated African languages of resistance much more powerfully than socially transformative class consciousness (Harries 1994: 222.) The formation of an African industrial proletariat was slow and uneven and did not replicate Fordist mass production. Racially segmented consumption constrained domestic demand, so that large scale mechanization had to wait until the manufacturing boom of World 60

War II (Alexander 2000.) In the end, however, the growing dependence of white-owned industries on black workers and perceived threats of labor radicalism shaped social policies in a racially hierarchical direction that belied the normative universalism of the dignity of work. A nascent social legislation catered primarily to the needs of poor whites, mostly Afrikaners moving into low-skill occupations. Public programs and housing schemes boosted white working class respectability as a political response to the dangers of white pauperism and indigence, which the state regarded as social diseases opening the way for undesirable mixtures of European and native poor (Lange 2003.) Against the threat of cross-racial class solidarities, the state fostered the convergence of white labor and white capital around a populist sense of common interest (Martinot 2003: 86.) Racialized social policies were, in the words of liberal critic Margaret Ballinger, the formative force in standardizing relationships of black and white in this country (cit. in Giliomee 1992: 630.) Social legislation was in the end decisive in constituting whiteness as an intersubjective category defining how a white person by virtue of being white ought to live in comparison to non-whites (Giliomee 1992: 630. Own emphasis.) But ensuring the respectability of white workers was also the upper limit of social policies, which eschewed redistribution and universal welfarist provisions and endorsed from the beginning free-market neoclassical economics and budgetary austerity. By the 1920s, local relief schemes for poor whites had lost ground as, in the words of a contemporary observer, they led to demand relief as a right so that it becomes a habit for the poor to expect help in every difficulty (Jensen 1928: 36) to public works programs (Bozzoli 1981: 79.) Following the model of the 1918 Factories Act, contributory social insurance was strictly tied to employment status, while until 1946 retirement benefits were almost entirely based on industry- and employer-based plans. Under the control of craft unions, they covered a minority even of white 61

workers (Duncan 1995: 74.) The 1919 Public Health Act, in force until 1977, endorsed private company-based medical insurance, rejected fiscally-funded national healthcare, and limited the role of the government to emergency care (Union of SA 1936.) Whites were, for sure, the intended beneficiaries, but mostly on condition of being employed (Verhoef 2006.) The 1920 Housing Act provided state subsidies for whites-only municipal housing to claimants with jobs, rather than generically addressing poor whites (Lange 2003: 96.) The dominance of market ideology as a discourse of white social advancement became possible once white workers were politically protected from black competition (Chanock 2001.) Racialized social programs encouraged the industriousness of the white proletariat, especially Afrikaner lower classes with a background in farming or urban self-employment recalcitrant to wage labor (Swart 2000.) State intervention prioritized the recruitment of white workers in government jobs and their access to superior education, better training, higher wages, and career paths, rather than social provisions across the board (Terreblanche 2002: 270-75.) The labor market was, in the words of the 1908 Transvaal Indigency Commission, the battleground for the ultimate struggle for economic superiority over the native (cit. in van der Horst 1942: 179.) Work-centered social policies structured white workers vocabularies of social justice away from state handouts and towards better occupational opportunities as the exclusivist reward of whiteness (Roos 2005.) The unequal citizenship established in the workplace extended to urban planning the attendant distinction between respectable white workers and culturally alien native laborers. The 1922 Transvaal Local Government (Stallard) Commission recommended that Africans be defined as temporary sojourners in white cities, with residence conditional on their contract of employment. Following on Stallards footsteps, the 1923 Natives (Urban Areas) Act envisaged a systematic urban racial segregation to support governmental plans to upgrade 62

white working class neighborhoods into middle class suburbs and enforce slum clearance by deporting Africans from low-income mixed communities to peri-urban native locations. To the Africans the policy defined the city as a place of institutionalized social precariousness, as temporary residence rights and meager economic opportunities relied on highly insecure, poorly remunerated jobs. The aim was to make sure that urbanization did not enable expectations for equal citizenship rights.1 The rise to power in 1924 of the Pact government, a coalition between the Afrikanerdominated National Party and the English-speaking Labour Party, heralded a renewed governmental commitment to job creation for whites through support for domestic industries and large state-owned corporations, like steel producer ISCOR and the electricity company ESKOM. A policy of civilized labor perfected the superior employment status of white and, to a much more limited extent, colored workers, made white labor unions dependent on state protections, and further entrenched the role of employment as the guarantor of racially unequal citizenship. The Pact marked the political ascent of South Africanism (Bozzoli 1981; Dubow 2006) as a form of white colonial nationalism centered on domestic manufacturing and commerce, in alternative to the Afrikaner nationalism of agrarian origins and the previously dominant, miningbased British imperialism. South Africanism remained committed to racial segregation to protect its white worker constituencies, but also saw regularly employed Africans as a crucial market for national industries. If African workers had to be partially incorporated as consumers, they could no longer be only cheap labor; as a manufacturers journal argued: If Henry Ford can pay a man five dollars a day for screwing nuts onto bolts, surely native labourers can be paid up to 10s a day for working jackhammers (cit. in Bozzoli 1981: 196.) Besides, consumerism and upgraded social standards could respond to demands for meaningful social rights by nascent 63

African workers organizations, chiefly the Industrial and Commercial Workers Union (ICU,) launched in 1919. Work remained the foundation of social hierarchies, which were no longer, however, only intended as racialized cultural distinctions but also as socio-spatial categories dividing Africans between a permanently urbanized minority and a majority of rural dwellers and temporary visitors to white cities. The state imagined that for the former wage labor could be a path of promotion from native to worker, with lifestyles approaching European standards, but still without equal citizenship and political rights. The shopfloor was, on the other hand, one of the few places where blacks could access very limited social provisions. The 1914 Workmens Compensation Act, for example, covered African workers. For rural Africans, instead, the 1927 Native Administration Act systematized the native reserves as spaces to separately reproduce a workforce deemed superfluous for the urban economy except on a temporary basis. South Africanism developed the normativity of work from earlier moral emphases on dignity to a new paradigm of socioeconomic progress for blacks and whites alike. The state presented wage labor as a modern alternative to dependency, on state-funded programs for whites and on unwaged subsistence activities for blacks. Noncontributory social assistance of a decommodified type or not linked to employment status appeared late, was highly racialized, and targeted only specifically vulnerable subjects. Old-age pensions were introduced in 1928, following the recommendations of the Pienaar Commission: they were means-tested rather than universal, and covered only whites and coloreds defined as deserving poor, or with no other means of subsistence (Meth and Piper 1984; Seekings 2007.) 2 Confirming the ideological boundaries of South Africanism, lawmakers rejected the proposal in the third report of the Pienaar Commission for a national insurance scheme covering short-term unemployment and including urban African 64

workers (Union of SA 1929: 24-25.) The commission was nonetheless adamant that no benefit could be extended to the rural African, supposedly content with the bare necessities of life (Union of SA 1929: 24.) South Africanism fine-tuned the boundaries between inclusion and exclusion into flexible social hierarchies, where blunt racial and cultural scaffolding did not entirely disavow the myth of social mobility through work. As a mode of colonial governmentality it combined racially discriminatory legislation with assumptions on the nature and conduct of abstractly defined humans (Lentin 2004.) In the end, South Africanist discourse heightened the contradiction between the universalism of employment values and the material inequalities they preside over. The 1924 Industrial Conciliation Act denied trade union and collective bargaining rights to Africans by excluding them from the definition of employees. The 1925 Wage Act allowed the government to set minimum wages to prevent the hiring of cheap black labor in industries with a white demand for low-skill jobs. The 1926 Mines and Works Amendment Act allowed for color bars in skilled mining occupations. The image of work as the force capable to turn natives into civilized individuals did not depart from colonial paternalism and white stewardship, albeit it infused them with claims to socio-scientific truth. Governmental commissions of inquiry increasingly resorted to expert advice blaming poverty on cultural and mental factors the stagnation of African societies, the lack of acquisitiveness in traditional farming that could be vanquished by a modern homo oeconomicus equipped with work discipline, individual ambition, and the mastery of scientific rationality (Wylie 2001.) The 1932 report of the Native Economic (Holloway) Commission specifically deplored African refusal of waged work as a barrier to progress and modernity:

65

When the raw Native has enough for his wants he stops working and enjoys his leisure. . . . He must learn to school his body to hard work, which is not only a condition for his advance in civilization, but of his final survival in a civilised environment (cit. in Ashforth 1990: 84-85.) In the end, the narrative of labor as a universal path to self-improvement justified the enduring subjugation of Africans by postulating their continuously unaccomplished transition to the colonizers image of Man (Young 2004: 160-62.) Apart from its penchant for social engineering, however, the South Africanist ideology of work also resonated in diversifying modes of African discourse. Responding to industrialization, black trade unions, middle classes, and community leaders articulated their own expectations of working-class respectability (Goodhew 2000.) The demands of the colonized interrogated the universalist values of the colonizer and tried to take advantage of their contradictions. For relatively affluent and educated African elites the discipline of work was a moral device to control unruly youth by depicting aversion to employment as a symptom of low self-respect leading to crime, alcoholism, gambling, and marital infidelity. In the 1920s and 1930s the ANC was politically moderate and led by middle-class notables who saw racial discrimination as a betrayal of the modernizing promise of the Empire. Opposition to segregation by trade unions like the ICU focused not only on class antagonism but also on the claim that modernized, enterprising Africans could socially advance to their rightful place as equals in the colonial order. The ICUs positions reflected a Victorian social evolutionism where economic activity, for example through the organization of cooperatives, was a condition for self-reliance and access to rights as imperial citizens (Champion 1927.)

66

Work-centered social policies provided therefore the state, African elites, labor unions, and political organizations with a terrain of negotiation and mutual recognition underpinned by assumptions on a morally sound social order and hostility towards resistance to proletarianization (Cobley 1997.) Thus, for example, government officials and ANC leaders like Abner Mtimkulu shared the view that urban recreational and welfare associations were needed to divert working class leisure from vice to preparedness for work.3 But the emphasis on wage labors respectability was not limited to elites, middle classes, and organizations: it also surfaced in the contested moral arguments of precarious workers themselves. An African casual worker and member of Johannesburgs Basotho MaRashea (the Russians) gang, recounted in an interview the contempt held in the organized underworld for the boTsotsi, young free-ranging criminals whose work is not to work (cit. in Guy and Thabane 1987: 440.) In many cases the ANC and law enforcement officers were equally alarmed by the tsotsi as an element disruptive of urbane political interactions and refractory to the social norms of production and patriarchal authority (Mager and Minkley 1993.) The nascent African nationalism mediated between aspirations of popular emancipation and ideas of entrepreneurialism and self-help, which were not merely the reflex of middle-class values, but also objects of opportunistic appropriation with which the colonized disputed the colonizers civilizational narrative (La Hausse 1993.) An equally powerful impulse came from independent African churches, especially the Zionist movement, which celebrated hard work and self-sacrifice as tools of salvation for the poor (Kiernan 1977.) Such early contiguities in the celebration of work by otherwise conflicting political actors would reverberate in the imagination of resistance politics and postapartheid democracy.

67

The Hopes and Disappointments of an Inclusive South Africanism

The collapse of the Pact and the rise to power in 1934 of the United Party (under the leadership of Jan Smuts from 1939) marked the triumph of the South Africanist paradigm of white patriotism and modernization (Dubow 2006: 221-27.) Delivery for poor whites while forestalling blacks claims to equality remained the governments priority. Policymakers recognized, nonetheless, the existence of stably urbanized African workers, apart from the rural temporary visitors they had previously envisaged. Official agendas, therefore, shifted towards experiments with the stabilization of urban working classes, which echoed broader continental colonial debates. The United Partys version of South Africanism incorporated to some extent a liberal criticism of segregation and civilized labor policies, not so much in the name of equal rights, but because racial privileges for white workers allegedly undermined the work ethic of urban, civilized Africans. Their legitimate aspirations to European standards were, in this view, unduly frustrated by blanket bureaucratic intrusions (Brookes 1927; van der Horst 1942.) Facing rising African workers mobilization in a context of wartime manufacturing boom and labor shortage, the UP combined traditional colonial paternalism with state interventions aimed, with the help of the social sciences, to stabilize and control urban waged employment, a task to which traditional laissez faire appeared inadequate. The states discourse prefigured an evolution from native to African worker verging on the recognition of limited social citizenship rights on a nonracial basis (Ashforth 1990: 127,) which were nonetheless contested within the white establishment. For Afrikaner nationalists and most Smutss collaborators, the mix of social proximity and cultural difference between Africans and Europeans made the extension of citizenship to the natives a scary prospect, heralding unpredictable claims and 68

conflicts. Critical voices saw, conversely, citizen as a step logically following native and employee in the evolutionary trajectory the colonial state and the labor market steered for the modernized African minority. Conservative opinions dominated the massive work of the Carnegie Commission of Investigation into the Poor White Problem. Its final report in 1932 merged philanthropic arguments with the American-style social science progressivism in which the commissions leading intellectual voice, Ernst G. Malherbe, was schooled. The report rejected, therefore, welfarist redistribution and extolled labor market participation, education, and training as safeguards for the self-preservation and prestige of the white people (cit. in Giliomee 1992: 642.) Far from advocating deracialized social provisions, the Carnegie commission exhorted whites, and to a smaller extent coloreds, to seek training for productive occupations as the best way to forge their temperament as active individuals. It also warned of the degenerative effects of expectations for social rights, which it saw as the prelude to civilizational descent to the level of the natives (Wilson and Ramphele 1989: 145, 296; Seekings 2008.)4 Subsequent social programs expanded redistribution, but remained strictly means tested, focused on vulnerable recipients outside the labor market, and continued to exclude Africans and Indians. Programs for blindness, child maintenance, and disability introduced in the second half of the 1930s covered only whites and coloreds, apart from some minuscule financial assistance for indigent Indians and blind Africans (Union of SA 1944a: 6, 13.) In 1937 a Department of Social Welfare was inaugurated by spinning out of the Department of Labour the social work functions of unemployed rehabilitation, but the new bureaucracy was not responsible for redistributive programs like the old age pensions. The new departments mission, imbued with conservative Afrikaner morality and Calvinist religious prescriptions, was, in the words of prominent 69

sociologist J. L. Gray, to instill the Puritan exaltation of work and the virtues of self -help (cit. in Seekings 2008: 533) in maladjusted, jobless whites always tempted by idleness. During the 1920s and 1930s the states opposition to social assistance programs for the working-age, able-bodied unemployed resulted in a typically bifurcated social citizenship paradigm that continues to this day. On one hand, social insurance in the form of retirement, unemployment, and medical benefits pertained to employer-funded schemes or other contributory programs based on waged work. As such, it benefited permanent over temporary workers. In 1937, a new Unemployment Benefits Act introduced a state-funded unemployment insurance program that ended up providing short-term benefits to only 88,000 employees, a tiny fraction of which were high-income Africans (Meth and Piper 1984.) On the other hand, noncontributory, decommodified public social assistance was never intended as an alternative to wage labor as it did not cover employable workers unable to find occupations. Strict means tests, small budgets, and the definition of target populations as economically inactive but deserving poor attached a heavy stigma on noncontributory grants, which reinforced the symbolic association of waged work with virtuous citizenship. Government policies mostly helped white workers out of the economic depression of the 1930s. In 1943, only 4 percent of all public expenditure in social assistance was directed to Africans (van der Berg 1997: 487.) Soaring wartime manufacturing production, however, absorbed growing contingents of black workers, including low-skill Africans. The African working class enjoyed, as a result, rising wages and bargaining power. The government of Jan Smuts, purportedly a Keynesian sympathizer, became amenable to the idea that a living wage for long-term African employees could facilitate the stabilization of non-white urban labor and counter the threat of militant unionization. It was not a radical turn towards the welfare state, but 70

rather an approach dictated by expediency and hostility to black workers organizations; in no way it questioned institutionalized racial segregation (Nattrass 2005.) Smutss ideas departed, nonetheless, from Stallards natives as temporary urban sojourners, and placed the question of African access to housing and social provisions at the core of the political contestation between the ruling UP and the mostly Afrikaner National Party (NP) as they neared the national elections of 1948. A government-appointed Social Security Committee issued in September 1943 a report strongly influenced by Lord Beveridges ideas and the nascent British welfare state experiment that clearly departed from the conservative approach of the Carnegie commission. The report endorsed a social security system based on redistribution and decommodification to cover nonproductive periods of life (Union of SA 1944a: 6,) including unemployment in working age, to last as long as the need lasts (Union of SA 1944a: 23.) It went as far as to recommend elements of nonracial universal benefits: it proposed to extend noncontributory old age and disability pensions to urban Africans and advocated new contributory national programs for the elderly, the disabled, the unemployed, and family allowances. Africans were expected to constitute one quarter of all recipients (Union of SA 1944a: 36-47.) Such proposals still excluded residents of the reserves, reflecting the conventional view that [they] have shelter and can eke out an existence so that they do not need the elaborate cash benefits indispensable for a civilised community (Union of SA 1944a: 6.) The amount of benefits would have, moreover, discriminated by race since official stereotypes considered the average African as a person with essentially frugal needs. Only civilized Africans, defined by income, were supposed to claim the same benefits as Europeans (Seekings 2005: 46-47.) Despite such a persistently hierarchical imagination, the 1943 report on social security represented an important departure from the hegemonic, work-centered view of social citizenship. Its universalist overtones, moreover, tried 71

to legitimize waged work among the Africans not just in terms of an abstractly defined dignity, but on the concrete terrain of rights. At the same time, the caveats that limited citizenship to a stabilized, civilized urban condition served to align African aspirations and desires with waged employment. The 1944 Pension Laws Amendment Act extended old age pensions to African men and women, but otherwise the government quickly jettisoned the most far-reaching recommendations of the social security report, which faced strong opposition inside the Smuts cabinet itself. Large British companies, the Afrikaner middle class, white workers, and farmers regarded social provisions for blacks as an unsustainable fiscal burden and a magnet for further African migration from mining and agriculture to the cities (Meth and Piper 1984: 8.) The 1945 white paper on social security retained the proposal of expanding unemployment insurance for African workers, but curtailed projected funding to the point of eroding much of what Africans gained from inclusion in the old age pension program (Duncan 1995: 79.) The deracialization of state pensions for the elderly was, in the end, minimal: in 1948, only about 200,000 Africans could claim them (Iliffe 1987: 141.) Legislation passed in 1946 established the Unemployment Insurance Fund (UIF) as a program to cover short-term unemployment with contributions from employees, employers, and the state. It covered black and white employees, but only in permanent jobs, and excluded mineworkers and farmworkers (Duncan 1995: 79.) The NP of Daniel F. Malan, supported by Afrikaner workers and middle classes, opposed UIF coverage for Africans. Couched in traditional anti-dependency, work ethic, and individual responsibility arguments, the NPs campaign impacted on the governments decision to suspend the implementation of the UIF a year after its introduction. For the UP, the move was an attempt to defuse touchy controversies on deracialized social programs in view of hotly contested elections. 72

No better fate awaited the far-reaching recommendations in the 1945 report of the Gluckman commission, which proposed a national health insurance system to rein in escalating private medical costs and defined healthcare as a fiscally funded universal social right (Union of SA 1944b.) The ANC, black trade unions, and even some white workers organizations had strongly mobilized in support of a public healthcare system. Smuts, however, chose to shelve the Gluckman report, worried of the possible differences and divisions (cit. in Pillay 1995: 77) with doctors, businesses, and NP voters, all strongly opposed to redistributive solutions. Far from laying the foundation of a unique and exceptionally generous system of social citizenship, as Jeremy Seekings (2005: 50) wants his readers to believe, Sm utss reformism was soon aborted by its own government, which ultimately preferred time-honored violent methods to social legislation as a means to deal with black working-class militancy. The ferocious police repression of the 1946 African mineworkers strike showed how little the segregationist state was prepared to risk with piecemeal social measures designed to deflect, not accommodate, the claims of the disenfranchised. Despite the limits and contradictions of governmental action, liberal-leaning politicians like Margaret Ballinger and Smutss minister Jan H. Hofmeyr thought that the official discourse of social advancement for modern Africans could be a possibility for a more inclusive South Africanism (Dubow 2005.) White liberalism also deferred, however, to white fears of, in Hofmeyrs words, the black mans numerical superiority and the menace presented by the black mans lower standards of living. Equal citizenship, he continued, was ultimately synonymous with the mixture of the races, a revolting prospect (Hofmeyr 1936: 30.) Better was for him a restrained liberalism pursuing realistic goals and content to hasten slowly along paths knowledgeable experts indicated (Hofmeyr 1936: 30.) One of these was what 73

Hofmeyr termed constructive segregation, combining the recognition of the sociocultural chasm separating Africans and Europeans with policies to make sure that native reserves were economically adequate, but not autonomous enough to deny labour for the white man as a necessary element in the economic structure of Bantu life (Hofmeyr 1936: 33.) Their ambiguities notwithstanding, liberal views of an inclusive South Africanism clearly resonated in African nationalist opposition to segregation. Equal social rights and the deracialization of social programs were as integral as universal political citizenship to the ANCs expectations for postwar democratization.5 The partys 1943 African Claims explicitly framed black demands for healthcare, education, welfare, and landownership in the moderately progressive language of the Atlantic Charter adopted two years before (van Niekerk 2003: 36364.) The ANCs elites cultivated their own South Africanism, which seemed nonetheless to echo the ideas of the government and white liberals in its conviction that African urban labour must be stabilized.6 By stabilization the ANC surely meant something drastically different from its opponents, namely the abolition of the migrant labor system and African access to economic opportunities. Yet, well into the 1940s African nationalist leaders were still regarding workers rights not as the object of social conflict, but as a goal to be shared with white reformists for the sake of ending racial bitterness and antagonisms which are undermining all the ideals of South Africa, namely Democracy, Christianity, and human decency.7 The UPs wartime agenda of labor stabilization faced the dilemma, common to social reforms in other colonial governments, of balancing universalism and difference, inclusion and stratification, the recognition of rights and the disciplining of claims. Even before the rise of apartheid, however, it was clear that the white-ruled state would not relinquish its segregationist outlook or its framing of domination over African societies in terms of tutelage and trusteeship. 74

The 1948 report of the governments Fagan Commission declared that the natives were not to become citizens even as it explained that urban Africans on their way to stabilization were no longer natives, but workers entitled to appropriate benefits and living standards (Ashforth 1990: 132-39.) Once again, the states discourse of waged work alluded to universal citizenship rights while materially restricting rights to selected few in a society where employment inequalities cut through hierarchies of race, residence, and occupation. Compared to the ambiguities and uncertainties of the Smuts administration, the NPs program of apartheid (separateness) was an all-out offensive against proposals to relax racial segregation in jobs, social provisions, and residential rights. The NP waved in front of its low- and middle-income constituencies the threat of oorstroming (inundation) of white South Africa caused by black access to citizenship rights and welfare benefits. To Smutss elusive stabilization, Afrikaner nationalism opposed the appeal of whiteness as a condition facing an existential threat that only rigid racial segregation could repel. The NPs anti-welfare ideology of work was not a universalist device but rather boosted white South Africas imagined last stand, which involved crude stereotypes of blacks allegedly inbred dependence on state handouts. It also, however, exalted the virtues of employment across the racial spectrum, presenting waged work as beneficial and civilizing, and social programs as detrimental and dehumanizing, for whites and blacks alike. In 1946, J.G. Strijdom, a future prime minister, wondered: Is it not a fact that natives only work to supply their immediate wants, and if you grant them old age benefits and other benefits you would only make them lazy? . . . They only work when starvation stares them in the face. . . . There are a large

75

number of Europeans to whom that applies as well (cit. in Meth and Piper 1984: 9. My emphasis.) With the NPs victory in the 1948 elections, governmental practices shifted from South Africanist pseudo-civilizational discourse to Afrikaner nationalist views of essentialized racial, ethnic, and cultural distinctions. Continuities as well as differences, however, underlie such a symbolic turning point. The NP government inherited from its predecessor a work-centered imagination of social relations. Its translation into practices of citizenship, however, would bring traumatic ruptures for the disenfranchised majority.

Apartheid Social Engineering and the Coercive Enforcement of Wage Labor Discipline

To address the expectations of its working class constituencies, the NP government did not primarily resort to redistributive programs, which became increasingly residual and underfunded. Like previous administrations, it rather relied on whites-only jobs, union rights, training and education, and designed a new system of Bantu education to form Africans into low-wage employees (Seekings and Nattrass 2005: 128-141.) Import substitution industrialization also deepened state protection for domestic industries. By 1970, the contribution of manufacturing to the gross domestic product exceeded 30 percent, more than mining and agriculture combined (Feinstein 2005: 144, 180-84.) Economic interventionism complemented fiscal thrift and a revived laissez faire in social policy. The apartheid regime presided over a racialized welfare system where as whites moved upwards into protected high-wage jobs, middle-class incomes, and private benefits dwindling social spending provided meager programs for blacks, which mostly advantaged coloreds and 76

Indians, leaving very little to Africans. The stigma attached to government grants reinforced the racially-defined second-class citizenship of many black recipients. Or else, programs ceased altogether: in 1949 the government excluded, without opposition from the UP, low-income Africans from the UIF presented as a cause of idleness and higher unemployment with the result that the number of African recipients plunged from 140,000 to 1,500 (Meth and Piper 1984: 17.) Only in 1967 did average African wages reach the UIFs eligibility threshold (Nattrass and Seekings 2000: 15.) Since less whites needed by then unemployment insurance, the state also stopped contributing to the UIF, leading to a dramatic decline in payable benefits. In general, while white workers had access to meaningful protection from risk and retirement income, wage labor did not have the same function for Africans, but rather amplified their uncertainty: the 1956 Pensions Act excluded low-skill, migrant, contract, hourly- and weeklypaid employees, meaning the vast majority of black workers, from employer-based retirement coverage. By the mid-1980s, total retirement benefits for whites were, as a share of national income, three times the size of all public provisions paid to three times more numerous black recipients (Kruger 1992: 30.) Similar patterns were observable in private medical insurance. In 1962 the government confirmed its rejection of national public healthcare and, since private plans mostly admitted high-income whites, by 1989 Africans were less than 4 percent of recipients of employer-based medical coverage (Price and Tshazibane 1989.) Black workers, instead, crowded poorly equipped public facilities limited to primary and emergency care. The core principle of apartheid was separate development, enshrined in the Promotion of Bantu Self-Government Act of 1959: citizenship for the natives, now redefined as Bantu, was superseded by belonging to culturally defined and bureaucratically sanctioned tribes. African residential rights in white cities were concomitantly restricted. If the UP government 77

had imagined that wage labor could uplift natives to the full status of workers, the apartheid project reversed the process and turned the Bantu into migrants not only Stallards temporary sojourners, but actual non-citizens and foreign visitors. Native reserves were therefore reorganized into Bantu homelands, run by allegedly traditional African chiefs under the supervision of a state apparatus separate from the ordinary bureaucracy. Prime Minister Verwoerd compared white South Africa to a workplace where African workers had no rights to claim. For him the country outside the homelands was European-owned property where natives were allowed to stay just like labourers on a farm (cit. in Legassick 1974b: 20.) The majority of South Africans were destined to experience work and citizenship as disconnected and mutually excluding spatial entities. Social spending for Africans was overwhelmingly directed to the homelands. Apartheid social policies thus deepened the commitment of the former government to the betterment of African cheap labor reservoirs (Union of SA 1944c.) Conversely nonracial redistributive programs, which Smuts had already abandoned in his final years in power, became utterly marginal in the NPs agenda until the late 1970s. The new regime particularly disliked noncontributory grants covering Africans, like old age pensions. In the apartheid fantasy of replacing Africans formal citizenship with tribalized identities, the state aimed to downsize what was left of nonracial provisions for the elderly and, in the words of an official in 1955, to evolve a system whereby we reinstate the natural obligations of Bantu authorities and Bantu culture in regard to their old people (cit. in Seekings and Nattrass 2005: 133.) By the end of the 1960s, the highest amount of African old-age pensions was only 13 percent of average white ones (for Indians and coloreds it was 41 and 47 percent respectively.) By 1958 Africans were 58 percent of all recipients, but cashed only 19 percent of the benefits (van der Berg 1997: 486.) 78

To turn white South Africa into a precarious place of employment for African workers, influx control legislation regulated black urbanization and discouraged defections from brutally exploitative conditions in the mines and the farms. Section 10 of the 1952 Native Laws Amendment Act established that Africans could permanently live in an urban area only by birth or through continuous residence or employment. So-called Section 10 rights, therefore, excluded migrant workers, subject to renewing their employment contracts in their respective homelands. Permanent urban residents could instead aspire to relatively stable prospects, as the regime initially devised an urban labor preference policy aimed at prioritizing African township residents for recruitment into local jobs. New urban African townships were established following the Group Areas Act of 1950, which required municipalities to segregate neighborhoods according to the four official, racially defined population groups. The government forcibly deported Africans still living in mixed neighborhoods to new model townships, as their planners styled them. Publicly funded housing projects in the townships were also intended to replace old native locations, which, despite their poverty and squalor, had often been vibrant places of labor and political activism. An important aim of the urban labor preference was to ensure the supply of contract migrant workers for mining and agriculture. The Bantu Labour Act of 1964 organized a dedicated infrastructure to hire African migrants by establishing labor bureaus in the homelands to act as intermediaries between local governments and households providing labor, and white employers requesting it. Individual reference books, or dompas, certified the employment and residential status of Africans, subjecting violators to arrest and deportation to their putative homelands. The Bantu Labour Act and subsequent government circulars, however, threatened with deportation also African permanent urban residents refusing waged 79

employment (Terreblanche 2002: 322.) The act was, finally, a cornerstone in the gendered geography of apartheid as it curtailed the ability of African women to gain permanent urban residence, even when trying to join male relatives. It was another attempt to confine women to reproduction in the homelands or specific urban jobs like nursing or domestic service. In 1970, only 3.6 percent of employed women had manufacturing jobs (Berger 1992: 227, 252.) By 1981, less than 15 percent of all manufacturing employees were black females, which mostly worked in historically highly feminized sectors, as in the case of colored women in the Capes clothing and textile companies (Berger 1992: 254.) Deborah Posel (2005) has described apartheid social policies as a racial modernist project replacing, in Foucauldian terms, the biopolitics of the self of the Smuts era, centered on individual work discipline, with a biopolitics of the population concerned with the spatial control and planning of aggregates and flows. Despite its accentuation of racial divides, the NP regime did not repudiate the idea of work as a force of progress and respectability for modern African individuals. The model township was a laboratory where public amenities, healthcare facilities, infrastructures, schools, and social workers ideally allowed the reproduction of a disciplined African working class in place of the unruly, unhealthy, potentially subversive lumpenproletariat of the old locations. To be a permanent township resident meant having a regular job and a family, while fixed-term migrant workers lived in separate, tightly policed, ethnically segregated, single-sex hostels. The township was therefore, Posel continues, a point where work and urban family norms intersected and mutually reinforced as imagined institutions of social stability. If the average African male was no longer deemed to be a fully fledged worker, let alone a citizen, urban Africans were nonetheless expected to behave as responsible employees and heads of families to match their residential status. The states focus on the family 80

responsibilities of African workers conveniently sidetracked claims for welfare handouts. It also gained, however, the support of African community leaders and elders concerned with controlling loose women and anarchic youth. Apartheid policy discourse fantasized that boundaries drawn through socio-spatial planning could flawlessly translate into objectively defined human collectivities and scientifically predictable individual conducts. Classic scholarly critiques have, paradoxically, reinforced such a functionalist understanding of the system. Structural Marxist authors looked at the apartheid labor regime mainly as a device to compress African workers wages and claims by offloading part of their costs of reproduction onto what was left of the subsistence economies of the homelands (Wolpe 1972; Legassick 1974a.) Subsequent reassessments argued that the collapse of the homelands made them unsuitable for reproduction, and saw influx control legislation as a mechanism to channel labor flows into different economic sectors, limit African urbanization, and stabilize skilled black workers in the cities (Hindson 1987.) More recent studies have, however, de-emphasized the states capacity to plan and harness social dynamics, and rather saw apartheids institutional interventions as reactions, often incomplete and contradictory, to ordinary peoples strategies and subjectivities. For Yann Moulier-Boutang (1998: 640-44) influx control was a response to African workers attempts to refuse work in the mines and the farms and flee towards better living conditions in the cities. Such observations resonate in the poignant verses of worker poet Alfred T. Qabula (1984: 49) evoking his escape from Carleton mine, a place of suffering, with its compounds, its violence, its homosexuality, a place crawling with the spirits of unappeased dead miners and workers. The place of gold, dagga [marijuana,] drink and oppression. Spatial segregation, Moulier-Boutang continues, translated therefore the hierarchical social ordering of work into the juridical categories of waged urban 81

employment, semi-servile mining and agricultural labor, and homeland subsistence economy. The wages of black urban employees were precarious not because complementary income was available from the homelands, but because black workers in the mines and the farms were even more insecure, unprotected, and unfree, and regarded competition for low-wage urban jobs as an uplifting prospect. The aim of influx control was not to stabilize permanent urban workers, but to pitch different exploited groups against each other and generalize black precariousness across the occupational spectrum. Refusal of work was not, however, limited to mining and agriculture, but was widespread in the black townships as well. Posel (1991: 82-90; 158-64) powerfully showed how the avoidance of factory jobs by township youth doomed the states urban labor preference to collapse within the first decade of apartheid. In 1962, the report of the governments Botha Commission raised the alarm that young labor market entrants were refusing waged occupations and preferred what the horrified writers described as idleness, parasitism, crime, and vice aided by unwaged support networks. Echoing the expert knowledge of the time, the report made social maladjustment and moral pathology two sides of the same coin: By the time they reach working age, they have either developed into a type that refuses to work, or by virtue of their instability and untrustworthiness, have become unemployable (cit. in Seekings and Nattrass 2005: 170.)8 Few years earlier, the manager of an urban compound for migrant workers had similarly complained: The detribalized group has today become a problem. He (sic) is the young, semieducated, arrogant, demanding and wont-work type. He is difficult to handle because he is very prone to disobedience and has, invariably, no inclination to work unless forced to do so (cit. in Posel 1993: 420.) 82

Section 10 rights, Posel (1993: 420) concludes, were intended to create a subservient and disciplined urban African workforce, but they ironically ended up being used as a weapon to resist exploitative labor. The provisions based on the 1964 Bantu Labour Act allowing for deportation from the city to punish work avoidance had little effect and could not save the urban labor preference. In the end, urban employers increasingly resorted to recruiting migrants, reputedly more compliant as they depended on the employment contract for permission to reside in the township. Migrant labor, therefore, was not just a tool of the state to make black workers cheap, but came to occupy a central place in the urban landscape of apartheid due to African grassroots subversion of wage labor discipline. The unionization of migrant workers in the 1970s would bring such unintended consequences to haunt apartheid dreams of social stability. In response to the apartheid states radical attempt to decouple work and citizenship in African lives, opposition movements elaborated a narrative that placed waged work at the core of resistance and social redemption. By the late 1940s the ANC was no longer led by moderate elites but had become a mass organization advocating popular mobilization and civil disobedience for equal rights, deracialized citizenship, universal social provisions, resource redistribution, and the nationalization of strategic assets. Yet, the ANC did not abandon its longstanding emphasis on self-reliance, entrepreneurialism, and independent economic initiative. Rather than preaching resistance to proletarianization, it imagined wage labor, once juridically free and unfettered by racial domination, as a condition of individual empowerment, even without a social transition beyond capitalism (Cobley 1990: 170-72.) The partys 1949 landmark Programme of Action combined claims to social inclusion and political representation with a powerful emphasis on economic rights. Establishing national industries and enterprises were there on a par with the struggle for workers protections.9 83

ANC activists were also increasingly involved in labor organizing. The ANC-allied South African Congress of Trade Unions (SACTU,) established in 1955, was as active in workplace battles over wages and working conditions as in campaigning for the ANCs program of political change. For the Communist Party of South Africa (CPSA) the black labor movement was, moreover, a channel to influence the analyses and strategies of the ANC. The alliance between the two organizations was for the CPSA the outcome of an ideological trajectory started in the mid-1930s with the adoption of the Third Internationals line of the popular fronts. The CPSA saw the trade unions as vehicles to build a nonracial alliance of black and white workers, turning wage labor from a reality of racial division into a horizon of popular unity. Communists, however, did not only praise wage labor as a potential outlet of revolutionary politics, but also described it as a pedagogical and moral force in terms that reflected the utterances of social reformers or bourgeois nationalist elites. Official statements of the CPSA praised working-class organizations as barriers to the increase of delinquency amongst the youth, especially the nonEuropean youth, whose instincts could be more productively directed towards the struggle for work.10 The CPSA was banned in 1950, and had to briefly dissolve in a climate of heightened repression. Communist activists, however, kept working in the ANC, ANC-aligned organizations, and the labor movement, and they greatly influenced the spirit and the wording of the 1955 Freedom Charter, the historic program of the ANC-led Congress of the People (Lodge 1983: 69-74.) The Charters demands for work and security included universal social provisions such as unemployment insurance, sick leaves, maternity leaves, and a national minimum wage to reward the right and duty of all to work. Eventually, as Cobley (1990: 200 01) suggests, even if the CPSA conveyed progressive and redistributive ideas to the middle-class leadership of the ANC, this latters values and intellectual orientations continued to extol 84

individual opportunities and economic activity, to which the CPSA provided solace through its Stalinist glorification of hard work. The violent repression of the anti-apartheid opposition, especially after the 1960 Sharpeville massacre, drove the ANC into the underground and in exile and marked the virtual collapse of black trade unionism for a decade. The rebirth of an independent black labor movement, discussed in the next section, responded to and amplified the socioeconomic crisis of apartheid and the contradictions of its labor regime. In time, resurgent black workers organizations would produce their own redemptive narrative of waged work, creatively connecting to the legacies of older movements.

Black Workers Struggles and the Redemption of Wage Labor, 1973-1994

The 1970s were for the apartheid regime years of economic, social, and political instability. Growing international isolation, the decolonization of Southern Africa, and the global energy crisis externally impacted on a withering mode of accumulation reliant on foreign commodity markets, imported capital goods, and a domestic demand mostly buoyed by a small white minority (Fine and Rustomjee 1996.) Black cheap labor had for long boosted the profits of white companies, but decades of inferior education and training and low wages enforced through state repression had also suffocated the contribution of non-white South Africans to capitalist growth both as consumers and as much needed skilled workers. Blacks, of course, suffered the most from economic crisis, as rising inflation and collapsing living standards deepened poverty levels in urban townships and rural homelands. In 1973, a strike wave propagating from factories in the Durban area began the resurgence of 85

independent black trade unionism after a decade of repression. The Durban strikes were not politically motivated and early unionization mostly addressed, even when surviving antiapartheid activist networks were involved, bread-and-butter concerns. Workers did not see the decision to join a union as an ideological statement, but as a response to a constant degradation of employment that, even before 1973, had become intolerable. In 1977 the government appointed the Riekert commission to investigate problems with manpower utilization and labor control. The commission heard employers complaints that young African township residents refused to work in the factories and used their Section 10 protections to be more choosy and work-shy (RSA 1979: 169) than migrants. The commissions report commented that a permanent urban resident who did not face the risk of deportation would rather remain without work than fill a vacancy not to his [sic] liking (RSA 1979: 169.) The document blamed a township subculture of prostitution, gambling, and smuggling for providing township youth with resources to escape wage labor, forcing companies to rely on older employees from the homelands, depicted as more precarious, therefore more reliable. Compliance is not, however, to be confused with consent or satisfaction. For migrant workers the contract of employment was necessary to legally reside in the city, but many shared the feelings one voiced in an interview (Webster 1985: 272): I would run too if I had somewhere to run to. Social studies of black labor struggles in the 1970s and 1980s usually see trade unions as distinct collectivities manifesting a superior form of consciousness in a drastic rupture with the lull of the 1960s (Friedman 1987; Baskin 1991; Seidman 1994.) Such analyses largely neglect the connections and continuities between black workers organizations and rich traditions of resistance to and subversion of wage labor. The refusal of work by township youth was, after all, decisive in companies decisions to hire migrant workers. Migrants, conversely, provided most 86

members to the new unions, like the Metal and Allied Workers Union (MAWU,) initially concentrated in urban manufacturing industries. Mandlenkosi Makhoba joined MAWU while working in a foundry in the East Rand, where he had moved from Mahlabatini, in rural KwaZulu. His published memories revealed how the experience of the factory gave a new concreteness and visibility to oppression. As economic necessity made escape impossible, the union provided a rare opportunity for autonomy and emancipation: We had no voice. We could not make our complaints heard. That makes a person very angry. And so my fellow workers and I decided to do something about this. We began to stand together and build workers strength and unity in the foundry. Our combined strength would be a very loud and confident voice (Makhoba 1984: 21.) According to early surveys of independent unionism, the radicalism of workers wage demands served to indicate more the extent of their rejection of the system than the nature of their realistic expectations (Fisher 1978: 215.) Through militant industrial actions migrant workers rebuffed the image of backwardness, docility, and subservience that the gaze of the state and capital impressed upon them (Webster 1985: 272.) Interviews conducted in the early 1970s revealed that strikes were not just weapons for tangible gains or the power of an abstractly defined proletariat, but acts to restore the humanity of workers against waged work as a daily reality of violation and abuse (Webster and Kuzwayo 1975; Webster 1979.) Such sentiments were not confined to the workplace: MAWU members cited the fight against job losses and the need for higher wages as main reasons to organize,11 but they refused to see the union as a purely productivist organization and supported, for example, the unionization of unemployed workers.12

87

Class identities provided migrant workers with a language to name domination and resistance in the workplace, but they also grafted the discourse of union membership onto complex pre-existing cultural formations infused with rural legacies of resistance to proletarianization (Sitas 1983; Bonnin 1987.) Christian and millenarian themes of popular justice loomed large in the migrants militant, populist workerism where the church of the union (Sitas 1983: 417-18) was the harbinger of the new day. Workers similarly signified their unions claims through moral economies that retained vital connections with non-market forms of income generation and sharing (Ngoasheng 1989.) Despite its economic decline, family homeland agriculture still buffered migrants against the insecurities of wage earning. To successfully organize migrants, the unions had to be not only effective in the workplace, but also capable of wresting workers identities from ethnic loyalties towards elderly conservative leaders, homeland chiefs, and African company supervisors, or izinduna (Sitas 1983.) The independent unions did not dismiss the migrants discursive formations and rural networks, but creatively harnessed their underlying imagination of justice to present workers organizing as its extension and incarnation in the urban environment. The unions critical commentaries of the wage relation used oral traditions enacted by izimbongi poets (Bonnin 1994,) which instantiated workers claims to control over their daily existence. Rather than operating as an activist in the tradition of left vanguardism, the unions shop steward had to perform as an imbongi to become conversant with a deep-seated grassroots deprecation of waged employment. The worker poet Alfred Qabula compared his MAWU membership to his familys exodus to the forest during the 1960 Pondoland rebellion. The union represented a similar travel to a place sheltered from the factorys harms (Qabula 1989: 68-69.) The success of labor organizations greatly depended on their capacity to re-elaborate the past to make sense of present struggles. In the migrants hostels, 88

the unions moral umthetho (law) articulated collective solidarity because it resonated more powerfully with a pre-capitalist sense of dignity than old ethnic identities by then deeply compromised with the apartheid labor regime. At the same time, the independent unions were laboratories for activists and intellectuals to recast wage labor as a signifier of social emancipation. White left academics and students had an important role in the resurgent labor movement. The wages commissions, established in 1971 by liberal and Marxist members of the National Union of South African Students (NUSAS,) helped start in 1974 the Durban-based Trade Union Advisory and Co-coordinating Council (TUACC,) to which MAWU belonged.13 Dissidents of established white unions and survivors from the SACTU era started a benefit fund assisting TUACC in launching other unions. TUACC unions immediately adopted a policy of worker control and elected leadership of production employees. White officials often controlled, however, appointed positions, including organizers and general secretaries. In the early years, MAWU and other organizations witnessed tensions between militant rank-and-file members and wages commissions activists who supported benefit societies arguing that black workers were not ready for shopfloor unions.14 Most white unionists eventually went along with African workers support for unionization, but often retained the impression that modern proletarian characteristics were at odds with backward migrant workers values (Hemson 1978.) White cadres rejected analyses of oppression and social antagonism that prioritized racial dynamics. For them, organizing at the point of production fostered inclusive identities because they identify capitalism, rather than whiteness, as the adversary. Status privileges and educational background gave nonetheless whites a disproportionate presence in leadership positions, and greatly conditioned the decision by the early unions not to participate in broader social and political struggles, which, particularly 89

after the 1976 Soweto revolt, white officials regarded as unrealistic in a context of intensified repression. Most emerging unions endorsed a nonracial line where, to quote from a document of the time, most black South Africans are workers. We believe, therefore, that to understand the problems facing black South Africans we must begin with the labour situation. It is the situation in which there is the greatest potential for forging new organisations through which blacks can reclaim their human dignity (cit. in Theron 2007: 3.)15 In this view, only by embracing the identity of waged workers could blacks start to distinguish the early glimpses of freedom, whereas claims to racial justice would be, instead, reactionary and divisive. Not only did the language of white union leaders relegate refusal of work to backward pre-capitalist cultural formations, but it represented wage labor as the true condition of selfrealization for politically-minded human beings.16 The wages commissions deployed the neutral language of the social sciences, in the form of studies on poverty and inflation, to articulate African workers demands in the nonracial terminology of an imagined universal working class (Davie 2007.)17 Richard Turner (1978,) a philosopher assassinated by apartheid operatives in 1978, argued in his very influential analysis infused with Sartrean existentialism that nothing like production struggles and shopfloor unions could school black workers into the emancipative values of nonracial, participative democracy.18 Activists from the black consciousness tradition also worked with early labor organizations, like the Black Allied Workers Union (BAWU.) Despite their limited success, they criticized TUACCs white leaders for obscuring the relevance of racial privilege and oppression behind the ideological screen of a nonexistent commonality of interests between black and white workers (Molamu 2002.) The black consciousness view of 90

anti-racist struggle emphasized blackness as a modality of desire, assertiveness, and revolutionary organization. It did not, however, discard class as a category to understand black oppression and the intersections of capitalism and institutional racism. White TUACC officials, conversely, used class analysis as the bridge between a normative universalism of work and a nonracial imagination that limited the significance of vastly different material conditions and opportunities between them and black union members (Buhlungu 2006.) Nonracial independent unionism consolidated eventually its hegemony with the merger, in 1979, of TUACC and other local organizations to form the 40,000-strong Federation of South African Trade Unions (FOSATU.) The new federation became the bulwark of industrial unionism, plant-based organizing, a class discourse of worker control, and labors independence from political organizations. South Africas white establishment eventually chose to address the growth of the independent unions by resuming the idea, abandoned in Smutss time, of stabilizing urban African workers through labor and social reform. Major industrialists no longer desired a rigidly racialized labor regime. Harry Oppenheimer, at the helm of the giant Anglo-American Corporation, spoke for many when he declared that cheap untrained, undifferentiated migrant labour (cit. in Seidman 1994: 130) was no longer conducive to economic growth. In an isolated country facing turmoil and in need to reduce its dependence on imported technology, companies wanted social stability, black middle-class consumption, and the removal of racial barriers to training and skills. Capitalist interests, through think-tanks like the Urban Foundation, claimed autonomy from the embrace of the state in their demands for deracialized urban labor markets, better housing and living conditions in the townships, and opportunities for black private

91

enterprises. They did not go, however, as far as advocating equal political citizenship; autonomy from the state was not meant to become confrontation with the apartheid system. The government eventually envisaged a comprehensive reformist agenda. The report of the Wiehahn Commission, published in six volumes between 1979 and 1981, recommended that African workers be legally defined as employees, which would allow them to belong to unions registered for the purposes of collective bargaining. The Industrial Conciliation Amendment Act of 1979 and the Labour Relations Amendment Act of 1981 followed such recommendations, even if migrants had to fight to access collective bargaining rights that the legislation initially intended only for permanent workers. Statutory racial job reservations were also abolished, but the new laws forbade union support for political demands and unprocedural strikes. In the intention of the government the native had finally become a worker, but at the price of relinquishing any further claims to citizenship. The Riekert report, released in 1979, extended labor reforms to an ambitious view of socio-spatial restructuring. Its technocratic phrasing represented African urbanization as an inexorable process driven by quasi-hydraulic dynamics of population flows that the state could scientifically harness for optimal efficiency (Ashforth 1990: 204-05.) The report argued that bureaucratically rigid racial categories were no longer economically functional and imagined to replace them with free markets as more legitimate determinants of social stratification. The hierarchical principle shaping the urban space had consequently to be no longer race, but class. Africans with regular waged employment would become urban insiders, with rights of residence, property, and business regardless to Section 10 provisions. Those not needed by city-based employers, or without jobs and accommodation, would be outsiders liable of expulsion and deportation. The report advocated gradations of citizenship for Africans based on the employment contract rather than on formal discrimination 92

by race or gender. It therefore partook of a liberal imagery of social differentiation created by the operation of economic forces that the state merely ratified. Hard work and productive employment laid therefore in official discourse the avenue for everyone to claim specific citizenship rights. In a paradoxical similarity with many in the white left, the government used nonracial arguments to resurrect the universal normativity of work, albeit not for purposes of equality and solidarity but to legitimize inequalities and hierarchies. Particularly notable was the way in which the Riekert report dealt with refusal of work by township youth, as it anticipated the economic liberalization of the 1980s and 1990s. Some employers had recommended the privatization of municipal township housing, the reduction of government grants, and new restrictions on unemployment insurance to increase African workers sense of responsibility and prevent their defection from employment. The report commented that there is some merit in the argument . . . that urban blacks should themselves bear a greater share of the total financial burden in respect of housing, services and transport so that they could hardly afford to stay unemployed (RSA 1979: 169.) The commission hoped therefore that economic necessity and market-based policies would succeed in imposing the discipline of work where state coercion had failed. The Wiehahn and Riekert reports expressed a quest for political legitimacy and stability that delegated the solution of South Africas social question to corporate capital. Their constant emphasis on free labor and free enterprise offered black workers a model of industrial citizenship requiring responsible and cooperative trade unions. Tamed and depoliticized working-class identities would also help confine claims to the productive realm, instead of spilling into civil, political, and social conflict. The governmental imagination of universal human beings defined by their employment situation, and not by racial, sexual, or cultural 93

criteria, placed work as the only permissible avenue to citizenship and an imagined, atomized individual imbued with market rationality as its sole beneficiary. A 1975 report on retirement pensions was straightforward: The government is opposed to the Republics developing into a welfare state as it is understood and manifested itself in other parts of the world, but it is committed to a policy which is aimed at the independence of the individual and of the community (cit. in Anonymous 1981: 37.) Social policies after Wiehahn and Riekert made the labor market central in determining access to housing and social provisions. As unemployment rose and occupations were more unstable, they also increased the insecurity of black workers across the board. Generalized precariousness aided by market liberalization provided the state with a new modality of labor control. The program of orderly urbanization, announced in 1985, led to the abolition of influx control and passbooks the following year. African urban residential rights became dependent on formal housing and employment, but the border between urban insiders and outsiders which included casual workers, migrants, and squatters was fluid and porous. The government started to privatize and re-commodify (Mabin and Parnell 1984) municipal council housing by converting residents rents into mortgage payments.19 State rhetoric was now preaching homeownership, self-help, and financial responsibility to African families, but mortgages and utility fees in privatized housing tied residents to their workplace, exposing the jobless to the risk of default, eviction, and the loss of residential status. Decades of underfunding for urban accommodations and infrastructures had created dramatic housing shortages, and many evicted families moved to unauthorized rental shacks in the backyards of formal dwellings, or became squatters in urban outskirts. After the abolition of influx control, growing waves of impoverished migrants also 94

moved from the rural areas to urban informal settlements. Employers saw in the governments policy of labor discipline through generalized insecurity a possibility to recruit a highly flexible workforce. The powerful Steel and Engineering Industries Federation of South Africa (SEIFSA,) for example, petitioned the government to allow, despite the institutional aims of putting township residents into formal jobs, shorter migrant workers contracts (Kraak 1993: 18.) After prolonged debates, FOSATU unions chose to exploit the opportunities provided by the Wiehahn reforms and registered in terms of the new legislation. By gaining legal recognition, they could strengthen their workplace presence but, to the governments chagrin, FOSATUs participation in the statutory industrial relations system did not tame labor mobilization. Fresh institutional recognitions allowed indeed black workers organizations to extend their claims from production to reproduction and citizenship. Trade unions with different ideologies shared a view of the working class as the main agent of social transformation. In the 1980s, however, intense strategic discussions arose on how far the unions involvement in political resistance should go. Unions depicted by their opponents as workerist, mostly FOSATU affiliates, had a guarded attitude towards the ANC and were afraid of becoming mere appendages to nationalist movements in exile. FOSATU defined its struggle mostly in class terms and suspected domestic political antiapartheid organizations of reproducing the class domination of bourgeois elites, which moreover did not disdain to take advantage of the business opportunities in apartheid reforms (Foster 1982.) FOSATU held, in Ivor Chipkins (2007: 88) words, a proletarian rather than popular image of the working class and warned that unless labors political organizations were fully independent from the liberation movement, they would merely abolish the legal structure of apartheid while subordinating workers to a new majority-based nationalist regime.

95

For a younger group of unions, which critics branded populist or community unions, political domination shaped the conditions of the black working class as much as workplace exploitation. They advocated, therefore, labors involvement in national liberation politics and participation in community struggles and alliances with local movements or civics. Prominent among them was the South African Allied Workers Union (SAAWU,) started in 1979 and mostly active in the East London area and the neighboring Ciskei homeland. SAAWU defined itself as a general and nonracial union, even if it originated from the dissolution of the black consciousness-orientated Black Allied Workers Union. Rather than confining its activities to the workplace, it also mobilized through township mass meetings, which for its leaders provided the magnetism necessary to galvanize workers in their social milieu (Bonnin et al. 1996: 157-58.) SAAWUs leader, Sisa Njikelana (1984: 30,) explained that black workers were not only union members, but also shared national oppression and its resultant limitations, insecurity and deprivation with every class and group within the black community, including black business owners despite their position as exploiters. Since precariousness is a defining condition of black South Africans lives, and the workplace is a central institution of social insecurity, SAAWU saw the boundaries of production as impediments to labor emancipation. As an official argued: Transport, rents to be paid, are also worker issues. I see SAAWU as a trade union: theres no doubt about it. The problems of the work-place go outside the workplace. If you are underpaid it goes at home or the community (cit. in Maree 1982: 38.) SAAWU, like most community unions, joined the ANC-aligned United Democratic Front, founded in 1983, and adopted the Freedom Charter as a set of political demands (Baskin 1991: 96

91-108.) The UDF paid homage to the leading role of the working class in the struggle, but its leaders, like Terror Lekota, echoed the ANCs moral narrative of wage labor as selfless, ascetic dedication to liberation in an ecumenical national project purified of class resentment: It is my opinion that, in the final analysis, what the workers want is not riches as such. What they want is the worth of their labour and a fair share of the wealth of their country. They want to be able to live like human beings and not like groveling slaves. They are not in search of fantastic sums of money for which they have no practical use. They lack that poisonous yearning of some people to accumulate millions of rands with the sole purpose of decorating themselves. What they want is the right to work, in conditions which are conducive for their health and for wages that will enable them to provide for their families (Lekota 1986: 203.) In the deepening social crisis of the 1980s, FOSATU and community unions alike were involved in struggles over housing, municipal services, and social benefits. In 1981 FOSATU successfully struck against the governments plan to enforce the compulsory preservation of retirement contributions, which would have made impossible for workers to cash benefits from provident funds in the event of job losses (Roux 1981.) For precarious employees accumulated benefits are crucial to surviving between occupations as unemployment insurance provides only scant protection for a limited time. For the government, however, the withdrawal of benefits was a potential burden on the state (RSA 1966: 24) as it could erode retirement provisions and increase claims for noncontributory, fiscally funded old-age pensions. The 1981 strikes were initiated by community unions, and when FOSATU joined them it mobilized community support (FOSATU 1981.) The campaign had powerful redistributive themes as it demanded worker 97

control of pension funds and their investment into socially desirable areas,20 rather than in companies supplying the apartheid state. FOSATUs new emphasis on issues of social reproduction was also evident in struggles over maternity rights and benefits as the growing casualization of employment affected the rising number of black women in low-end, insecure manufacturing and service jobs.21 By the mid-1980s, the members of many FOSATU unions had pushed their otherwise reluctant leaders to enter the fray of social conflict and antiapartheid political mobilization. The federation endorsed campaigns against orderly urbanization, residential evictions, rising rents, and the continuous denial of black political rights. Labors redemptive discourse found its eventual outlet in the battle for equality and decent living standards. The aspiring citizen articulated by union struggles challenged the production-focused worker imagined by the state. Workers narratives of resistance to the eviction of illegal urban dwellings revealed how union solidarity addressed broader social instability while questioning the appalling decay of township housing. A MAWU shop steward recommended as a political consideration22 to define as shacks all African accommodation because overcrowding and physical collapse blurred all distinction between formal and informal housing. Another added: Now recession has cursed our one time lovable and acceptable houses to ukhukhu [chicken coops], but as long as the workers live and organise their house will play the tune and capitalists will have to listen to and dance. Hope to see you again, our dear mjondolo [dwelling].23 A third shop steward elaborated on the political implications of local livelihood struggles: This issue is not concentrated on shacks only. Its a broad issue. Its like an octopus, its got tendons, it reaches far. It covers, amongst others, community 98

councilors and the council, their so-called houses. It covers the private sector and covers the government. It covers you and me.24 Even if MAWU initially regarded itself primarily as a workplace union, it eventually had to recognize the political implications of occupational and social instability among its members, migrants and permanent township dwellers alike. The organizations charismatic leader, Moses Mayekiso, concluded that at that point MAWU was no longer a union . . . its a movement of workers fighting for liberation (cit. in Sitas 1985: 42.) Scholars have defined the involvement of organized labor in social and political struggles as political unionism or social movement unionism (Lambert and Webster 1988.) For Gay Seidman (1994) social movement unionism was the logical outcome of essentially structural circumstances. In her view the combination of authoritarianism, industrialization, and the urbanization of a semi-skilled African proletariat forced shopfloor unions to evolve towards a politics of citizenship and demands of participation in political decisions but also access to social and economic resources (Seidman 1994: 40.) The argument regards the unions convergence towards the positions of the ANC as linear and unproblematic. It therefore greatly simplifies actual processes of labor politicization and their ideological dynamics. Von Holdt (2002) warned that social movement unionism is shorthand for highly specific and uncertain processes enabled by a unique intersection of state racism, workplace despotism, and labor identities shaped by highly complex popular and non-class themes. The unions political profile was often the product of contingent leadership responses to grassroots expectations for social radicalism. Labors involvement in community politics was also highly situational and dependent on local interactions, rather than reflecting the impending hegemony of a nationalist political agenda or, as Seidman has it, fully fledged programs for citizenship and democracy. The 99

forces driving the unions beyond the workplace and into social conflict had rather to do with the challenges of daily survival, the threats of layoffs, the fragility of housing conditions, and rankand-file quests for a more decent, less abusive life. In important and strategic ways, however, the unions presence in social movement politics questioned the image of the nonracial, universal worker dear to labor leaders and intellectuals in the FOSATU tradition. Once the struggle moved outside the workplace, the abstractly normative identities of the employee and the union member ceded ground to the multiplicity and concreteness of workers desires. In many FOSATU affiliates, black workers yearning for radical politics explicitly attacked white leaders and cadres more attached to a production-centered meaning of labor activism. In Port Elizabeth, labor and civic organizations denounced FOSATUs autoworkers union and its white and colored leaders for denying support to civic groups facing government repression (Southall 1985.) In 1983 MAWUs branch in the strategic East Rand region split and the newly established United Mining, Metal and Allied Workers of South Africa (UMMAWOSA) lambasted the bureaucratic dominance of white intellectuals stifling members demands to join community struggles against evictions.25 The 1982 FOSATU congress confirmed the federations policy of nonalignment with political organizations, but delegates from the Paper, Wood and Allied Workers Union (PWAWU) caused controversy with their position that workers are Africans first.26 The leaders of another affiliate, the Sweet, Food and Allied Workers Union (SFAWU) argued that only through social and political engagement could labor organizations lead movements where union members were a small minority.27 Failure to do so, SFAWU continued, would doom organized labor to irrelevance, similarly to European and American trade unions, which while seemingly militant,

100

inevitably end up mediating conflict between capital and labour, and leaving the political struggles of workers to political parties that are not controlled by workers.28 The concept of social movement unionism provided labor leaders and intellectuals with a unifying narrative to rationalize and make sense of highly localized and contested processes. It did so, however, by eliding and obscuring the rank-and-file criticism of the universalist imagination of work widely held in the top echelons of major unions. Proponents of social movement unionism tried to paper over rifts and potential splits in FOSATU by closing ranks around the ideology of working-class nonracialism. In 1983, FOSATUs secretary general, Alec Erwin, delivered a paper that endorsed labor-community alliances as a way to minimize potential divisions along opposing political loyalties and reject conventional views of FOSATU as a federation staunchly folded on the workplace.29 He envisaged labor as part of a popular front against the regime, while remaining alert to the presence of class enemies in the community groups. Had FOSATU insulated itself from social struggles, he concluded, it would have been at risk of destroying the slow and hard work of building up a national trade union movement, particularly where gigantic achievements of non-racial solidarity have been won. In the mid-1980s the economic and political crisis of apartheid entered its terminal phase. The official rate of unemployment, which probably underestimated the actual figure, shot up to one quarter of the economically active population. The NP government adopted a course of economic liberalization and privatization that decisively departed from classic apartheid protectionism and intervention. Companies adapted by trying to regain profit margins and tame labor militancy through layoffs, automation, and more casual employment. Trade unions of all tendencies were by then regularly taking part in social protests that in many areas, such as the Eastern Cape and the core industrial region of the PWV (Pretoria-Witwatersrand-Vereeniging,) 101

escalated into full-blown township insurgency. In September 1984, the Vaal Triangle, south of Johannesburg, saw the beginning of a movement to boycott utilities and mortgage payments, which by May 1986 had reached Soweto and the East Rand (Chaskalson, Jochelson, and Seekings 1987.) The UDF stepped up campaigns of peoples power aimed to make townships ungovernable, to which the regime responded with intensified repression and virtually uninterrupted states of emergency. In 1985, at the conclusion of a long merger process, most of the countrys main union bodies, including FOSATU and SAAWU, gathered to form the 450,000-strong Congress of South African Trade Unions (COSATU.) The new federation aimed to transcend old divides between workerists and populists and provided a middle ground combining the leadership of production workers in the FOSATU tradition with the explicit political alignment alongside the ANC-led national liberation movement (Baskin 1991.) Defining itself as a socialist force to advance working class interests in the struggle for democracy, COSATU adopted in 1987 the Freedom Charter as a guiding document. The labor federation was on its way t o a formal alliance, officially announced in 1990, with the ANC and the SACP, as the Communist party had clandestinely reconstituted itself. The birth of COSATU was a success for the leadership project of social movement unionism. The federation and its affiliates like the new National Union of Metalworkers of South Africa (NUMSA,) into which MAWU merged in 1987 joined political protests and mobilizations opposed to the privatization and commodification of housing and services. COSATUs social agenda was, however, mostly focused on defending its members from the devastation wrought by unemployment and economic decline. While it channeled its radicalism into demands for political change and avowed socialist rhetoric, COSATUs social movement 102

unionism attuned labors imagination of class struggle to the imperatives of economic recovery and employment growth. Therefore COSATU inherited FOSATUs demands for a living wage, which the latter defined, however, as what workers themselves decide they require to live [and] what workers are able to get from their managements,30 while the former mainly translated into job creation.31 The living wage campaign did not result in any sustained mobilization, and was in the end marginalized to a mostly symbolic status. COSATUs demands for universal living unemployment benefits were also comparatively subdued, similarly to its short-lived attempt to organize unemployed workers.32 COSATU, therefore, maintained commitments to redeem wage labor, but it bent them towards aspirations of socioeconomic reconstruction under a future ANC-led government, which muffled criticisms of waged work and productivism. The redemption of wage labor and its subversion, once intimately joined, were now increasingly disconnected. Conversely, the social movements of the 1980s drew inspiration from the successes of trade unions, but they also radicalized workers practices and demands in the shopfloor. In response, COSATU-aligned scholars, like Karl von Holdt (2003: 114,) bemoaned that the unruliness and disorder of township youth politics spilled over into the workplace. As apartheid drew to an end, labor activists were lured into an imagination of work as the repository of disciplined identities, national renewal, and unperturbed productivity. COSATUs political shift to nationalism was not, however, easy and painful lacerations accompanied it. The federations pro-ANC profile alienated migrant workers in the hostels of the Witwatersrand, the early militant core of independent unionism. As Ivor Chipkin (2007: 147) commented, social movement unionism upheld a certain image of the black worker as a citizen in waiting, but only as part of a specific national-democratic popular project. Older migrants, 103

however, were reluctant to join overtly political campaigns, which meant the risk of dismissal and deportation, and kept an image of the union as primarily concerned with healing the indignities of production. They resented the city as a place of money, materialism, and possessive competition, and felt that young township activists, who often looked down on migrant workers as traditionalist and backward, had overstepped by dragging the unions into supporting the ANC. Despite deep social proximities and daily interactions between township and hostel residents, stereotypes of city-born youth as disrespectful, corrupt, and undisciplined abounded among migrants. The alliance of social movement unionism and nationalist politics unsettled a delicate equilibrium between these different sections of labors rank-and-file. The resentment lingering among many migrants from Zulu-speaking areas opened a political vacuum that was partially filled by the chauvinist message of the Inkatha Freedom Party (IFP,) a staunchly pro-capitalist, anti-ANC organization professedly representing Zulu authenticity (Mamdani 1996: 252-84; Chipkin 2007: 132-42.) As the empathy of COSATU unions for the precarious predicament of migrant workers wore off, migrants turned to a conservative political identity. In places where the frictions between the IFPs traditionalist nationalism and the ANCs modernist nationalism were more strident, like the East Rand, escalating violence plunged whole communities, by the early 1990s, into a state of virtual civil war. The legalization of the ANC in February 1990 marked the beginning of the negotiations towards the first democratic elections as most political leaders strove to avoid descent into unstoppable violence. With racist authoritarianism gone, the popular imagination of wage labor redemption became inseparable from the reconstruction of the countrys capitalist economy in the stormy waters of free-market globalization. Soon after the unbanning of the ANC, left union organizers started to complain about a technocratic mindset gaining hold in COSATU, where 104

young officials concerned with productivity increases began to replace older stalwarts steeped in struggle and resistance (Marie 1992.) COSATU gained access to complex negotiations with business representatives, which made it more receptive to experts advice than to the influence of elected plant representatives. The ANC-led alliance, finally, recast COSATU members as patriotically responsible workers, while labor radicalism gave way to demands deemed feasible by African nationalism in power.

Conclusion

On the surface, late apartheid reforms and labor stabilization in colonial Africa had a lot in common and failed for similar reasons: they could not gain legitimacy and ended up being used by disenfranchised workers to raise new radical claims. The social policies of the late 1980s, however, were a foretaste of times to come as they confronted black militancy with increasing commodification and precariousness of employment. The deracialization of labor markets and urban planning proposed free labor as a signifier of citizenship while embedding it into institutions of unfreedom to channel otherwise deeply reluctant populations into waged work. In the end, the unions became more politically assertive and joined township insurgencies. Their focus, however, was less on revolutionary transformation than on protecting their members in increasingly difficult battles for survival. As COSATUs struggles combined workplace defensiveness and political radicalism, a new imagery of the working class as a functional component of the national body emerged. Once politically marginalized, the ANC could now deploy its idiom of hard labor as self-reliance, probity, responsibility, and class cooperation to

105

rephrase the redemption of work. Coming full circle back to Cecil Rhodes dream, the dignity of work signified African desires away from the criticism of wage labor. The apartheid regime failed to convince blacks that wage earning could underwrite a not merely coercive form of governmentality and discipline. The ANC embarked, while ascending to power, on the same task but with a popular legitimacy unparalleled in South Africas history. The fact that apartheid ended in a negotiated transition rather than as a cathartic social transformation left the project of redeeming labor incomplete and open-ended. But the ANCs abandonment of socialist rhetoric and alignment with macroeconomic moderation also highlighted how ambiguous and malleable the discursive relations of work and citizenship have been in South Africas history to begin with. In few other contexts has work provided such a contested point of intersection for profoundly dissimilar imaginations of the human and the citizen, from the racial states disciplinary fantasies, to democratic nationalist protestations of universal rights, to ordinary visions of activity liberated from the capitalist workplace. Perhaps the inability of resistance to wage labor to hatch unambiguously alternative meanings of work can help explain why the massive revolutionary forces energizing twentieth-century South Africa were in the end, and with relatively little trauma, reabsorbed in a postapartheid official imagination that centered citizenship around productive economic activity. The next chapter will discuss the ways in which the post-1994 government has conceived the relations between work and citizenship as a crucial device to navigate the contrasting currents of political liberation and economic liberalization. On its way to win the 1994 democratic elections, the ANC could glimpse the contours, elusive for all previous rulers, of a new South Africa where a deracialized capitalist labor market could be legitimately compatible with discipline and governability instead of paving the road to the revolution. The capacity of state 106

imagination to mold social processes should not, however, be overestimated. The new government would still face radical labor constituencies, massive poverty, and vast inequalities, which would demand fundamental change, redress, and delivery of social services. Even if job creation became a master signifier of left imagination, widely different ideas on the type and nature of jobs to be created would persist. The troubling similarities between the work-centered citizenship discourse of the ANC and the racial state would, finally, expose a certain hollowness in the postapartheid project, which alternative modes of signification can constantly penetrate and pry open.

107

2 The Work-Citizenship Nexus of Postapartheid South Africa

Now one of two things must take place. Either you must do something, or something must be done to you. (Herman Melville, Bartleby, the Scrivener)

Resistance is Futile. The Governance Project of the ANC in the New South Africa

Ufilumuntu, ufilusadikiza (the man is dead, but his spirit is alive) was a rallying cry of marching workers in the Durban strikes of 1973. With the elections of 1994, the spirit of the worker was reborn in the body of the citizen. Deracialized citizenship constituted the new South Africa as a realm not just of rights but also of responsibilities and obligations attendant to full membership of the nation. The embrace of citizenship provided the ANC-led government with a powerful tool to discipline popular expectations for radical change into a governable institutional project. Left critics have often complained that new policies of liberalization and fiscal discipline betrayed and sold out the ideals of the liberation movement on the altar of the neoliberal Washington consensus. The basic contention of this chapter is rather that not only global ideological paradigms but also the discursive legacies of a long history of negotiations and contestations between nationalist elites and the racial state explain the shifts in the ANCs postapartheid orientations. Central in this regard is the idea of the citizen as an abstract individual primarily relying on economic initiative and the labor market as objective sources of socioeconomic emancipation. COSATUs official scripts translated old motives of wage labor 108

redemption into the image of a universal, nonracial worker claiming rights as a matter not just of constitutional protections, but also of resource redistribution. The government of the ANC, to which COSATU is allied, endorsed and celebrated the centrality of the worker as a builder of postapartheid citizenship while simultaneously vacuuming it of its most contentious desires. The substantial investment on work as a signifier of citizenship weighed heavily against redistribution and decommodification, depicted social conflict as pathological, and consigned popular resistance to history. While celebrating the contribution of grassroots radicalism to liberation, the new rulers moved to tame and defuse it as inimical to investment and growth. The invocation of social inclusion in the new democratic constitution provided radical desire with surrogate satisfaction. Bolstered by the advice of experts rapidly coming to terms with internationally hegemonic neoliberal ideas, the new South Africa finally joined a postcolonial predicament where, as Fred Cooper (2000: 111) wrote: The powerful universalism of free labor ideology . . . has confronted Africans and their descendants with a compelling set of images: the idea of a self-motivated, self-disciplined worker, making choices in a labor market and thereby determining the welfare of his or her family. It has been hard for laborers to be cultural relativists, to insist that punctuality, diligence, and obedience were peculiar European folkways. The power to label in international debates has by and large gone the other way, and it has been the work habits of the African that have been labeled peculiar, while market-discipline has been discussed as an acultural, universal norm. COSATU, which at the end of the 1990s had almost two million members, was in a contradictory bind: it often deprecated the abandonment of revolutionary or even radical ideas of 109

transformation but also increasingly relied on its alliance with the ANC and its institutional presence to influence social policies. The legitimacy of the postapartheid order heavily hinged on the combination of juridical equality with a free capitalist labor market untarnished by racial favoritism. Not only did it provide channels to discipline popular claims; it also allowed the government to translate its concerns for the poor into a commitment to job creation, a theme dear to the ANCs labor allies. Job creation displaced redistribution as an imagined solution to the postapartheid social question but, to the extent that it interrogated the role of the state in fostering employment, it also opened new terrains of contestation over what type of jobs were to be created, and the protections and benefits to be attached to them. Intense labor mobilization accompanied the postapartheid transition. In 1988 nationwide strikes defeated the governments attempt to introduce a restrictive amendment to the Labour Relations Act. As a result the unions mostly represented by three federations, COSATU, the National Council of Trade Unions (NACTU,) and the mostly white Federation of South African Labour Unions (FEDSAL) negotiated with the state and business the historic Laboria Minute of September 1990. The document extended basic labor rights to previously excluded employees, such as farm workers, domestic workers, and public servants, and established the principle of tripartite negotiations between unions, government, and employers over all future changes to labor legislation. In 1991, workers strikes against the introduction of a new value-added tax (VAT) led to the formation of the National Economic Forum (NEF,) a tripartite corporatiststyled organ to negotiate social and economic policies. With the beginning, after February 1990, of the official negotiations for a democratic South Africa, however, power relations in the ANC-SACP-COSATU alliance shifted. The 110

alliance, formalized on June 27, 1990, was aimed at fundamental political and economic transformation of South Africa into a unitary, democratic and non-racial country based on the demands of the Freedom Charter.1 In its original spirit, it was supposed to operate on the basis of consensus, while respecting the political independence of its constituent organizations.2 In the political negotiations, in which COSATU members were part of the ANC delegation, the ANC acted nonetheless as the leader of the alliance. Even agreements reached at the NEF, to which the ANC declined to participate, were subject to the decisions of a future ANC-led government. The campaign for the 1994 elections, finally, sealed the politically dominant role of the ANC by placing many COSATU and SACP leaders in the partys lists of candidates. COSATUs political role declined as its societal influence was rising, a paradoxical predicament that exposed the federations many vulnerabilities. Economic crisis and corporate restructuring threatened union members with layoffs and unemployment as the ideological alternative of state socialism was in global decline. The ANC, now as the government in waiting, expected from COSATU a more moderate profile, shifting from resistance to socioeconomic reconstruction and the demobilization of radical expectations. The first half of the 1990s were violent times characterized by multiple conflicts chiefly opposing supporters of the ANC and the Inkatha Freedom Party (IFP,) or residents of formal township housing, informal settlements, and migrants hostels vying for scarce urban land while the outgoing NP government attempted to condition the transitional political process with the use of armed vigilantes. Public concerns and fears coalesced around a longing for order that stigmatized all types of militancy and social antagonism. The media amplified the alarms of government and opposition politicians, liberals and African nationalists alike, which stigmatized the disruptiveness of unauthorized strikes and the youths undisciplined and riotous behavior. They saw there propensities for unspecified, 111

noisy, potentially criminal anger, disrespect for authority, breach of organizational discipline, and immature claims for social entitlements at variance with the requirements of employment and production. The institutional sphere of the transition admitted only consensus, compromise, and negotiation as legitimate arenas of politics, therefore signifying contestation as deviance (Marais 1993.) Young township militants, or comrades once the heroes of the 1980s insurgent experiment with localized peoples power and prefigurative democracy (Suttner 2005) became in everyday imagination dangerous extremists, ominous shadows of a possible anarchy to come. Of particular concern was the continuing boycott, well into the 1990s, of rents and utilities fees, with which residents of many black townships reclaimed access to decommodified housing, water, and electricity. In 1993, only one third of African township residents were paying their bills; in 1994, they had dropped to one fifth (Seekings 1997: 150.) The ANC was now preoccupied with the fiscal consequences of non-payment for the future local governments it would run. It tried to put an end to the practice through agreements with local civic organizations, but boycotts continued, often in open defiance. Amidst the uncertainties of the transition, the ANC and the NP shared the aim to discipline the unruly social practices of the poor and reestablish authority as decisive matters of political stability. In a 1992 interview, Nelson Mandela lamented that we are sitting on a time bomb. . . . Their enemy is now you and me, people who drive a car and have a house. [Their enemy] is order, anything that relates to order, and its a very grave situation (cit. in Saul 2002: 41.) Well before 1994, as the crisis of apartheid became irreversible, the ANC in exile had reformulated the meaning of liberation into rolling back popular uprising and making democratic South Africa a governable space. In the mid-1980s ANC leaders had started to meet in exile with business representatives pushing for reform at home. Thabo Mbeki was by then the leading 112

intellectual voice among those who wanted the ANC to marginalize socialist orientations and redefine the liberation movement as a representative of multiple class constituencies, including black business, as long as they supported the antiapartheid revolution as an ANC-led effort (Gevisser 2007: 464.) Mbekis position, on the other hand, harked back to themes of selfimprovement through hard work and entrepreneurial initiative, which never lost their significance for the middle-class leaders of African nationalism and their supporters in the black business community (Gevisser 2007: 583.) Mbeki inherited, as an image running deep throughout the history of the ANC, a profound belief in self-reliance as a correlate of political emancipation as well as individual virtue. In its attempt to restore order, the ANC could count on its relationships with organized labor as a significant liberation political dividend (Hirsch 2005: 6.) The alliance entered the transition with a program of growth through redistribution, a Keynesian paradigm with which COSATU tried to salvage its socialist aspirations. It emphasized the satisfaction of basic needs, advocated a strong state role in the economy, including the nationalization of strategic assets, and envisaged a growth path driven by domestic demand and labor-intensive jobs. Among its practical proposals were universalist social policies such as a national fiscally funded pension system to overcome the fragmentation and inadequacies of the existing UIF and old-age state pensions.3 A summit in Harare in April 1990 christened growth through redistribution as an alliance principle, but five months later a follow-up meeting (Harare 2) of ANC-aligned economists dropped the idea in its final document, which also omitted to endorse nationalization, a crucial component in the ANCs historic manifestos. When the deputy head of the ANCs Department of Economic Planning, Tito Mboweni, presented the Harare 2 document at a COSATU policy workshop, he faced virtually unanimous outrage and condemnation. Despite his 113

attempt to explain the abandonment of nationalization as a tactical move as it is a word that brings images with it and we have been under intense pressures from companies who [ sic] want to invest,4 union representatives rejected what they saw as a conversion to macroeconomic conservatism. NUMSA decried that the ANC was falling under the influence of corporate business and jettisoning redistribution. Not more effective were Mbowenis remarks that, given the balance of forces, the unions role was to educate workers to the need of broad social consensus, rather than rally around slogans like nationalization. A delegate from the post office and telecommunications union, POTWA, concluded that it was not unreasonable for th e masses to regard ANC statements as a sell-out position.5 The leadership of COSATU, however, was by then substantially involved in a governance conversation centered on institutional partnerships, which re-elaborated images of the universal waged worker as the carrier of a general mission of social emancipation. Many union activists entered a specific temporality marked by immediate needs of socioeconomic reconstruction rather than remotely future revolutionary prospects, and yearned for deideologized, pragmatic alternatives to clear-cut polarizations between state and market. A technical vocabulary of functional differentiation and consensus-seeking started to replace views of structurally antithetical interests and power relations. NUMSA organizers Adrienne Bird and Geoff Schreiner (1992,) for example, argued for lean state regulation within a broad civil society consensus steered by social contracts and the participation of all stakeholders. Similar sentiments inspired the view of ANCs Tito Mboweni (1992: 26) that a co-operative partnership between labour and capital is a crucial determinant of stability and international competitiveness. The effacement of conflict by governance in alliance debates elicited union cadres acceptance of the need to sacrifice demands and calibrate expectations for the wellbeing 114

of the nation, especially the poor, the unemployed, and the marginalized. For Ebrahim Patel, general secretary of COSATUs Southern African Clothing and Textile Workers Union (SACTWU,) productivity increases, not workers needs, ought to be the legitimate foundations of wage demands if South Africa wanted to join a global modernity where protection, state regulation, and nationalisation give way to efficiency, self-regulation, [and] worker empowerment (Patel 1994.) Union-aligned intellectuals refined the discursive coincidence of empowerment and work by spurring organized labor to repudiate anticapitalism and embrace a post-adversarial philosophy committed to productivity, profitability, and growth (Kraak 1992: 417.) Workers propensity to strike, and not the material realities motivating them, came to be represented as the problematic legacy of apartheid (Grossman 1997: 162.) The editor of the South African Labour Bulletin, Karl von Holdt (2003,) decreed that violent conflict was the main obstacle for the unions in their task of restoring established interactions between labor and capital. The risk otherwise was, he continued, that radical political styles of township youth bent on rent boycotts and direct action could infect the disciplined, responsible behavior of respectable rank-and-file workers. Democracy, citizenship, work, and production became inseparably linked, almost interchangeable terms in the imagination of the unions. COSATU supported a massive research effort ending up in the Industrial Strategy Project, whose final report powerfully resonated with the modernist fantasies of late twentieth-century postfordism. In its salient arguments it saw redistribution and social justice as endpoints in a high road to growth that, departing from emphases on domestic demand, would put citizens to work through export-driven industrialization and flexible manufacturing (Joffe et al. 1994.) Technical idioms emphasizing skills and technologies resignified workers essentially as resources to be optimized, which provided a new incarnation to the unfulfilled dreams of Wiehahn and Riekert. 115

The ANC presented itself as a pragmatic organization, devoid of ideological dogma in its creative pursuit of functioning, consensual solutions, thus rephrasing the social into a set of objectively definable technical challenges. The approach echoed the governmentality precepts of the World Bank and the International Monetary Fund (IMF,) which Mbeki and other ANC leaders met in 1992. The occasion symbolized the rise on the international stage of Trevor Manuel, then the head of the Department of Economic Planning, as the ANCs leading economic strategist. Manuel wanted to appear an innovative maverick, impatient with received ideological keywords and disdainful of what he saw as the populist extremism of the unions rank and file (Gevisser 2007: 668-70.) His ostensible pragmatism found a congenial counterpart in the economic planners of the late NP government, particularly finance minister Derek Keys. The two concurred on the need to contain labor radicalism and adopt a prudent budgetary course (van der Westhuizen 2007: 242-44.) South Africas financial problems and ongoing negotiations with foreign banks, finally, facilitated the convergence of the ANC and the NP around the imperative of social stability (Allen 2006.) The ANCs social policy positions initially committed the new democratic government to universalism and a guaranteed and expanding floor of economic, social and educational rights for everybody (African National Congress 1991a.) Progressive taxation, egalitarian access to municipal services, and cross-subsidization from wealthy to poor communities figured prominently in ANC documents. Its economic policy programs supported a developmental state within a mixed economy through the participation and empowerment of labor and civil society organizations (African National Congress 1991b: 4-5.) The ANCs socioeconomic strategy for its 1991 national conference argued that the new state must accept ultimate responsibility for the provision of welfare, including a national retirement scheme and 116

compulsory universal unemployment insurance (African National Congress 1991b: 6-7.) But by then growth through redistribution had given way to the view that redistribution needs to take place within a framework of responsible fiscal and monetary discipline (African National Congress 1991b: 7.) Fiscal discipline required constraints to public spending as necessary to attract investment and successfully drive the country back into international markets. The crucial Policy Guidelines for a Democratic South Africa corralled redistribution inside macro-economic balance, including price stability and balance of payments equilibrium (African National Congress 1992: D1.3); the states responsibilities in the provision of attractive and competitively priced goods and services were now shared with a dynamic private sector (African National Congress 1992: D1.5.) The ANC warned that generalized social provisions could degenerate into a handout approach to be countered by the importance of the family as it is understood within the social and cultural norms (African National Congress 1992: I.4.) Universalism and decommodification were, finally, overtly downplayed as social provisions were stratified according to employment. Wage earning and related benefits would therefore cater for the needs of the employed, while state-funded safety nets would cover only the deserving poor, whose ability . . . to contribute to society through work or other ways is beyond their control (African National Congress 1992: I.5.) Claiming adherence to the Freedom Charter, COSATU was demanding a new, universal social security system integrating the existing, separate programs of noncontributory social assistance and contributory, employment-based social insurance. The ANC, dramatically reversing its own previous positions, posited nonetheless that the two systems would remain separate, as under apartheid (African National Congress 1992: I.6.) The centrality of the wage as the normal form of inclusion foreshadowed the subject of postapartheid 117

governmentality as a citizen claiming rights on the basis of economic initiative rather than social solidarity. The normative emphasis on the family as it is understood, furthermore, naturalized and objectified the gendered divisions between male breadwinners and female caregivers (Sunde and Bozalek 1995.) The emerging policy discourse prescribed access to a wage, despite the material devastation of migrant labor and precarious work, as the primary condition for a respectable male citizen to take care of himself and his dependants. By now, beneath the official alliance between ANC and COSATU lingered a growing discrepancy in their meanings of the relations between work and social emancipation. For the latter the creation of waged employment had to salvage and update old promises of dignified jobs and social citizenship rights. By disguising radically redistributive demands into slogans of job creation, productivity, and workplace cooperation COSATU also tried to regain popularity among ANC constituencies worried of the impact of labor radicalism on unemployment, crime, and social decay. In the run-up to the 1994 elections, rank-and-file labor expectations remained high, similarly to those of township residents boycotting rents and fees. A survey of COSATU members revealed more urgent concerns with social redress than with the ANCs policy priorities. Far from being limited to wages and working conditions, workers social citizenship imagery radically questioned the ability of the capitalist class to deliver (Ginsburg et al. 1995: 71.) The most widespread demands were, in order of importance, better housing, education, healthcare, safety, electricity, water, land, and public transport. Almost three quarters of respondents said they were ready to extraparliamentary direct action in case the new democratic government failed to meet their expectations (Ginsburg et al. 1995: 71.) Even if for the leaders of the ANC waged employment was central to social inclusion, it did not necessarily guarantee social equality; it was alternative, not complementary, to resource 118

redistribution and decommodified social provisions. Compared to COSATUs claims, for the ANC the moralization of the working class (Jones and Novak 1999: 116,) or the creation of positive attitudes and dispositions towards work, gained precedence. Both perspectives claimed the same roots in liberation politics and the loyalty of laboring people, but configured divergent subjectivities. The transition made clear that, as Cooper (1996: 14-15) wrote about African postcoloniality, the working class is constituted by imaginative projects as much as social relations. The ANC won the elections of April 1994 with 62 percent of the vote. Its campaign posters crying for Jobs, Jobs, Jobs! vindicated a mandate for productive employment as the core of national reconstruction. Employment policies, however, could also be a tricky terrain of negotiations, as a study of the political attitudes of South Africas new elite concluded: Politics, for the present, has demobilized the poor or rather it has mobilized them for electoral purposes while containing pressures for social transformation. But transformation has been promised indeed, oversold by the ANC. The party therefore finds itself playing incompatible roles, simultaneously promising transformation and containing pressures for transformation while it seeks to tackle poverty through the pursuit of economic growth (Kalati and Manor 1999: 118. Own emphasis.) As the ANC ascended to political rule, nation building replaced national liberation. Relieved of their potential for social disruption, old social justice catchphrases found a new life in the ANCled alliances experiment with progressive governance (Piper 2006.) Here, as Alan Hirsch, a prominent ANC economist and policymaker, argued, it is the states job to underwrite the improvement in the quality of life of the poor and to reduce inequalities, but with a firmly 119

entrenched fear of the risks of personal dependency on the state and of the emergence of entitlement attitudes (Hirsch 2005: 3.) William Walters (2004) proposes a definition of governance with features that well apply to the ANCs postapartheid political philosophy: First, the democratic state shares its regulative powers with multiple, external sources of authority, primarily supranational processes and institutions. Therefore entities like international markets, investor confidence, and global macroeconomic constraints acquire, not differently from the age of the Riekert commission, the solidity of objective truths and invariable laws unaffected by political contestation. Second, the diffusion and dispersion of the sources of authority takes place not only globally, but also at the sub-national level. The ANC-led alliance, therefore, presents itself as a progressive combination of socioeconomic interests capable of optimizing the self-regulation of civil society. Contrary to the imagination of the late NP state, the government does not direct social processes but manages adaptive dynamics. Nonracial democracy, corporatist institutions, and a liberal constitution ideally provide arenas where otherwise conflictual interests can balance themselves around national obligations of productivity and competitiveness. In response to the ideological dichotomy between private-based and state-driven solutions, the ANC invokes the flexible pragmatism of public-private partnerships and the participation of social actors as stakeholders. Third, citizens of the new South Africa are human resources to be maximized, trained, and educated. In 1998 the ANC government commissioned a report into poverty, which opined that public social provisions are not, as the unions argued, means of redistribution and social redress, but safety nets aimed at human development to make individuals capable to contribute to growth according to their full human potential. The document regarded social 120

policy as a biopolitical tool enhancing the health and skills of the population as the wellbeing of human capital is decisive to eliminate political and social unrest (May 2000.) To be virtuous citizens, the poor should take individual responsibility for their own inclusion instead of noisily clamoring for social equality. Fourth, social conflict becomes a disruption to be minimized. Postapartheid governance regards societal pluralism and popular participation as assets only when they can be harnessed in institutionalized interactions aimed at building consensus. Tito Mboweni (1992: 26-28) argued that in a democracy labor and civil society should uphold imperatives of productivity and competitiveness as a matter of free choice, but added that the government is responsible for aligning freedom with the rationality dictated by uncontroversial economic processes. To be conducive to harmonious nation building, social standards and objectives must not be the result of social contestation. They should rather be translatable into what the ANCs parlance designates as delivery indicators, or the technical quantification of housing, education, health, welfare, and municipal utilities compatible with given resources. The governments claim to objectivity is embodied in its stated mission to ascertain what works, whereas conflicting social forces are allegedly bound by their narrow self-interest and ideological loyalties. As a corollary, the pragmatics of postapartheid governance can seamlessly combine free market and welfarist methods by presenting policy choices as responses to specific case-by-case challenges (Piper 2006: 142.) The governmental profession of non-ideological, solution-oriented approaches facilitates a managerial relationship with customers in opposition to a political one with citizens (Li 2007.) Fifth, the ANCs image of South Africas social question downplays the role of economic processes as causes of poverty and inequality, which tend increasingly to be depicted as external 121

ailments that strike individuals and groups predisposed by specific misfortunes or lack of initiative, morality, or temperament. Not only does this discursive mode present economic growth as the unquestionable panacea to poverty; it also dismisses and deprecates redistributive claims often characterized as entitlement mentality as pointless annoyances, when not contributing factors to poverty itself (Rojas 2004.) Instead, the ANC took to praise the resilience and coping abilities of poor individuals and families. The above mentioned government report made poverty alleviation a matter of strengthening the ability of poor people to fight poverty by building their assets (May 2000: 12.) The individual initiative the state aims to enable, therefore, is a specific economic enterprise. To such ends public healthcare, education, or limited cash transfers can be efficiency-seeking and compassionate programs, but so are cutbacks on all redistribution that can generate dependency. Many left commentators have echoed labors allegations of an ANC sellout, arguing that the 1990s marked a sudden, unexpected turn away from a socialist tradition towards neoliberalism. In some analyses, ANC leaders driven by contingent conveniences were simply lured and duped into accepting the agenda of the IMF and the World Bank (Bond 2005; Klein 2007: 194-217.) Framed in almost conspiratorial tones, this line is, however, unconvincing. Entrepreneurial views of black social emancipation have always been much thicker than socialism in the ANCs ideological tradition. For Alan Hirsch (2005: 34-36) socialist positions were indeed exceptional in the history of the ANC, and mostly belonged to the period of exile, when alliances with left regimes were vitally important. As a polemical label, neoliberalism simplifies a complex bricolage in the ANCs postapartheid public discourse, which incorporated elements of macroeconomic conservatism, state-regulated developmentalism, and

socialdemocratic welfarism. The mix was indeed crucial to convey the message that, to defuse 122

social contestation, policies must be driven by practical sense, not ideology. Most importantly, the thesis of the ANCs sudden surrender to global neoliberalism overestimates the importance of contingency and underestimates how the ANCs appropriation of free-market economics trailed a process of capitalist upgrade that dates back to the 1970s (Marais 2002: 84.) As Kurt Weyland (2006) observed, bounded rationality and past choices often drive policy adoption in transitional contexts, and NP technocrats had time to acquaint the incoming elites with the rules of the game during South Africas prolonged political change. Rather than being upset by abrupt breaks the social policy imagination of the South African state seems to be caught up, across radically different regimes, in a sort of cognitive long duration. The focus on the poors individual initiative has recursively appeared with renewed allure especially in moments of contested transformation, from Smutss South Africanism, to the technocratic modernism of the apartheid regime, and the myth of self-reliance in African nationalist discourse. All of them praised economic participation as much as they feared social disorder and despised policies inducing dependency. The ANCs association of emancipation with self-reliance finally authorized fiscal discipline and the centrality of market forces by selectively connecting them to a history of progressive struggles, which effectively foreclosed competing narratives (Crush 1995.) The discourse of liberation elevated the market to the socially equalizing role of measure and reward of hard work regardless to race, gender, and ethnicity, a symbolic asset obviously unavailable to old regimes. Similarly to what Cooper (1996: 382) noticed for African decolonization, the ANC validated its embrace of globally hegemonic ideas by vindicating the cultural uniqueness of South Africas development trajectory. The rigid racialization of the previous order had condemned its work-centered discourse to illegitimacy. Once in power, the ANC could count 123

instead on what postcolonial scholars identified as the ideological malleability of nationalism, which can be derivative of Western-inspired themes (Chatterjee 1986) while opportunistically drawing norms of order and responsibility out of local cultures and practices (Chatterjee 1994.) After 1994, a democratic constitution, a deracialized labor market, and governmental support for black entrepreneurs promised a new age of legitimate social stability. The natives finally became, as Chipkin (2003) put it, citizens in rights. To also be, he continues, citizens in practice, capable of actually exercising their new rights free from poverty and need, however, they were required to shed dependency habits, moderate their claims, defer social expectations, suppress resentment over class inequalities, and place work at the center of responsible conducts. Employment promised enterprising citizens a way out of marginality and an alternative to enduring inequalities by race and class. Such a massive symbolic investment in work as the foundation of democracy was nonetheless destined to clash, as the next section shows, with the harshness of the postapartheid world of production.

The Changing Face of Precariousness

After apartheid, waged employment remained remarkably vulnerable and unstable for most South African workers. Economic liberalization made work, rather than a remedy to poverty and inequality, a contributing factor in their reproduction. Insecure occupations and meager wages blurred all boundaries between inclusion and exclusion, which in the normative imagination of policymakers clearly divided those with jobs and those without. Similarly to Deborah Jamess (2007: 254) comment on another major area of disappointed expectations, land reform, the

124

connection between work and citizenship has thus begun to unravel since the former no longer promises a source of unity on which the latter can be based. In the 1990s and 2000s, employment has mostly expanded in the informal economy or in the most precarious and contingent formal occupations (Hirsch 2005: 172.)6 After 2002 economic growth picked up and net job creation has occurred, but economist Miriam Altman (2006) estimates that, while mostly capital-intensive industries like automotive, basic metals, and chemicals expanded, new jobs piled up in low-skill services like domestic and personal care, retail, or outsourced company operations like call centers where a newly empowered black business class started to invest and consume. Most urban unemployed lack qualifications to compete for scarce highly skilled, highly remunerative jobs. Even if absolute joblessness, which overwhelmingly affects Africans, is largely an urban phenomenon, low-wage informal occupations are mostly concentrated in rural areas, where they contribute to place the local black population at the bottom of the countrys income ladder (Hajdu 2005.) Across very different local labor markets, the youth are the most permanently excluded from wage earning: in the mid2000s, 65.8 percent of the unemployed aged 25 to 34 had never held a paid occupation (Bhorat and Oosthuizen 2005: 46.) Most of them fall under the official definition of discouraged jobseekers people who say they want a job but do not actively look for it a new governmental shorthand for a socially problematic population evoking past metaphors of largely juvenile restlessness and ungovernability. At the end of 2008, the unemployment rate was 23.2 percent of the economically active population, similar to the 1995 levels, but official measurements no longer include discouraged jobseekers, which would push unemployment to almost 30 percent (Statistics South Africa 2008b.) As under apartheid, the measurement of unemployment remains a problematic, politically loaded exercise. The definition of employed 125

used by Statistics South Africa, for example, includes occupations like beg[ing] money or food in public or catch[ing] any fish, prawns, shells, wild animals or other food for sale or family food (Clarke 2004: 569.)7 In the 1990s scores of factories and mines closed as the economy liberalized and apartheid-age protections ceased. Regions of heavy manufacturing, the historical core of independent unionism in the 1970s and 1980s, were hit particularly hard by economic decline. More than 10 percent of manufacturing jobs were lost between 1996 and 2000 alone, with metal and engineering industries shedding as much as 30 percent (Statistics South Africa 2000.) The vulnerability and precariousness of employment in South Africa, however, are not confined to the lack of formal jobs. They are rather the expression of an increasingly hierarchical world of work, where a shrinking minority holds formal, stable occupations with benefits. In the words of a doyen of South African labor law, Halton Cheadle (2006: 698,) the growth of flexible, casual, and contingent jobs has been so steep that some have questioned whether these forms of labour should continue to be called atypical. For him, casualization has eclipsed the concept of fulltime work defined by common law traditions of the employment contract. Even many workers enjoying the protections of unionization earn wages that often barely cover basic necessities. To the inadequacy of wages contributes the commodification of life as rents and mortgages escalate, access to water and electricity depends on cost recovery from end users, and healthcare and retirement provisions require increasingly rare full-time occupations. In practice, the wage ceases to function as an alternative to poverty. Miriam Altman (2006) defined the working poor as employed persons earning less than R 2,500 per month, which is approximately the individual threshold for income tax exemption. In her estimate, 65 percent of South Africas waged population live in poverty.8 Half of these, especially in 126

agriculture, domestic services, and construction jobs, earn less than R1,000 per month (about four US dollars per day.) But it is workers with earnings between R1,000 and 2,500 (which include more stable, unionized employees) who most frequently report chronic food shortages, as households below R 1,000 are more likely to access limited means-tested state grants. In 2006 the Cape Town-based Labour Research Service found that average minimum wages in manufacturing stood below 80 percent of a living wage calculated according to essential consumption items for a family of five. Researchers concluded that so-called formal employment . . . is unable to support a household (Elsley 2007: 82.) Between 1995 and 2003, employed persons earning less than two dollars per day have more than doubled, from 900,000 to two million (Casale, Muller, and Posel 2005: 18.)9 The collapse of employer-funded benefits mirrors poverty wages, as companies have used casual and temporary employment to cut retirement and medical coverage. In 2007, 42 percent of the formally employed had no retirement provisions (Madonsela 2008: 24.) The HIVAIDS epidemics has undercut the life expectancy of male and female workers to under 60 and 65 years respectively, which makes the retirement age with full benefits impossible to reach in about 80 percent of company-based funds (SANLAM 2004.) In 2008, the wage replacement ratio of South African retirement payments was less than 30 percent, compared to 56 percent in OECD countries and the 40 percent the International Labour Organization (ILO) deems the acceptable minimum.10 Wages fail to enable a decent life in the post-employment age not only due to inadequate coverage but also because non-permanent workers usually collect their accrued benefits every time they lose their jobs. On the healthcare front, in 2006 only 27.2 percent of the employed and 13.5 percent of the self-employed had employer-funded medical insurance. If covered dependants are counted, only 14 percent of the South African population had access to 127

healthcare services delivered through employment, down from 16 percent in 2000 (Council for Medical Schemes 2007: 51.) The universalist postapartheid rhetoric of wage labor as a condition of dignity and citizenship is belied by the daily duress of work, albeit no longer determined by racial legislation, but by the contract of employment. The casualization of work has slowly undone the gains and expectations of three decades of workers struggles, as COSATU itself recognized: The sub-contracting, casualising and division of workers is an attempt to deny workers the very citizenship rights that democracy promises them: the right to organize and to engage in collective bargaining and the right to work in fair and decent conditions. It is the re-emergence of a new form of apartheid employment strategies. It undermines COSATUs project of extending democracy and the rights of citizenship into the economy and working life. (COSATU 1997a: 127.) Karl von Holdt and Eddie Webster (2005) depict the labor market of postapartheid South Africa as a three-layered onion. On the outside are the unemployed, between one third and one quarter of the economically active population. In the center are the core formal sector workers, the half of employed people with stable wages, benefits, and collective bargaining provisions. Already in the 1980s most workers in this layer were Africans benefitting from late-apartheid labor reforms to gain access to skilled positions (Crankshaw 1997.) Upward mobility, however, failed low-skill employees, whose wages were globally outcompeted by other emerging economies. In the expanding middle layer of the onion, therefore, is a precarious workforce comprising 30 percent of wage earners in casual jobs or domestic services and a further 20 percent in informal, subsistence, or survival activities. African workers are the overwhelming majority in these

128

sectors, which also provide most employment opportunities for women (Kenny and Webster 1999; Grest 2001; Kenny 2005.) Precarious work is, nonetheless, a diversified reality with a variety of often overlapping employment relationships (Clarke 2006.) Casual workers can be hired for as short as a day; perhaps they apply for jobs at the workplace door, or rely on connections among friends and family.11 Workers defined as temporary have more standardized fixed-terms contracts and fully fall within the legal category of employees, which requires at least one hour of paid work per week. Externalized workers are hired as independent contractors, or through temporary employment agencies (or labor brokers.) The Basic Conditions of Employment Act (BCEA,) the fundamental legislation on working conditions, does not apply to most of these workers as it generally considers them self-employed (Clarke 2004.) Finally, informal workers are employed in undocumented enterprises and survival activities such as street vending, spaza shops, and small repair businesses. Late-apartheid policies supported African micro-enterprises as a way to nurture a possibly collaborative black entrepreneurship and outsource production from unionized factories, but the trend did not stop with democratization. Of the 1.4 million jobs created from 1995 to 2003, more than half were in the informal sector, domestic work, and subsistence farming (Casale, Muller and Posel 2005: 18.) Informal workers are totally uncovered by labor legislation, which apart from the BCEA includes provisions on union rights and collective bargaining in the Labour Relations Act, regulations on qualifications and training in the Skills Development Act, and measures to overcome workplace discrimination in the Employment Equity Act. During the 1990s the restructuring of production and technological innovation went hand in hand with the proliferation of precarious jobs. Devey, Skinner, and Valodia (2006) 129

documented the expansion of informal employment by defining it according to ILO criteria as work in the absence of legal protections, written employment contracts, and statutory benefits, rather than the conventional, narrow definition as work under unregistered employers. In this light, formality and informality tend to overlap as the contours of the permanent contract of employment become hazy and indistinct. About 44 percent of informal workers (80 percent of which have no written employment contract) are in permanent relations with their employers, while 16 percent of formal employees are not. More and more informal workers are hired as subcontractors by formal enterprises, even in manufacturing sectors where the externalization of functions was once limited. Almost 90 percent of informal workers, but also one third of formal employees, have no company-based retirement coverage. Finally, 44 percent of formal workers, but only 8.4 percent of informal ones, are members of trade unions. Labor organizations have serious troubles recruiting casual or informal employees. Not only does the vulnerability of such workers deter their choice to collectively organize, but in the tradition of industrial unionism, prevalent in South Africa, union membership depends on continuous employment. Unions are indeed generally reluctant to devote resources to organizing contingent workers who are unable to pay regular dues (Appolis and Sikwebu 2003.) The unions often face, therefore, a dilemma: casualization undermines them, but their organizational tools are inadequate to contest it (COSATU 1999a.) The informalization of work hardly reflects the sense of initiative that market ideologies often associate to the concept. The labor market is not here merely a space of opportunities regulated by the invisible hand of global competition. Rather, its physical fluidity congeals in specific configurations of gender and geographic inequality that reproduce and transform longstanding regimes of labor mobility. The end of apartheid did not mean the eclipse of migrant 130

labor, even if it took the suffix system away from it. As unionized male workers kept losing their jobs, female family members relocated from impoverished rural areas to informal trade, domestic work, and low-end retail jobs in the cities (Posel 2003.) The space of the household extended to connect economic activities on a national scale while it internally conflated production and reproduction. Dependence on multiple precarious jobs, inadequate social grants, and the threat of the escalating HIV-AIDS epidemics pushed poor communities into a type of entrepreneurship finalized not to accumulation or upward mobility but to the mobilization of evanescent resources and social networks to survive (Hosegood and Timaeus 2006.) Women have reclaimed new roles in the daily management of household production and reproduction, as they control migration remittances, apply for state grants, earn informal income, and decide how to stretch it across extended kin relationships (Lemke et al. 2002.) The consequent threat to male authority over the allocation of family resources becomes a source of anxieties that can ignite the latent resentment of precarious employment, redefining domesticity a terrain of tension and violence. A leading factor of employment insecurity is the subcontracting of work through labor brokers. When workers are hired through temporary employment agencies, their conditions are no longer determined by a contract with their employer, but by a commercial relationship between the employer and the labor broker (Cheadle 2006.) Work ceases to be an arena of competing claims and becomes a service that labor brokers render to workers as their clients. To employers, labor brokerage allows to recruit workers not covered by company benefits, while many brokers operate as small enterprises, which legally entitles them to exemptions from statutory basic conditions of employment.

131

The proliferation of labor brokers whose number increased maybe sixfold from 1990 to 2004 (Theron, Godfrey, and Lewis 2005: 18) accompanied democratization. In 1973 the Department of Labour in the NP government argued that workers hired through brokers were still to be legally considered as employees of the contracting firm.12 In that decade employers and white unions preferred nationally centralized bargaining to undermine the independent unions demands for workplace-based negotiations. Established interests feared therefore that labor brokerage could lead to downward wage competition and the erosion of collective bargaining institutions. Metal and engineering employers asked the Wiehahn commission to recommend outlawing labor brokerage entirely.13 In the early 1980s, as the independent unions grew stronger and started to use centralized bargaining to their advantage, employers reversed their approach, pushed for contractual decentralization, and started to regard brokers as the employers of the workforce they provided.14 It was, however, in the 1990s climate of employment flexibility, mass layoffs, and crisis of centralized bargaining that labor brokerage boomed. Approximately six hundred brokers were recorded in the metal and engineering industry alone, half of which unregistered in violation of the law. According to union lawyers they conspired15 with employers to hire a fluctuating workforce for longer hours than regular employees and without benefits and security. Employers associations boasted that with labor brokers they could finally make sure that the unions are kept off the end-users back as L[abor] B[roker] workers seldom threaten to strike.16 Political democracy and nonracial market discipline relieved labor brokerage of many unpalatable associations with apartheid-age union bashing and resurrected it under the aegis of entrepreneurial spirit. By the early 2000s, the Metal and Engineering Industry Bargaining Council reported that widespread abuse and contractual violations linked to the use of labor 132

brokers could seriously damage collective bargaining.17 Employers found that the old adage of If you cant beat them, join them is becoming more and more prevalent and is undermining the equity on which the employer/employee relationship in industry is based.18 In the postapartheid employment crisis, material insecurity is only part of the picture. More broadly, features that in past labor struggles gave work a certain thickness as a catalyst of social imagination the contract of employment, the ability of workers to negotiate their efforts and rhythms, the workplace as a place of identity and solidarity, the masculine rhetoric of breadwinning seem now to evaporate. For many young workers, who did not experience the earlier generations redemptive mythology of work to begin with, it was difficult to celebrate organized labor as a progressive social actor without holding back guarded cynicism. The idea of dignified wage labor sounded increasingly hollow and distant in daily survival struggles haphazardly patching together irregular jobs, social grants, and economies of smuggling and counterfeiting, as in the petty theft of power cables and construction materials that Ari Sitas (1998) sardonically called new hunting-gathering. Yet, work gained a solid, unambiguous position in the moral and political self-representation of the new democracy and in the ANCs discourse of an active, self-regulated civil society. Divergences and contrasts between the ANC and its union allies would not question a postapartheid institutional discourse that, as the next section elaborates, made the citizen-worker a subject of governance built on fiscal discipline, market-based self-reliance, and employment-centered social cohesion.

Building the Patriotic Worker: The Democratic Constitutionalization of Wage Labor

133

When, in April 1994, Nelson Mandela became the first South African president elected by universal suffrage, organized labor encapsulated its expectations in an acronym: RDP, or Reconstruction and Development Programme. The RDP was the political and electoral manifesto of the ANC-led alliance, and it catalyzed the redistributive demands of COSATU and the SACP, which had many of their leaders run as ANC candidates. The process leading to the RDP started in January 1993 with COSATUs proposal to the ANC for a reconstruction accord. An initial document went through various drafts and long discussions in a top-level committee of the alliance, which referred to the September 1993 COSATU congress a substantially different version from the one the union federation had initially endorsed. The document adopted at the congress, which became the eventual RDP, hardly resembled the original labor-driven strategy, being rather a leadership compromise among the three components of the alliance (Lodge 1999.) As such, it internalized the basic tenets of the ANCs emerging governance discourse: fiscal discipline, macroeconomic stability, restraint in workers demands, and a reduced emphasis on redistributive programs. The RDPs accents on meeting basic needs and attacking poverty and deprivation (RSA 1994a: 2.2.2) portrayed the social challenges of democracy in consensual terms and a language of moderation and affordability. The central institution of social cohesion was there the labor market, a concept that the document shot through with powerful moral overtones. The desirability of specific policies depended not on their capacity to modify power relations and resource allocation, but on the extent they fostered proper behaviors among otherwise troublesome target subjects. The RDPs recurrent use of the word excluded indicated, unless referred to citizens with special vulnerabilities, the unemployed. Responses to poverty and social marginality relied therefore on the self-activation of the individual as a function of the 134

competitiveness of the national economy and an agent relieving the state from onerous claims. Echoing the third way thinking of Britains New Labour or the Clinton administration in the US, the RDP saw the state as creating opportunities for all South Africans to develop their full potential (RSA 1994a: 2.2.4.1) by becoming employable. The policy priorities of the new democracy would be job creation, productivity and efficiency, improving conditions of employment, and creating opportunities for all to sustain themselves through productive activity (RSA 1994a: 2.2.4.2.) The RDPs conflation of social emancipation and the labor market aptly illustrates Timothy Mitchells (2002: 118) point on social policies as tools that normatively format social relations under the pretense of merely representing their objectivity. The stress on what is feasible within allegedly given market constraints turned the South African social question from an arena of needs, grievances and conflicts (Steinmetz 1993: 2) into a matter of order and discipline. The RDP discussed social security at length, but not as a way to decommodify social provisions or reduce household dependence on low wages. It rather made the aim of social programs to enhance individual initiative while offering residual protection against market risks. Specific unproductive populations, or the poor, the disabled, the elderly and other vulnerable groups (RSA 1994a: 2.2.4.4) could legitimately claim public social support. The able-bodied would not, however, deserve it without first exercising their ability to mobilise sufficient development resources with the state intervening only where necessary (RSA 1994a: 2.2.3.) The RDP posited economic participation as the morally sound alternative to welfare dependency: Although a much stronger welfare system is needed to support all the vulnerable, the old, the disabled and the sick who currently live in poverty, a system of 'handouts' for the unemployed should be avoided. All South Africans should have 135

the opportunity to participate in the economic life of the country (RSA 1994a: 2.3.3.) Jobs were imagined as the unquestionable solution to poverty and inequality, rather than part of the problem, as many workers experienced them. The RDPs epistemic framework confirmed the separation, of colonial and segregationist origins, between social assistance largely public, means tested, and targeted at unproductive citizens and social insurance, largely private and dependent on regular employment. The dichotomy ran throughout the document. With regard to housing, the RDP prioritized private financing for low-cost accommodation, confining public subsidies to means-tested interventions for the poor (RSA 1994a: 2.5.14.) For healthcare, the programs proposal of a public national health system was limited to primary care and the immediate provision of free services to children under six, the homeless and, over the longer term, pregnant women (RSA 1994a: 2.12.6.) In curative medicine and hospital services the RDP endorsed the dominance of for-profit private insurance and employment-based schemes (RSA 1994a: 2.12.11.1.) COSATU continued, nonetheless, to see the RDP as its brainchild and chose to contest its meanings and interpretation within the alliance. Labor could now count on the protections of the 1995 Labour Relations Act which endorsed centralized collective bargaining as a principle an unprecedented institutional presence, and the ability to influence policy in corporatist-styled tripartite bargaining organs. The newly established National Economic Development and Labour Council (NEDLAC,) which replaced the National Economic Forum, aimed to build consent on proposed labor and social legislation on its way to parliament. COSATUs choice to fight for the heart and soul, as many of its officials put it, of the alliance was surely in recognition of the profound links between the labor federation and the national liberation movement. It was also 136

shaped, however, by the perception that organized labors influence on the post-1994 government was a significant exception to the trend of neoliberal structural adjustments sweeping Africa and most of the developing world (Adler and Webster 1999.) COSATUs vision of corporatist policymaking recalled Henri Lefebvres (1977) idea of a state mode of production, whereby institutions like NEDLAC do not merely rubberstamp market-based solutions but assert their autonomy as sites where broad societal interests relate and organize growth and productivity through legitimate accords. It represented, in short, the materialization of the old workerist dream of shaping politics not as a separate realm, but as an obvious extension of production dynamics. Union leaders thought that, by donning the ethically superior mantle of producers, South African citizens could still reverse the tide of liberalization and build a democracy based on the indissoluble bind of work, redistribution, and rights for all. COSATUs hopes for the implementation of the RDP were unequivocal: The private sector should not be allowed to escape its commitments to redirect investment towards the poor. . . . Wealthy South Africans are using the policies of nation-building and reconciliation to clamour for their privileges to be left untouched. They fail to see that the stability of the new democracy cannot be built on squeezing the poor. There is one important source of income to assist in creating greater equity in society, and that is to tax the wealthy. Indeed, modern fiscal policy has developed to achieve this aim (COSATU 1996).19 NEDLAC gave COSATU the opportunity to exploit fissures in the government and build support among ministries averse to liberalization.20 Policymaking by societal agreement proved to the unions that work-based citizenship could still be a universal, progressive project. NEDLACs self-image, however, rather resonated with the ANCs emerging style of governance, for which 137

there is no other alternative (NEDLAC 1995: 3) to minimizing contestation among stakeholders willing to consider the possibility of making short-term tradeoffs (NEDLAC 1995: 3.) Limits to a possible progressive usage of NEDLAC were that economic policies fell beyond its scope and the unemployed, the informal sector, and nonunion workers lacked a meaningful voice in its bargaining processes. The government often claimed to speak in the objective interest of the poor and excluded, which allowed it to push labor towards unfavorable compromises for the sake of more disadvantaged citizens. NEDLAC was indeed a channel to defuse radical labor demands for the sake of social stability (Marais 2001: 231.) In the end, market discipline prevailed over the redistributive pretensions of the RDP, whose importance was mostly restricted to official rhetorical gestures to assuage organized labor. In late 1994, the governments RDP White Paper stated that only fiscal discipline and public spending containment could achieve the social objectives of the RDP (RSA 1994b.) The rise of Thabo Mbeki, first as Mandelas deputy and, from 1999, as president, mirrored the conservative turn in economic policy. In August 1995 Mbeki took charge of a cabinet committee aimed at a national growth and development strategy. Among its priorities the committee emphasized establishing appropriate social safety mechanisms to minimise the adverse effects of structural adjustment, stabilisation or reform programmes on the workforce, especially the vulnerable, and for those who lose their jobs, creating conditions for their re-entry through, inter alia, continuing education and training (Office of the President 1995.) The emphasis on education and training retrieved a priority of the old NP regime and married it with the active labor market policies interventions to facilitate inclusion through work as an alternative to universal decommodified provisions which were gaining ascendancy in the 138

parlance of international organizations. President Mandela lent his enormous prestige to the ANCs market discipline. In his Parliament opening speech of 1995 Mandela warned that the government literally does not have the money to meet the demands that are being advanced and criticized a culture of entitlement [which] results in some people refusing to meet their obligations such as rent or service payments. Mandelas final words sounded like a rude awakening for labor: Mass action of any kind will not create resources but would rather meet its match in the government we lead (cit. in MacDonald 2006: 90.) The modernist, egalitarian project of liberation politics had morphed into a mode of governance reflecting a postmodern preoccupation with the orderly administration of differences determined by naturalized market forces (Hong 2005.) The ideology of work played complex roles. It was essentially pedagogical and taught individual citizens to accept their station in life and count on their productive efforts for any further advance. But, as decent jobs with benefits were clearly not the norm, it also enabled, in Giorgio Agambens (1998: 26) terms, the production of a biopolitical body where the state controls the original inclusion of the living in the sphere of the law. Whereas the national liberation struggle had represented the oppressed as biopolitically homogeneous due to the states control of flows and reproduction, the democratic government could use equal market opportunities and the image of a caring state to legitimize social exclusion and manage differential inclusion. Those without decent jobs remained, in Agambens terms, bare life, or a surplus of bodies relying, to become real citizens, on chances for a good life to be grasped on the labor market. Thabo Mbekis view of social justice placed a particular emphasis on the removal of racial discrimination in economic activities and private enterprise. By insulating race as a signifier of emancipation from socioeconomic power relations, the government disabled the 139

criticism of market forces in popular condemnations of poverty and inequality. The promotion of black corporate entrepreneurship was a cornerstone in Mbekis policy philosophy (Mbeki 1999.) The states redistributive prudence was thus counterbalanced by its activism in reshuffling the racial profile of the elites (Terreblanche 2002: 38.) Policies of black economic empowerment geared state contracts and patronage towards black-owned companies and induced established white corporations to deracialize their executive boards. Mbekis idea of African renaissance, Saul Dubow (2006: 270-71) writes, justified a corporate-driven modernization project by the selective appropriation of indigenous knowledge within a rhetoric of independence and sustainability. A state-dependent black middle class (or black diamonds in media parlance) embodied the ANCs imagination of a patriotic bourgeoisie and gradually rose to prominence over organized labor both as a partner in governance and a role model of social uplift. Responding to accusations of political cronyism, the ANC presented black economic empowerment as a socially responsible partnership between state and capital aimed at poverty eradication through job creation. Continuities with the old NP governments strategy to promote a black bourgeoisie were, however, evident and not merely at a discursive level. The National Business Initiative a corporate umbrella body established in 1995 under the auspices of the ANC government was funded by the Urban Foundation, the main reformist, free-market thinktank during the NP regime (Taylor S. 2007: 172-73.) Black economic empowerment rephrased in the idiom of popular and democratic capitalism the model of citizenship based on individual acquisitiveness that the previous government had tried without success to legitimize. Iconic in this regard was the launch, on June 16, 2005 (the anniversary of the 1976 Soweto revolt,) of the South African version of the TV show The Apprentice, conducted by long-time ANC stalwart turned mining tycoon, Tokyo Sexwale, under the tagline the new revolution. Trade unions 140

criticized the excesses of greed of the new elites and the rising corporate influence on the ANC, but their indignation suffered from many weaknesses. First, the new business-friendly climate benefited not only the elites, but also many small-scale black entrepreneurs, which struck a chord with the unions productivism and job creation demands. Second, union pension funds and investment companies played an important part in many black empowerment deals. Third, Mbekis African renaissance skillfully tapped left sentiments by appealing not just to corporate rationality, but to self-reliance and economic independence as building blocks of the ANCs tradition (Gevisser 2007: 633.) Despite its racial overtones, finally, black economic empowerment squarely fell within the ANCs nonracial discourse as it presented black access to economic opportunities as a necessary step to an inclusive social order (Tomaselli and Shepperson 2001.) The ANC could therefore easily dismiss as divisive all left criticism of class privilege and argue that marketbased empowerment was not an elitist idea but the signifier of a cohesive national citizenship organically embracing the workplace rank-and-file, the small township entrepreneurs, and the pinnacles of corporate boardrooms. There was much more in the pro-business shift of the ANC government than a sudden capitulation to the global seductions of neoliberalism. It was rather an articulate project of modernization centered on removing the fetters of apartheid-age monopoly capitalism, reassuring white elites by defending existing property relations and encouraging racial solidarity between black capitalists and black poor by erasing the interface of race and class from the definition of social antagonism (Marais 2002; MacDonald 2006.) It would also have been, however, a highly abstract, technocratic, possibly unstable project without the repertoire of ideas, emotions, and loyalties that the ANC constantly drew from its ideological tradition of liberation as an enterprise of self-improvement. 141

Trevor Manuel, appointed minister of Finance in 1996, best personified the shift towards corporate capital in the interventionism of the ANC state. Contrary to the participative spirit pervading corporatist bargaining, Manuel believed economic policies to be technical issues decided by the governments superior expertise. For him, liberalization was a matter not of political contestation but of adaptation to objective global best practices. Mark Gevisser (2007: 663-64) describes the close partnership between Mbeki and Manuel as rooted in a shared combination of pragmatism, appreciation of self-reliance, aversion to depending on foreign finance, and the belief that state intervention makes markets work for the greater good. The 1996 Growth Employment and Redistribution (GEAR) macroeconomic strategy was the signal achievement of the Mbeki-Manuel association. The broad outlook of GEAR came from a panel of mostly orthodox free-market economists, with a decisive input from the World Bank, under the auspices of Mbeki and the deputy minister of Finance, former FOSATU union leader Alec Erwin. Manuel made GEAR the ANCs official economic strategy without negotiating it with the alliance partners, SACP and COSATU, whose leaders were just perfunctorily consulted (Gevisser 2007: 665.) GEAR aimed to enforce the RDP by stating that sound macroeconomic indicators precede, logically and sequentially, social delivery (Hirsch 2005.) Its main goals were financial stability, low inflation, and fiscal discipline, and to achieve them GEAR mimicked neoliberal structural adjustment programs by subordinating redistribution to strictly defined growth and budget deficit targets. The strategy proposed corporate tax incentives, wage restraint, and labor market flexibility as necessary to attract investor confidence and spur job creation through globally competitive production. GEAR reclaimed for the government a role of economic modernizer based on its knowledge of economic forces, which cast organized labor as a sectional interest perennially in 142

danger of falling under the sway of ideologically-motivated populist extremists (Goetz 2000.) Fiscal conservatism also undercut labors ability to push for redistributive social policies through NEDLAC. As employers took advantage of GEAR to unilaterally restructure and lay off workers, corporatist institutions could hardly guarantee an equitable societal consensus. The SACP and COSATU rejected GEAR as neoliberal, a criticism to which Thabo Mbekis oft quoted utterance, just call me a Thatcherite (Gevisser 2007: 666,) seemed to provide solace. SACP ministers, including Erwin, were nonetheless tasked with implementing GEAR, and critical voices within COSATU were smothered by the lack of a comprehensive alternative, the federations continuous commitment to the alliance, and the closeness of some of its leaders to the Mbeki circle (Buhlungu 2005: 710-11.) COSATU had indeed initially commended GEARs homage to education, infrastructural development, and job creation.21 South Africas first democratic constitution, adopted in 1996, reflected the market discipline and political moderation informing the transition. Chapter Two of the Constitution (the Bill of Rights) upheld equal social rights including healthcare, housing, welfare, social security, employment equity, and labor organization as well as civil and political liberties (RSA 1996.) The document formally bridged the gap, which late apartheid reforms found insurmountable, between workplace fairness and universal citizenship (Cooper 2004.) Constitutional provisions committed the state to the progressive realization of rights and obliged the government to take reasonable steps to that end. A newly established Constitutional Court made claims to socioeconomic provisions justiciable (Liebenberg 2001.) As a counterweight, however, the phrasing of the constitution emphasized access to rights rather than the enforcement of rights, which offered the government a loophole to limit redistributive policies as private and civil society actors can also enable access to rights. 143

Two further important qualifications impinged on the role of public policies. First, government interventions can be deemed reasonable if they take place within resources made available by the budget, in which the Constitution refused to interfere (Steinberg 2006.) The document could thus pitch the larger needs of society as a constraint on the specific needs of particular individuals.22 The universalism of rights referred, in unmistakably biopolitical terms, to abstract humanity in order to categorize actual claims and legislate their validity and exclusions. In practice, the Constitutional Court used budget considerations to reject the notion of a minimum core of self-standing social rights even when it ruled to provide social relief, as in the celebrated 2000 decision in Government of the Republic of South Africa v Grootboom on the access of homeless citizens to accommodation (Bollyky 2002.) Second, the Constitutions liberal restraint from intervention in private property and economic relations further dampened its redistributive potential. As a result, universal socioeconomic rights enforced state accountability while containing social antagonisms and transformation (Davis 2001.) By marrying liberalism to the socially inclusive view of African nationalism the Constitution blessed GEARs injunction for the poor to orientate their desires to the market without subverting the prevailing socioeconomic order. Constitutional discourse reinforces a governance project encouraging citizens with jobs to restrain their wage demands for the sake of the unemployed poor and these latter to restrain their redistributive demands to encourage private job-creating investment. Hard work is the point where individual uplift, household stability, and national prosperity converge and connect. The success of South Africas democratic experiment ultimately defines its active citizens as either employees or employees-in-waiting and equates their patriotism to laboriousness, patience, and moderation. The new public morality appealed to labors symbolic repertoire of heroic, 144

muscular, selfless dedication to the struggle, with the difference that this time struggle no longer meant class antagonism, but the reconstruction of the nations capitalist economy. As a delegate to the 1993 congress of COSATU put it: We used to have stayaways without pay. We as workers should take the lead in sacrificing (cit. in von Holdt 1994: 41.) At the top of state power, finally, and amidst a frequent use of revolutionary jargon to endorse GEAR, the ANC held a vision of political institutions steeped in the modernist imagination of postcolonial developmentalism. Claiming technical expertise and a prescient vision of the wellbeing of the national body, the government assumed as in the crucial 1996 document The State and Social Transformation the prerogative of defining workers legitimate interest and ensuring the centrality of the continuing and special role of the progressive trade union movement and its leadership to the mobilisation of the black workers to understand and adhere to the broader objectives of the process of democratic transformation, in their own interest. The instinct towards "economism" on the part of the ordinary workers has to be confronted through the positioning of the legitimate material demands and expectations of these workers within the wider context of the defence of the democratic gains as represented by the establishment of the democratic state. If the democratic movement allowed that the subjective approach to socio-economic development represented by "economism" should overwhelm the scientific approach of the democratic movement towards such development, it could easily create the conditions for the possible counterrevolutionary defeat of the democratic revolution. The task of educating the working class on the need to correctly balance short term material gain with the longer objective to build sustainable economic growth and a secure democracy, is 145

not a task to be left to the progressive trade union alone. It is the task of the unions as representatives of the working class but it is also the task of the entire democratic movement as the custodian of the aspirations of the democratic majority which includes the black section of the working class (African National Congress 1996: 6.9-6.12.)23 The early labor policies of the ANC government built on GEARs principle of regulated flexibility, or the attempt to enhance flexible employment while recognizi ng basic protections and rights to workers and unions (Cheadle 2006.) The government-appointed Labour Market Commission warned that flexibility could not just be synonymous with precarious jobs and supported fiscally funded social safety nets, but only as a targeted tool to improve human resources and encourage active jobseeking, rather than in terms of universal social provisions, which the commission deemed unfeasible. Most commissioners preferred public works projects to social handouts for the working-age unemployed even if they recognized that wages paid in work-based schemes, like the Community-Based Public Works Programme, were below poverty levels. The consensus was to limit state grants to citizens unable to work, and even then the cochairperson of the commission, former unionist David Lewis, proposed to use social programs as a quid pro quo for wage moderation.24 Trade union submissions to the commission largely focused on employment creation and almost entirely neglected social provisions for casual workers or the long-term unemployed.25 In the end, the recommendations of the commission mostly revolved around active labor market policies, education, and training. The final report portrayed employment as absolutely good and desirable in itself. Its expansion was a self-explaining, self-standing goal, to which considerations of wages and social provisions were fully subordinated: 146

While employment creation must form the heart of any strategy designed to alleviate poverty, higher wages and more generous transfer payments insofar as these are feasible would also reduce poverty, provided higher wages do not lead to a counter-productive decrease in employment (Department of Labour 1996: 67.) As a principle, the report supported minimum wages by industry, but it also favored an active governmental role in exempting sectors and regions from collective bargaining depending on their productive requirements. For small and medium enterprises, whose workers already lacked protections, it foresaw a less onerous schedule of minimum conditions (Department of Labour 1996: 35.) In 2005 the ANC would propose to consider as small and medium and dispensable from contractual minima all companies with less than two hundred employees, a suggestion tantamount to ending collective bargaining in South Africa (Godfrey, Maree, and Theron 2006b.) The Labour Market Commissions report claimed to eschew the ideological polarization between unregulated markets and rigid state interventionism, presenting instead its idea of regulated flexibility as a practical third way. The commissions quest for expediency gave, however, the government power to define localized states of exception and exclusion from stable jobs with benefits, an approach that Marlea Clarke (2005: 238) calls regulated precariousness. During the 2000s, the concept of regulated flexibility decisively inspired South African labor legislation, for example the 2002 amendment to the Basic Conditions of Employment Act, which lacked a definition of employee to unequivocally cover labor brokerage, independent contractors, and informal workers. COSATU was not in principle opposed to regulated flexibility, which it saw as a more advantageous negotiating terrain than unfettered neoliberal flexibility. The federation was also, however, substantially unprepared to counter the possible 147

symbolic associations of the word flexibility, which started to resonate across the media and the academe as a shorthand for deregulation (Bezuidenhout and Kenny 2000.) The unions vocal opposition to casualization and calls to ban labor brokerage betrayed a general strategic vacuum. A 1996 COSATU document candidly complained that there is a worrying trend of the use of the word flexibility. . . . We dont understand what this means, but we know this to be a word by business when they want to seek gaps to exploit vulnerable sections of workers (cit. in Clarke 2005: 253.) COSATU developed a more structured opposition to the ANCs policies under the leadership of Zwelinzima Vavi, general secretary from 1999, the same year Thabo Mbeki rose to the presidency. Vavi was not among the union leaders in Mbekis circle, which often branded his critique of the presidents unilateral and technocratic governance as nave populism. His polemics reflected nonetheless a growing rank-and-file disappointment at the creeping centralization of power in the presidency, which foreclosed spaces of debate and societal bargaining (Gumede 2005.) Much social and labor legislation, for example, was discussed in the Millennium Labour Council (MLC,) which Mbeki started under the auspices of the ILO and staffed with labor and business representatives of his choice. NEDLAC, which Mbeki tended to consider too adversarial, was further confined to fine-tuning technical details (Jansen 2001.) By the end of the 1990s, COSATUs opposition to GEAR had strained its relations with the ANC, as the unions vowed to greet labor market flexibility with blood in the streets. 26 The ANCs response, contrary to its usual consensus-seeking aplomb, was equally blunt. At the 1998 congress of the SACP Mbeki fulminated against those who consider themselves to be the very heart of the left that, in pursuit of an all-consuming desire to present themselves as the sole and authentic representative of the progressive movement, seem so ready to use the hostile 148

message of the right and thus join forces with the defenders of reaction to sustain an offensive against our movement (Mbeki 1998.) If left critics were now labeled as extremist and traitorous, unions failing to follow the governments line became targets of overt threats. The ANC leadership has considerable leverage over COSATU officials as it provides a necessary link to resources and careers in the private sector or in government. Grassroots conflicts, however, proved more elusive to control. In 2000, in what the National Intelligence Agency saw as a threat to national security (cit. in Buhlungu 2002: 195,) the employees of the Volkswagen plant in Uitenhage struck against a contract signed by the company and NUMSA. Mbeki took the Parliaments stage for a stern comment on the events: Jobs, a better life for our people in the context of a growing economy, and our standing in the eyes of the investor community cannot be held hostage by elements pursuing selfish and anti-social purposes (cit. in Buhlungu 2002: 194.) Rising workers mobilization mirrored the deterioration of employment. Working days lost to strikes jumped from 650,000 in 1997 to 3.8 million in 1998, settling to 2.6 million in 1999 (Godfrey 2007: 40-41.) Workers anger towards liberalization was fueled by the perceived insensitivity of their political leaders. In a 2000 interview minister Manuel declared: I want someone to tell me how the government is going to create jobs. Its a terrible admission, but governments around the world are impotent when it comes to creating jobs (cit. in Marais 2002: 92.) President Mbeki made a similar remark to his biographer, Mark Gevisser, in 2004, when the economy was growing at a 4.5 percent annual rate after four years of continuous expansion:

149

You can have a ten percent growth rate and it is not going to solve the problem of unemployment, because we have people who are unemployable (cit. in Gevisser 2007: 690.) The recognition by political leaders that waged employment faced a crisis impervious to positive macroeconomic cycles only added to the workers skepticism at GEARs technocratic modernism. By blaming South Africas social crisis on unemployable people, moreover, the ANC inadvertently admitted the contradiction that, despite its rhetoric of work as the embodiment of universal values, actual employment dynamics were reproducing a hierarchical social order. The governments response was to further stress the universalism of work in ideal and normative terms, which implicated individuals and households in their failure to overcome material inequalities. In 2003, a presidential report outlining the progress of the first democratic decade viewed South Africa as a society where two economies asymmetrically coexisted. The document defined the first economy as an advanced, sophisticated economy, based on skilled labour, which is becoming more globally competitive (Office of the President 2003: 97.) It includes, therefore, the minority with access to stable employment in registered enterprises as fully included, active citizens. The second economy, conversely, is a mainly informal, marginalized, unskilled economy, populated by the unemployed and those unemployable in the formal sector . . . [which] with the enormity of the challenges arising from the social transition . . . risks falling further behind, if there is no decisive government intervention (Office of the President 2003: 97.) State officials who contributed to the report clarified that government intervention would prioritize the reduction of dependence on public grants via work-based programs focused on temporary employment, training, and microcredit (Hirsch 2004.)

150

The two economies metaphor addresses the contrast between normative universalism and practical inequalities by recasting wage labor as the foundation of a legitimate and objective social hierarchy, scientifically observable in the productive status of each citizen. Inequalities within waged employment do not, moreover, detract from the overall imperative of upward labor market mobility. The expressions informal, marginalized, and unskilled opposed to advanced, sophisticated, and skilled suggest a diagnostic and therapeutic contrast between social pathology and normality and fold factual measurement and idealtypical categories into a narrative of labor as virtuous citizenship. The coexistence in the ANCs definition of employment of two distinct, not necessarily congruent, meanings one as a norm, the other as a material reality sanitizes wage labor from the social disruptions of its own decline. In a reminiscence of colonial modernity, the first economy evokes the old fully stabilized African urban insiders as opposed to the troublemaking masses of casual workers and informal squatters. The coloniality of postapartheid fantasies of work ultimately underscores wage earning as the civilized and efficient lifestyle towards which the unruly space of Otherness (Wynter 2003) of backward unwaged multitudes must tend to overcome laziness and aversion to modern production discipline. South African academics influenced by the concept of chronic poverty have, conversely, criticized dualistic analyses and their underlying idea of wage labor as an unquestionable vehicle of social inclusion (Meth 2003a; Du Toit 2004.) They rather argue that low wages entangle the poor in a permanent cycle of deprivation, collapsing exclusion and inclusion into a vast area of precariousness where the informal and casual jobs of the second economy are produced by restructuring and layoffs in the first. The two economies thesis reaches, nonetheless, far beyond the labor market as it feeds a normalizing narrative steeped in 151

specific injunctions. As oppressed, the poor had legitimate grievances and conflicts; as patriotic citizens, they must patiently and enterprisingly seek their way into the first economy. Sociologist Jeremy Seekings (2006) well encapsulates such zeitgeist in his longing for a normal South Africa finally freed from the disruption, the restlessness, and the revolutionary passions of the past, where the downtrodden, especially the youth, move on as responsible jobseekers living ordinary lives. It is an apparent paradox of South Africas history that even in times of acute social traumas and political antagonisms ruling elites and their opponents have often held similar fantasies of order and normality. The postapartheid official imagination is no exception, but it also delegated such fantasies to the work-citizenship nexus precisely when work could hardly avoid a precarious life for those relying on it. As the next chapter will show, this contradiction would eventually haunt an increasingly beleaguered governmental discourse.

Conclusion. Disciplining Citizenship

The first two decades of democratic government have faced an unresolved contradiction: The deeper waged employment has decayed into a condition of precariousness and immiseration, the more prominent work and job creation have become as governmental responses to social problems. Many policy analysts have warned the ANC that the crisis of work is not limited to unemployment, but questions the validity of wage labor as a mechanism of social integration. Postapartheid realities of work hardly warrant the conclusion that the urban industrial working class does not represent the poor (Nattrass and Seekings 1997a: 466) because, as an allegedly relatively privileged social sector, it has done quite well since 1994 (Seekings and Nattrass 152

2005: 374.) A vast research project outlining employment scenarios up to 2024 concluded that even if the unemployment rate fell by half, to 13 percent, the share of South Africans living in poverty would decline by only 15 percent, to 35 percent of the population. The director of the project, economist Miriam Altman, commented: I was shocked because I thought if you halved unemployment, you would halve poverty.27 Past the initial shock, Altman (2007a: 18) concluded that the link between work and well-being in South Africa seems tenuous. As Baudrillard (1983: 65) wrote, work as an institution that heralds the rise of the social is producing and destroying the social in the same movement. Yet, the postapartheid state has not sought alternative ways of living decently regardless to employment status. It rather assumed economic participation and decent life as coincidental in a resolutely self-validating argument. The discursive centrality of wage earning powerfully reterritorialized the desires of the poor within the labor market while simultaneously satisfying the states goals of social stability, capitals concerns with productivity and industrial peace, and labors anxieties over unregulated flexibility. Through its subordination to the imperative to work, citizenship has operated as a disciplinary construct. As Foucault (2001) would have it, the centrality of wage labor has provided the episteme, a modality of state knowledge enunciated as moral-behavioral prescriptions, for the governmental dispositif, or the actual, mundane practices that regulate the production of society. The work-citizenship nexus remains, nonetheless, exposed to contestation. While sharing the governments glorification of employment, organized labor is subject to the claims of a precarious rank and file still waiting for work to deliver stability, redistribution, and social rights. Work in the end reflects the ambiguous predicament of what Etienne Balibar (1994) called the citizen-subject, a specifically modern form of political subjectivity simultaneously 153

characterized by emancipation and subjection. In South Africa, labors precarious liberation continuously disturbs disciplinary practices with a multitude of unruly, unpredictable longings. The unions capacity to represent a changing world of work seems, conversely, more unequivocally in trouble. The South African working class remains fairly deeply organized, with a union density oscillating between 35 and 40 percent of the employed population. As the largest union federation, COSATU had in the late 2000s about 1.8 million members, but since the mid1990s it has mainly grown thanks to the massive recruitment of public sector workers previously mostly excluded from unionization.28 In the private sector, restructuring and casualization have cost the loss of 150,000 members between 2000 and 2005 (Vavi 2005.) But quantitative indicators are only rough reflections of the cultural metamorphosis of the labor movement. Many union activists of older generations have spearheaded liberalization policies as members of ANC-led cabinets. Postapartheid labor leaders have gradually seconded the ANCs progressive governance modality celebrating post-adversarial restraint. Exemplary are the words of COSATU mineworkers leader, Gwede Mantashe, in support, as a member of the 1996 Labour Market Commission, of employer representative Andre Lamprecht: South Africa Inc. . . . must square off against the rest of the world, not each other.29 Rather than a surrender to neoliberalism, however, the unions ideological trajectory reveals continuities with past emancipative visions. The hegemonic nonracial tradition of trade unionism had insulated the claims of a universal worker from antagonisms of race and gender. Its productivist imagination glorified wage labor as a creator not just of commodities but of identity and solidarity. With the end of antiapartheid resistance and the decline of socialist alternatives, the ANCs ambition of building the nation as a self-reliant collectivity of producers was the only surviving ideological tradition the unions could fall back to. 154

The institutions of the new democracy tried to organize the coexistence of disparate ideological legacies of the liberation struggle by blending elements of socialdemocratic universalism, corporatist interest representation, and liberal constructs of economic relations. Their point of convergence was what Hardt and Negri (1996: 88-93) termed the constitutionalization of work, or a sturdy link between the codification of labor rights and the actual policies that enforce a conflict-free labor discipline. It is a technology of citizenship that promises what all previous modes of governance failed to achieve: to defuse the threat restless and precarious multitudes pose to the socioeconomic order and rally around the mystique of production the anarchic youth recalcitrant to work, the insubordinate residents who refuse to pay rent, and the ANCs own claim-making popular constituencies. As Cooper, Holt, and Scott (2000: 16) wrote: a claim made in terms of citizenship ratifies an elites image of itself even as it potentially constrains that elites options. The grammar of citizenship seems to perform the same task in the ANCs mode of governance, providing otherwise insubordinate imageries with an outlet in the plea for the government to deliver. The 1996 Constitution defined a set of rights as a terrain of demands that did not destabilize existing economic processes and social relations. There is a correspondence between the universality of constitutional rights and the abstract impersonality of the worker as the subject fully authorized to claim them. Both enable what Jean and John Comaroff (2004a: 192) called the negotiation of incommensurables because they allow, in a society characterized by diverse cultural practices, a common language of demands that makes the desires of the ruled recognizable and manageable for the rulers. South Africa does not escape, the Comaroffs continue, a predicament where the global spread of neoliberal capitalism has intensified the grounding of citizenship in the jural and a ubiquitous language of legality and constitutionalism 155

acquires an almost magical capacity to accomplish order, civility, justice, and empowerment, and to remove inequities of all kind (Comaroff and Comaroff 2004a: 192.) Not only has the centrality of work in postapartheid public discourse decisively contributed to such a widespread displacement of desire, but the attendant politics of rights mystifies the structural hierarchies of waged employment, according to a paradox well pinpointed by Wendy Brown (2002: 430): [Rights] must be specific and concrete to reveal and redress . . . subordination, yet potentially entrench our subordination through that specificity . . .; they emancipate us to pursue other political ends while subordinating those political ends to liberal discourse. The liberal understanding of constitutional rights, moreover, faces emancipative politics with a dilemma Brown (1995: 115) again identifies: What happens when we understand individual rights as a form of protection against certain social powers of which the ostensibly protected individual is actually an effect? It remains nonetheless to be seen to what extent deeply ambiguous relationships between normative discourse and material realities of work can flawlessly signify citizenship rights. The next chapter will look at how, in the end, the crisis of waged employment can fracture official imagination allowing for cracks in the anti-dependency rhetoric of production, which eventually test the stability of the work-citizenship nexus.

156

3 Contesting Commodification. Social Policy Debates in the Crisis of Waged Employment

The question which stirs us . . . is not the well-being human beings will enjoy in the future but what kind of people they will be. Max Weber (1994: 15)

Introduction. Governing in the Shadow of Precariousness

The ANCs model of governance aims to be inclusive and representative. Commissions of inquiry, public hearings, and consultative processes provide platforms for stakeholders to enact competing visions of society, legitimation strategies, citizenship claims, and expectations on the roles and status of productive and unproductive members of the nation. At this level, the state charts its progress against constitutional promises of rights and grand programs of delivery, business tries to ennoble corporate profits with social responsibility, labor demands its policy payoffs for past and present struggles, and marginalized communities seek recognition and relief for their plight, finally speakable in a language of common humanity. The previous chapters hinted at the malleability of the ANCs ideological approaches in response to such diverse constituencies. The flexible geometries of postapartheid social policy rest, nonetheless, on an axiomatic of rule premised on two unquestionable principles. First, social issues are not, by their very nature, ideologically contentious and only need solid, objective data, 157

technical expertise, and creative methodological assemblages. Second, poverty coincides with unemployment and waged work identifies social citizenship. Work-based citizenship is the point of intersection for imperatives of economic growth and processes of biopolitical reproduction, welfarist constitutional obligations and the primacy of individual rights, ideals of empowerment and norms of self-reliance. It is also where the ANC government addresses a crucial challenge of the modern nation-state: how to ensure the compatibility of universal citizenship with the inequalities arising from market relations (Dean 1999: 50-53.) The primacy of the work-citizenship nexus in defining postapartheid political subjectivities, however, also makes the relationships between economic initiative and dependency a terrain of contestation, which is the object of this chapter. Virulent opposition to welfare dependency has become a distinctive characteristic of the ANC government. Yet, multitudes of impoverished, overwhelmingly black households have come to depend on social grants for their livelihoods. Targeted to those unable to seek employment mostly the youth, the elderly, and the disabled social grants carry for the ANC a significant return in terms of electoral support, despite the recurring budgetary concerns of policymakers. At the same time, the governments vehemence against welfare entitlements reaffirms the obligation for the employable including unemployed, precarious workers, and the working poor to earn social protection through work, not redistribution. A government study on work incentives portrayed not working as a morally and psychologically debased human condition marked by stigma, degradation, isolation, boredom, the denial of opportunities and social exclusion (Department of Social Development 2006a: 9.) Yet, its list of the costs of working did not include low wages, focusing instead on transport, child care issues and the opportunity cost of time [and] the loss of unemployment assistance. Academic debates have similarly celebrated the capabilities of the 158

poor in order to present non-work related programs as inducements to idleness.1 Union leaders greatly contributed as well to the new intellectual climate with a single-minded focus on job creation and heroic glorification of work. By the first half of the 2000s, however, the decline of decent jobs, despite robust economic growth, and widespread protests at the governments lack of delivery dramatically placed the idea of work-based social inclusion at the center of contestation in the policy process. New ideas of non-work related universal grants tested the boundaries of official discourse and new approaches to poverty and employment would emerge in the ANCs response to South Africas social question.

Social Policy as a Technology of Self-responsibility

As under the apartheid government, in democratic South Africa the employment status determines citizens access to social provisions. The social security system is still characterized by the drastic separation of its two main components, social insurance and social assistance. Social insurance is largely contributory and relies on occupational schemes, like retirement pensions, medical aid, and unemployment insurance. Equally shared contributions from workers and employers usually provide such programs with a measure of decommodification. Social insurance is also largely privatized and, with exceptions like the Government Employees Medical Scheme (GEMS,) established in 2005, and the Government Employees Pension Fund (GEPF,) established in 1996, the state is virtually exempted from contributing to it. Government contributions to the Unemployment Insurance Fund (UIF) administered by the Department of

159

Labor but funded by employers and employees ceased after legislation inspired by the 1979 Wiehahn report included most African workers in the program. The UIF provides, in any case, only very limited short-term relief, covering termination, maternity, and illness, for temporary unemployment. The program is also, in line with ILO recommendations, punitive as it cuts benefits to workers who fail to actively seek employment or lose their job due to misconduct, including certain types of strike. The 2001 Unemployment Insurance Amendment Act brought for the first time domestic and seasonal workers under the UIF, which continues however to exclude independent contractors, casual and informal workers, government employees and, of course, first-time jobseekers.2 The government has opposed employer-funded unemployment insurance for precarious workers, as it would constrain labor flexibility. The UIF commissioner commented: As much as wed like to regulate the market, we also dont want to kill jobs. . . . I think that is where the tension is between us and labor, because traditionally we come from the same history and I think the expectations they have is that well be sympathetic to their reasons and demands, but we are with the government now, you see? Lets take the whole issue of casualization. In terms of labor, what you must do is to legislate so that it is not attractive for an employer to casualise, right? . . . Now, our approach is we are in a globally economy now, whereby we cannot act as if we are in isolation from what is happening in the world.3 In the end, the UIF underwrites a notion of unemployment as a temporary accident of life, rather than an enduring condition, which explains why workers losing their jobs have continued to seek longer term relief by withdrawing their retirement benefits (RSA 1992; Meth, Naidoo, and Shipman 1996.) 160

Retirement provisions are entirely privatized and mostly controlled by large financial corporations. In its last days, the NP government boosted private retirement schemes by moving government employees pensions from a pay-as-you-go to a fully-funded scheme, which ties benefits to the financial performance of the fund. The ANC government endorsed this shift, and the unions acquiesced with it, as the private retirement industry is a major business for black economic empowerment, where the unions pension funds play an important role (Hendricks 2006.) Incidentally, the conversion to fully-funded public employees pensions required a massive injection of highly leveraged state finances, which led to a fourfold increase in government debt from 1989 to 1996. GEAR and related policies added such burdens to the financial constraints necessitating restraint in social expenditures. Private retirement schemes face labor with a significant contradiction: unions demand universal benefits from the state, but their members continue to access them selectively through negotiations with private companies (Barrett 1989.) The dilemma has hampered labors capacity to develop coherent proposals on social security. Instead, as NUMSA officer Dinga Sikwebu argued, trade unions have paradoxically abetted a labor-driven privatization of social welfare, which reinforces accusations to the working class of being an elite.4 Moreover, most rank-andfile workers are members of provident funds, retirement schemes that allow the collection of benefits before retirement, for example in the event of job loss. The matter is thorny for many unions, afraid that their members can be lured into accepting layoffs by the prospect of large provident fund payments they can use to repay household debts. As long as wages are inadequate for daily survival, the unions face explosive consequences: The general squeeze on workers leads them to see retirement money as a source of relief for their general problems. . . . [This] is becoming, from a survivalist 161

point of view, a big problem for the union. . . . I can understand a worker that gets a report from the fund saying he has accumulated R120,000, and there isnt an effective bursary scheme to get his kids to school, and they are being financially excluded. Or maybe his house is going to be taken because of the policies of the government on housing. Why should he spare money for when he retires when there are so much more pressing things to deal with now? . . . You cant survive with your wages, so what do you do? If you go to the loan shark you sink into indebtedness and the only way to get out of that is to get retrenched and take the package. This has created a lot of instability for the union, its a time bomb for us.5 In the 1990s, the income of most retirees was below the poverty line, and for 40 percent it was below the value of the state old-age pension (RSA 1995a: 18-19.) In 2000, only six percent of South African workers were financial independent at retirement (United Nations Development Programme 2003: 43) and one third continued to work afterwards (Falkena and Luus 1999.) Moreover, the dominance of private provisions and the lack of state programs enabled employers to shift retirement risks onto their employees through defined contribution plans, which limit the liability of companies and make benefits dependent on the funds market results (Andrew 2004.) The uncertainties of life have also led many workers to prefer defined contribution to defined benefit schemes. The latters returns are more secure as the funds rules determine company liabilities, but they also reward seniority and stable employment. Defined contribution plans, conversely, are riskier but carry higher returns for short-term employment, which is often the consequence not only of layoffs but also of the impact of HIV-AIDS in shortening the life expectancy of many workers well below the retirement age that makes defined benefits viable.6 162

Faced with the inadequacies of a retirement system that adds to the uncertainty of workers lives, the national Treasurys main concern is that the withdrawal of benefits from provident funds makes retirees rely on public old-age pensions. The consequent policy philosophy remains the one charted in 1995 by the governments Smith Committee, which argued that all workers should build savings for retirement instead of expecting the handout of state pensions (RSA 1995a.) The mutually reinforcing inadequacies of unemployment and retirement benefits undermine the idea that wage earning can lead to a stable life after work. As old age becomes fraught, for those who can reach it, with uncertainties and anxieties, limited social assistance benefits provide the main relief for non-working people. Social assistance is noncontributory, means tested, and administered by the Department of Social Development (formerly the Department of Welfare and Population Development.) Full fiscal financing makes social assistance grants entirely decommodified, even if, through the means test, they are intended to target low-income claimants. Social assistance stands to social insurance in a relationship that mirrors the hierarchical nature of work-based citizenship. Being socially insured means receiving the superior benefits that go with full economic participation. The socially assisted condition, instead, brands a population incapable of self-responsibility, where the states tutelage sorts, through the means test, the deserving from the undeserving. The most important social assistance provisions are the State Old Age Pension (SOAP) for women and men of respectively 60 and 65 years and older7 the Child Support Grant (CSG) for primary caregivers of children up to 15 (in 2008,) and disability grants, payable also to many HIV-positive patients. To receive the CSG an applicant must not only pass the means test but also not refuse employment without good reason (RSA 1998a.) The disability grant, on the 163

other hand, is the only noncontributory, fully decommodified, not work-related social grant claimable by working-age South Africans. Otherwise, those older than 15 and not yet in their 60s are entirely excluded from social assistance. As a result, unemployed workers running out of UIF benefits, first-time jobseekers, informal workers, and many casual employees do not receive any social security provision whatsoever. For them precariousness of work goes hand in hand with a virtually total commodification of life; the more they need public social protection, the less entitled they are to claim it. Unemployed and casual workers are often able to stay above sheer destitution only by sharing government grants paid to elderly, disabled, or child-rearing relatives (Sagner and Mtati 1999.) As work becomes more insecure, workers depend increasingly on their elders: In 2007 a South African low-income household had 3.8 members on average, but households with SOAP recipients had 4.6 (Bester et al. 2008: iv.) The end of apartheid, the deracialization of social assistance, and the gradual rise in the eligibility age for the CSG since it was first introduced in 1998, have boosted the number of social grants recipients. From 2.5 million in 1997, they have soared to more than twelve million in 2008, or approximately one quarter of the population.8 Whites received half of all social expenditures in 1975; by 1993, they claimed only one sixth (Seekings and Nattrass 2005: 151.) Sharing multiple grants within a household helps alleviate poverty and, for as many as two and a half million South Africans, social grants make the difference between survival and starvation (Meth 2006: 432.) Considered individually, however, grants are meager. In October 2008 the monthly SOAP was set at R 960 (96 US dollars,) the CSG at R 230 (23 US dollars) per child, and the largest disability grant was also R 960. Only by the mid-2000s have increases in state grants started to keep pace with inflation (May and Hunter 2005: 121.) With the rising prices, in the second half of the 2000s, of food and basic necessities, the value of grants declined again. 164

The stern gaze of the state and the stigma of public opinion hover above social grants claimants. Popular media constantly parade South African versions of the welfare queen stereotype. Newspaper stories abound on the unregulated sexuality of young girls who become pregnant to access child support grants to buy clothes or mobile phones or feed their drug habits.9 Minister of Social Development, Zola Skweyiya, offered to the opprobrium of Parliament cases where a mother who drives a BMW comes to fetch a grant and then uses it to have her hair done (cit. in Di Lollo 2006: 16.) Other officials gossiped of people with HIV refusing treatment to lower their CD4 count and become eligible for disability grants, or even cases of self-inflicted HIV infection (Nattrass 2007: 185.) In presenting the 2004 national budget, minister Manuel justified a moderate raise in the disability grant with the argument that more generous increases could be a disincentive for people to look for jobs as being disabled could suddenly appear relatively lucrative.10 The governments own statistics deny generalized perverse behaviors by grant claimants. Yet, officially sanctioned rumors weave a narrative of immorality to welfare fraud, which shores up the residualist view that, to avoid undermining the value of work, social grants must be confined to specific, socially approved individual malfunctions, and not act as measures to alleviate poverty and unemployment (Department of Social Development 2006a.) Official concerns that already small social grants could run out of control at any time hardly reflect the truly universal and also surprisingly generous welfare state Jeremy Seekings (2005: 49) imagines as a distinctive aspect of South African exceptionalism. He nonetheless aptly exemplifies the mainstream commonsense that poor people expecting more from the state are either ungrateful or inconsiderate: Although South Africas public welfare system is exceptional, South Africans typically take it for granted, contemplating

165

extensions of the welfare system as if such initiatives were commonplace around the world (Seekings 2005: 49.) As government interventions for the poor, social assistance grants remain largely secondary to work-based programs, education, and training (Meth 2007b: 93-94.) The governments 2001 human resource development strategy, called A Nation at Work, anointed as its organizing principle employment, defined as the full range of activities that underpin human dignity by achieving self-sufficiency, freedom from hunger and poverty, self-expression and full citizenship. Nationhood and productive citizenship are inter-dependent, and it is in this sense that we speak of a nation at work for a better life for all (RSA 2001a: 5.) It was a rallying cry for the total mobilization of the citizens material, symbolic, and emotional assets to wards work as the frontline in the war on poverty; welfare dependency became tantamount to national treason. Government planner Alan Hirsch (2004: 11) proclaimed that government does not want to build conditions of dependency or entitlement attitudes. South Africa cant afford to become a granny state. The expansion of social grants programs has been, nonetheless, remarkable. The SOAP, currently paid to three quarters of old-age South Africans, is indeed approaching universal coverage. Left critics who, like Patrick Bond (2005,) see post-1994 South Africa as an unrelenting slide towards market liberalization fail to account for such a divergence from canonical neoliberal prescriptions. Perhaps growing grant coverage is to some extent an unintended policy effect as deepening poverty allows more and more families to pass the means test. Treasury officials grew thus alarmed at the unsustainability11 of grants. Other government bureaucracies, particularly the Department of Social Development, have, however, praised the social impacts of grants and even mooted the abolition of the means test.12 Contrasting policy 166

positions reflect the dilemmas of the ANC for which, despite their financial burden, social grants are highly functional to control the political loyalties of poor and potentially restless communities. Through public assistance state sovereignty politicizes social reproduction by welding the social security of the poor to the security of the state. Local patronage networks, especially in rural areas where tribal chiefs act as enablers and facilitators, bind ANC politicians to claimants for whom grants are often the only monetary resource available (Niehaus 2006a.) It is indeed not infrequent for ANC candidates to downplay, in election times, the partys anti-dependency tirades and boast social grants among voters who expect them as rights. James Ferguson (2007) argues that the grants operate as a technique of poverty alleviation that transcends the conventional opposition of welfarism and neoliberalism as they recognize that precariousness and informality are inevitable features of postapartheid society and try to incorporate them in its system of governance. In the process, social grants define a grey area of second-class citizenship where the borders between inclusion and exclusion fade and overlap. They institutionalize able-bodied members of recipient households as individuals for which the compulsion to compete for uncertain job opportunities must never fail. The South African system of social grants, in other words, generalizes and organizes precariousness while containing its undesirable social consequences. It also makes the uncompromising condemnation of dependency and entitlement particularly powerful by associating it not to cold market rationality, but to the image of a caring and compassionate state. The pedagogical function of fostering positive orientations to work permeates postapartheid social policy. The healthcare system is largely privatized and vastly unequal. In the mid-2000s per capita private curative care expenditures affordable for less than twenty percent of the population were seven times higher than in the public sector (Leon and Mabope 2005: 167

33.) A small minority of workers can afford private care through company-based health insurance. The vast majority, and the virtual totality of unemployed and precarious employees, rely on depleted public facilities, which deliver free primary healthcare but limit free hospital care to low incomes or specific needs, mainly children under six and pregnant women. Even for workers with health insurance, companies have responded to escalating medical costs and the HIV-AIDS epidemic by confining their contributions to low-coverage schemes, expanding preauthorization for treatment, or replacing insurance altogether with lump sum cash payments (Old Mutual 2005: 49-51.) Post-1994 healthcare policies bowed to corporate resistance to fiscally funded national health insurance. In 1997 the government rejected proposals for a public system even if funded through payroll taxes, which would have covered ten million low-income citizens and precarious workers.13 Postapartheid policymakers followed instead in the footsteps of their predecessors by explicitly recommending employment-based care to contain state funding for public hospitals (RSA 1995b: S.23.) A universal system financed through redistribution would have been, conversely, an encouragement of an entitlement, as opposed to an insurance, mentality (RSA 1994c: 57.) The 1996 Constitution defines healthcare as a right, which the ANC government has tried to enforce by equalizing access to private insurance, establishing minimum benefits packages, and eliminating discrimination based on pre-existing conditions. The private sector has, however, generally managed to avoid servicing low-income markets.14 To access decent healthcare amidst policies that encourage behaviours that will prevent people ultimately relying on the state (Department of Health 1997a: 7,) a regular job remains the necessary, albei t by no means a sufficient, condition. Housing programs have also revealed deep continuities with late-apartheid policies, which celebrated homeownership and individual self-help as hallmarks of respectability and 168

social inclusion. The ANC government has eschewed large scale public housing projects and tried to attract private finance for low-income accommodation by devising partnerships with labor and civil society organizations (Pottie 2003.) Policymakers defined this a residualist and incremental approach to housing delivery, aimed to provide security of tenure and access to basic services as well as possibly a rudimentary starter formal structure to the poorest of the poor (Department of Housing 1994: 5.3.7.) In practical terms, the policy meant for the unemployed and low-income workers, largely living in self-built informal housing or backyard shacks, an updated version of the apartheid-age site and service programs, which merely provided some land a tap, a toilet, and at times a basic foundation and a minuscule room (Morris 1998: 766.) Upon such rudimentary beginnings residents were supposed to build formal homes with their own resources. Self-help ideology thus purportedly reconfigured the homeless as partners, not claimants, of policies. For a government consultant, low-income citizens would access housing by providing their labour on a voluntary basis, generally without remuneration (cit. in Pottie 1999: 94.) To enable self-help and home ownership the government deracialized public housing subsidies that in the past had predominantly privileged whites. A new subsidy program has provided homebuilders with starter finance to attract mortgage lenders, which the apartheid regime had failed to bring into black townships due to social unrest and utilities nonpayment (Bond 2005: 122-51.) For the purpose, ANC policymakers even vowed to indemnify investors against politically inspired risk, meaning boycotts and demonstrations by the ANCs own supporters.15 Subsidies alone would be in fact too small to build habitable structures; as the Housing minister put it, they will seldom, if ever buy the kind of house people imagine when they turn to daydreaming.16 Their amount was set on a sliding scale according to earnings, and 169

was payable only to indigent households. Almost one quarter of beneficiaries had in the end no income or survived on state grants, but also many formal and informal workers turned out to be poor enough to claim more than one third of individual subsidies, which comprised one sixth of all subsidies (Public Service Commission 2003: 86-87.) By limiting public interventions and emphasizing individual resources, the subsidy program confirmed and deepened social and spatial housing inequalities, which now depended on income and occupations rather than race. As new black elites started to move into affluent, previously white-only suburbs, a minority of black workers with stable jobs and benefits could own newly privatized former council houses in the townships, while the majority of unemployed and precarious workers ended up in informal and self-built, often subsidized, dwellings (Royston 2002.) To support market discipline, the corporatization of municipal services turned water and electricity utilities in many cities into income-generating companies relying on user payments to cover costs (McDonald 2002.) Residents who failed to pay for their rising bills could be disconnected or evicted. The Department of Housing defined as affordable monthly rates for full services (piped water, waterborne sanitation, and basic electricity connections) that could take as much as half the wage of a casual worker (RSA 1995c: 6.2.2.) A rhetoric of individual empowerment disguised the centrality of market forces as housing access was not a goal of public policy but a function of employment and income. Responding in Parliament to the observation that housing is a right, not a commodity,17 the Housing minister and prominent Communist leader, Joe Slovo, warned that the government would not keep listening to excuses18 from poor residents unable to pay for bills, rents, and mortgages adding: Let me make very clear that people who refuse to respond to [our] very fair offers will be occupying accommodation which neither belongs to them nor to which 170

they have any right. We are determined for the rule of law to take its course in these instances. We are saying to the communities that it is in their interest to support the rule of law, because only then can they hope to attract housing investment on a sufficiently large scale.19 ANC politicians recursively used revolutionary and marxisant jargon to argue for market discipline and public cutbacks; one parliamentarian defined the post-1994 housing policy as a dialectical unity20 of government subsidies and corporate finance. For minister Slovo it was a way to end the undeclared war between communities, the state and the private sector by giving the poor opportunities of individual market participation, instead of making them passive recipients.21 Even left leaning civil society organizations argued that subsidies should avoid substandard housing by reducing recipients dependency on the state22 and promoting the ingenuity and commitment communities demonstrated, even when confronted with limited resources.23 Confirming the emphasis on personal responsibility in housing policy, the governments Masakhane (lets build together) campaign was a massive media and symbolic effort to convince residents in low-income communities to pay their bills and cease boycotts in order to attract private development. In the end, ownership of a formal home has remained unreachable for most unemployed and precarious workers. Their only alternative has often been, therefore, to spend government subsidies to build with inferior construction materials and in frequent violation of standards poorly designed shanties that rapidly deteriorated or collapsed. In 1999, only 30 percent of subsidized houses complied with building regulations (National Institute of Economic Policy 2000: 30.) Presented as an opportunity for the former disenfranchised to enrich citizenship with the pride of property, the subsidy program ended up generating widespread disillusionment 171

among its intended beneficiaries for whom it often just meant leaking tin roofs and cracking walls (Zack and Charlton 2003; Magasela et al. 2006.) Research has indeed often shown that workers would rather prefer controlled and subsidized rental accommodation to elusive homeownership prospects that only add to their precariousness.24 In conclusion, a social policy discourse infused with anti-dependency morals and images of individual initiative tried to bring to life a virtuous individual citizen at once worker, consumer, and owner as the imaginary partner of the state. Wage labor was its pivot, despite the inability of employment to underwrite decent incomes, social security, retirement provisions, healthcare, utilities, basically functioning housing, or elementary necessities for the large majority of wage earners. The second-class citizenship of precarious workers mirrored indeed the second economy official metaphor of their productive functions. They survive through relatives paltry social grants, overcrowded and underserviced public healthcare, and subsidized shack-like accommodation, which materialize the Malthusian rhetoric of perversity (Hirschman 1991) used by politicians and experts to depict decommodified social provisions as burdens on the state and pathologies corroding work ethic. Economist and government consultant Servaas van der Berg (1996: 13,) for example, presented wayward welfare laxitude almost as a viral contagion, where social grants leaking from the poorest to the not-sopoor cause perverse incentives to work search behavior. Belying the constitutional universalism of rights are representations of the non-working poor as an alien threat to the body politic, for which self-help is a psychological as well as social treatment. Social assistance as a mere lifeline limited to the near destitute is all that remains of old popular demands for universal decommodification, operating mainly as a therapeutic signifier of the victimhood and suffering of entirely objectified recipients. Results can be disquieting: a disability grant claimant once told 172

a researcher: Yes, I like this HIV/AIDS because we have grants to support us (cit. in Nattrass 2006: 13.) Casual and irregular workers are, in the end, suspended between the opposite polarities of official discourse: not vulnerable enough to legitimately claim social grants, and yet unable to rely on the wage for a decent living, their precariousness originates in the labor market, but social policies extend it to all ambits of life. Yet, the incongruities of social policies predicated on waged work while few have stable jobs with benefits can test even the most impervious paradigm. In South Africa, the precariousness of work would eventually reopen the contestation over the meaning of postapartheid social citizenship.

Laudable Citizens and Silly Fools. Work, Families, and the Developmental Social Welfare Idea

It is not surprising that welfare-to-work, or workfare, experiments in Europe and North America have influenced South African debates on employment-centered social inclusion. For the first minister of Welfare in the ANC-led government, NPs Abe Williams, work was a moral, not just social, corrective to poverty: We must at all costs avoid a dependency syndrome. As President Bill Clinton said, Social welfare must create a second opportunity, it must not be a way of living.25 As a word, workfare has never been popular in South African policy debates. The 1996 Labour Market Commission rejected it as implying coercion and stigma and deemed it irrelevant to a country with millions of long-term unemployed and few available jobs. Many experts, however, 173

recommended social benefits conditional on the work effort rather than just aimed at victims of circumstances (Olivier 2000a: 29.) COSATU itself cautiously endorsed, in the name of job creation, a proposal by minister of Labor, Tito Mboweni, to bind unemployment benefits to their claimants use of job placement services (COSATU 1999b: 6-7,) a position that even the Labour Market Commission rejected. The labor federation, nonetheless, continued to oppose as workfarist the governments focus on active labor market policies in alternative to income support provisions (Coleman 1999.) The Department of Labour (1999: 33,) on the contrary, preferred retraining and job placements, albeit often at a lower pay, and flirted with the option of relaxing collective bargaining regulations to facilitate hiring. Soon after it rose to power, the ANC-led government defined the fight against perverse uses of social grants a priority and launched a campaign against welfare fraud. In 1996 an official report defined fraud as not only intentionally criminal or deceptive behavior but any occurrence when money is paid to people who should not be in the system in the first place (Department of Welfare 1996a: Ann.C, 7.4,) including, therefore, recipients no longer passing the means test. In response, a national system of biometric surveillance, of which South Africa provided then one of few examples in the world, was put in place. Its technologies and procedures derived directly from apartheid-age counterinsurgency programs (Breckenridge 2005.) By then, Geraldine Fraser-Moleketi, a Communist Party official and Williams former deputy, had become the ANCs first minister of Welfare. She shared her predecessors aversion for the culture of entitlement and framed it within the RDPs and GEARs aspirations that active labor market policies would replace pensions as the mainstay of social security for poor people.26 She proposed to link social security grants to developmental programmes so that recipients would be empowered to become self-supporting.27 The ANC was particularly 174

concerned that low-wage workers were not saving enough for their retirement, preferring to rely on the states SOAP. The chair of the parliamentary Portfolio Commit tee on Welfare, Cas Saloojee, complained that the SOAP was becoming unsustainable and destabilizing,28 and suggested a national contributory insurance scheme to make the poor fund their own unproductive life, confining public old-age grants only to people who cannot fend for themselves.29 Such early debates generated the concept of developmental social welfare, with which the ANC accommodated its social policy approach to the self-imposed resource constraints and imperatives for market-driven growth in the RDP and GEAR. Developmental social welfare was powerfully shot through with the ANCs contempt for decommodified social provisions and its suspicions on the alleged lifestyles of their recipients. Social grants were, therefore, to be limited to special needs while job creation, not direct income support, would address the plight of the unemployed. An early testing ground for the approach was the introduction in 1998 of the Child Support Grant (CSG,) the brainchild of a government committee chaired by academic Frances Lund. Following the recommendations of the Lund committee, the CSG initially set at a mere R 100 per month (the committee had recommended R 75,) and only for children aged 0 to 6 replaced an apartheid-age State Maintenance Grant, substantially more generous but also racially skewed in favor of colored and Indian families as part of the National Partys network of patronage (Vorster, Rossouw, and Muller 2000.) The Lund committee worked in a governmental climate shaped by GEARs draconian fiscal thrift and injunctions like the ANCs Cas Saloojees to challenge the beneficiaries [of child support] to participate and work in some socially needed project.30 The ANC totally opposed the extension, even in a downsized form, of the existing State Maintenance Grant, which Saloojee deemed as unsustainable as funding a 175

tenth province.31 Its eventual replacement with an extremely modest CSG was the first case of elimination of a major program in the history of South African social security (Makino 2003: 12.) Faced with the governments prohibition of programs increasing social expenditures, the Lund committee, by admission of its chair, didnt argue very strongly on grounds of social justice [but] on grounds of what practically works.32 The replacement of the old maintenance grant with the CSG inflamed nonetheless the unions and civil society organizations, despite Fraser-Moleketis intimation that there is no turning back (cit. in Calland and Taylor 1997: 212.) COSATUs chemical workers union called the CSG a return to apartheid (Zain 2000: 15,) while the federation branded it a recipe for disaster (COSATU 1997b.) Reflecting with hindsight, Frances Lund (2008: 116) commented that there were moments of real despair and sadness when the compromises seemed too severe to live with. Social welfare activists considered therefore a success that a child support program had survived at all, considering that the government was probably favorable to its total suppression.33 The Lund report had, however, ideological as much as practical importance. Its main emphasis was on the parents sense of financial responsibility as a way to minimize redistribution, treading the culture of payment path indicated by the Masakhane campaign for municipal utilities: [Masakhane] has had to contend with years of conditions during which a culture of resistance to an illegitimate government condoned - and even lauded - nonpayment. A similar campaign to build a culture in which parents accept their responsibility towards their children, and in which those who pay are regarded as

176

responsible and laudable citizens rather than weak and silly fools, is urgently needed (Department of Welfare 1996b: 5.3.2.) The report could thus present the material limitations of grants as a positive pedagogical function. In praising individual responsibility against dependency, it fantasized that the nuclear family could powerfully motivate welfare recipients to move into jobs, no matter how poorly paid: A concern with social security benefits is that they may reduce the likelihood of people seeking jobs or other ways to provide for themselves. . . . In a country with such high unemployment rates, a simplistic answer to this question is: 'but there are no jobs to be had anyway.' This is so, but labour market patterns are changing internationally, with more and more people (especially poorer people) making their own jobs, or engaging in a range of productive activities, some waged and some unwaged, to put together a living. South African wages for women, and for women in small towns and rural areas, are so low that the value of the State Maintenance Grant may commonly be higher than the lowest local wage (Dept. of Welfare 1996b: 3.5.4.) After Lund, the denunciation of maintenance-oriented social programs (Midgley 1997: 180) became crucial to the ideology of developmental social welfare. South African proponents of the paradigm were strongly influenced by the capability approach popularized by Amartya Sen (1985) and Martha Nussbaum (2000,) which prioritized the aptitude of individuals to achieve wellbeing through the functional use of material, psychological, and social assets. There were, however, also discrepancies. Nussbaum (2003,) for example, intends wellbeing as a condition to be achieved across the life course and does not prescriptively subordinate it to work. Not only 177

does the capabilities approach positively regard entitlements, anathema to the ANC, but it also refuses to equate human capabilities with the production-focused concept of human capital, dear to ANC ideologues (Patel 2005: 104.) In 1997 the government issued after a vast process of civil society consultation presented as a break from apartheid-age expert-driven policymaking34 the White Paper on Social Development. The document aimed to devise a comprehensive social assistance to those without other means of support, such as a general means tested social assistance scheme (RSA 1997: 7.26b.) Warning that in view of fiscal constraints, it is not possible for the welfare function to grow in real terms in the medium term (RSA 1997: 8.11) and the high expectations of many people for social delivery cannot be fully met in the short-term (RSA 1997: 1.9,) the White Paper supported selectivity and residualism. Public programs would target the poor, those who are vulnerable and those who have special needs as not integrated into the mainstream of society (RSA 1997: 8.4.) Self-reliance was the guiding concern; the states role was to provide citizens with an opportunity to play an active role in promoting their own wellbeing. Redistributive interventions were legitimate only as investments which lead to tangible economic gains (RSA 1997: 1.8.) Otherwise, more stringent and appropriate means testing and eligibility requirements would divert people from the welfare system (RSA 1997: 6.22) into employment. The document unreservedly praised work as the antidote to exclusion and did not recommend any specific social protection for the working poor. Only after intense pressures from the unions (COSATU 2000a: 4) it included a vaguely phrased commitment to comprehensive social assistance for everyone without means of support. The white paper imagined individual responsibility and families with regular breadwinners as pillars of self-empowering collectivities. The framework combined modernist 178

ideas of economic activity with an allegiance to African cultural values, primarily the concept of ubuntu as communal support and reciprocal obligation (RSA 1997: 2.24.) The position of ubuntu was, however, contradictory: as a compassionate ideal rooted in cultural specificity and authenticity it evoked local traditions in opposition to the bureaucratic abstractions of Europeanstyle welfarism. For this purpose, however, it had to become ancillary to the equally Western, neoliberal idea that personal responsibility is enabled by social spending cutbacks. The white paper invested the family with functions akin to what Jacques Donzelot (1979) identified as moralization, normalization, and tutelage and critics branded as gender-neutral, functionalist and moral terms (Sevenhuijsen, Bozalek, Gouws, and Minnaar-McDonald 2003: 306.) Its glorification of the family as the basic unit of society meant, in fact, a nuclear family revolving around the roles of male breadwinner and female caregiver. The ANC would subsequently stress the ethical dimensions of its familist discourse. A 2005 draft policy document tasked the family with rebuilding the moral fibre of the nation in a mor al regeneration movement (cit. in Sewpaul 2005: 315) that educates the youth to work and independence. The emotional appeal to the resilience of an idealized family erased the relevance of social processes and power relations. The draft document continued: The policy must empower families to have resilience in dealing with challenges facing them, and be able to access and utilize resources. The value of self-reliance over dependency and learned helplessness must be enhanced. Families should therefore be capacitated regarding socio-economic issues without creating dependency. The family should restore its pride and dignity in order to reverse dependency and the displacement of family responsibility (cit. in Sewpaul 2005: 318. Own emphasis.) 179

The intellectual architect of the white paper was Leila Patel, a social work professor and, from 1995 to 1999, the Welfare departments director general. Patel argued for a family-centered and community-based social policy to propagate entrepreneurial values with which

disadvantaged groups could overcome their passivity (Patel 2001.) She actively promoted the sub-contracting of welfare services to companies, nongovernmental organizations, and community associations (Department of Welfare 1999: 19-25.) With the help of non-state actors, she wrote, the poor would become empowerment partners in the privatization of state assets and potential shareholders . . . without making capital contributions (Patel 2001: 39, n.11.) As the feared antithesis of the enterprising poor, the passive one was supposedly the pernicious product of T.H. Marshalls income-based redistribution (Patel 2005: 96-118.) For Patel the white paper responded to the degeneracy of traditional welfarism with a productivist, asset based . . . active social citizenship rooted in personal lifestyle decisions in which expectations about entitlement are backed by co-operation and self-restraint (Patel 2005: 99.) Patels defense of market discipline, however, eschewed a mere identification with neoliberalism. From her activist past and social work practice she instead resurrected and conveniently refashioned progressive-sounding cultural repertoires in a legitimating role. Developmental social welfare became therefore the reclamation of an authentically African, nonEurocentric way of caring, based on reciprocity and community development. Patel found ammunitions for welfare cutbacks in the idea that in pre-colonial times, the welfare needs of individuals were met through the wider society, and communalism, co-operation and mutual support by individuals and the social group were highly developed (Patel 1992: 34.) Colonialism disrupted African self-reliance, dignity and respect of tradition not through capitalist market relations and wage labor but by imposing curative, institutional, and 180

bureaucratic European welfare models (Patel 1991: 105.) GEARs caps on social expenditures, conversely, promised to restore the authentic development which has grown out of the real conditions and traditions of life in South Africa and the right and duty to work in the 1955 Freedom Charter (Patel 1991: 124.) South Africas alternative welfare delivery would replace the bureaucratic straitjackets of European welfare culture with flexible societal partnerships based on participation, volunteerism, and fees for services (Patel 1991: 109.) Universalism found no place in Patels diagram. For those unable to lift themselves by the proverbial bootstraps there were only affirmative action in the private sector and selective social policies in the public (Patel 1991: 260.) A remarkable discursive turn made corporate governance and managerial rationality the modern embodiments of timeless African community traditions. The echoes of liberal multiculturalism as it mobilizes progressive themes of authenticity, difference, and place-specific customs to disable the alleged suffocating homogeneity of redistribution (Kymlicka 1998) were clearly audible. The discursive shift did not, however, happen in a void, and Patel probably overlooked how the pure and authentic African traditions she reclaimed were in fact indelibly inscribed in the imagination of colonial developmentalism. In South Africa the social policies of segregation and apartheid, often supported by African elites, were hardly curative and institutional, and their bureaucratic deployment rather endorsed a pseudo-traditionalist imagination that closely resembles Patels, and for quite similar purposes. White minority regimes imagined, in their own way, selfsustaining and static indigenous communities to argue that, even when becoming modern, African workers could not aspire to European social standards. To avoid sharing the fruits of its obsession with work discipline and productivity, the apartheid state fantasized the native poor as naturally predisposed towards pre-capitalist affiliations. Its stereotypes of African communal 181

solidarity, informal safety nets, income pooling, and microcredit counteracted demands for universal social programs (Webster 1984.) For the Africans designated as permanently urbanized, colonial governmentality recommended the Western-styled nuclear family mostly to shift the burden of care from the state to patriarchal breadwinning and unpaid female housework (Ferguson 1999: 170-77.) In the end, Leila Patels developmental social welfare strikingly evoked Britains late Colonial Secretary, and former trade unionist, Arthur Creech Jones, for whom African progress depended upon the development of a sense of community obligati on and social responsibility and service (Lewis 2000: 316.) Colonial rule and free-market ideologies of individual initiative share a common penchant for the rhetoric of tradition and authenticity when it serves to contain popular claims. To comfort the white papers compassionate and culturally sensitive advocacy of lowwage employment, social surveys regularly indicated that the poor wanted work, and would do almost anything to obtain it. Respondents, nonetheless, recognized unemployment as the result not of failing characters and morals, but of socioeconomic processes that could only be countered by public interventions, progressive taxation, and more generous social programs. Tellingly, the poorest did not seem particularly fond of job training (Community Agency for Social Enquiry 1998; Noble, Ntshongwana, and Surender 2008.) Armed with developmental social welfare, the ANC could now, however, discredit the very idea that political liberation entailed a structural change in the distribution of resources: The idea that the world now owes me a favour because I was a victim of apartheid oppression may well be understandable, but it simply confirms and continues a cycle of dependency (African National Congress 1997.)

182

As social claims were portrayed as inimical to authentic African work values, the labor market was no longer, as Somers and Block (2005: 276) put it, an external economic reality but an immanent source of ethical conduct. Prominent political commentator Xolela Mangcu (2002: 95) reflected a general intellectual mood by contrasting assets, or the human capital of the poor as a guarantee of sound citizenship, with income as the negative signifier of redistributive demands with debilitating political effects. Moral discourse, policy rationality, and socioscientific assessment congealed in what Sanford Schram called the psychic life of welfare: Welfare must of necessity continually distinguish itself from the specter of the other the bad welfare, we might say that haunts it. Welfare must continually prevent itself from appearing to pose a threat to the dominant norms of work and family. It must continually prove itself to be the good welfare that promotes work and family norms, or it will lose legitimacy and disentitlement politics will accelerate (Schram 2000: 160.) Developmental social welfare was concerned with, to use Mitchell Deans (1999: 162) formulation, how to govern through the autonomy of the governed when they are no longer virtuous, or how to orientate the poors free choice towards work and away from the legacy of antiapartheid claim-making politics. To turn the poor into self-empowered partners the state had to punitively rein in the entitlements that could lead to their self-representation as recipients. The underlying social imagination is, in the end, contradictory: in it the poor, by their very nature, simultaneously yearn for work and drift to avoid it. COSATUs initial reply to the white paper on welfare shared many of its ideological assumptions, including the centrality of active labor market policies, concerns with dependency, and the idea that social assistance should privilege deserving poor on a means tested basis. It 183

rejected, however, the separation of social insurance and social assistance, and supported a universalist approach, which included generalized public income support for the unemployed (COSATU 1996.) According to the federations parliamentary officer, Neil Coleman, however, the unions disliked the notion of developmental social welfare as an appropriation of a developmental discourse to serve conservative ends and legitimize the idea that social security depends on community self-help, which is ludicrous. The fact is that one of the tragedies of the first term of government was that people can be attracted by these discourses, it is an emperor with no clothes sort of phenomenon. . . . They were trying somehow to pass the burden onto [the poor] to develop their own social security networks, which was part of a broader economic package to divest the state of responsibility and to shift the burden from the state to poor communities.35 The unions opposition was, nonetheless, stifled by their single-minded focus on job creation. In 1996 the three main labor federations, COSATU, NACTU, and FEDSAL, jointly attacked the ANCs liberalization policies but, while demanding fiscally funded public works and laborintensive housing projects, they hardly mentioned social security (COSATU, FEDSAL, and NACTU 1996.) In 1997 COSATU appointed a panel, the September Commission, to conduct a broad organizational and strategic reassessment. The federation was by then conversant with the work of ILO analysts, especially economist Guy Standing, who, working as advisors for the government, stressed the link between the growing casualization of employment and the inadequacies of social provisions (Makino 2003: 14.) COSATU grew impatient with the ANCs social policy framework

184

which as advanced by certain technocrats, is interpreted to mean that communities should in the first instance look after their own, purportedly to avoid dependency. This approach arguably contributes to an abdication by the state of its social security responsibilities, and a deepening of poverty and dependency (COSATU 2000b: 3.) COSATU demanded instead a social wage, a term that gradually replaced the time-honored living wage, now considered to be too narrowly focused on monetary quantities (NUMSA 1996: 14.) The federation defined the social wage as an economic floor below which no South African should sink. The social wage comprises direct income transfers (such as social security and retirement benefits) and social subsidization of the costs of basic needs (such as housing and health) (COSATU 1997a: 61-62.) Egalitarian redistribution would allow people to function effectively as part of society (COSATU 1999c,) increase domestic demand, and contribute to economic growth, avoiding a skorokoro (broken down) scenario of economic stagnation and lack of social delivery (COSATU 1997a.) By covering costs of household reproduction, the social wage would also partially relieve employers of the wage burden of hiring unskilled labor (COSATU 1998.) In the end, the goal was to increase labor market participation, not alleviate dependence on waged jobs. Sure enough, COSATU wanted new jobs to be decent, secure, and with adequate provisions (COSATU 1997c.) As waged employment declined, the unions tried to speak to the plight of precarious, unorganized employees rather than closing ranks around their due-paying, regularly employed members.

185

The social wage revived, in COSATUs intentions, redistribution as a counterweight to the ANCs corporate and managerial drift. It also recognized that employment-based social provisions contradicted labors call for universal coverage and eroded the ability of workplacebased identities to address the socioeconomic challenges of democratization. Awakening from the shock of economic liberalization, labor leaders concluded that to salvage wage labor as the core of a general democratic imagination of citizenship they had to directly contest governance and assert their visibility in society at large. COSATUs Neil Coleman commented that: Despite the propaganda that you see in the media about COSATU being a narrow, economistic worker elite blah-blah-blah, the reality is that we could have adopted a very different strategy. We could have negotiated with the private sector, for health care benefits, for housing benefits funds, and retirement funds. But instead on all these issues . . . we are interested in a public strategy which is based on employment, and in a policy position that the social wage must extend across society and get to unemployed people, to the poor, and so on.36 COSATUs policy proposals, backed by left-leaning academics, started to tackle the inadequacies of jobs in ensuring decent lives. The federation, however, fell short of problematizing the glorification of employment in the ANCs policy discourse, which would have implied uncomfortable questions for the unions own heroic narrative of the universal worker as the harbinger of freedom. In the end, COSATUs demands principally revolved around social spending for job creation, including fiscally funded temporary public works (COSATU 2004.) The federation had, moreover, no desire to antagonize the ANC, whose leaders were already tagging the unions as ultra-left (COSATU 2002.) By presenting its demands as basically compatible with the ANCs framework of poverty alleviation COSATU 186

prioritized its allegiance, under the banner of the national democratic revolution, to the ruling party, which the unions still considered as essentially a working-class movement albeit momentarily enthralled by corporate capital and unaccountable technocrats (COSATU 2000b.) In October 1998, the government convened at NEDLAC a presidential jobs summit with the aim to strengthen the employment impact of existing policy initiatives and introduce new feasible initiatives while avoiding to try and bring about major changes to the governments [sic] policy (RSA 1998b.) The government wanted a renewed societal consent for active labor market policies, training, and job placement, with an eye to strengthening smallmedium enterprises and special community-based public works. Rejecting the unions demands, the ANC did not commit new funds to employment policies but expected the unions to rally behind the Masakhane campaign and organize members to pay for rents, services, etc. The three labor federations submitted twenty-one joint programs for job creation (COSATU, FEDSAL, and NACTU 1998.) Three of them hinted at the provision of decommodified social services. The first proposed a state-run program of rent-stabilized housing in exchange for labors support of the Masakhane campaign (COSATU 1999d.) The government continued, however, to prefer incentives which enable private sector delivery and private ownership (RSA 1998b) to public rental housing. The second program (named support measures for the unemployed) requested the government to resume contributing to the UIF. The third program, the twentieth in the unions list, proposed a basic income grant for poor South Africans, particularly for the unemployed. The unions presented the Basic Income Grant (BIG) as consistent with the promise of a comprehensive social security system contained in Section 7.26 of the welfare white paper. They set the proposed BIG at R 100 (approximately 10 US dollars) per month to be paid to every South African citizen without means test or employment 187

requirements. Individuals with monthly incomes above R 3000 would repay the BIG through their taxes, guaranteeing the progressive and redistributive nature of the program. With the BIG, for the first time the South Africa labor movement demanded not just relief for the jobless, but a universal income transfer of a decommodified nature, not dependent on the recipients ability to sell their labor power. The proposed BIG was, nonetheless, quite modest and in no way could be thought of as a meaningful income replacement. The unions, however, saw R 100 a month as a winnable demand, a possible breach in the wall of governmental reluctance.37 For COSATU the BIG was not an alternative to the wage, nor was it meant to question the dogma of employment-based social inclusion; it would rather stimulate domestic demand and encourage the desperately poor to shake off their alleged apathy and look for jobs. As a palliative to rampant unemployment, moreover, basic income could forestall possible popular resentment at the ANC and its trade union allies.38 The amount the unions proposed for the new universal grant was indeed so tiny that it hardly corroborates the claim by Jeremy Seekings and Nicoli Nattrass, for which they provide no evidence, that labors BIG idea was a self-interested attempt to defend the privileges of union members by offloading the maintenance costs of their poor relatives onto the government (Seekings and Nattrass 2005: 393.) Dismissing the unions motives as substantially duplicitous and heartless does not help understand the uses of the basic income grant in the interactions between COSATU and the ANC. By concentrating on the feasibility of its proposal COSATU tucked the social wage into the alliances debates, where it sought the mobilization of ideological support39 against what it perceived as a slow, inexorable neoliberal drift. While it pushed the BIG in NEDLAC and Parliament, the federation tried to build supportive civil society alliances. NEDLAC alone was, in fact, a slippery slope as Treasury representatives actively 188

stonewalled any proposal for social programs that implied increases in public spending.40 Business delegates shared the governments commitment to fiscal restraint, which in NEDLACs consensus-seeking atmosphere stacked the deck against radical programs of comprehensive social inclusion. Sometimes the Treasurys position overtly clashed with so-called delivery departments, like Labor and Social Development, more open to redistributive interventions. 41 The governments reception of the BIG was, however, also constrained by its priority on the reform of unemployment insurance, also under consideration at NEDLAC.42 Initially the two negotiating processes were separate and uncoordinated,43 and in the end labor representatives were more urgently concerned with the practicalities of unemployment insurance than with unfamiliar, hazy ideas of universal basic income.44 Despite the unions ambitions to build a broad social mobilization in support of basic income, COSATUs Oupa Bodibe resignedly concluded that in the end labor relied too much on parliamentary work and NEDLAC.45 COSATUs tactical limitations, however, had probably to do with the timidity and uncertainty of its proposals. The weight of its productivist discourse did not allow radical alternatives to work-centered policies to take off. COSATU continued to regard contingent and unprotected workers as the victims of liberalization, the socially marginal to be brought back to the fold of the real, stable, unionized proletariat. But wage labor as a condition of stability was vanishing, and claims for universal basic income have therefore underscored a critique of the postapartheid ideology of work and production. COSATUs cadres, instead, chose to see the BIG as facilitating entry in the labor market and participation in the economy.46 For them the involuntarily unemployed, rather than the working poor, were the main beneficiaries of universal basic income. Mimicking the anti-welfarism of the ANC, COSATU distinguished indeed

189

between good income programs enhancing peoples employability and opportunities for selfemployment (COSATU 2000c: 9.5) and bad ones creating disincentives to work: Such an approach takes forward the developmental and redistributional economic logic of the RDP focussing [sic] on redistributing wealth and opportunities and bringing people into economic activity. This approach aims to fundamentally restructure social relations rather than being a purely welfare driven approachwhich partially alleviates but also institutionalises poverty, by perpetuating marginalisation and dependency (COSATU 2000c: 4.5.) Other social actors, however, approached universal basic income less deferentially. In 1998 the South African National NGO Coalition (SANGOCO) convened poverty hearings aimed at making the lives of the poor more visible in policy debates. The final reports of the hearings brought poverty back from the abstraction of statistical indicators into harrowing stories of physical violence against women and vulnerable workers, polluted neighborhoods, collapsing subsidized houses overflowing with human waste, public hospitals without doctors or medicines, deaths from preventable diseases. Many respondents said they could not cash grants they were entitled to because they had no money to travel to distant government offices, or had their applications frustrated by racist and incompetent officials demeaning them with the language of stigma and contempt. The hearings generally followed the script that poverty coincides with unemployment, but they also had to recognize the impact of precarious jobs and poverty wages, which again and again surfaced as major determinants of social collapse (Liebenberg and Pillay 1998.) Public perceptions of widespread poverty and vulnerability among the employed population eventually influenced labors representation of the BIG. The 1998 Jobs Summit had 190

not taken any commitment, but agreed to further debate the idea within the notion of comprehensive social security of the 1997 white paper. Two institutional processes dealt with basic income. In March 1999 NEDLAC established a task team with representatives from government, labor, business, and the civil society to discuss social support for the long-term unemployed falling through the cracks of the existing safety nets (NEDLAC 1999a.) Few months earlier the government had started its own interdepartmental task team with representatives of various state agencies relevant to social security (RSA 1999.) The NEDLAC team accepted the need for sweeping reforms, encompassing employer-based benefits, social assistance, healthcare, and housing. It was, however, reluctant towards universal basic income, preferring instead a means-tested dole for the long-term unemployed to increase their economic activity and try to reduce dependency on the state through promoting self-sufficiency (NEDLAC 1999b.) The approach took a page from the agenda of business representatives, for whom any guaranteed minimum income could not be more than a workfare-type measure in a work orientated contributory system of social protection where great importance is attached to work generally for reintegration into society and economic independence.47 The government, on the other hand, favored income grants only for citizens below the poverty line, not a universal BIG (Makino 2003: 15.) The new director of the Department of Social Development, Lucy Abrahams, frankly explained that, in times of fiscal austerity, social security is more about economic realities than social realities and grants which are neither necessarily developmental nor sustainable (NEDLAC 1999c) had to be struck down. The final reports of the government and the NEDLAC teams diverged with regard to basic income. The former, released in July 1999, complained that the lack of employment-based private social insurance among low-income South Africans was a burden on fiscal resources 191

(RSA 1999.) It therefore recommended an extension of public provisions, including state funding for the UIF, universal health insurance with a minimum package of benefits, and mandatory, contributory retirement pensions to alleviate claims for state old-age grants. It finally proposed to replace the means-tested SOAP with a universal, contributory old-age program. The document preferred to the basic income grant a monthly benefit for households in extreme poverty. In the end, the governments report rejected universalism, confirmed the role of private provisions, and stratified claimants into three brackets, which mirrored employment hierarchies. The poorest of the poor would receive residual public assistance with no contributions; middle income strata would access contributory, state-subsidized moderate but reasonable benefits; and high income groups would enjoy well regulated private provisions of their choice. The market emerged, once again, as a natural law, not to be tampered with; the states role was to assist those at the bottom so that they would not lose all initiative and hopes in the system. Public assistance for the working poor would relieve employers of healthcare and retirement costs, make low-wage jobs more attractive, and sanitize their second-class citizenship status. The report of the NEDLAC team, issued in March 2000, did include a universal basic income grant among its proposals (NEDLAC 2000.) In the end, organized labor, directly represented in the NEDLAC process, managed to place the BIG idea on the agenda of a state institution. By then, moreover, the concept of universal basic income was energizing the imagination of civil society. In 2001 a BIG Coalition was formed with the participation of more than thirty community organizations, NGOs, religious associations, social movements, research institutes, and COSATU. For the idea that social citizenship could rely on universal, decommodified programs, promising times seemed to be dawning.

192

The Wage-Income Relationship is Breaking Down: Basic Income and Contested Decommodification

In January 2000, the minister of Social Development, Zola Skweyiya, emphatically declared that the country was facing a deep social crisis, . . . a time bomb of poverty and social disintegration [which] has the potential to reverse the democratic gains made since 1994. He continued by introducing a ten-point programme of action that in many ways departed from the existing developmental social welfare approach: Our social policies assume the ability of families and communities to respond to the crisis. Welfare has proceeded as if these social institutions are fully functional and provide the full range of social support that is required to restore the well being of people. Such a business as usual approach cannot continue (Department of Social Development 2000.) Two months later, the cabinet appointed the Committee of Inquiry into a Comprehensive System of Social Security for South Africa, chaired by Prof. Viviene Taylor, a long-time welfare activist and former ANC official. The 18-member Taylor committee included six academics, mostly jurists and economists, various government representatives, and the director of COSATU-aligned National Labour and Economic Development Institute (NALEDI.) Among the experts providing research inputs was ILO economist and basic income grant proponent, Guy Standing. Public hearings were convened for civil society organizations, which nonetheless resented their lack of direct representation on the committee.48 The Taylor committee investigated, for the first time in the countrys history, ways to design an integrated social policy framework out of fragmented programs of social assistance, 193

social insurance, and healthcare. The committees mandate excluded, nonetheless, housing and municipal services. Among the intended beneficiaries of the exercise were the working-age structurally unemployed, the working poor, and unprotected employees who were falling through the cracks of existing programs. The gaze of social policy was no longer concentrated on vulnerable sectors outside the labor market; the Taylor committee broadly addressed the relationships between work, citizenship, and social integration. As it dared to acknowledge that employment might indeed cause social vulnerability, the committee was also less obsequious than its predecessors towards the governments self-imposed fiscal constraints. It indeed questioned both the naturalness of market forces and the vaunted technical neutrality of the state. It called the governments priority on budget austerity disturbing in view of the most dramatic socio-economic challenges in South Africas history (RSA 2001b: 9) and added: Government has effectively de-prioritised social sector spending over the medium-term . . . despite improved fiscal capacity . . . reflecting the position that increased economic growth should not be translated into additional public goods and services (RSA 2001b: 8.) The committees chair, Viviene Taylor, criticized the community survival and coping mechanism discourse49 that defined the 1997 white paper, and admitted that economic participation is in itself no guarantee of self-sufficiency and independence. In a subsequent contribution Taylor (2007: 16-17) directly attacked the governments work-centered imagery, and its ideological commitment to a residual (safety net) welfare system, rather than social protection as a right of citizenship. She argued that anti-dependency fulminations, particularly by the Treasury, succumb to the influences of former welfare states of Europe that are curbing welfare spending under conservative neo-liberal state regimes. 194

The basic income grant idea was from the start a major topic of debate in the Taylor committee. Expert opinion was generally sympathetic to the grant, but its supporters had rather different views on its meanings and purposes, reflecting diverging options in international debates. Shared by most basic income theories is the realization that, even if employment is inevitably necessitated by economic processes, it does not necessarily secure relief from poverty. Conservative and free-market theorists define basic income as a production-orientated active labor market, or basic capital, policy (Cunliffe and Erreygers 2004.) By alleviating most pressing needs, basic income induces the individual to seek employment. For others basic income is primarily orientated to subsistence, not production, and it addresses the social inequalities determined by labor market hierarchies. Grants should therefore be universal and independent from the individuals conduct towards work; they actually allow recipients to reduce their dependence on increasingly unprotected jobs. The position, however, redefines rather than criticize the centrality of work in the imagination of citizenship. For Standing (2002) universal grants constitute a transition from the old welfarist idea of the social wage, still focused on monetary provisions linked to employment, to a social income that encompasses cash and non cash transfers from public, civil society, household, and community networks. The aim here is to salvage the connection between wage labor, rights, and human dignity whereas active labor market policies are at risk of merely forcing people into low-wage jobs. A universal basic income, Standing continues, would translate the responsibility to labour into a real right to work, a condition of self-realization not merely imposed by economic necessity. For other theorists (van Parijs 1995; Gorz 1999) basic income would thus operate as Karl Polanyis double movement, saving capitalism from the social consequences of unfettered market forces.

195

A growing scholarship has, however, argued for universal basic income as a critique of wage labor. Claus Offe (1997) views employment in globally liberalizing societies in an inherent contradiction: economic necessity and social policies make it central to peoples lives, and yet it is materially unable to ensure stability and wellbeing to most. On similar grounds, some support basic income as a citizenship income, not merely a response to labor market failures or an inducement to economic participation, but a new terrain of claims and political imagination (Hardt and Negri 2000: 401-03; Monnier and Vercellone 2007.) As Foucault (2008: 223) wrote, a worker is not merely an object of economic analysis, but a sentient, acting, signifying being for which the wage is an income, not the price at which he sells his labor power. Precarious employment fundamentally alters the relationship between wage and income; the wage, or the payment for a clearly delimited portion of the workers time, becomes unreliable as an income to satisfy needs. Transient workers are therefore forced to act like what Foucault (2008: 226) termed entrepreneurs of themselves, developing across their lifetime, in the workplace and at home, behavioral, communicative, and cognitive skills needed to face employment risks, compete in uncertain labor markets, and cope with frequent jobs changes. To the extent workers acquire such capacities outside the conventionally defined working hours, employers can appropriate them at no cost. A universal citizenship income would, therefore, simultaneously compensate value-creating activities across the lifespan and the broad social fabric, and diminish the compulsion to work for wages. It would, therefore, reduce the weight of the workplace in defining the meaning of human activity and reconfigure unemployment and precarious employment from realities of domination and powerlessness into conditions of alternative political imagination. As wage and income become mutually antagonistic terms, Negri (2005: 250) concludes, struggles for a dignified life and the 196

independence of workers self-valorization cannot be confined to production-based demands for decent jobs. The basic income debates in the Taylor committee showed a penchant for its conservative definition as an active labor market intervention (Barchiesi 2007b.) Universalism was less immediate a concern than avoiding inappropriate groups from receiving existing grants. 50 A pervasive fear that any redistributive grant, no matter how small, could weaken work ethic a feeling well summarized by the rhetorical question of committee member, and ANC parliamentarian, Mike Masutha: If you have all these nice social benefits, where is the incentive to want to go back to work?51 paralyzed discussions. In the absence of a conversation on decommodification, the committee fell back to employment as the ultimate condition of social inclusion. The research submitted to it almost unquestioningly accepted market rationality as the basis of individual behavior. Arguments reminiscing colonial paternalism often represented the poor as usually incapable of virtuous self-regulation, whose character the state must shepherd with tough love and work ethic lest sloth and laziness gain the upper hand gnawing at the foundations of the national edifice. Some committee members were mostly concerned with the fiscal burdens of a universal grant. Academic Pieter Le Roux, for example, opposed funding the grant with income taxes as they would unfairly advantage those too poor to pay for them while penalizing corporations and high-income earners. Instead he proposed using the value-added tax, which hits low and middle incomes the hardest, so that the grant would actually be no longer universal and only the truly destitute would gain a net benefit. It was a version of social solidarity in which the poor are primarily responsible for funding the survival of the very poor and thus stimulate the entrepreneurialism of the disadvantaged (Le Roux 2001: 28.) Other conservative supporters 197

of the basic income grant in the Taylor committee wanted to set the payable amount at a level low enough that recipients would still accept jobs at paltry remunerations. Statistician Anthony Asher (2001: 67-68) preferred a job creation programme paying low wages to a basic income grant. His approach reflected the persistence and transformations of the schooling the natives bodies to hard work image discussed in the second chapter, whereby allowing a weakening of individual resolve to produce would herald a social apocalypse with maladjusted youth at its center. Asher (2001: 66) lamented as causes of South Africas social problems sexual promiscuity, wasteful consumption, and limited respect for the rule of law by young men . . . inadequately socialised as they have been brought up without role models or discipline in their families, schools or at work. State grants would only contribute to social and moral decay as they penalize formal jobs efforts and undermine production leading poor people into a downward psychological spiral that undermines ambition and wisdom while they wait for others to help and to lead them. Low-wage jobs, conversely, provide a way out by strengthening peoples moral beliefs as a remedy to self-destructive behavior and moral failure. Blaming the social condition of the poor on their moral, psychological, and behavioral deficiencies was surely a minority view in the Taylor committee. Its members shared, nonetheless, a justification of social grants mostly on pedagogical grounds, as incentives to hard work and enterprising spirits. Law professor Marius Olivier recognized that changes in the world of work are more important factors of social insecurity than the poors behavioral weaknesses. Therefore he argued that social solidarity necessitates redistributive programs, but their priority should be to actually and actively encourage labour market participation as part of a movement to active labour market policy measures stimulating adherence to the work ethic imperative (Olivier 2000b: 35. Own emphasis.) 198

Labor-friendly contributions to the committee did not substantially contest the workcentered imagination of mainstream experts, even if they relatively stressed the significance of the basic income grant as a means to reduce poverty and uphold constitutional social rights. The Economic Policy Research Institute (EPRI) completed a massive investigation, explicitly inspired by Guy Standings work, on the projected impacts of basic income. EPRIs analysis, which also backed the demands of the Coalition for a Basic Income Grant, praised grants-based programs not so much as tools for equality and decommodification, but fiscally affordable, developmental incentives to economic activity and high-risk/high-reward job search (Samson et al. 2004: 4.) Guy Standing, on the other hand, conveyed to the committee the opinion that, as a matter of human nature, the vast majority of people without work, and able to work, do in fact want work, since work provides at least part of an individuals social and economic identity (Standing 2003: 16. Own emphasis.) Without distinguishing between work and wage labor, he characterized employment in terms of natural law rather than social relations. Standing further elaborated, in a paper written in an imaginary future where South Africa has won the soccer world cup and the basic income grant has long been introduced: Compared with means-tested schemes, a universal grant actually provided a greater incentive to search and to take jobs, particularly low-wage jobs or lowincome, own-account activities. Since these were precisely the kind of new work opportunities that were emerging, this was crucially important (Standing 2003: 13. Own emphasis.) The theorist that most influenced left support for basic income, therefore, argued that in South Africa universal grants should not be alternative to precarious employment but inducements towards it. 199

COSATUs own submission to the committee stopped far short of seeing basic income as a redistributive alternative to the wage. The federation prudently saw the proposed grant as a relatively small sum benefitting not only the unemployed but also the working poor by lay[ing] the foundation for more productive and skilled communities where the poorest households have a little cash, which could contribute to their economic potential (COSATU 2000d: 3.2.) Committee members close to the labor movement were indeed more daring and hinted at the decommodifying potential of social grants in reducing the exposure of the unwaged to market relations. For the director of NALEDI, Ravi Naidoo, early discussions on universal grants at the 1998 Jobs Summit were largely a P[ublic] R[elations] exercise, which nonetheless underlined the growing discrepancy between wage and income. A progressive agenda, he continued, would require not only decent work with adequate wages and benefits, but also decent income sustained by public, decommodified social provisions.52 University of Natal professor Charles Meth, a fierce critic of official anti-dependency rhetoric, urged the committee to take a stand against workfare, which he called racist and sexist.53 Punitive and stigmatizing views of social grants were, however, deeply engrained. A submission by the Development Bank of Southern Africa, eventually rejected, supported in-kind payments and food vouchers instead of cash.54 When Meth asked what research was available on possible incentives to work for unsocialised, dangerous unemployed, a Treasury representative provided the following example: I mean, the biggest thing in there is actually the American literature on the incarceration of that population [amused reaction.] I am, no, I am serious, I kid you not. . . . The business of welfare to work as it is practiced in Britain [has] a much more integrated system, rather than the kind of fragmented thing that you 200

have in the United States. . . . The point about that system is that it has coerced people into. . . it has done away with the entitlement system of benefits and it has coerced them into a system into which they only receive benefits if they enter and they go through the stages of the programmes. But the point about it is that it is hugely expensive.55 As the custodian of the governments macroeconomic conservatism, the Treasury nagged the committee with cost containment preoccupations and, as an official from the Department of Social Development remarked in veiled criticism, a narrow concern with the grants potential in reducing crime and increasing work-seeking behavior.56 The Treasury, after all, had notified the committee in advance with a move that, Meth said, stinks to the core57 that its support for social security reform would depend on its projected costs, rather than on the jobs it expected to create.58 Viviene Taylor chose in the end to delay the publication of the committees report, completed in July 2001, by six months to pressure the government to make its costing strategy public rather than posthumously veto the reports recommendation.59 Taylors insistence for an urgent social policy overhaul probably shaped the wording of the final report, which was more progressive than frequent overtones in committee debates. The report endorsed a concept of comprehensive social protection, which is broader than the traditional concept of social security, and incorporates developmental strategies and programmes designed to ensure, collectively, at least a minimum acceptable living standard for all citizens. It embraces the traditional measures of social insurance, social assistance and social services, but goes beyond that to focus on causality through an integrated policy approach

201

including many of the developmental initiatives taken by the state (RSA 2002: 41.) Social protection entailed three interconnected, mutually reinforcing components: income, capabilities (including healthcare, housing, and education,) and assets (both income generating and social capital.) The concept restated wage labors role in avoid[ing] dependence on cash benefits (RSA 2002: 41.) Commodification and decommodification were never mentioned, while social citizenship was only once, in passing. The report, however, also questioned enduring tropes of the official work-citizenship nexus. As a principle, it proposed social provisions to be universal-as-possible, non-work related, and not primarily dependent on an availability to pay (RSA 2002: 42.) It consequently rejected workfare and residualist welfare as ways to divide deserving and undeserving poor. Little was left of the dominant emphasis on targeting and special needs in the 1997 white paper. Also a novelty was the reports recognition that poverty and inequality in South Africa are rooted in the labour market (RSA 2002: 25) where the wage-income relationship is breaking down (RSA 2002: 32.) The argument failed, however, to interrogate casual and precarious employment, as it cursorily blamed working-class poverty on persistent racial inequalities, widening gaps in African wages, and the adverse conditions of specific sectors like domestic and agricultural employment (RSA 2002: 56.) It conversely insisted that manufacturing and public employment played a progressive role, albeit only relatively privileged workers reaped its benefits. The two-fold nature of the South African economy (RSA 2002: 35) meant, the report warned, that many unprotected poor, unemployed, and informally employed may never have stable jobs. Policy interventions and social grants would, nonetheless, help uplift such strata to make them 202

actively pursue gainful opportunities. The reports espousal of the dual economy image aligned it on many issues with the prevailing work-based discourse of citizenship. A major concern was that the lack of effective social provisions for the majority could inflate the unions wage demands, thereby undermining workers commitment to labor flexibility (RSA 2002: 31.) To encourage risk-taking and self-reliance (RSA 2002: 61) among the poor, the document endorsed a universal monthly basic income grant of R 100, even if it admitted that the conditions for its immediate implementation do not exist (RSA 2002: 62-63.) Capitulating to the governments financial stringencies, it recommended, therefore, a phased introduction of the program, to take place from 2004 to 2015, with due attention to issues of fiscal feasibility. A more urgent social imperative were, on the other hand, training and active labor market policies including labor-intensive public works, which the report supported while warning of their likely low wages (RSA 2002: 73.) Compared to the disciplinarian tones of the 1997 white paper, which mostly cast the working-age unemployed as undeserving poor, the Taylor report encouraged all the poor, regardless of their employment status to become, in Foucaults sense, self-entrepreneurs (Ferguson 2007: 82.) The white paper talked of responsibility as an ethical quality, the Taylor report emphasized risk-taking as an economically functional conduct. The rationale for social programs, the report argued, could not imitate Eurocentric welfare ideas derived from experiences of quasi-full, stable employment unrealistic in South Africa. It should rather encourage informal and contingent occupations. To exorcise the dependent, lazy poor, the Taylor report conjured up the rational, utility-maximizing one, confidently entering the market as a proper homo oeconomicus. For the business magazine Financial Mail, this harks to an interpretation of egalitarianism we support: that all should enjoy equal opportunities.60 Echoes 203

of colonial governance, this time the market as the arena where African individuals can escape stagnation and test their mettle and civility, proved once again hard to dispel. The Taylor reports disavowal of basic income as a decommodifying intervention eventually revealed a major ambiguity in its view of precarious employment, simultaneously a cause of poverty and the most sustainable way out of it for those prepared to take the risk. The Basic Income Grant Coalition, where COSATU was a major player, hailed the Taylor report, calling it an important form of social citizenship, possibly a central component of the democratic States contract with the people (Coalition for a Basic Income Grant 2003.) The governments response to Taylor was, conversely, rather mixed: enthusiastic on its active labor market recommendations, favorable on spending more on existing grants and extending public healthcare and retirement provisions, relentlessly hostile to the basic income grant and any non-work related benefits for the unemployed. The ANC tried to bury basic income immediately, and COSATUs pressure barely managed to keep it on the alliances agenda. The conclusions of the ANCs 2002 national policy conference approved Taylors broad fram ework of comprehensive social protection, but did not mention basic income grants. They praised, instead, public works programs offering temporary jobs for the sake of pride and self-reliance of communities (African National Congress 2002.) For the first time, on the other hand, the Treasury recognized, in its 2002 Medium Term Budget Policy Statement, social spending and socioeconomic rights as developmental interventions and not merely a fiscal burden (Herzenberg and Mzwakali 2003: 9.) Treasury statements, however, strongly opposed also by invoking the work of scholars Servaas van der Berg and Jeremy Seekings, for whom South Africans already had an exceptional welfare state any new redistributive program (BIG Financing Reference Group 2004: 22.) 204

In 2003 NEDLAC hosted a Growth and Development Summit, which resurrected discussions on basic income, but also revealed a gradual closure of the debate around the imperatives of growth and employment. The most tangible consequence of the summit was the launch, in 2004, of an Expanded Public Works Programme (EPWP,) with which the government gained the support of civil society organizations, pushing demands for a basic income further to the sidelines. The employers clamored for work-based public programs in their staunch opposition to universal cash transfers (Business South Africa 2003.) By then the ANC leadership, especially the office of president Mbeki, and many government departments had also unleashed a virulent ideological offensive on the basic income grant (Di Lollo 2006.) Recurring themes were the association of work with honor and dignity and the moral stigma on anyone claiming handouts. The debate seemed to run full circle back to a bygone social imagination that opposed work-shy natives and hard-working citizens as categories fusing socio-scientific and ethical judgment. Despite the Taylor committees attempt to represent basic income as the midwife to the poors entrepreneurial industriousness, the ANC perceived it as the latest manifestation of an insidious, perennial temptation to work avoidance that could lead to social and economic disaster. Few months after the publication of the Taylor report, Treasury minister Manuel lambasted the basic income grant as populist and unsustainable (Makino 2003: 19.) Claiming to quote Cicero as a precursor of republican virtue, Manuel later added that people must again learn to work instead of living on handouts.61 In his 2003 annual State of the Nation address, president Mbeki (2003a) made clear that his administration wanted to reduce the number of people dependent on social welfare, increasing the numbers that rely for their livelihood on normal participation in the economy. This time Mbeki did not depict those 205

incapable of normal participation, or the second economy, in mere functionalist terms, as uneducated and unskilled. They were also carriers of negative features, people subject to the social ills associated with poverty and human despair, such as particular crimes, including murder, the abuse of women and children, and other crimes against the person, as well as alcohol and drug abuse (Mbeki 2003b.) As not working steered the poor away from virtue, Mbeki defined the governments caring duty as changing the lives of those of our people who were previously described as surplus people by convincing them to enter the labor market in the spirit of letsema [cooperative work], vukuzenzele [arise and act] and the new patriotism (Mbeki 2003b.) The government spokesperson and ANC ideologue, Joel Netshitenzhe, moved the polemical barrage on a principled ground, arguing that the ANC had, compared to the Taylor committee, a different philosophy centered on the opportunity, the dignity and the rewards of work (cit. in Coleman 2003: 122.) COSATUs Neil Coleman responded to Netshitenzhes philosophical outlook which literally replicated Cecil Rhodess celebration of work as dignity discussed in the first chapter that in a context of mass unemployment and generalized precariousness offering the dignity of work as opposed to social grants is a cruel illusion (Coleman 2003: 122.) Viviene Taylor retorted to Netshitenzhe that this kind of aspersion contributes significantly to the development of the dependency stereotype (Taylor V. 2007: 19 .) It was, however, clear that at stake for the government was, more than the basic income grant and its fiscal implications, the very legitimacy of a discourse that suppressed the possibility of any universal social provision independent of employment. According to the minister of Trade and Industry, Alec Erwin, the problem with the [basic income grant] is not the money but the idea (cit. in Hart 2006: 26.) 206

The Department of Social Development, which sponsored the Taylor report, continued to defend the basic income grant and, warning on excessive optimism about public work programs, argued that formerly dismissed as mere populism, (modest) redistribution is increasingly being recognised as the most effective way to reduce poverty. . . . Even the World Bank has begun to acknowledge this (Department of Social Development 2002.) The department, however, joined the rest of the government in its aversion to handouts. The minister in charge, Zola Skweyiya, was the only member of the Mbeki administration to support something like the basic income grant.62 He remained, nonetheless, convinced that universal cash programs are a measure of last resort as they could promote dependency instead of self-reliance.63 His department favored, therefore, placing child support and disability grant recipients in half jobs where they could migrate from welfare benefits.64 By 2006, the focus of the department had shifted to ensure that all South Africans enjoy the dignity of work (Department of Social Development 2006b: 11) by overcoming individual barriers to employment. Moving people from welfare to work, the ANC explained, would restore the dignity of beneficiaries, whereas the basic income grant would neither have the broad or deep impact on poverty eradication nor the broad mobilization of resources to alleviate poverty (African National Congress 2007a: 3.) The developmental state the party imagined repudiated the ideological background of the welfare state because only employment would allow, in terms strikingly resembling the 1979 Riekert report, outsiders (people who face barriers) to become insiders in society (African National Congress 2007a: 3.) By early 2007 Mbeki had definitively discarded the basic income grant as tantamount to abandoning the poor.65 The governmental alternative to basic income was the EPWP, a fiscally funded public works program to develop, maintain, and repair infrastructures, public amenities, and the 207

environment. Mbeki launched it with the aim to reduce the number of those dependant [sic] on social grants . . . by engaging in gainful economic activity and exercising their right to human dignity (Mbeki 2003c.) As a new national program, it built on a set of locally implemented community-based initiatives, in existence since 1994, like the Zivuseni (uplift yourself) project to build schools and clinics in the Gauteng province. Fascination with international experiments from India, Mexico, and Brazil combining work incentives, conditional cash grants, and behavioral guidance for the poor motivated support for the EPWP (Samson et al. 2007.) Similarly to such experiences, the EPWP regards temporary jobs as opportunities for skill formation and economic participation. It also grafts onto them the ethical imperatives of volunteerism, self-sacrifice, and deferral of financial gratification the government propagated in its letsema and vukuzenzele campaigns (Twala 2004.) The EPWPs ambitions clashed, however, with the practical results of its implementation (Altman and Hemson 2007.) By its fourth year of operation, the program had created 220,000 full-time equivalent jobs, or only one third of its stated target. The overwhelming majority of the programs beneficiaries remain in temporary occupations, nearly 90 percent of the affected households stay below the poverty line, and almost all the programs remunerations go into basic necessities like food and clothing. Between 2005 and 2007 as the department of Labor determined for special public works programs exceptionally low minimum wages, eliminated paid leaves, and curtailed extra pay for Sunday work (Godfrey and Clarke 2002: 6-7) the average wage of EPWP jobs fell by 40 percent. The programs wages had in fact to be miserable enough to attract citizens with no income whatsoever, and not workers already employed in lowearning occupations.66 The EPWP amounted, in practice, to the elevation of precarious, unprotected employment as a national tool of social inclusion. Viviene Taylor (2007: 18) 208

dismissed the EPWPs principle as any work is better than social security support. Ironically, she continued, the program deepened dependency on public spending, proving that the government has no problem with handouts when they reinforce the ideological centrality of work. The defeat of the basic income grant eclipsed the option of reconciling the normative universality of work with the material inequalities it produces. The Taylor commission tried to bridge that gap, where the 1997 white paper remained stuck, by recalibrating the values of employment into a narrative of entrepreneurialism accessible, with the support of the state, to all citizens. Any poor, employed or unemployed, receiving grants or not, could be a microentrepreneur. Through a decommodified universal income grant the state would help everyone develop human capital, no one would be undeserving. In the end the government feared, however, that the mere enunciation of decommodification in any of its social policies would subvert the official intimation to work. Scare at the old ghosts of refusal of work eventually won the day as the ANCs rhetoric often became unusually aggressive. As Charles Meth (2008: 17) sadly commented: because South Africas elite believes that social grants cause dependence, massive surpluses that could be extracted from the rich are converted into 4x4s and gold estates. The left and its labor supporters are also, however, responsible for the fate of basic income. As Ferguson (2007: 79-80) commented, they never transcended the confines of liberal governmentality, which, in Foucaults meaning, accepts redistribution only as a stimulus to the competitive initiative of the individual. The ultimate convergence of state discourse and its productivist critics does not, however, necessarily instantiate compliant and predictable subjects. Sometimes ordinary voices may even complicate the very production process of official imagination. In November 2002 COSATU brought Lundi Rasmeni to testify before the Taylor 209

committee. Mr. Rasmeni was a worker and welfare activist from the Western Cape, where he supported eleven people on a monthly salary of R 3,000, which in Seekings and Nattrasss (2005) statistical judgment would be in a relatively prosperous middle class earning bracket. Despite the fact that many members of the Rasmeni family also received social grants, the total household income barely covered food and basic supplies, which regularly dwindled well before the end of the month.67 Undercooking food, he said, was a way to make it last longer. Taking issue with the theory of welfare dependency, he reminded the committee that under the old regime politicians used to say: die kaffers kan nie regeer nie [blacks cannot govern themselves.]68 He then asked: What is the government thinking about the youth of the 1980s whom contributed to the liberation of this country, sacrificed their education and their childhood? Some of these young adults didnt finish their high school and some of them had the opportunity to go to university or technikon [technical college] but dropped out due to financial difficulties. Some of them were employed but retrenched and now are unemployed because companies are downsizing to make more profits. Postapartheid democracy would indeed unequivocally respond that it regards the descendants of those youth as the new generation that, unless forced into employment, cannot govern itself.

Conclusion. Precarious Employment as the New Peoples Contract?

It may seem paradoxical that, by the end of the 2000s, the South African social citizenship framework looks like, despite the resoundingly coherent principles inspiring it, an idiosyncratic amalgam of often unintended elements. The governments fiscal thrift has reproduced a society 210

where the richest 10 percent control half the national income and the poorest 60 percent receive only 15 percent. Policymakers routinely deplore the perils of entitlement mentalities and welfare dependency, and yet one quarter of the population (three quarters of the elderly and two thirds of children under fifteen) is on social grants that absorb almost ten times the budgetary allocations for work-based poverty relief (Human Sciences Research Council 2003a.) The constitution emphasizes redistribution and social justice, yet social grant recipients live well below poverty levels, while the government continues to oppose the measure, a universal basic income grant, with the greatest antipoverty potential (Meth and Dias 2004.) The ANC centers its social policy philosophy on employment, promising a peoples contract to create work and fight poverty in its 2004 electoral banners, but presides over a society where decent jobs are disappearing and declares that, in the end, it is not the governments task to make them reappear. The government asks for social solidarity in opposition to unbridled free markets, and yet it lets labor market hierarchies determine the legitimacy of social claims. Politicians, academics, and opinion-makers reassure that, in the words of an ANC parliamentarian, people want jobs, not handouts. When they see the government making new grants, they feel as though their hands are being tied.69 Yet, they are bothered that a universal allocation as small as ten US dollars per month can drive millions out of jobs into idleness.70 In a country, finally, where stable jobs provide a decent life to perhaps less than one third of the population, the left and the labor movement continue to celebrate productive employment, justifying social grants with the argument that they put people to work, rather than reducing the economic compulsion to earn wages. The combination of apparently incongruous practices in the ANCs techniques of government has been, nonetheless, remarkably successful in propagating certain normative constructs. First and foremost it powerfully transmitted the idea that benefiting from social grants 211

does not exempt unemployed able-bodied individuals from acting as jobseekers exercising responsibility, patience, moderation, and active risk-taking regardless to the actual availability of decent jobs. No government check comes without a stern reminder to avoid the temptation of entitlement culture attached to it. The seeming contradictions of postapartheid social policy ultimately fit the governments view of the poor as a Janus-faced creature, constantly lured into laziness and sloth, but also in possession of a natural economic ambition that the state has the duty to nurture and guide. The inhabitants of the second economy, are, in particular, the modern incarnations of colonial images of casual workers as menacing others, which the idioms of production and the statistical categories of the social sciences allow to capture and normalize into poverty lines, means tests, and gradations of merit. In their material practices and daily lives the poor may well be barely visible and governable; but the state cannot allow the poor to exist as a category challenging institutional truths, chiefly the unyielding compulsion to work, lest it jeopardizes the stability of market-based democracy. Through work, the poor are imagined as active partners in the orderly government of the postapartheid social question. Conscious of the mistakes of African postcolonial developmentalism, the ANC argued that the ultimate power of the state is its capacity for self-limitation, which transfers governmental authority to the civil society instead of acting as an all-powerful instrument protecting and delivering to a passive populace (African National Congress 1998: 6.) The state reclaims, nonetheless, control over the elaboration of national imagination. The role it plays in this regard is recombinant more than generative: it revives and reassembles distinct strands of South Africas modern discursive heritage, from Anglo-Saxon (mean-spirited and coercive) individualism (Meth 2004: 14) to Calvinist ideals of hard work, from the African nationalist iconology of the industrious poor to Keynesian 212

redistributive rhetoric, from socialist slogans of social provisions as rights to religious definitions of them as charitable interventions. Mbeki added, on his part, a touch of new Labour workfare, with which he was conversant in his frequent appearances in third way global elite forums. Entrepreneurialism as an emerging grammar of social advancement rallies behind the ANCs mastery of the state as the overarching social organism (African National Congress 1998: 2) old and new corporate elites, unionized workers, the poor, the casuals, and the unemployed. The privileged can use the purported social mission of accumulation and acquisitiveness to shelter their own selfishness from possible criticism. Mbeki himself (2006: 1416,) reassured that, in spite of widening inequalities, the deification of personal wealth cannot be the distinguishing feature of the new citizen of the new South Africa. Economic participation, he added, was not the same as personal enrichment at all costs, and the most theatrical and striking public display of that wealth. It was rather part of an RDP of the soul, a step towards social cohesion and human solidarity, which the wealthy and the poor alike should embrace to avoid a Sgudi Snais (life good for some) scenario characterised by conflict between those who have-a-lot versus those who have-a-little.71 Despite the eclipse of basic income and decommodification, the Taylor committees image of the poor as a rational individual optimizing resources and sagaciously managing risk has gained traction in recent social policies. In his 2007 State of the Nation address, Mbeki announced plans for a widespread reform of social security, including a retirement program for all workers funded by payroll taxes and a wage subsidy payable to employers to facilitate youth labor market participation (National Treasury 2007b.) The project would, moreover, control social spending by encouraging low-wage workers to save for old age and attract rural social grant beneficiaries into poorly paid urban jobs (National Treasury 2007a: 9-11.) COSATU 213

opposed the wage subsidy proposal as a way to create more unprotected, downgraded employment, and renewed its demand for a basic income grant as an alternative. 72 The Department of Social Development has also, however, differed from the Treasurys aim to downsize social programs by increasing precarious jobs, proposing instead a universal old-age pension pegged to inflation (Samson 2007.) Many labor scholars argued that expanded social provisions can indeed earn workers consent for flexible employment without loss of social guarantees a concept often termed flexicurity, or a move from passive social protection to the social management of risk (Benjamin 2006: 39.) South Africas employment problem is, in such views, that labor rights are so detailed that they become unenforceable, making the implementation of flexibility illegal in the form of widespread violations of rules on working time, paid leaves, and minimum wages. As a solution they propose looser constraints to precarious employment. To counterbalance the consequent diminution of equality and social justice, basic state provisions would help the poor empower themselves on the market. Labor policy was not the only realm resonating with what Trevor Manuel praised as entrepreneurial ambitions.73 In 2003 Mbeki mooted a new social health insurance system combining primary healthcare via general taxation and the extension of private care to low-wage workers through payroll taxes and a public risk equalization fund. The proposal enhanced the role of work-centered provisions even if increased public spending would support the risk management strategies of precarious workers. Some even saw it as a possible gateway for national public healthcare (Botha 2008.) In 2004 the Department of Housing announced a new strategy (Breaking New Ground,) which admitted that the emphasis on low-income homeownership in the existing subsidies led many recipients to live in peripheral slums. As an alternative it proposed public funding for a range of solutions including inner-city rental, which 214

organized labor advocated (Department of Housing 2004.) The new program also aimed to encourage informal residents to convert their shacks and labor power into marketable equity to access public housing funds. Evanescent survival resources could thus ideally provide a sense of empowerment and, in the words of Housing Departments officials, psychological ownership.74 In the end, however, workers with a regular income, rather than casuals and the unemployed, benefitted from the program. After years of stagnation, private development has shifted towards the formally employed but dried up in marginal neighborhoods, where publicly subsidized accommodation has also dropped for the three quarters of households with low or no wage income (Rust 2006.) Steven Friedman and Ivor Chipkin (2001) raised apt questions for the ANCs entrepreneurial approach to liberation: To what extent are the deracialization of economic power and state fantasies of solidarity between new black elites and the black poor conducive to socially inclusive citizenship? Does black economic empowerment contribute to the stigma of those who cannot or do not want to empower themselves through work? Despite their pretensions as avenues to job creation, programs promoting black corporate ownership have turned into devices for aspiring investors to leverage their way into boardrooms rather than start new companies. If anything, postapartheid entrepreneurship has destroyed decent jobs instead of creating them. In 1992, each million Rand of gross domestic product corresponded to 7.5 fulltime occupations; by 2000, they had dropped to less than five (Pollin et al. 2006: xxiii.) Friedman and Chipkin (2001: 48) observe that it is at least possible that democracys survival depends more on its ability to continue offering citizens intangibles than its capacity to reduce inequality. The question then becomes how stable political survival can be if such intangibles hinge on promises of employment, which is clearly declining as a social reality. How much 215

social injustice and harshness can the rulers symbology of liberation accommodate before it loses credibility? Which alternative imageries and practices can precarious work produce to replace it? Official imagination has exalted ideal employees as embodiments of civic virtue, but actual workers have become redundant and disposable. Sociologist Andries du Toit (2005: 16) responded to Jeremy Seekingss insistence on low-wage jobs as the panacea for the poor that it is probably dangerous to equate access to a job any job as an escape from poverty. The evident inadequacies of the existing social citizenship arrangement makes even erstwhile supporters wonder: Should it continue to be based on the manifestly incorrect premise that all able-bodied adults can support themselves through work, or should it be redesigned to address the large hole . . . through which so many unemployed people are currently falling? (Nattrass 2007: 179.) Of course, through such holes also fall the growing numbers of working poor who haunt the social foundations of the postapartheid experiment. From the next chapter, this books viewpoint will thus shift from official discourse and institutional contestation to how ordinary workers signify labor, citizenship, and liberation.

216

4 The Changing World of Work in Gauteng

Texaco, the oil company which used to occupy that space and which had given its name to it, had left aeons ago. It had picked up its barrels, carted off its reservoirs, taken apart its tankers' sucking pipes, and left. Its tank truck sometimes parked there, to keep one foot on the dear property. Around that abandoned space are our hutches, our very own Texaco, a company in the business of survival. Patrick Chamoiseau (1997: 24)

Introduction. Dreaming of Modernity in the Place of Gold

Of the nine provinces constituted in the post-1994 reshaping of South Africas administrative map, Gauteng (seSotho for place of gold) is the smallest but most populated and economically important. In the mid-2000s, the size of Gautengs economy was more than one third of South Africas and 10 percent of the whole African continent (African National Congress 2004: 5.) Most of the provinces surface is what under apartheid was known as the PWV (PretoriaWitwatersrand-Vereeniging) area, at the center of which is the Witwatersrand, a vast urban region of 7.5 million inhabitants with the city of Johannesburg in the middle, flanked by the West Rand and the East Rand, named in 1999 the Ekurhuleni Metropolitan Municipality.

217

Historically, the Witwatersrand is South Africas productive heartland. Since the 1880s its transformations have reflected the countrys capitalist development, from the early imperial hegemony of mining magnates, the Randlords, to import-substitution industrialization during segregation and apartheid, to the financialization that accompanied the democratic transition. The spatial organization and physical appearance of Gauteng are largely produced by colonial urbanism and differential racial and gender strategies to recruit, house, and control waged labor. As apartheid declined and faded into representative democracy, the removal of racial barriers and widening rural-urban inequalities drove waves of African domestic migrants seeking elusive job opportunities into peri-urban informal settlements. Once whites-only leafy suburbs became destinations for new black upper and middles classes, while former black townships still convey, despite a tumultuous social differentiation and the upgrade of commercial and residential spaces, the dull homogeneity of apartheid modernist planning. More than any other process or feature, waged work defined Gauteng. Not only did its labor regime acquire a peculiar, uncompromisingly functional demarcation of production, reproduction, and consumption underpinning the regions stark geography of class and race. The elites perceptions of Gauteng have always been mediated by dreams of industrialism and fears of its social consequences. Early Cape liberal and Rhodess minister, John X. Merriman, regarded Johannesburg, its Stock Exchange, and its prostitutes and prize fighters, with a combination of fascination, horror and disdain (Beinart 2001: 62.) Later on, new socioeconomic scares the unauthorized proliferation of female-headed households or the township youths refusal of work bothered the rulers confidence in a linear, cumulative, plannable progress. Under democracy, the leaders of the newly established Gauteng province have tapped the postindustrial parlance of the day to fantasize of a global city-region driven by hi-tech exports, 218

knowledge industries and creative classes (Mahlangu 2007.) Troubling the new dreams of modernity are marginalized multitudes with a fragile place in the global city narrative and more mundane concerns with unemployment, poverty wages, and lack of water, electricity, housing, paved roads, and refuse collection. The normative order of the global city, centered on private enterprise and corporate investment, instantiates at the local level the unequal citizenship articulated by the two economies discourse in the national sphere. Gauteng, and the Witwatersrand more specifically, are ideal places to study the postapartheid promise of work and its disappointments on the ground. This chapter introduces the social, productive, and employment background for my two subsequent case studies of manufacturing workers in the East Rand and municipal employees in the City of Johannesburg.

City of Industry. The East Rand/Ekurhuleni and the Promise of Work

The names of some East Rand industrial areas I researched New Era, Vulcania, Ergo Road speak of the sense of efficiency, possibility, and technological might that accompanied industrialization. The names of the black townships containing their workforce Labore, Geluksdal (valley of luck,) Katlehong (seSotho for place of success,) Tsakane (xiTsonga for happiness,) Thokoza (isiZulu for be grateful,) and Duduza (isiZulu for comfort) suggest, conversely, the intention of apartheid planners to embellish wage labor with a promise of civility, prosperity, and order. Michel de Certeaus (1984: 115) observation that narrative structures have the status of spatial syntaxes seems here to gain validation in reverse: signs attached to spaces materialize official narratives. The promises of waged work have, however, signified the East Rand as a place not only through state planning and discourse, but also across generations 219

of black workers struggles. The independent trade unionism of the 1970s and 1980s had in the East Rand its stronghold. Workers conferred their own modernist template upon the region as a terrain of solidarity, militant claims, and connections to community struggles, proving Evelyn Nakano-Glenns (2002: 2) point that labor and citizenship are brought into the same frame mostly at the level of locality. The East Rand comprises a group of towns Alberton, Germiston, Boksburg, Brakpan, Benoni, Kempton Park, Springs, and Nigel merged since 1999 into the Ekurhuleni Metropolitan Municipality. The region owed its early economic growth to mining, but between the 1950s and the 1970s it experienced a rapid industrialization, which made it South Africas workshop. Local industries benefitted from apartheid-age policies of import substitution, investment in infrastructures, especially railways, and state-owned steel and energy corporations. The manufacturing share of the local workforce rose from 27 percent in 1950 to 52 percent in 1970 (Cockhead 1970.) By the end of World War II, rapidly urbanizing African industrial workers had outnumbered African miners (Nieftagodien 2006: 93-95.) Metal and engineering industries rose to prominence: by 1976, of the regions almost 200,000 African manufacturing employees, 130,000 were metalworkers (Sitas 1983: 3-4.) The East Rand produced durable goods for a burgeoning white consumerism thanks to capital-intensive production, cheap energy, and abundant raw materials. The precarious lives of black workers reflected the interpenetration of workplace hierarchies, segregation policies, and spatial stratification (Webster 1985.) As semiskilled African males steadily moved into the employment of mass-producing companies, they also gained permanent urban residence. Unskilled, stressful, and dangerous occupations, which black township residents refused, went, instead, to migrant workers housed in separate hostels and 220

compounds run by local governments or private companies. Migrants experienced the despotism of wage labor as a continuum extending from production to accommodation; the hostels were enclosed disciplinary institutions under the systematic surveillance of African supervisors (izinduna) and armed guards (the blackjacks.) Rural affiliations and imithetho (norms and moral codes) remained valuable assets to make sense of the alienation of capitalist work and urban life. The imagination and strategies of migrant workers also creatively reshaped the symbolic and material interactions of urban and rural. The decisions to stay in the city, cultivate the status of breadwinning masculinity, or channel social claims through a trade union depended on factors like the economic conditions of rural homelands, the size of ilobolo (bride payment,) or chances to meet women from the same ethnic group (Bonner 1993.) In migrants oral narrative and literature, wage labor occupies a liminal space shot through with gendered ambiguities: male workers could resent its harsh and hazardous conditions, but also rationalized them into a test of masculine strength, prowess, and cunning allegedly needed to succeed in an adverse urban environment (Shanafelt 1988). Early African resistance in the East Rand opposed, inside and outside the shopfloor, working for wages. Formal housing in what were then known as urban black locations did not keep pace with population growth, therefore land invasions and squatter settlements proliferated. African women moving to the cities in defiance of the authorities eschewed wage labor and opted for informal economies of beer brewing, small subsistence agriculture, and sex work. In the 1930s, the white administrators of Brakpan defined Africans preferring casual work to formal occupations as a dangerous class of habitually unemployed mostly made of maladjusted youth and women without legal residence with their unbalanced minds, unnatural surroundings and idleness (cit. in Sapire 1987: 370.) In the wake of wartime industrial 221

expansion, the rapidly growing African proletariat became a breeding ground for nationalist and socialist politics, which responded to worsening living conditions, but also repositioned wage labor as a foundation of patriarchal order and modern claims to emancipation (Sapire 1993.) African male breadwinners, the ANC, the Communist Party, and the segregationist state shared the glorification of wage earning in vocabularies of social control directed at rebellious youth and wayward women. In the East Rand as elsewhere, apartheid urban planning tried to control the reproduction of black waged labor through technocratic social engineering (Bonner and Nieftagodien 2001: 45-65.) Upon recommendations of the Mentz commission (1951-1955,) model townships were established in the late 1950s and the early 1960s; Daveyton (Benoni,) with its standardized, electrified, family houses was their prototype. Residents of old African locations were forcibly removed to the new townships, as Payneville gave way to KwaThema, Stirtonville to Vosloorus, KwaDukathole to Katlehong, Charterston to Duduza, and Brakpan Location to Tsakane. The destruction of the locations was part of the states suppression of historical hotbeds of urban radicalism, thus the new townships were also designed to divide and moderate black political expressions. In the 1970s and 1980s, the East Rands African manufacturing workers were the core of independent trade unionism. The FOSATU tradition of unions like the metalworkers MAWU was particularly effective in organizing migrant workers. By establishing their moral authority in the hostels the unions caused, as Ari Sitas (1983: 243-44) argued, a cultural revolution that resignified wage labor into images of collective power, autonomy, and claims to social justice. The unions focused their organizing efforts on large, mass producing companies, which had, however, an uneven penetration. The small size of South Africas largely white consumer mark et 222

allowed in fact for the continuous strength of small enterprises in specialized, craft, or customized (jobbing) productions (Kraak 1987.) The independent unions lacked the resources and the full-time personnel to organize across such a fragmented landscape. Therefore they relied on the self-organization of rank-and-file workers, which became protagonists of their own recruitment. Organizing from below reinforced migrant workers sense of controlling their unions. For many of them, cast at the margins of modern humanity by the state and township residents alike, it meant that organizations centered on wage labor could, after all, deliver a redemptive promise of dignity and power. During the East Rand strike waves of 1981 and 1982, migrant workers, often acting spontaneously and outside union discipline, were particularly militant. In the Wadeville (Germiston) industrial area, at the center of the strikes, FOSATU-organized factories rose from two to twenty-three in 1981 alone.1 To manage such a rapid expansion, FOSATU unions established shop stewards councils, later to become union locals, which gathered representatives from organized factories by industrial areas (Baskin 1982.) Even if the councils were mostly focused on workplace issues rather than sociopolitical claims,2 they helped cement a shared awareness of local problems and common languages of struggle (Ruiters 2000.) Confronting the states smooth productivist modernism, the imagination of labor counter-represented the East Rand as a geographical space of contestation. As the living conditions of black workers worsened amidst the social and economic crisis of apartheid, trade unions demands transcended the boundaries of production. Labor struggles were not, however, aligned with liberation movements in exile. Well into the 1980s, workers claims were rather concerned with daily issues of reproduction, residential rights, transportation costs, and rising rents. The labor reforms of the 1970s and orderly urbanization in the 1980s 223

opened spaces of social contestation to counter escalating unemployment and layoffs. Meanwhile, the townships of the East Rand became focal points of the resurgent domestic opposition movements and their strategy to make urban areas ungovernable. Most black local councils collapsed, their members were physically attacked, and boycotts routinely affected schools, shops, and public transportation. Housing and municipal services were major terrains of conflict. By the late 1980s rising rents and fees had led to generalized payment boycotts. Confronted with a permanent housing shortage, backyard rental shacks and informal shantytowns expanded exponentially. Often their dwellers were not rural migrants, too poor to pay rent or build shacks, but unemployed and precarious workers moving between cities in search of survival opportunities (Frankel 1988.) Tsakanes population grew from 31,000 in 1980 to about 90,000 in 1988, half of which living in shacks, two thirds without electricity.3 In KwaThema almost 80,000 people lived in informal settlements.4 The average occupancy per dwelling reached 7.8 in Thokoza, 9.4 in KwaThema (Pillay 1984) and 13.04 (equivalent to five square meters per person) in Wattville.5 Competition for land often erupted in violent skirmishes. Daveytons population doubled from 1961 to 1982, but formal dwellings increased only by one third (Malinga 2000: 162-215.) Black politicians serving in the townships municipal council tried to salvage their embattled position by fomenting conflicts between township residents and illegal squatters. In 1986 squatters moved to the new, nearby settlement of Etwatwa, which rapidly boomed to a population of 160,000 with almost no municipal infrastructure.6 Local governments started indeed to respond to land invasions with mass evictions. A critical site was the township of Katlehong (Germiston,) which the urban engineering of racial modernism had imagined as a garden township with homes rivaling those in European suburbs where native workers could learn the virtues of 224

homeownership and industriousness.7 Between 1974 and 1981 the authorities allowed, in the absence of new housing construction, temporary informal settlements8 but, as their population soared, in 1983 the municipality started razing squatter dwellings and remove their residents. The evictions met a determined popular resistance that brought together community movements and trade unions. Among the evicted were many migrant workers members of FOSATU, which perceived their forced removal as a direct political attack because everyone knows that rural people are most militant.9 The rank and file urged union leaders to join township mobilizations in defense of working-class communities under threat. Collaborations between organized labor and social movements have led several scholars to regard the East Rand as a laboratory of social movement unionism (Webster 1988; Seidman 1994.) As I have argued elsewhere (Barchiesi and Kenny 2008,) however, the social activism of the East Rands black workers responded to situational and contingent conditions, which the generic concept of social movement unionism tends to overlook. Visions of social emancipation enabled by wage labor emerged tentatively from daily survival struggles and inchoate interactions with communities and state agencies, rather than reflecting a necessary alignment with national liberation politics. Unionized workers realized that precariousness at the point of production had to be ultimately counteracted in the social fabric. As MAWU proclaimed: When our members are dismissed they lose their accommodation (bed) in the hostels. . . . Also, those that are retrenched are being kicked out of their houses for not being able to pay rent. Those workers who stay in the shacks are being harassed day and night and their houses (shacks) demolished.10 East Rand workers entered alliances with civics and community organizations with a strong sense of their own specificity and a deep-seated suspicion of black nationalist township leaders, 225

which they often criticized as petty-bourgeois politicians without clear democratic mandates (Swilling 1984) or haughty urbanites who already secretly despised migrants.11 The unions often took the initiative. In Springs, for example, FOSATU activists did not merely join social movements, but played a leading role in establishing the local branch of the East Rand Peoples Organisation (von Holdt 1987.) Shop stewards saw themselves as representing the mission of the productive class to lead and not be compromised by businessmen and intellectuals with no feeling or understanding of the down trodden masses.12 Union meetings were for them venues to forge a new language of claims linking production and reproduction: what happened in these meetings is that workers would start raising other things that affect them, issues such as incredible water bills, unannounced rent rise, unrealistic rental accounts received by some of the people, and also Putco bus problem i.e. shortage of School buses, bus routes, behaviour of the bus drivers and Coupon System.. . . These problems were always raised by workers as prime consent [sic] demanding immediate attention. [Own emphasis]13 To counter the danger of bourgeois hegemony disguised as national struggle, the Springs shop stewards continued, the union had to encompass the community. The focus remained, nonetheless, on practical issues more than ideological pronouncements. Rather than on abstractly defined class identities, grassroots claims to guide social mobilization relied on the awareness of labors strength in pushing for material improvements. At the same time, the political implications of their activism were clear to the unionists in Springs. We should not be evasive on the slogan MAJORITY SHALL RULE,14 they concluded, but added that the majority were the working people. 226

Participation in social movement politics did not, however, forestall divisions among East Rand workers. Many Zulu migrants, the independent unions erstwhile militant backbone, were alienated by COSATUs alignment with the ANCs tradition, which they saw as political adventurism by self-serving township-based cadres (Chipkin 2007: 139-48.) On top of worsening living conditions, migrants felt that COSATU was neglecting them and dismissing their sacrifices for the sake of an alliance with the ANC that took resources away from labor organization and exposed the unions to unnecessary risks of repression (Sitas 1996.) The ethnic Zulu nationalism of the Inkatha Freedom Party (IFP) exploited these growing chasms to extend its appeal to the East Rand hostels. The IFPs politicization of migrant workers against the ANC, COSATU, and the UDF contributed, by the early 1990s, to engulf Thokoza, Katlehong, and Vosloorus in persistent violence with thousands of casualties (Bonner and Ndima 1999.) With hindsight, many COSATU leaders regretted the federations blunders with migrant workers, but they failed to connect them to the realignment of workers struggles alongside African nationalism, which would discipline local labor demands once the struggle was won.

Economic Restructuring and Employment Decline: The East Rand in Transition

Following COSATUs adoption of the ANCs program, East Rand workers massively participated in political strikes, but their militant visibility contrasted their deepening economic vulnerability. Employers imposed layoffs, technological change, and work reorganization without consulting the unions, which perceived their control of workplaces slipping away due to increasingly opaque economic forces. The recession of the 1980s was followed by liberalization and restructuring in the 1990s, which confirmed the decline of East Rand manufacturing 227

(Rogerson 2006.) Five towns out of nine recorded negative economic growth from 1981 to 1991, and seven had negative or zero growth between 1991 and 1993 (Kok 1998.) By the end of the 1990s, manufacturing was no longer the regions dominant economic activity. Industrial decline was particularly harsh in strongly unionized heavy industry sites like Germiston and Springs (Barchiesi and Kenny 2002.) In the generally depressed investment climate that followed the 1994 elections and the introduction of GEAR, capital flows in Gauteng privileged FIRE (finance, insurance, real estate) and high technology sectors in the Johannesburg-Midrand-Pretoria corridor. In the East Rand/Ekurhuleni area, instead, industry gave way to low-wage retail and services (Urban-Econ 1999.) The devastating impact of post-GEAR restructuring led to approximately 80,000 net job losses in the region from 1988 to 1999. During the 1990s the metal and engineering industries, once the core of the East Rand economy, laid off almost 200,000 workers nationally as total hourly employment fell from 385,000 to 255,000.15 In 1999 services and retail reached 44.3 percent of the East Rands total employment, while manufacturing dropped to 32 percent from the 1970s shares of 50 percent and above (Barchiesi and Kenny 2002.) Supermarkets and shopping malls extensively hired casual and temporary workers. As Bridget Kenny (2003; 2004) has shown, the precarization of service employees recentered their identities and claims from idioms of citizenship and social change towards contingent survival imperatives. High unemployment contributed to the vulnerability of casual wage earners: in the first five years under GEAR, unemployment in the Ekurhuleni municipality soared from 32.2 to 40.4 percent, compared to overall Gauteng figures of 28.1 and 36.5 percent (Development Bank of Southern Africa 2005: 167.) In the large southern townships (Kathorus, Daveyton, Duduza, Tsakane, Etwatwa, and Wattville) the official unemployment rate jumped from 43.5 percent in 1996 to 52.2 percent in 2001.16 228

Employment decline reflected national trends and was less a manifestation of deindustrialization than of labor-saving corporate strategies in an increasingly competitive environment; when South Africas economy partially recovered in the early 2000s, lost stable jobs did not come back (Machaka and Roberts 2006: 126.) From 2001 to 2006 industrial employment grew nationally by seven percent, but five out of seven new net jobs were not permanent (COSATU 2007a.) In the East Rand, the recovery did not mean a return to the old industrial economy.17 By 2005 the contribution of manufacturing to Ekurhulenis jobs had further fallen to 25 percent and the sector provided only one quarter of the areas gross value added, compared to a combined 53 percent for finance, commerce, and services (South African Cities Network 2006: 3-14.) Wholesale and retail trade had become the main job creators and their employment share nearly equaled that of manufacturing (South African Cities Network 2006: 3-28.) Unemployment in the East Rand slightly receded to about 30 percent, but it never went back to pre-GEAR levels and remained almost five points higher than in neighboring Johannesburg (South African Cities Network 2006: 3-16.) The informal sector, moreover, expanded faster than the formal one, with growth rates of respectively 3.64 and 2.12 percent for the period 2001-2004 (South African Cities Network 2006: 3-18.) East Rand companies followed national trends of technological innovation, decentralization, subcontracting, casualization of work, and labor brokerage (Bezuidenhout 2004a.) Large unionized concerns downsized and outsourced production to smaller players, either companies specialized on specific segments like palletizing, repairs, tool-and-die making, or informal businesses for low-overhead operations like recycling or certain types of assembly. Small companies carved up markets created by the decentralization of large ones, of which they became suppliers in a cutthroat process that a NUMSA organizer called competitive 229

outsourcing.18 Abuses and exploitation of a deunionized, flexible workforce inevitably followed, as the same labor organizer reported: One night I just went to the township, I went to a certain house, of course I knew what I was looking for. I bought one bottle of beer, and there were people busy doing [a telephone appliances company]s job. Its a manual job, very, very easy. They assemble small pieces of rubber putting them in the phones, stuff like that. They work at home, then [the company] comes and collects, paying them very poorly. Theyre taking advantage of the social situation of the workers. These people, due to a poverty situation, are compelled to accept that kind of conditions.19 COSATU unions have tried, with some success, to negotiate casualization. As early as 1999, NUMSA won contractual clauses to permanently hire casuals in the engineering industry.20 On the unions demand to prohibit labor brokerage, however, progress has been far more limited (NUMSA 1999.)21 In general, NUMSA recognizes that organizing fixed-term workers is inherently problematic: Well, you can organize them if you wish, but unions are much different from, say, a burial society or an insurance company, whereby you got an insurance only if you have a problem. With the union its completely different, you build a relationship over time. Some of the issues workers are enjoying today have been fought for many years, even the recognition of the union has been an uphill battle. That is the problem in organising fixed-term contracts. You can organise them but in a union you build up for a future, you must have an agreement. If you dont

230

have an agreement contract workers are terminated when their contract expires, so that is what makes it difficult to organise in those industries.22 Organized labor has suffered greatly from employment changes: the national union density in formal non-agricultural sectors has declined from 36 percent in 1997 to 32 percent in 2003, but it would be 26 percent if largely nonunion agricultural workers are counted (Braude 2004.) Only in 2006 has unionization returned to mid-1990s levels (NALEDI 2006.) In 2000 NUMSA had 25,000 members in its major Wits East region which covers the southern part of the East Rand or 26 percent fewer than in 1996 and a 45 percent drop since 1989.23 The union density among South African metalworkers fell likewise from 48 percent in 1997 to 42 percent in 2003 (Braude 2004.) NUMSA lost 26,000 members nationally from 200,000 to 174,000 between 2000 and 2003; when it went back to 217,000 in 2005, union density in metalworks had further dropped to 37 percent (NALEDI 2006: 63,) another evidence that newly created jobs were mostly precarious and unorganized. Deepening social inequality and vulnerability accompanied the erosion of stable jobs and working class solidarities. In the mid-2000s, more than one third of Ekurhulenis households had annual incomes of less than R6000, or 500 US Dollars (Development Bank of Southern Africa 2005: 167); 30 percent lacked formal shelter, the second highest figure among large cities (Pillay, Tomlinson, and du Toit 2006: 11,) and more than 40 percent of dwellings had no running water (South African Cities Network 2006: 38.) When I conducted my research, less than one percent of employed workers received housing subsidies, as the average monthly income of beneficiaries was a meager R 542.24 The earnings of the near totality of the working poor were too high to qualify for public housing finance and too low for mortgage lending.25 The luckiest

231

found family accommodation in proper brick houses, but many lived in shacks, either self-built or rented backyard rooms. As in most of the country, Ekurhulenis local government has abdicated from the role of direct provider of municipal utilities, which would require cross-subsidization from wealthy, still largely white suburbs to mostly black and poor townships. The corporatization of water and electricity demands entities delivering services, often council-owned companies partnering with international investors, to generate income by recovering costs from end users. In poor communities, basic services are commodities, not fiscally subsidized entitlements. Payments are often no longer based on periodical meter readings; residents must instead purchase water and electricity in advance and pre-paid devices monitor and end their provision. The governments Masakhane campaign tried to enforce users payments with the rhetoric of personal responsibility. Many Ekurhuleni residents, however, responded to increasing costs and diminishing access by illegally bypassing prepaid meters to plug water and power grids; those still with metered consumption refused to pay municipal bills. In 1997 a wave of rates boycotts started in the East Rand townships of Tembisa, KwaThema, and Tsakane. Mass demonstrations and attacks on municipal offices revealed the political dimension of the events, where local community movements, often even branches of the ANC-allied South African National Civic Organisation, were actively involved (Barchiesi 1998.) In 2000, more than 90 percent of residents in Katlehong, Thokoza, and Vosloorus had utilities arrears (Johnson 1999: 14.) They continued a history of protest that had once severely disrupted apartheid-age local governments; democracy could not stem them despite the ANCs fulminations at the culture of non-payment. Persistent poverty and the precarization of employment told in the East Rand a story that resonated throughout the country. A vast multitude of producers was stranded between the 232

decline of apartheid-age, state protected industrialism and GEARs dreams of a globalized, service-orientated economy that gave them little prospects and almost no rewards. Unlicensed trades, sex work, and illegality address then the pressing imperatives of survival among those the old order termed surplus urban dwellers and the new regime designates as the second economy. The governments normative centrality of labor finds a powerful ally in the total commodification of life, which makes the impact of unemployment utterly devastating, turns survival into desperation, and reveals the ruthless material compulsion underneath self-satisfied official celebrations of the dignity of work. Ekurhuleni is an experiment with translating the two economies theory into a spatial practice of government. When South Africas system of urban administration was restructured, the metropolitan area was not anchored to a pre-existing municipality but amalgamated many separate towns and cities. Signification and ideology gave it, nonetheless, particularly thick boundaries. The idea of Ekurhuleni replaced certain elements of the East Rands old geographical coherence manufacturing industries, insurgent townships, and radical workers organizations with the market-driven social stratification that inspired post-apartheid urbanism, for example in the 1998 White Paper on Local Government (Bond 1999.) The differential speed and depth with which industries adapted to economic liberalization recast local inequalities. Few benefited from global competition in knowledge-based services and high-technology productions. Many were reconfigured as a precarious population of unemployed and casual workers whose propensities to social instability are kept in check by survivalist selfentrepreneurialism. Local planners have indeed embraced inequality as economically functional to their modernist narrative where global city-regions drive the nation to progress and wealth by 233

successfully attracting investment and accumulation. Ekurhulenis professed role as a laboratory to articulate the technically advanced cognitive capitalism of the elites with the backward and disorderly survivalism of the poor compensated for its lack of historical rationale. The leaders of the metropolitan municipality envision their mission, accompanied by the ubiquitous adjective developmental, as fostering a diversified local economy able to meet local needs, support sustainable development and adapt to changes in accordance with global demands and shifts and a skilled community exhibiting capabilities in self-reliance, innovation and continued reskilling to meet the needs of a growing economy (Ekurhuleni Metropolitan Municipality 2007: 15.) Growth would, conversely, serve the purpose of creating sustainable opportunities for people from the 2nd economy to engage in the first as equal participants (Ekurhuleni Metropolitan Municipality 2005: 6.) As in the imagination of national rulers, the metaphor of the second economy evokes a highly contradictory mix of disquiet and possibility. In it, the inequities of liberalization determine potentially destabilizing needs, but also the necessary idea that only micro-enterprising self-discipline and consumer wisdom, not public services, can lift the anonymously redundant multitude to the ranks of national citizens. The role of locality as a space where state normativity is translated into the immediate task of governing a precarious world of work is similarly apparent in my second case study of Johannesburg municipal workers.

Johannesburg Municipal Workers and the Corporatization of Local Service Delivery

South African black municipal workers have confronted the apartheid labor regime more harshly than manufacturing ones and faced tougher obstacles to unionization (Barchiesi 2007c.) Residential segregation and dispersed production sites have for long hampered collective 234

solidarity in a largely migrant workforce recruited from the homelands and confined in municipal compounds. Apart from the subordinate juridical status of migrants, African municipal workers faced, even after the labor reforms of the 1970s, legal prohibitions for public employees to unionize, strike, and collectively bargain. Access to such rights in the 1990s finally allowed mass unionization in the sector. Under apartheid, workers organizations were delayed and tentative, but provided nonetheless connections between municipal workplaces and the emerging black labor movement. The conditions of city council employment, moreover, revealed specific possibilities as workers claims expressed the simultaneous position of producers of services and citizens using them. Labor struggles drew therefore strength from the moral vocabulary of the community to transcend the stifling boundaries of the workplace. Since its early days Johannesburg has witnessed African municipal workers attempts to unionize. The citys initial tumultuous growth as a territory of corporate conquest and relentless exploitation overlooked public infrastructures and encouraged the militancy of workers providing municipal services in appalling conditions. Already in 1918 the bucket boys, mostly migrants from Transkeis amaBhaka population hired to collect night soil in the absence of public sanitation, struck for higher wages (Beinart 2001: 102-03.) The still moderate ANC supported them, but it would be a long time before stable unions emerged. Historically, the productive texture of Johannesburg is more diversified and fragmented than the East Rand, its workers are less subject to Fordist manufacturing, and its African entrepreneurial layers are more robust, thus a comparable geographic conduit for labor solidarity was missing (Mabin 2007: 45.) Significant attempts to unionize Johannesburg municipal workers were ultimately unsuccessful. In July 1980, 12,000 African migrants, out of a 14,000 black workforce in the Johannesburg City 235

Council (the name of the white municipality at the time,) embarked on a wage strike that was crushed by police violence, thousands of dismissals, and the deportation of many strikers to the homelands (Keenan 1981.) In 1982 the city council stopped hiring migrant workers and then, as a further impediment to unionization, covered half of its headcount with casuals recruited among local residents. In 1986, the main union in the sector, FOSATUs Transport and General Workers Union, organized only one tenth of the citys employees (Markham 1987.) A decisive stimulus to the national unionization of municipal workers came, instead, from the opposite side of the country, where the Cape Town Municipal Workers Association (CTMWA,) an old white-dominated, bureaucratic, and conservative union, had fallen since 1964 under the control of young radical organizers committed to internal democracy and leadership accountability. With the rise of the independent unions in the 1970s, the CTMWA became an outspoken critic of apartheid and denounced the established white labor organizations, while it successfully unionized Cape Towns colored municipal workers (Buhlungu 2001: 79-82.) Even if it was not part of FOSATU, the CTMWA shared its emphasis on nonracialism and workplace organizing, which it nonetheless combined with a consistent support for community struggles.26 For the CTMWA labor-community alliances meant the fulfillment of [workers] rights as citizens27 through a fight not only for the economic betterment of workers, but against Apartheid and political oppression28 and the rejection of an economistic view of trade union unity.29 After the formation of COSATU in 1985, various trade unions, inside and outside FOSATU, merged into a new national organization of municipal workers, the South African Municipal Workers Union (SAMWU,) launched in 1987 as a COSATU affiliate. SAMWU inherited from the CTMWA, which by far contributed most of its members, a peculiar blend of 236

workplace-based and community-orientated activism rooted in an image of itself as a representative of citizens as well as workers (Ernstzen 1991.) SAMWU, therefore, pioneered opposition to privatization among COSATU unions by mobilizing against plans by the late NP government to corporatize municipal utilities. In 1988 the union inaugurated its antiprivatization campaign, which targeted in particular the subcontracting of local services or their commercialization into income-generating entities. SAMWUs insistence on public provision and regulation demanded, instead, wealthy white municipalities to fund services for poor black townships (Barchiesi 2007c: 58-60.) The democratic transition and new labor rights facilitated SAMWUs membership growth, which made it one of COSATUs fastest expanding affiliates as public sector unions substantially increased their weight in the federation. At its inception in 1987, SAMWU had 16,000 members, overwhelmingly Cape Town colored employees inherited from the CTMWA. By 1993, members had grown to 73,000, one third of which were from the Transvaal province, centered on the PWV (future Gauteng) area. By 1997 membership had soared to 120,000, one third of which from Gauteng alone. In 2003, SAMWU was the largest local government union in South Africa, organizing 55.9 percent of the total workforce in the sector and more than 70 percent of Johannesburg city workers, while nine South African municipal workers out of ten were by then unionized (SAMWU 1997; 2003.) Together with new opportunities, however, SAMWU faced the fresh challenges of economic liberalization. The ANC-led government outlined its framework of municipal restructuring in the 1994 Reconstruction and Development Program and the 1998 White Paper for Local Government. Their stated aim was to merge former racially segregated city councils into larger, integrated municipalities suited to close huge gaps in access to services. After GEAR, 237

however, the national government shrank direct funds for municipalities and encouraged cost recovery policies (McDonald 2002.) A mode of managerialist governance averse to state intervention and favorable to cooperation with the private sector, which hegemonized national social policy, took hold in municipal planning as well. The World Bank was an important player and it regarded the newly democratized Johannesburg as a laboratory for entrepreneurialism in service delivery (Chandra and Ahmad 2000.) The quest for thrift, efficiency, and self-reliance promised to shelter the allocation of local resources from competing political claims, but also paradoxically tasted like late-apartheid municipal liberalization and its predilection for the corporate bottom line. Post-1994 local governance could, however, pivot on a political mandate of redress for disadvantaged communities, and used it to devise diffuse and participative processes for civil society actors willing to mutate from resisters to stakeholders. The approach reminded what Barbara Cruikshank (1999) called new contractualism, or the institutional representation of marketization as a deal between service providers and users defined as freely choosing, apolitical individual consumers rather than citizens with collective claims. Municipal government upholds the constitutional imperative to enable rights not by actually providing services but by unlock[ing] the developmental potential of its citizens (Parnell 2004a: 3) and their capacity to seize market opportunities. Whereas old imageries of citizenship rights gave rise to expectations the state could hardly manage, the resignification of the citizen into a consumer devolves governmental tasks to the individuals economic self-discipline. Similarly to employment-based social benefits, commodified municipal utilities hinge on the personal management of social risk through market processes that define actual social protection as a relation of negotiated intersubjectivity (Dean 1999: 173-84) between the individual and corporate capital. The states 238

imagination of its subjects thus posits the responsible customer in the community as the counterpart of the active jobseeker in the labor market. In the case of Johannesburg, the social challenges of the transition were massive. In the early 2000s, one third of its economically active population was unemployed and 17 percent of its inhabitants lived in informal dwellings (Everatt, Gotz, and Jennings 2005.) One half of households were in deep poverty, with incomes below R1,600 (170 US dollars) per month, and 18.5 percent had no income at all (Palmer Development Group 2004: 30.) One third of African households had no running water, and half were without flush toilets in their dwellings (Beall, Crankshaw, and Parnell 2000: 113-15.) In Soweto, the countrys largest former African township, 30 percent of the population stayed in backyard accommodations, half of which shacks, and a further 14 percent resided in hostels or informal settlements (Morris et al. 1999.) An aspiring black middle class and corporate elite acceded to affluent lifestyles and consumption patterns either in formerly white neighborhoods or in newly prosperous enclaves of black townships. For the poor, including a constant inflow of domestic and foreign immigrants, waged employment has largely ceased to signify feasible prospects of advancement or even the possibility to negotiate inadequate public policies and eroded formal modes of social life (Tomlinson et al. 2003.) Much of the inner city has become a loosely regulated zone of trade, barter, money laundering, and counterfeiting amidst ageing, unrepaired apartment blocks interspersed with incongruously gentrified pockets of cafes and art galleries. A haphazard semilegal entrepreneurial texture connects and overlaps with gangsterism, vigilantism, unfettered real estate speculation, undocumented migration, and national and international crime syndicates. It also, however, provides services and protections to a precarious multitude, invisible to official regulatory agencies, unmarked in institutional territorial grids, unable to claim constitutional 239

rights, and redundant on the labor market. Employment strategies are then subordinate to modulated transactions among discrete population groups (Simone 2004b: 419,) forcing residents to continuously reinvent the everyday, improvise livelihoods, and cultivate multiple identities to plug survival networks revolving around skills, religious affiliations, or ethnonational identities. Johannesburgs official narrative of postapartheid transformation was encapsulated in its city development strategy for an African world-class city (City of Johannesburg 2001: 14759.) In 2002 a broad strategic blueprint, called Joburg 2030, imagined that by the third decade of the century Johannesburg would complete its journey from a mining/manufacturing economy to a service-based economy where inefficiencies and diseconomies would be ironed out and people and infrastructures would work seamlessly together to increase profitability. Instead of an obsolete, slack, and grievance-prone proletariat the labour force will be dominated by white and blue collar workers with a culture of numeracy, technology, and high service standards. In a telling revelation of the sources of its fantasy the document concluded: While there will still be some poor people, poverty will have been reduced . . . and the quality of life of most citizens will have more in common with citizens of San Francisco and London than with those in developing country cities (City of Johannesburg 2004: 332-333.) The obligatory boosterism of countless planning documents has since then testified to the elites psychic need to exorcize their anxieties about Johannesburgs possible future as an informal, rusty, chaotic third-world metropolis (Murray 2008.) The discursive production of a globalized Johannesburg summons, instead, an alternative universe of shiny and efficient postindustrialism. Such a representation surely risks dismissing the living conditions of the majority as abnormal and pathological. It also, however, resignifies precarious employment into 240

two contrasting normative narratives. On one hand, informality denotes an unruly, potentially threatening otherness to be kept at bay with ubiquitous gated communities, high walls, electrified fences, armed checkpoints, and surveillance cameras. But informality can also indicate a virtuous path to accumulation, which finds solace in clichs from the citys colonial origins as a rugged frontier promising endless rewards for daring entrepreneurs (Beavon 2004.) In the end, Johannesburgs promise welcomes and even celebrates the excluded, but only if they relinquish redistributive demands and frame their self-expression as victims of past injustices channeling expectations for redress into acquisitiveness, consumption, and cultural recognition. A photographic volume on South Africas urban shantytowns, under the indicative title Shack Chic, celebrated the triumph of artistic tenacity over adversity (Fraser 2003) and used the poors aesthetic sense to sanitize and depoliticize their material oppression. A special issue on Johannesburg of the journal Public Culture argued that the contingency, indeterminacy and provisionality of the postapartheid urban condition condemned analyses of inequality, power, conflict, and class to irrelevance as gloomy remains of left elitism (Mbembe and Nuttall 2004: 366.) As a reality of experiment, artifice, and opportunity Johannesburg disclosed to the authors a distinctively linear trajectory, where workers, the poor, criminals, and illegal immigrants recede and civic-minded public intellectuals of all races, as well as highly skilled migrants, jet-setters, and a new black elite unstoppably turn the city into the great shopping mall for most of sub-Saharan Africa. Michael Watts (2005: 184) powerfully criticized this approach as an updated progressive optimism that, by superseding the political with notions of freedom and democracy that emanate from fashion and advertizing, cannot go beyond a desperate search for human agency . . . in the face of a neoliberal grand slam.

241

At a practical level, behind Joburg 2030 was a comprehensive restructuring exercise, called iGoli 2002 (iGoli, seTswana for city of gold, is one of the citys popular nicknames,) which created a unified municipality, or unicity, to combine eleven previously segregated local authorities. The new Johannesburg encompassed vastly unequal spaces: the former white city councils of Johannesburg, Sandton, Randburg, Roodeport, and Midrand, black townships like Soweto, Alexandra, and Lenasia and, at its southern end, the vast informal settlements of Orange Farm. A single municipality was established in 1995 and four years later it became the Greater Johannesburg Metropolitan Council (GJMC,) the predecessor of todays City of Johannesburg Metropolitan Municipality. Facing a budget crisis verging on bankruptcy, the council had declared a state of fiscal emergency, while the Gauteng provincial government forced upon it a three-year freeze on capital investment and new employment and required the restructuring of local services into public-private partnerships and independent business units. The first local democratic elections in 1995 consecrated the ANCs control of the city council. Counting on a close to two-thirds electoral majority, the party reshaped Johannesburg along the lines of its free-market agenda. In February 1999, the city council appointed a select committee of municipal officials, the Transformation Lekgotla (council,) which joined a four member management team to steer the restructuring process. The two organs had nearly absolute executive powers and were not subject to the oversight of elected representatives. Despite the ANCs praise of civil society participation, the imperative of administrative efficiency and a selfimposed fiscal emergency defined local reform as a centralized, technocratic, and politically unaccountable exercise. World Bank advisors were involved since the early stages of the process,30 and an organization review coordinated by the business consultancy giant

242

PriceWaterhouse & Cooper endorsed the corporatization of municipal departments. Anticipating workers disquiet the review warned: The trade unions will offer resistance to efforts which look like a privatization. Furthermore, employment risks or lose [sic] of pension fund advantages in mergers will engender great disquiet and opposition (Greater Johannesburg Metropolitan Council 1998: 8.) At the beginning of 1999, the chair of the management team city manager, ANC politician, and former activist Khetso Gordhan presented the iGoli 2002 plan, which the city council ratified on March 16. Up to that point, neither the unions nor civil society organizations had been asked to comment on the plan; the municipality started consulting them only after its approval (Thobejane 2000.) IGoli 2002 eventually opted not for full-scale privatization but for corporatizing and commercializing municipal activities. It transferred service delivery responsibilities from municipal departments to autonomous business units, or utilities, agencies and corporatized (UAC) entities operating as registered independent companies but with the council as their sole shareholder. Three UACs, designated as utilities, were put in charge of income-generating services water and sanitation, electricity, and waste management. Agencies are responsible, instead, for nontradable services like roads, stormwater, parks, and cemeteries. The municipality is no longer supposed to subsidize the budgets of the UACs, except for agencies, which do not generate income through direct user payments. UACs, conversely, can enter service delivery contracts, monitored by the council, with private entities. Households defined as indigent, initially estimated at a mere 20,000, would receive a modest lifeline of free water and electricity (Greater Johannesburg Metropolitan Council 1999a.) In its methods and objectives the plan was 243

in many ways reminiscent of late-apartheid local government privatization. It used the fiscal crisis to shape public perceptions and mobilize consent for drastic technocratic solutions, supposedly speedier, less controversial, and more efficient than democratic deliberation. On such premises, it could then propose the subcontracting of services, the elimination of public crosssubsidies, the reduction of the councils redistributive functions, and an emphasis on user payments as a condition to finance the expansion of utilities. Institutional amnesia on the historical role of private accumulation in the citys social disparities allowed the council to glorify market discipline as its modern day savior. According to municipal consultants, marketbased utilities could optimally insulate payments for services from political interference and give local politicians a powerful weapon against property invasions, rent boycotts, mortgage defaults, and illegal water and electricity connections (Govender and Aiello 1999.) The ANC intervened decisively in support of iGoli 2002 and suppressed dissenting voices like the Pimville (Soweto) councilor, Trevor Ngwane, expelled from the party for criticizing the plan. COSATU denounced the Johannesburg municipal restructuring as an act of neoliberal privatization, and SAMWU made it a central target of its antiprivatization campaign. 31 The municipal union did not object to the need of restructuring to tackle the citys fiscal crisis, but it feared that the new corporatized entities would lead to layoffs, the employment of casual labor, and the erosion of workers benefits. SAMWU resented the undemocratic process of iGoli 2002, a plan that in the style of old apartheid politicians denied a transformation of the city in a way that will benefit the poor.32 Its opposition resurrected the discourse of labor-community links to reinforce workers claims with the plight of the poor and their frustration at the lack of progressive change:

244

Whilst acknowledging the crisis, we are firmly of the view that national government has a moral (and indeed constitutional) responsibility to ensure that local authorities are able to meet their service delivery and staff obligations. . . Our members have shown great loyalty, commitment to service delivery and patience in continuing to work notwithstanding the uncertainty (SAMWU 1998.) Despite labor opposition and ongoing negotiations, in September 1999 the municipality unilaterally decided to proceed with the implementation of iGoli 2002. SAMWU went on strike against the plan the following month, trying to revive social movement unionism by gathering the support of the Johannesburg branch of the Communist Party, local ANC dissenters, student groups, and various anti-neoliberal civil society groups, citizens forums and crisis committees. Social movement actors, largely composed of unemployed and social grants recipients, often preferred direct action to institutionalized processes and supported tactics like unauthorized water and electricity connections. By seeking broad social alliances, SAMWUs national leaders hoped to rescue labors political role and counterbalance losses on the terrain of workplace relations. As its general secretary, Roger Ronnie, explained: People have lost a lot of confidence in the ability of political parties to give expression to their needs. We will take up their struggles independently of parties. We need to build strong alliances with community forums.33 In May 2000 SAMWU and its civil society allies formed an anti-iGoli forum out of which, in September, came the launch of the Anti-Privatization Forum (APF,) a coordinating body of community and political organizations opposed to the neoliberal policies of the ANC government. The alliance between SAMWU and social movements mirrored a radicalization in

245

workers opposition to iGoli 2002, which culminated in protracted strikes during November and December 2000. As negotiations lagged, however, various factors decisively undermin ed SAMWUs stance (Barchiesi 2007c: 63-64.) The ANC gained a resounding victory in the December municipal elections, also thanks to COSATUs support. Faced with the municipalitys unilateralism, SAMWU lacked resources and skills to meaningfully negotiate a highly complex restructuring process. As a COSATU affiliate, SAMWU was bound to support the ANC-led alliance, and a strike by its members in the countrys largest, and ANC -run, municipality was potentially embarrassing. The alliance eventually weakened SAMWUs stand against privatization, and brought internal divisions to the surface. COSATU leaders did not favor labors closeness to movements like the APF which they regarded as ideologically antigovernment and intervened to moderate SAMWUs radicalism. The APF, on the other hand, disliked the unions privileged status as employees representatives in negotiations with the city council, arguing that iGoli 2002 was not a mere labor relations issue, but affected fundamental matters of social rights and service delivery on which local communities demanded a meaningful political space (Parnell 2007: 149.) SAMWU itself was divided; its Johannesburg organizers were reluctant to prolonged confrontation as they hoped to gain concessions on their members jobs and benefits, while the national leadership was more focused on the unions principled war on privatization. In the end, despite SAMWUs stated opposition, iGoli 2002 was implemented on schedule and the new corporatized UACs started operations on January 1, 2001. SAMWUs avowed roots in social movement unionism were ultimately unfruitful in its challenge to market-based municipal restructuring. What worked under authoritarianism proved ineffective within democratic governance. If anything, the iGoli 2002 saga left SAMWU with a 246

profound confusion over whether organized labors task is to contest social policies or manage them in the hope of decent deals for the rank and file. As in the East Rands own painful transition, the workers sense that identities built around waged employment can inspire a progressive imagination of citizenship came out profoundly shaken. SAMWUs Johannesburg organizer, Hlubi Biyana, seemed to admit that much when he complained that his role in the restructuring process was merely rubberstamping and that there is a clear limit to what unions, as reformist organizations, can hope to achieve: We have a responsibility of defending and protecting our members. . . . We would like a situation where even if there are corporatized entities it is still possible to strike an agreement that says that the council guarantees jobs.34 Even more than in NUMSAs case in the East Rand, SAMWUs dilemmas in Johannesburg are amplified by far-reaching changes in the composition and the discourse of its membership. The city councils exceptional emergency powers and its employment freeze from 1997 to 2000 have resulted in a more contingent workforce than ever before. In the utility historically employing most fixed-term and casual workers, waste management, their share has jumped from 36.6 percent of the total headcount in 1998 to almost 50 percent in 1999 (Greater Johannesburg Metropolitan Council 1999b: 8.) SAMWU, like most unions, does not represent these employees: less than ten percent of its members are in fixed-term contracts, while casuals are practically absent.35 An organizer argued that, even if recruiting nonpermanent workers is theoretically desirable, in practice that would be a high-cost operation with poor returns from an income point of view (cit. in Bonner 2004: 8.) More importantly, however, SAMWUs opposition to casualization could sit uneasily with local communities, especially poverty-stricken shantytowns, where casual jobs in refuse collection are often an important source of income. 247

Johannesburg was the main, but by no means the only, laboratory for managerialist, market-orientated, and business-friendly local government restructuring. Similar processes across the country have dramatically strained the unions membership and representative capacity. In the early 2000s SAMWU was losing members for the first time after thirteen years of uninterrupted growth, and the overall unionization rate for local government fell from 55 to 51 percent between 2002 and 2003 (Braude 2004: 56.) In Johannesburg, most SAMWU members were dispersed across the new UACs, which became their new employers. By 2005, only 6,000 of the 25,000 members in the Johannesburg branch were still directly employed by the city council. The managers of the new companies have taken advantage of the newly divided workforce to weaken its negotiating power and step up their opposition to collective bargaining (Mongale and Sekaledi 2005.) In an attempt to allay workers fears and elicit their trust, the municipality and the UACs agreed to three years job security for their employees and promised to hire long-term casuals into permanent positions.36 In exchange, however, managers demanded more flexible schedules and heavier workloads to cut labor costs. In the case of waste management, the new corporatized utility, called Pikitup, placed casuals in its permanent payroll but at lower wages than existing employees even if it projected a gradual equalization and it confirmed its freeze on new permanent hires. It finally subcontracted its fleet to a private company, which provided fewer trucks than the city council, thereby increasing the workload of refuse collection teams (Samson 2004: 119-25.) Together with new employment hierarchies came highly unequal services: regular curbside waste disposal remained largely limited to affluent suburbs; in poor communities and informal settlements Pikitup continued the longstanding practice of contracting collection to small, often unregistered local operations recruiting a totally casualized and unprotected workforce (Smith and Morris 2008: 434-36.) In few cases, 248

as in Soweto, subcontracting took advantage of government-funded public works like the Zivuseni scheme of the Gauteng province which hired at the lowest possible wages. In Orange Farm and other informal sites volunteers worked without monetary pay and protective equipment. The city council has especially targeted women, traditionally at the bottom of the municipal employment ladder, for such schemes as they associate waste management with stereotyped images of feminine work as selfless community care (Samson 2005: 9-18.) As labor battles became more defensive, the secretariat report for SAMWUs 2003 congress noted a growing lack of appeal for militant and political unionism, to which workers increasingly prefer legal support and service organizations aiming at protecting jobs and benefits (SAMWU 2003: 6.) The precariousness in the social conditions of Johannesburgs municipal workers is, similarly to their East Rand comrades, not just a product of material changes, but also reflects the symbolic effacements of a global city narrative that either relegated labor struggles to invisibility or represented their plight as an oddity from a bygone age. SAMWUs appeal to workers organization as the point of intersection of workplace demands and citizens claims was defeated not only by the ambiguities of the labor-ANC alliance, but also by its imagination, which recodified dignity and rights as the results of a specific mode of productive discipline. The City of Johannesburg has been far suppler and more successful than the unions in internalizing and normalizing the mutations of the world of work and building a vision of modernity and progress out of precarized employment. The councils 2006 Growth and Development Strategy confessed to a certain overoptimism in the original Joburg 2030 document and its view of full employment through high-end service jobs. As a corrective it embraced the ANCs freshly published two economies scenario. The new blueprint maintains therefore an unforgiving view of global urban competition and warns that Johannesburg will 249

suffer from the rise of cheap-labor, high-productivity East Asian metropoles. In this brave new world, it predicts, only the well-educated will now have prospects for accessing the economy (City of Johannesburg 2006: 28.) As a consequence, Johannesburg would share the priority of national institutions to make the second economy a governable space by nurturing the human capital of would-be business people that have not yet come to fruition, or which are struggling to stabilise. To dispel accusations of cold-hearted market rationality the city government reaffirmed its commitment to aid resilient and proactive entrepreneurialism. The cast of characters in the new municipal script of social achievement abundantly draws from popular managerial lore the masculine largely self-trained part-time IT guy forging innovative opensource software applications in his garage as well as from NGO-type, and equally gendered, self-help clichs the womens cooperative whose members may or may not be able to get it together to react to a US tourists enthusiastic response to their beaded dolls (City of Johannesburg 2006: 30.) The citys strategy closely reflects the Taylor reports emphasis that poor citizens are economic assets to be optimized even if full-time waged employment let them down, and to that end entrepreneurial ethics should be combined with minimal social provisions. Confronted with mounting opposition to municipal privatization, in 2001 the national government helped municipalities introduce a lifeline free basic services mainly water and electricity and partial write-offs of unpaid rates for indigent households (Modsell 2006.) The civil society, however, usually considered the level of free services inadequate. The South African Human Rights Commission (2004: xii-xiii) showed that the statutory amount of free basic water, six kiloliters per household per month, fell far short of the fifty kiloliters recommended as a minimum standard by the World Health Organization. The APF and other social movements have also 250

criticized the free basic services policy as marketization by stealth because it allows municipalities to impose the installation of prepaid devices as a condition for residents to benefit from the program (von Schnitzler 2008.) The combination of entrepreneurial rationality and basic safety nets in the restructuring of the urban space conferred, nonetheless, a physical materiality to the hierarchies of income and employment that the states citizenship imagination naturalized at a discursive level. South Africas articulation of national social policy and local planning confirms Aihwa Ongs (2006: 7) observation that market-driven strategies of spatial fragmentation respond to the demands of global capital for diverse categories of human capital, thus engendering a pattern of noncontiguous, differently administered spaces of graduated or variegated sovereignty. It is, nonetheless, legitimate to question how effectively can the elites fantasies of resourceful and enterprising poor govern the zones of disaffiliation (Murray 2008: 31) determined by the collapse of waged employment and the hollowing out of social support networks.

Conclusion. Invisible Workers and the Discursive Production of Postapartheid Spaces

The postapartheid reconstruction of the Gautengs urban economy has replicated a national policy imagination that turns the poor into democratic subjects not as claimants but as empowered agents of their own uplift through economic activity. It has also mirrored the modulations of official discourse, where the focus of economic participation has shifted from jobseeking to entrepreneurship. Meanwhile, a large portion of the local working class, once a stronghold for independent unions and their expectations, has become superfluous and redundant. In their attempt to spin the adverse fate of waged employment into the uncertain 251

promises of urban globalization, municipal institutions have obscured the conditions of precarious labor and narrowed its possibilities down to market competition. The two economies metaphor has joined narratives of local specificity to resignify the resourcefulness of informal and casual workers in a distinctively antipolitical sense. The normative centrality of work has therefore emerged from economic uncertainty stronger than ever. The restructuring of work has also closely interacted, as a modality of market discipline, with the commodification of life and basic services. Local governments have claimed a superior knowledge of the goals and the techniques of development to gear social relations according to perspectives where, as Beauregard and Tomlinson (2007: 248) summarize, conflict is erased (or hidden,) government is represented in managerial terms, citizens are conceived as customers, and grassroots organizing is discouraged. The relationships between local governance and organized labor have placed the unions in a thorny position, where they cannot expect obvious gains neither from traditional oppositional politics nor from attempts to engage and shape institutional processes. The transcendence of this binary contrast, which Oldfield and Stokke (2007) advocate, has surely not resulted in what they anticipate as a post-adversarial win-win progressive framework where work and benefits are protected, universal access to services is guaranteed, and civil society influence is entrenched. Rather, the political subordination of trade unions in the ANC-led alliance and their discursive and analytical subjection to the productivist imagery of the government adjusted worker subjectivities to the requirements of market-driven governance. The problems organized labor encountered in Johannesburg and the East Rand confirm the dilemma, surely not peculiar to South Africa, of trade unions playing a decisive role in democratization and yet failing to influence its outcomes in a context of nationalist party 252

dominance. The fading roles of the unions as vehicles of solidarity, proponents of social visions, and organizers of lived experiences makes it particularly important to look at the meanings workers attach to their own precariousness and how they contest the official imagination of the work-citizenship nexus. The two final chapters will turn to this task. The next one will deal with the workplace as a site where East Rand manufacturing workers and Johannesburg municipal employees experience and respond to their deepening uncertainties.

253

5 Translation Troubles. Signifying Precarious Work on the Shopfloor

. . . The body itself is a script or perhaps one should say a ceaseless inscribing instrument. Gayatri C. Spivak (2000: 14)

Introduction

The social precariousness of work, this book has emphasized, is determined by adverse employment changes as much as by normative governmental assumptions that make work central in the citizens conduct. So far I am, however, still dealing with what the introduction identified as the official top-down, so to speak mode of signification of the work-citizenship nexus, or the plane of official discourse, institutional imagination, and policy rationality. Postapartheid governance, aided by the unions own normative universalism, has represented the worker as a malleable and adaptable subject, apt to waged employment under a variety of conditions, but also ready to respond to the lack of waged employment through enterprising economic initiative. State institutions and labor organizations have jointly built, even while contesting policies and political spaces, the working class as a desired subjectivity (Chakrabarty 1988) and the employee-as-citizen as the universal agent of popular sovereignty. The introduction also warned, however, not to overstate the case for governmentality and invited to contrast official representations of the laboring subject with the workers own significations of employment changes. The problem becomes, then, one of commensurability and 254

translation; it requires, in other words, an analysis of how consonant the states work-citizenship nexus is with the values of its intended subjects, and how smoothly it can be rendered into their everyday discourse. The relevance of the task is underscored by South Africas peculiar postcolonial condition where, even if the state imagined capitalist employment as a social norm of modernity, working for wages has historically provided an object of intense contestation rather than a shared horizon of inclusion. In postcolonial societies, as Dipesh Chakrabarty (2000: 17) remarked, the problem of capitalist modernity cannot any longer be seen simply as a sociological problem of historical transition . . . but [is] a problem of translation as well. It is worth reminding that difficulties with placing the compulsion to work at the heart of conducts and values in twentieth-century Africa has a lot to do with the fact that even when trade unions, labor law, productivity regimes, and welfare systems enabled collective claims they did not, contrary to the West, translate wage labor into an idiom of citizenship. In the postcolony, as Gyan Prakash (1995) argued, citizenship hardly defines the social as a smooth space of commensurable claims and responsibilities. Chapter 3 and 4 have emphasized how the ANC government responded to the precariousness of work by grafting onto its original celebration of wage labor a portrait of the casual, informal, and fixed-term worker as an empowered micro-entrepreneur. This chapter and the next will highlight the limitations of official visions in shaping material practices, and will argue that precarious work determines a space of incommensurability and untranslatability that ultimately defy the work-citizenship nexus as a modality of government. Beyond the horizon of the policy discourse, precarious work is far from homogenous; the variety of its material experiences and cultural formations refutes indeed the states ambition to ontologically define labor as an abstract precondition of human dignity. Not only does the gap between the official 255

imagination of work and its mundane dimension undermine liberal and nationalist mythologies of progress. It also questions left idealizations of class solidarity, at least to the extent their desire for wage labor obscures the hiatus between the few who, as Eddie Webster (2005) put it, can earn a living in paid employment, and the many who make a living by haphazardly improvising daily survival. If citizenship is, however, defined as a contested civilizational idea, shot through with inequality, exploitation, and violence, the view that juridically free capitalist labor necessarily evolves into universal values and rights can no longer hold. Abstract labor will then have to be analyzed again as a terrain of antagonism rather than assumed as synonymous with the common good. In Mark Holmstroms (1984: 231-63) classic research on Indias industrial proletariat, workers popular analyses suggested that work is by its very nature essentially insecure. Besides, skilled and unskilled casual workers had different perceptions of employment insecurity, both could differ from unskilled permanent employees, and ethnicity and gender affected meanings within each group. Precarious jobs and their fragmentation were represented not as external threats to allegedly normal stable and cohesive employment conditions, but obvious starting points for negotiation and contestation. Workers did not simply shun the protections of secure and decent jobs, but did not base their expectations on them either. Rather, to win fairness at work they deployed cultural themes external to the production of commodities. The discursive practices of those who are not fully incorporated into waged employment contain, even when they are used within the capitalist workplace, endless repertoires inconsistent with capitalist work discipline and the associated forms of governance. Workers can choose to identify themselves as poor rather than producers, if this helps mobilize the moral support of communities. Their militancy can be unorganized and unpredictable because low wages and hard 256

work do not merely infringe expected codes of workplace behavior and indeed may be even tolerable in the workplace as tests of masculine prowess but violate legitimate household needs, religious ideas, or the status and respect that go with age (Chakrabarty 1988: 189-93.) In a sense, rather than being just unimportant in workers claims, citizenship is too important to have its claims confined to an oppressive production environment. Chakrabarty (1988: 227) powerfully concluded that by centering workers demands on the relationship between work and citizenship the left makes them complicit with the rule of capital and aligns their desire with the nation-state as the dominant ideological discourse of our time. Attaching universal attributes of freedom and emancipation to commodity-producing labor would then ultimately be the limit to all politics and struggle. An alternative would be to emancipate the quest for a dignified life from citizenship claims that reproduce labors subjection to capital. To the extent precarious employment configures such a political possibility, it also subverts what Gayatri Spivak (1992) defined as the absolute translatability of desire into the normativity of work. Among workers in the East Rand and the Johannesburg municipality the postapartheid restructuring of production defeated what the politics of national liberation once promised as an exception among postcolonial societies, or the prospect of a truly inclusive democracy where work could guarantee dignity and rights for all. The insecurities of work have instead not only made employment contracts frail and hollow, but also determined, for permanent and fixed-term workers alike, an existential condition defined by Felix Guattari (cit. in Stivale 1998: 199) as systematic endangering. The workplace became a site where livelihoods are perceived as constantly on the line as a result not of individual failure but of deeper, menacing, barely categorizable socioeconomic forces.

257

Coping with Something Strange: The Disappointments of Workplace Transformation in East Rand Factories

The economic change that followed the first democratic elections was experienced by the labor movement both as a rude awakening and the continuation of familiar late-apartheid dynamics. COSATU alarmingly noticed: Something strange is happening in the new South Africa: the unions are being forced back by company restructuring under increasing competitive pressure from global markets. In company after company, unions are faced with employer initiatives to cut jobs, subcontract or outsource a range of functions and to employ casual rather than permanent workers (COSATU 1997a: 6.2.) For organized labor the ascendant discourse of globalization, which the ANC government embraced, encapsulated the permanence of apartheid-age tactics of union avoidance in the countrys new political dispensation. Ideological disputes over globalization, however, did not help rank-and-file workers allay their confusion as the ANC in power turned against its old working-class constituencies. In place of decent jobs for all as the expected social compact of liberation, the unsettling feeling crept in that the workplace was changing in unpredictable, terrifying directions. Unilateral restructuring defined an unforeseen limit to the political incorporation of the black working class. The East Rand plants I visited are, or in some cases were, located along a strip of old mining and manufacturing towns, stretching from Germiston to Nigel, at the southern reaches of the Ekurhuleni municipality. Surrounding them are the vast proletarian concentrations that once were the African model townships of Katlehong (Germiston,) Thokoza (Alberton,) Vosloorus 258

(Boksburg) collectively known as Kathorus KwaThema (Springs,) Tsakane (Brakpan,) Daveyton (Boksburg-Benoni,) Wattville (Benoni,) Duduza (Nigel,) and the former colored or Indian areas of, respectively, Geluksdal (Brakpan) and Actonville (Benoni.) The eastward drive along the N17 highway, leaving Johannesburg behind, discloses the complex stratifications of the areas manufacturing history, punctuated by hills of debris from long gone mines. Early sights are the imposing industrial establishments of the Wadeville (Germiston) industrial area. The pipes, chimneys, and workshops of the SCAW steel mill are visible from miles as a reminder of past dreams of heavy, large scale industrialization and a historic signpost for the independent unions struggle to assert themselves at the core of the apartheid work machine. As the travel proceeds amidst monotonous farms and scorched grassland, smaller factories take over and the tracks of industrialization become feebler. Approaching Brakpan, a town where mining remained prominent and manufacturing lagged well into the 1960s (Sapire 1987: 365,) the geography of production has completely changed. While shopping malls still clearly stand out, to find factories one has to exit the highway and drive into pockets like Vulcania, where a branch of Baldwins Steel is located tucked away along what could well be country roads. Baldwins was part of the Dorbyl group, which took it over in the late 1980s. The Brakpan branch specializes in cutting and selling structural steel largely to construction companies through Dorbyls national network. The merchant operation managed to survive restructuring in 1991, when the plant lost the production of flat steel to another branch, was momentarily scheduled for closure, and had its entire workforce dismissed. Union representatives explain the events as a corporate attempt to run away from strikes (Respondent 15.) Eventually, another branch was shut down. Baldwins Brakpan reflects the general condition of the East Rands metal and engineering industries, historically characterized by concentration, 259

vertical integration, and collusive price setting by suppliers, often formerly state-owned conglomerates like Iscor (since 2005 Mittal Steel South Africa.) Smaller players and downstream businesses have thus been forced to adapt by focusing on cost-cutting technologies and compressing wage costs while limiting investment in human resource development. In the sectors ruthless world of work, which persisted after the end of apartheid, the Dorbyl group seemed to be an exception as it adopted an ostensibly participative and unionfriendly approach. In 1992 a reshuffle of the top corporate echelon had seen the rise of a new leadership around CEO Bill Cooper, whose message of non-adversarialism and negotiated transformation stood in stark discontinuity with older days, when Dorbyl was one of the most staunchly antiunion employers in the generally repressive metal industry. The new management, conversely, mirrored the post-1994 governmental emphasis on seeking consensus as the foundation for economic success. In response to union radicalism it advocated a common vision that defies old narrow thinking and practices, allowing the empowering of the workforce to give everything theyve got and management to lead confidently, underlying that the time of fanagolos, dictatorship, double agendas, unnecessary confrontations and all those things has ended in 1994.1 Coopers tenure was punctuated by initiatives with a high symbolic impact, like a corporate truth and reconciliation commission (TRC) modeled after the national, globally famous one, from which it borrowed experts and personnel. The commissions aim was to build trust in the company by allowing workers to narrate, and requiring managers to recognize, past corporate abuses. The reports of the Dorbyl TRC catalogued the usual horrors of the apartheid workplace, with several accounts of white foremen physically and verbally assaulting black workers, or of blatant racial favoritism in promotions, overtimes, or company loans.2 By 260

consigning workplace authoritarianism to the past, however, the company absolved itself for continuing it in the present as a discussion of post-1994 restructuring was left outside the scope of the commission. A NUMSA organizer, on the other hand, justified such a decision by arguing that after the advent of democracy productive changes are driven by market forces,3 less liable to blame than overt racist despotism. Albeit it was a toothless body in terms of company policy, the commission aimed to benefit from the new national discursive paradigm of reconciliation to delegitimize labor conflicts. The main metal union, NUMSA, was then confronting a wave of restructuring in which notoriously union-bashing companies were emboldened by the imperative of globalization; for the union, therefore, Dorbyls symbolic opening to workers contribution felt like a rare soothing intermission. National union organizers looked at Cooper in almost messianic terms as he seemed to herald the conflict-free, mutually beneficial corporate patriotism that the ANC-led government was preaching. For NUMSA what was happening at Dorbyl was nothing short of keeping the emancipative promise of wage labor alive; it was the proof that, for once, industrial change and global competition could be worker-friendly. There would be no need for rank-and-file radicalism, which risked bothering the leaders of COSATU and the ANC, if the company spoke for the general interest. A union official commended Dorbyls CEO: Sometimes other employers called him a NUMSA executive. Hes viewed as a union man.4 Dorbyls profession of corporate enlightenment was starkly contradicted, however, by the quite different notes managerial posturing struck at subsidiaries like Baldwins. The crisis of the 1990s had shrunk domestic demand for steel products and faced Baldwins Brakpan and its 115 employees with growing uncertainties. The factory did not lay workers off, but shifted much of its focus to steel cutting, a relatively capital-intensive operation, which the management 261

envisioned as a lean, customized production done by a flexible workforce in a ferociously competitive environment of many rats and mice who operate on a small basis.5 Although layoffs were not scheduled when I visited the plant, they were routinely rumored to adjust the workforce to sudden demand shifts. The companys quest for nimble and versatile labor foreshadowed union members fears of being targeted for future downsizing. Some allowed me to interview them only if I waited outside the factory gates and drove them home so that they could escape the foremens gaze. The muscular talk of the managers seemed to confirm the workers trepidations: If we can avoid all this crap and this hassle with trade unions. We cant fire here theres a poor guy that starts a business and then he finds out that he cant fire, he has to deal with the unions, people with disabilities, the CCMA,6 the Employment Equity Act If we can get away with all that crap and employ a machine instead of having ten workers employed; that would make perfect business sense for the poor guy that is doing the business.7 In common with Dorbyls participative ideology, nonetheless, the managers in Brakpan held the view that unions are old-fashioned relics of apartheid-age disruptions. Democracy and globalization gave employers a chance, instead, to regain the upper hand. In response, workers disparaged the companys affectations of corporate sensibility with caustic comments. Many refused to participate in the Dorbyl TRC, dismissing it as mampara (buffoonery); older employees equated it with the liaison committees employers used in the 1970s to sidestep union representatives. The shop stewards were also disappointed by how their national leaders had bought Dorbyls narrative of transformation. The NUMSA shop stewards council for the group thus intervened by discouraging members from entering into [ sic] any 262

form of co-determination at Dorbyl plants.8 The union noticed, however, that threats of downsizing cowed rank-and-file resolve: attendance of meetings declined as participation in management-initiated forums increased, and employees were more available for longer hours and multiple tasks (Morapedi 1996.) Respondents reported indeed a deepening perception of precariousness underpinned by the intensification of work and their sense of becoming utterly disposable: One of those tricks used by the company, they want to reduce a number of employees and they have more work for each employee in order not to employ casuals, those are the reasons they are using. Employees that go are never replaced, that is another tactic they use. . . . Maybe the next time youll come back only not to find me, you never know, these days anything is possible. Im not saying Im leaving, but they are just hinting that anything is possible these days. Today Im here, tomorrow Im not here, after sixteen years. Q: Well, they cant fire people just like that A: They cant fire, but they make means that you lose your job. Q: What do you mean? A: Frustrate you, make you run around, make you feel lost, many things. They wont fire you just like that, no, they are actually even scared of the rules of the game, of the courts and all that (Respondent 2.) Managerial unilateralism is, on the other hand, visible to workers through multiple daily practices, the most resented of which is favoritism, often racialized, in task allocation and promotion decisions. The chairman of the small shop steward committee told me that growing workloads and racial discrimination drove a comrade and friend of his to suicide by hanging. The 263

persistently authoritarian politics of production defines a critical terrain of incommensurability between the democratic exaltation of work and the persistent salience of race: Here, you see, apartheid is alive, there are jobs which are made for white guys. Why dont they give people a chance? It is better to try and fail than to fail to try. We once had a meeting and I said it seems to me that black people are used as doormats, if you have a mat in the door someone will come with mud in their shoes and wipe his feet there, that is what we have here (Respondent 6.) A worker bitterly added: Affirmative action as introduced by our new government is applicable only to the other side of the coin [employers], but our side of the coin is not receiving anything (Respondent 4.) For black workers, both in weekly paid clerical and hourly paid production positions, flexibility is just another face of managerial arbitrariness, which contributes to their fragility by blurring qualifications and identities. An office-based dispatcher was required to double as a picker-loader of steel beams on delivery trucks. I met him in the plants shop-steward office; a poster hanging above his head read: God came and saw my work and he was satisfied. Then he came again and saw my salary, he turned away and bowed his head and wept. The requirements of modern management had for him an ancient flavor: Some of us blacks are not moving at all. I prefer to make myself as an example, even if I know six or seven. For sixteen years I have been staffing one area, only names change: today I am loader, next week I am an expeditor, next month Ill be running as a clerk, but the job is always the same, no money improvement, promotions are made to suit whites. Q: But your job description 264

A: I dont know whats my job description, I dont have one. I no longer know what answer I am supposed to forward to my children who ask dad, what job are you doing? I am supposed to be called multiskilled because I am on the floor, I do the loading, I am in the office, I run around, I do the administration, come back here, answer the phone. The guys at the machine, I dont think they have a job description either. . . . If one has to pick up those materials he must do as the boss says it should be done. He cannot think, hes not allowed to think (Respondent 2.) As he continues, his voice becomes agitated and his conclusion is a sad commentary on the fate of wage labors promise: I brought some white boys here that didnt know anything and now they are sales managers, they are part of the directors, and they have gone through me. When they came in they found me, I showed them how we work, and from this department they moved onwards. Boys! I mean boys. Im not undermining them, but today they are the senior guys, and I wont move an inch from where they left me behind (Respondent 2.) He showed me his pay slip indicating a weekly wage, after sixteen years of service, of R105 after deductions, which mostly go into a whopping R400 weekly payment for the company-based medical insurance. Even with two incomes in the family his wife worked in the Gauteng Provinces Health Department the total household income hardly covers food expenditures. In other companies besides Baldwins managerial rhetoric emulated the participatory and collaborative tones of postapartheid governance in the attempt to devise corporate competitiveness as a discourse of solidarity alternative to class. The past company parlance often vowed to transcend referred to workplace conflict, not the despotism of production. Black 265

empowerment deals and affirmative action programs which often recruited former union organizers in junior management positions contributed to such symbolic elisions by announcing the irrelevance of adversarialism in a postracial new age. The executives of the Germiston plant of Paperlink, a paper trading company part of the large Mondi group, vastly invested in company patriotism and experimented with teamwork, quality circles, and worker suggestion schemes. The overall framework often remained, however, steeped in a staunchly paternalist view of black workers advancement. A company official, for example, boasted as adequate financial reward for the employees commitment an outing for them, a day at Gold Reef City [a mining-inspired theme park] or a picnic with their family.9 Like at Baldwins, however, union members seemed to appreciate the considerate language of management and saw it as a break from Mondis otherwise obdurate antiunion stance, where, as a shop steward said, you could cry until the tears were finished (Respondent 141.) Union decline in the industry has been, however, traumatic. Paperlink is mostly organized by COSATUs Chemical, Energy, Paper, Printing, Wood and Allied Workers Union (CEPPWAWU,) a union born in 1999 from the merger of two historic bastions of independent trade unionism, the Chemical Workers Industrial Union (CWIU) and the Paper, Printing, Wood and Allied Workers Union (PPWAWU.) Before the merger the two unions had a combined membership of 93,000 (Buhlungu 2001: 129); by 2006, CEPPWAWU had dived down to 62,000 (NALEDI 2006.) The crucially important East Rand locals lost almost one quarter of their members, largely to layoffs, only in the year before the merger and half of those who remained had no collective bargaining.10 Already in the early 1990s PPWAWU complained that the large, vertically integrated conglomerates dominating the paper industry, Mondi, Sappi, and Nampak, were downsizing 266

their headcount through outsourcing and sidestepped the union by using Japanese participatory management techniques and coopting cadres and ANC members into low-level managerial jobs (PPWAWU 1993.) For CEPPWAWU, Paperlinks self-image as a company respectful of its workers merely continued such practices through what one organizer called a we are family mentality amounting to a very questionable relationship between management and shop stewards, which could instill in their minds the idea that there is no need for the organization.11 Corporate transformative pretensions mirrored, on the other hand, the unilateral implementation of outsourcing and multitasking. After the company contracted out its fleet, drivers were working on three-month terms, lower wages, more deliveries, and no medical benefits. Alternatively, workers from other departments had to double as deliverers, which generated substantial resentment: I have been abused so much, to such an extent, that I am no longer willing to help. What department do I work here? That is the first question I ask to myself. I have to work in dispatch, in admin clerk, receiving clerk, I dont have a license but I am driving the truck so I am not happy individually. I would leave this job immediately if there was something better (Respondent 141.) Heavier workloads and the disintegration of professional identities amplify the perceived insensitivity of the employers and make working conditions dangerous and intolerable: Nowadays when you are a driver, it is as if you are working on the mine. You see, now my leg is broken, they shot me in September, in Doornfontein, trying to hijack me. If you are a driver they can kill you any time, any time. They have hijacked me three times, and you must answer for it, they will ask you a lot of

267

questions, a lot of questions They think maybe you sold it. That is unfair and humiliating (Respondent 147.) A crucial terrain of labor struggles has always been the unions demand to abolish racial wage gaps by standardizing pay grades and remove obstacles to black workers promotions. Legislation and collective agreements have entrenched such gains but layoffs threaten to erode them as they stretched surviving workers on a broader range of tasks with variable remunerations. In the East Rands vestiges of large scale manufacturing skills have largely ceased to function as predictable grounds for rewards and identities. Cases in point were the two glass packaging factories I visited, Consol Glass (Wadeville) and MB Glass (Leondale,) both in the Germiston area. Their business is characterized by large workforce concentrations, vertical integration, and continuous, 24-hour production. Under apartheid the industry served different facets of social control by producing, among other objects, bottles for South African Breweries, a major contributor to consumption and fiscal revenues in black townships, or bulletproof screens for police and military vehicles. For CEPPWAWUs Wits region, which draws more than half of its members from companies with less than thirty employees, organizing large plants is strategically important. Market liberalization and competition from cheaper plastic materials led glass companies to automate production and slash employment. Smaller businesses thrived by taking over specialized cutting jobs, which further fragmented and downgraded workers conditions.12 PFG Glass retrenched almost 1,000 workers, half of its headcount, from 1995 to 1999.13 Between 1994 and 2000 Consol shed, on its way to be listed on the Johannesburg Stock Exchange,14 more than five hundred production jobs or two thirds of the total.15 Despite its vaunted participatory management style, it hardly consulted the union. By March 2008, MB

268

Glass was left with only 270 employees in its Leondale plant;16 when I visited it in 2000, there were about 600, 450 of which in direct production. In a sector with strong craft traditions like the glass industry, black workers once considered advancing into artisan jobs a lifetime achievement, the proof that the worth and dignity of human toil could be asserted even in the most unfair workplace environments. Restructuring was, however, turning artisan jobs into empty shells with formal descriptions no longer matching actual content. One African operator joined Consol in the 1980s; despite holding a high school degree and a four-year fitter certificate, he was hired as a laborer at one quarter of an artisans wage. For years he worked in what he describes as piccanin jobs under less qualified white employees, but eventually his stoic endurance reaped the fruits of political change when he became the plants first ever black certified electrician. He did not regard his new status as part of a labor aristocracy because he won it through militant union struggles and imagined it within a general movement of labor uplift. His expectations, however, proved shortlived as the qualifications required for the job faded while his wage stagnated: In this department I am still the only black fitter working with four whites. Besides, the job got deskilled now because machines are capable of doing their own diagnostics and they can be fixed without having to stop them, and major repairs are done outside. A fitters job is now mainly about inspection rather than fixing (Respondent 197.) In engineering companies craft positions held firmer but they also became more insecure as they are often outsourced to labor brokers. At Nigel-based Union Carriage and Wagon (UCW,) changing employment contracts visibly reshaped worker identities. Started in the 1950s by an Australian investor to manufacture locomotives, trailers, and wagons, UCW grew under 269

apartheid thanks to contracts from the rail parastatal, Spoornet, and the mass transit systems ferrying workers from black townships to white-owned workplaces. UCWs fortunes peaked in the 1970s and early 1980s, when trains were symbols, immortalized in Hugh Masekelas Stimela, of the alienation and oppression of migrant workers. MAWU organized the plants black employees, even if Afrikaner artisans continued to belong to the pro-NP Yster en Staal union. White workers recalled UCWs past as a company ruled by racial harmony, but the chair of NUMSAs shop steward committee retorted that racial domination started to collapse only with political democracy; otherwise it was cruel before. . . . This company was a volkstaat. . . . Black men werent even supposed to come to the office; your supervisor could come, not you, otherwise you were fired on the spot (Respondent 21.) Towards the end of apartheid, the liberalization of the taxi industry and municipal budgetary austerity meant hard times for UCW, which had to fall back to repair jobs and occasional export contracts. In 1996 the giant Murray & Roberts corporation took over the company, which embarked on massive restructuring to grasp new opportunities for global competition. UCWs focus shifted to subcontracts for multinational enterprises, which required a nimble, rapidly adjustable workforce thus facilitating the resort to labor brokers. From 1997 to 1999, therefore, permanent employment dropped from 800 to 310 in the 1980s it could be between 1,200 and 1,500 half of which white artisans. By early 2000 the headcount was scheduled to stabilize at 150, seventy percent of which hired through labor brokers. NUMSA accused the managers of targeting its members for layoffs while preferentially hiring unqualified whites for permanent artisan positions.17 The company, on the other hand, never negotiated retrenchments with the union, which according to a human resources manager and former NUMSA shop steward would have questioned perceived authority.18 Restructuring signaled 270

a reversal of fortune for UCW. In 2003 the company partnered with a black empowerment consortium led by the J&J Group, co-founded by former COSATU general secretary, Jay Naidoo. New public investment in rail transportation brought the first domestic contracts in twenty years; yet, by 2007, with production levels comparable to the golden days of the 1980s, UCW employed only 407 workers.19 Labor brokerage has profoundly destructured employment at UCW and managers used it to reshape workers orientations to the company and to their jobs. Personnel requirements were covered by FEDMO, a local employment agency started in the early 1990s by a Portuguese immigrant and former UCW foreman who once coached the professional soccer team Moroka Swallows before turning to labor brokerage when the economic downturn bit into permanent jobs.20 The 1990s were in the East Rand a tumultuous period of opportunity for the temporary employment industry as hundreds of agencies often fly-by-nights unlicensed operators or retrenched employees hiring small work gangs through personal connections sprang up. For NUMSA it was an ominous sign, also because the recruitment of strikebreakers was a large part of the labor brokerage business. FEDMO was a licensed broker and, as a remarkable exception, NUMSA managed to unionize it after protracted confrontations. It paid contractual wage rates, illness, retirement, and unemployment benefits, but not medical insurance. Self-interest mattered in this strategy because FEDMO, like many brokers in the engineering industry, mostly hired artisans and depended therefore on its reputation to attract highly qualified freelancers. FE DMOs manager visited the plant regularly, and the guarded friendliness of the shop stewards was tempered by their realization that he was astutely taking advantage of a highly detrimental fragmentation of the workforce. The broker, on his part, self-consciously cultivates the image of a socially 271

compassionate entrepreneur insisting that, while few other employment agencies provide his benefits, he is happy to sacrifice profit margins for the sake of building respectability and trust. While talking he showed me a leaflet from a competitor claiming that as contract staff are not unionised, a clients time is not spent negotiating with union representatives and work attendance of contract staff is not dependent on trade union attitudes and agreements. He did not approve of this antiunion stance, but insisted that for him FEDMO, not NUMSA, is the real representative of UCWs contract workers because he negotiates with the company on their behalf. Paternalism tinted with racial condescension transpired when he referred to his employees: When I started at UCW, I used to think and behave like all the other brokers and the union didnt like that, and I thought, This cant be real, why must I give you ten Rand more? If these people that are bigger than me are not doing it, I am not doing it. Then I started talking to the union. . . . Other companies think that if they use labor brokers they wont have to deal with the unions, to get rid of toyitoyi. I am the only one who allows the workers to belong to the unions, theres no other labor broker that I have seen doing that. And still our workers dont stand out of the gate jumping up and down. . . . I have no problems with blacks. I come from a football background, I was training the Moroka Swallows in the 1970s, I have been training black teams since 1972-73, I was one of the two only white coaches, so I know these people, I know how they are and I know how they look, and they respect me, they are the ones that make me what I am, and I respect them. So I have never had problems with what the unions want, otherwise they would have started dancing outside the gates. I know, Franco, that is not what you want to hear 272

Q: No, look, there is nothing that I specifically want to hear A: No, really, there have never been fights between us, I have come a long way with these people, I have always been with these people, many years.21 Labor brokerage separates hiring workers from employing them, translating the neat juridical category of the labor contract into something fuzzy and indistinct, which disorientates workers because not only the boss is not obviously visible, but also who is supposed to represent them is no longer granted. For FEDMO, a good broker effectively negotiates working standards and prevents illegality: I know lots of labor brokers outside and they dont basically care about the people, what will happen tomorrow to that guy. My people are people who have worked here before, we dont just put an advert saying that we want six artisans, I personally bring them here and because I bring them, I know what the person can do. . . . And let me tell you something: nobody comes to my door, nobody swears at me, nobody calls me names and I would like to think that I am still doing business ten years down the line. . . . Outside its very stupid, companies would just say I need a boilermaker, and thats it, and they wouldnt worry about provident fund. So I have to make R1.50 and hour while others make R10.50 because they dont pay benefits, thats exploitation. . . . So, you people looking at the labor brokers are quite right: labor brokers in general is probably the worst business around. . . . You cant imagine; you havent even started to imagine. In the end the employer is able to reassert, through contract hires and constant fears of layoffs, control over a highly skilled and unionized workforce, which used to strongly identify itself with the workplace: 273

Its bad for everyone. Now even foremen get retrenched. . . . It looks like a person is blown away, they are taken away from a family, you see. When we enter here we want to stay, because we get to know each other and its sad to see other people staying at home when theyve got a family. I also feel bad, really, because we have been together (Respondent 40.) FEDMO also recruits white artisans, but their positive opinion of UCW as one of them described it, a good company, a company that looks after its people (Respondent 37) reveals feelings that in the event of a temporary layoff it would be easier for them than for black workers to have their old jobs and wages back. Such a sense of stability is, for black employees, largely shattered. Some of them were laid off from craft positions and rehired soon after for additional, unskilled tasks carrying lower wages. The idea that craft and qualifications protect workers from market uncertainties has consequently collapsed together with expectations for racial fairness as foremen discretionally allocate tasks and remunerations in clearly biased ways: There is mone y for the black township, money for the colored township, and money for the white suburb (Respondent 29.) Old modalities of racialized precariousness did not therefore disappear but seem to have rather undergone a postfordist adaptation. The travails of black employees at UCW and Baldwins hardly match the image, conveyed by the states policy discourse and mainstream social sciences alike, of the prosperous middle-class lives of postapartheid workers acting as role models for the second economy. New insecurities rather overlap with the remnants of past practices of domination to question not only wage labors promise but also its hard-won gains and meanings. Sometimes, moreover, the disintegration of a social existence centered on employment can be sudden and catastrophic. Kelvinator South Africa was once a company, based in Alrode, near Alberton, manufacturing 274

electric household appliances. As such, it has accompanied South Africas path to mass consumerism and the excitements of its eventual deracialization. Established in 1929 as C.G. Fuchs, the enterprise benefitted from the war effort and apartheid-age industrial protection. It eventually started producing fridges with licenses from the United States (Westinghouse) and Sweden (Electrolux.) The vast Barlow Rand group, one of the pillars of apartheid manufacturing, absorbed it in 1978 and turned it into the Barlows Consumer Electric Products Group (Barcep.) Alrode was an early stronghold for MAWU, but in 1984 the unions East Rand branch suffered the breakaway of UMMAWOSA, which, criticizing the dominance of white officials in FOSATU, made headways in the area.22 In the Barcep plant, MAWU faced strong managerial resistance and won recognition only in 1983 (Forrest 2005: 242); as a result of the split, its membership, once almost fifty percent of the workforce, was almost erased.23 In November 1996, amidst growing competition, Barlows announced the intention to liquidate the plant, then employing almost 2,000 workers. A month later, however, the company was taken over for a symbolic R 1.00 by a consortium formed by NUMSA and Sovereign Capital Investment (SCI,) led by businessman Simon Koch, which owned a license from USbased Kelvinator to produce fridges and ovens in South Africa. Existing workers, defined founder employees, were given a 20 percent share of the company.24 NUMSA agreed to place their severance payments into a Kelvinator Employee Founder Share Trust, which, together with new capital injections from SCI, would restart the company as a going concern.25 Koch launched an ambitious expansion of Kelvinator, expecting a soaring demand for household appliances in black townships the democratic government planned to electrify. Under apartheid the townships serial, nondescript four-bedroom matchbox houses transmitted to the outside world their dwellers function as anonymous, fungible providers of labor power. Now, however, 275

domesticity was heavily invested with expectations of dignity, upward mobility, and, especially among young workers, conspicuousness. The possibility to buy good furniture and appliances ideally materialized the recovered worth of wage labor into everyday life and, in the case of women workers, allowed to renegotiate gendered household chores (Meintjes 2001.) Kelvinator recruited new employees and by the end of 1998 its headcount was almost 1,300 800 in direct production 840 of which belonged to NUMSA. In exchange for Kochs promise to save the plant, the union had agreed to vast concessions, including the layoff of 400 older employees, a no-strike pledge, and a two-year freeze at the industrys minimum rate of the wage increases of workers hired at the lowest grade. In practice, these newer workers would receive 25 percent less than founder employees in the same jobs. By de facto exempting Kelvinator from the engineering national wage agreement, NUMSA accepted to have its lowskill members in the plant split into two wage brackets, a potentially explosive situation. New hires were largely youth, mostly without previous work experiences, and they immediately disliked the deal, accusing the founders of having selfishly sold them out without consultation. Confronted to a union they considered insensitive, they resorted to tactics of refusal of work; theft, sabotage, and absenteeism became rapidly widespread. Moreover the companys expectations for postapartheid consumerist modernity were, similarly to the governments vision of progress driven by electrification, overly optimistic. Township residents were often too poor to pay for the rising rates of corporatized utilities and faced mass cutoffs as a result. Finally, trade liberalization and cheaper imports adversely hit the market. A competitor imported fridges from Swaziland, where hourly wages were allegedly one third of Kelvinators,26 and unsold stock piled up at the Alrode plant. By the end of 1998 the company was in stormy waters and looking for rescue (Barchiesi 1999.) Potential investors demanded an extension of differential 276

wages for founders and new workers, which NUMSA accepted, precipitating a total breakdown in relations between the two groups. Many founders also felt betrayed because Kelvinators crisis had denied their shares dividends, and they joined the newer workers militant demand for wage parity. In the end, NUMSAs shop stewards split: some accused the company to renege on its promises; others blamed Kelvinators problems on the undisciplined radicalism of younger workers. The former prevailed and, as the company started a fresh round of layoffs in mid-1999, NUMSA pulled out of negotiations.27 In September Kelvinator announced, without notifying NUMSA, its application for provisional liquidation. As a result, all 1,200 employees lost their jobs. When I visited the factory, NUMSA was trying to talk potential investors into a bid to save Kelvinator. Every Thursday, workers gathered in the courtyard, which they renamed Freedom Square, to hear reports from the negotiators. At the gates there were permanent pickets, often joined by wives and children. The workers desperate refusal to disappear stared directly at their broken dreams of an independent life: Before I used to say that if the company survives I will survive. Now that the company is dead I must see where to go. Its not easy to say that I am going to go back home and start to be dependent again on my parents, its not something I can do anymore. But, as it is now, I havent paid my rent for two months, and the guy is starting to get pissed off and said I cant stay there for longer than two mont hs. I am in deep problems in this moment, unless the company, of course, survives. Otherwise, I can go to the nearest lake and drown my problems. Q: Dont say that! [Everyone in the room laughs] 277

A: Just joking . . . but Ill have to go back to my parents and start depending again with my child . . . how can I do it? (Respondent 53) Prospects of falling back on the family are generally considered a mark of failure and humiliation, but they do not cherish memories of working at Kelvinator. The companys final years are rather described as a long, agonizing decay of erstwhile promises of empowerment, partnership, and decent jobs: I cant think of other things, I no longer have visions, plans and what-what. The treatment I received inside the company, I wonder whether we can have a sticker here where, apart from our value, it is written respect, honour and trust, that we respect one another whether you are a white man or a black man, and if we see one another so that you can accommodate me and appreciate me. I came to work here mainly because of the financial pressures that I had at home, but I was not feeling good, coming to work in this company meant dying mentally, because even our bosses, managers and foremen, they were not treating us well. Q: Would you say there was racism here? A: Yes, there is something like that inside the company. You feel very low, down, useless, hopeless, discouraged. When I had to attend to such and such a problem they shouted at me: Why? Why? We dont have people here! Production is slow! (Respondent 52) Eventually, South Africas leading company in the sector, Defy, bought Kelvinators assets and, in 2000, relocated them from the Alrode site to its existing plant in Ezakheni, a distant, semirural area near Ladysmith. Only 250 workers were rehired as rampant unemployment allowed Defy to slash one third of the old Kelvinator wage rates, which helped double the companys 278

operating profits and boost its market shares (Bezuidenhout 2004b.) The Ezakheni industrial park was established in the 1980s under the apartheid governments industrial decentralization program, which attracted businesses to low-wage African homelands. An old FOSATU stronghold, Ezakheni received from the ANC-run postapartheid local government incentives to lure Chinese capital. The new municipal administrators tried to sweeten the deal by warning COSATU to abandon radical mobilizations not to scare potential investors. Massive poverty, joblessness, and the cost of privatized utilities underpinned the residents cries for job creation and COSATUs local leadership ultimately chose to be loyal to the ANC and rein in rank -andfile militancy (Hart 2002: 269-87.) What was left of Kelvinators legacy of labor organizing mostly survived as unprotected employment and poverty wages. The Kelvinator drama was a further setback for workers hopes to benefit from a managerial discourse of company patriotism and corporate citizenship. The psychic shockwaves from the companys collapse were magnified among its employees by the faith they had put in Simon (Koch,) as many referred to him, the CEO as savior, a messianic visionary leader, or simply a good guy who got bad luck even if while it lasted everything was going first class (Respondent 55.) The managers explanations for the fall of the Alrode plant, conversely, echoed a governmental imagery sprouting from unreconstructed colonial roots. In their side of the story responsible and mature employees, available to cooperate for the common good, confronted unruly young troublemakers, inadequately socialized to the companys ethos and prone to wreak havoc in the shopfloor as they were doing in the townships. One executive said the new recruits were disruptive elements, not part of Kelvinator culture, who radicalized the union into causing divisions and conflicts that marginalized old guys, who were instead committed towards the company.28 Koch went further to blame a young bunch of guys, . . . disruptive 279

workers with nothing to lose. He said it was a mistake to hire black youth made restless to authority and discipline by their education: I wanted to give them a chance, but labor-intensive production was not suited for them.29 Rather than advancing a coherent counter-narrative, however, the union broke down. The laceration in labor solidarity over the meanings and advantages of workplace collaboration revived the old wounds of the MAWU split, when young workers also questioned the abstraction of production-based identities. The factory and workplace organizing thus simultaneously declined as signifiers of socioeconomic emancipation. Some younger workers who did not experience the heroic days of antiapartheid trade unionism and for whom entering the labor market usually meant a choice between unemployment and precarious jobs had distinctly disenchanted commentaries on the progressive aura of community with which management and the union surrounded the workplace: Sometimes [the factory] felt like a torture chamber, it felt very hard to report for work. I felt like, why am I going to that place? That place is boring me, I dont feel that place like home. . . . And he [Koch] was calling us a family, we are a family here! We are a family! Yes, a big, unhappy family (Respondent 51.) Municipal workers in Johannesburg, to which the next section will turn, also suffered the impact of the postapartheid transformation of work. Compared to the East Rands industrial proletariat, however, their specific location in governmental procedures and social reproduction would differently articulate workplace and community struggles.

280

New Canaan, New Egypt. Workplace, Community, and Identity among Johannesburg Municipal Workers

Visiting the waste management and roads services affected by the restructuring of the Johannesburg municipality gave me contrasting impressions. Local managers and technicians credited the city council with a modern and rational vision as seamless as the functionalist, unremarkable apartheid-age office buildings where I interviewed them. They glowingly pointed at walls bulging with colorful pie charts, diagrams, and projected efficiency gains, which revived the otherwise bare furniture. The smooth, alluring jargon of postapartheid public management flourishing with terms like streamlining, ring fencing, rightsizing, and income generation punctuated the optimism these administrators bestowed upon market forces after the lean years of austerity and budget cuts. They seemed to embody the municipalitys promise that restructuring would be driven by farsighted technocrats, not deferent to political pressure or mired in the lack of capacity and expertise within council structures (GJMC 1999a: 24.) Yet many of them, still mostly white in what used to be a reservoir of Afrikaner protected employment, belonged to the staff union IMATU, at the time allied to SAMWU in opposing the iGoli 2002 plan. Moreover, the closer I got to the frontline of restructuring, the roads and waste depots, the more ambiguous managements views became. Despite the councils promise of decent work in a world class city, some maintained a distinctly muscular anti-union language and most frankly admitted that corporatized utilities would be an opportunity to enforce employment flexibility, subcontracting, and fixed-term jobs. In the words of an executive manager, technical know-how, not the meddling of politicians and unionists, would take care of a key lack of work ethic 281

breeding people who are skimming from their backsides in the course of the day.30 Another executive, in the waste department, eyed the opportunity to get rid of expensive provisions on job security, healthcare, retirement benefits, and the statutory forty-hour working week, which he defined the biggest threat to our survival. Anticipating my question on possible union opposition to longer hours and lower remunerations he added: There are lots of people out there who need a job.31 Line managers were equally explicit. An official in the Central roads depot, in the inner city neighborhood of Benrose, was candidly enthused that corporatization would spell the end of union-protected municipal work, which in his view mostly benefitted the sick, lame, and lazy.32 The managers general view finally belied the municipalitys claim that fiscal austerity before iGoli 2002 was an exceptional emergency measure and rather saw it as establishing a trend to reverse workers power. As in private companies, the precariousness of employment, actual or threatened, was a weapon to reestablish authority in the workplace. In daily interactions with their employees, however, depot officials tried to allay their concerns, not without robust doses of paternalism and condescension or winking sexual allusions. The council hired, for example, a touring theater troupe to perform a roadshow, titled iGirlie 2002, which was playing for the workers at the Central roads depot when I visited it. Actors performing clichd stakeholders the socially sensitive male council bureaucrat, the cynical female street-sweeper eventually won to the cause, the doubtful but well-intentioned employees enacted the message that nothing had to be feared from transformation with lines like a computer cant clean the roads. SAMWU was caught unprepared as the employers had announced the performance as dealing with HIV-AIDS, but its protests were fruitless. The workers did not miss, however, the contradictions between iGoli 2002s stated rationale and the pronouncements of those tasked with implementing it, which only increased their already 282

considerable uncertainties. The prevailing feelings among the rank and file were fear of losing benefits and protections and anger at not being consulted. Distrust of the councils reassurance to maintain job security and employment was widespread and SAMWU was not spared as workers criticized its hesitations. Johannesburgs waste management is a prism that reflects social inequality into the geographical unevenness of consumption and amplifies it in the lack of uniform services across rich and poor neighborhoods. In white suburbs, neatly aligned garbage bins are regularly collected at curbsides; in historically black townships refuse is often disposed in gigantic open collective dumpsters, which regularly overflow and crawl with rodents (Barchiesi 2001.) Coping with an erratic service, residents illegally burn their waste, hence the acrid smell that is characteristically among the first sensory impacts on the casual visitor. The corporatized waste utility, Pikitup, started operations in January 2001 with the mandate to fix massive inequalities and restore financial viability. The white flight to the suburbs, which intensified after the desegregation of the inner city in the mid-1980s, had boosted the costs of the service and contributed to the councils budget problems. The iGoli 2030 plan aimed to redensify the inner city, but the suburban sprawl gained vigor as new black elites joined the old white ones in barricading behind fenced-off developments and boom gates that, with the councils authorization, closed public roads to traffic. Meanwhile, migrations from the countryside and the inadequacies of the housing subsidy program facilitated population growth in Johannesburgs seventy-three officially recorded informal settlements. The commercial waste from shops and restaurants in wealthy neighborhoods provide a vital source of income, so the budget cuts that preceded iGoli 2002 mostly affected poor townships.33 Services for low-income households declined even if, with only one third the 283

average waste of rich ones, they contributed twice as much to municipal revenues (Greater Johannesburg Metropolitan Council 1999a: 15.) In practice, poor communities financing the extension of refuse collection to the suburban sprawl were requested sacrifices to mend the fiscal problems caused by the sprawl itself. As an independent self-financing utility, Pikitup depends on user payments; iGoli 2002 terminated municipal cross-subsidies that once covered up to 40 percent of the waste management budget. As nonpayment is already rife in the townships, Pikitups profitability relies on compressing labor costs. Despite plans to recruit casual workers into permanent positions, the company continued to use various forms of contingent employment. Unskilled tasks like street cleaning or services in informal settlements were transferred to community contracting operations run by local businesses, usually connected to political notables, and paid by the council, usually with additional funding from private financial institutions, to hire unemployed residents. Strongly encouraged by most metropolitan municipalities, community contracting exemplifies the ANCs emphasis on the entrepreneurialism of the poor as an alternative to welfare dependency. It is also a source of patronage and legitimation for politicians and councilors in otherwise explosive social realities. SAMWU, conversely, has maintained a long-standing opposition to community contracting, regarding it as a way to undercut municipal workers wages through competition from nonunion casual labor. The waste management workforce is uniquely stratified by gender, age, employment contracts, and residence. In the townships the word amaBhaka, from the people where Johannesburg used to recruit its refuse collectors, has for long denoted a lowly social status (Samson 2004.) When the municipality ceased hiring migrant workers in the 1980s, African women took over their jobs, whose stigma assumed therefore a gendered dimension. Such 284

powerful symbolic associations reinforced the numerical preponderance of female workers in low-wage, insecure occupations, where male union members often share the employers stereotypes of street cleaning as an extension of putatively feminine domestic tasks. Many older migrants, however, are still employed in the service and live in municipal hostels. They can compare, at a biographical level, their modern, market-driven precariousness to the previous racial despotism, a narrative connection not so transparent to the many young casual workers hired during the postapartheid fiscal austerity. The council traditionally divided casual workers into two groups. The casual-casual gather every morning at the depots gates looking for a days work without benefits; management discretionally hires them without consulting the union or trucks them where workforce is short. The permanent casuals are those iGoli 2002 aimed to absorb into the permanent payroll; they are continuously employed but have no contract and can be hired for as short as one day per week. When the city froze new hires, as many as three quarters of a depots employees could be casuals with wages ranging from R 35 to R 45 per day, compared to R 60 for regular council workers.34 Jobs have become increasingly volatile, however, also among traditionally male, unionized municipal workers in relatively skilled occupations. The Johannesburg Roads Agency (JRA) started operations in January 2001 as an independent business unit replacing the old municipal roads department (Greater Johannesburg Metropolitan Council 2001.) In the years of austerity that preceded iGoli 2002, the capital budget for roads and stormwater drainage was basically erased,35 while the councils hiring freeze drove vacancies to 47 percent of the theoretical fully manned headcount (Greater Johannesburg Metropolitan Council 1999c: 17.) Once again, poor black townships, which already received less than one fifth of the overall budget, suffered the most as their roads remained unpaved and without drainage (Greater 285

Johannesburg Metropolitan Council 2001.) Otherwise, the path to the JRA was littered with the remains of a once proudly organized working class. Budget cuts meant that the few road repairs and new constructions still taking place were left to private contractors bringing their own trucks and equipment. Contracts could, however, be very expensive, thus costs were significantly reduced by hiring temporary workers to replace vacant municipal positions on augmented work schedules and with fewer machines and vehicles (Greater Johannesburg Metropolitan Council 1999c: 18-19.) As a result unfinished jobs and unrepaired damages kept piling up. The municipality defines roadworks as a non-income generating service, therefore, even if the new JRA can still count on council funds, its finances critically depend on cutting wage costs, which absorbed one quarter of the pre-iGoli 2002 operating budget. The agencys officials abundantly borrowed from the governments anti-entitlement entrepreneurial rhetoric when they argued that a profit-seeking work ethic will be easier to establish in a private company (Greater Johannesburg Metropolitan Council 2000,) and the JRAs style was unabashedly corporate and managerialist. In the end, even if they were presented as ad-hoc exceptional measures, the budget cuts of iGoli 2002 allowed the council to break SAMWUs permanent employment strongholds and hand to the new roads agency the fragmented, more compliant workforce it needed. Compared to the East Rands industrial workers, the lives of municipal employees are much more steeped in migrancy. Sixty percent of my respondents had moved to Gauteng from rural areas between 1968 and 1988. The older ones were mostly from the traditional reservoir of Transkei but the younger came from all over the former homelands. Almost all had their first jobs in local government. Seniority and continuity of employment facilitated strong interpersonal ties and deep shared memories (see table 5.1.) [TABLE 5.1 HERE] 286

Municipal workers attachment to their jobs is explained by the relatively good benefits healthcare, retirement, housing schemes, employers loans accompanying secure employment contracts. By eroding such conditions and undermining the union, iGoli 2002 disrupted one of the social locales that, in the postapartheid context, most identified work with stability. Market competition is less obvious a pressure for public workers than private ones, and liberalization did not gradually intrude in municipal employment in the form of inexorable, objective market imperatives. Rather, municipal workers understood iGoli 2002 as a traumatic, politically motivated break. They refused to see market rationality as self-propelled and selfexplanatory as they could regard it as the product of specific, intentional, unilateral institutional interventions. They had, so to speak, in sight the enemy that for the East Rand workers remained elusively behind the scenes. Corporate patriotism had no purchase in the waste and roads depots, and relations between employers and employees were starkly confrontational. Sure enough, a major casualty of this passage was the union, which once delivered not only material gains to its members but the meaning and pride of serving communities as full citizens and human beings. Workers responses to iGoli 2002 had, however, much less of the fatalism that I often detected in the East Rand factories; they rather compensated for the difficulties of collective organization by mobilizing a vibrant moral repertoire that positioned their claims within legitimate community needs. Johannesburg city workers represented the downgrading of their employment conditions not only as a blow for their families but as an unfair inhibition to their capacity to improve their neighborhoods: As workers our primary objective is to serve the community but the second primary objective of workers is to upgrade the standard of living of every individual. Rather than upgrading the standard of living of every individual we 287

are receiving the contrary status as workers. Now generally we are saying the performance with regard to the current government compared to the previous deteriorated. We were expecting under the new dispensation good quality performance in terms of serving communities but we are not getting that (Respondent 103.) Injuries to the material and symbolic connections between work and community reopened past wounds, which became tools to decode the present. Memories of the apartheid workplace are layered and ambiguous; their inhabitants are surely the baas boys, the impimpis (informers,) and foremen arbitrarily dismissing workers. There was also a union, however, to counter unfairness and abuse: The job now is heavy, more than before, since we have got this new city council. Before, in the apartheid system, it was heavy but not like now. The old city council was not right because of apartheid, but we were working alright. . . . Before we got training and with this new council we havent got training (Respondent 109.) Older workers evoked apartheid as an ambivalent before, a dark age but also a time of realization of a collective power that is now increasingly perceived with nostalgia. In contrast to industrial workers, who did not suffer the horrors of migrant labor and life in the compound to the same extent, municipal workers paradoxically had a much less grievous sense of how things used to be. Expressions like better, more manageable, or acceptable kept popping up in my interviews to describe how union struggles visibly modified working under apartheid. Thus one worker from uMzimkhulu (Transkei,) who spent four decades employed by the Johannesburg municipality while living in a dorm-like compound room shared with nine other 288

people, could tell me that he still preferred the way it was then. Compared to the evils they knew, iGoli 2002 was, in the words of a shop steward, abnormal, unusual, totally unacceptable (Respondent 87.) It announced a return to past domination but with none of the reassurances of unionization; a travel towards uncharted, menacing waters, rather than just back in time. It destabilized a world of work that workers eventually came to imagine as regulated by steady boundaries and predictable power dynamics. With fewer permanent employees, workloads increased and tasks became blurred, while money for functioning tools and safety overalls ran short. Even younger workers grown up in Johannesburg and not employed during the old regime internalized the historical temporality of their comrades. An African female street cleaner at her first job in the Avalon (Soweto) depot almost idealized the apartheid age as it receded in her direct memories: Now its worse even than when the Boers were sitting on our neck, worse than under apartheid. It was better before. . . . The people in the apartheid system were coming to the people, never mind they were doing wrong, but they were coming to talk to the people (Respondent 89.) Unsettling as conjuring up the ghosts of apartheid to exorcise the fears of iGoli 2002 may be, respondents were not merely comparing past and present. They rather selectively enacte d class historical memory into ideal normative schemes (Bauman 1982: 115.) The before they missed was not a chronological category but metonymically signified a certain balance of workplace power now under threat. The managers refusal to come and talk to the people was the ultimate insult because it combined an obsession with financial results with indifference to the workers humanity, which democracy promised to value. A waste worker used a powerful

289

metaphor to describe democratization as a liberating exodus halted and reversed by the precariousness of work: We are saying we are going back. Now we can say we voted for our government and we thought we were going to live better but because it is our government sitting there on top, they are making us suffer. . . . It is like coming from Egypt and now we are going back to Egypt. The old government was Egypt and we thought we were going to Canaan, but instead with this new [waste] utility we are going back to Egypt (Respondent 86.)

We Feel Sort of Redundant. Surviving the Flexible Workplace

Municipal and manufacturing workers alike saw the corporate pursuit of flexibility as an unfairly tipped scale: Before when we were working harder we were doing money, but now we are working faster and we are not doing fuck-all, because they are telling us that even if we are working harder the company is not making money! It is like we are going back to the dark years, where we used to have slaves; people are being enslaved in this company because of restructuring (Respondent 181.) Road workers take, within municipal personnel, particular pride in their skills, a feeling infused with the self-consciously masculine image of bending the natural environment through prowess and physical strength. Precariousness is here about material insecurity as much as the lingering feeling of becoming disposable:

290

We used to construct [roads] before, but now we are simply doing repairs, jobs have been contracted and since 1993 the construction of roads has been contracted out, that was the directive. When we were constructing we were sure that we were working but since then the capability of constructing roads has dropped and we feel sort of redundant, you know? We feel that we might lose our job because we are only doing maintenance but when we were doing constructions previously we had by all means the feeling that our job was secure. Since we are doing maintenance and cleaning there is a likelihood that we might be kicked out because we might be perceived as if we are idling, we are not working, and that is already happening: councilors are claiming that municipal workers are idling (Respondent 116.) It was not, however, abstract labor as such that became redundant, but workers accustomed to claim rights, stable income, and social provisions. Around them, the commodification of life and the loopholes of social policies have reproduced an impoverished semi-employed multitude with little or no social support and in great need of monetary resources. In the 1990s Consol Glass shed half its workers through outsourcing, especially by contracting its recycling operations to micro-enterprises or informal businesses. CEPPWAWU shop stewards, who had no access to these plants, told me they hired casual workers, rumored to be undocumented migrants from Zimbabwe and Mozambique, on one- to three-week contracts at half of Consols wages. Informal glass recycling also takes place in myriad sites across Gautengs townships and squatter settlements. In 2007 I visited one of them in a particularly poor section of Orange Farm. Described as a community project, it was supported, I thought paradoxically, by a local social movement affiliated with the Anti-Privatization Forum. Local residents, including teenagers, 291

worked all day in the dust and filth smashing bottles, for which they were paid by weight, inside a vast dumpster; their only tools were hammers and they had no protective equipment apart from rubber gloves. Workers perceptions of the decay of stable employment connected technical and organizational dimensions of workplace change. In the paper industry, brutal downsizing and the competitive outsourcing of operations deemed non core businesses eviscerated the unions in large companies.36 Layoffs were not only motivated by economic difficulties but were often integral to a flexibilization of the production process predicated on the weakening of workers solidarity. The Wadeville plant of Nampak Corrugated slashed its headcount from more than 800 to 260 between 1995 and 2000, mostly through new technology and partially through outsourcing, as production almost doubled thanks to multiskilling and work intensification for the remaining workforce.37 The plant chose to focus on manufacturing corrugated sheets and outsourced its non-core box-making operations to small companies mostly hiring temporary workers. In practice, such factories replaced Nampak Wadevilles unionized box makers using sheets Nampak itself sold them. Corporate restructuring, therefore, produced market competition rather than merely adjusting to it. Indicatively, after 1994 the Nampak group jettisoned its historical opposition to organized labor and joined the emerging mainstream strategy of union incorporation through corporate patriotism. Following the postfordist managerial orthodoxy of the time, it introduced just-in-time and total quality management with the stated aim to overcome adversarialism and the practical effect of softening labor opposition to impending restructuring (Sociology of Work Programme 1994.) By 1998 PPWAWU, as the majority union, was brought to its knees by membership losses and eventually signed a world class initiatives agreement committing itself 292

to global competitive benchmarks. The decline of labor radicalism compounded a growing loss of sense and identity among workers who were now turning towards a largely survivalist imagery: When workers on the floor every time see these retrenchments, there is no longer that feeling that when you were going to work you knew what you were doing. Now most workers just come to work for the sake of getting their pay because they know that they can be retrenched any time. Before there was a motivation to go to work, a pride, but now most workers are not happy, they just work for sake of the money (Respondent 170.) As customary coordinates of negotiation and conflict faded, and a permanent feeling of being disposable and redundant crept in, workers responded with a mixed of fatalism and rationalization, which, as in the case of this self-defined Africanist operator at Nampak, did not spare erstwhile militant imageries: If one is asking that question [how has the company changed over the past five years?,] one must bear in mind the situation of the country at that time compared to now. . . . We must take political considerations into question. Before we were not competing globally, now we are competing globally, there are more companies coming in and we have more competitors now, so machines have taken some of our jobs (Interview 179.) Feelings of being, to recall Guattaris expression, systematically endangered can make employees comply with corporate dictates for work intensification, which determine their own redundancy. For fear of being retrenched many Nampak workers budged over the companys requests for double shifts, or sixteen consecutive hours at work: 293

If a manager comes and says, Hey, you must work double shift, you cant say no, because they are scared and managers are taking advantage of that: if you dont want to work, the retrenchment is coming. They dont say that openly, but you can hear the tones, you know, even if they dont say it directly (Interview 170.) The threat of redundancies enforced all sorts of flexible working time arrangements in the companies I researched. In the glass factories, which usually work as continuous three- or fourshift operations, employees had different weekly schedules, alternating day and night shifts with very limited breaks, and they could be called in the middle of the night to cover for colleagues who could not keep up. Exhausting flexible time arrangements carry however, advantages for specific groups of workers, which further weaken collective solidarity. MB Glass, for example, had a quick change system, which required all employees to change shift two or three times a week, so that they worked three weekends a month, but also got two consecutive days off. The mechanism suited the many migrants customarily hired from KwaZulu-Natal, as it allowed them to visit their families every week, but was intensely disliked by local workers, who had to sacrifice most weekends and their regular family lives. The precarization of work would not probably elicit the employees acquiescence if, apart from causing insecurity, it did not somehow selectively respond to the needs of an increasingly fragmented workforce. Many artisans at UCW, for example, preferred working for the labor broker FEDMO, which did not exact healthcare contributions, therefore paying higher wages, and provided better access to paid overtime, for which regular UCW employees were subject to stricter and more expensive union rules. Workers families, on the other hand, may need cash

294

more than healthcare as debts to repay or prepaid water and electricity devices to recharge are more pressing than unforeseen future emergencies: You see, if you are unmarried and are not a husband your problems would be less than mine, especially on the finance. For those years when I was a permanent, I really didnt benefit anything, but I started to benefit when I started to become a contract, then I saw an improvement in my life, at home and in my life. I could save something like R500 a month while I am giving my wife enough money to support herself and her children, I could pay an equal share of the telephone bill and the electricity bill, and we still suffer. But before, when I was a permanent, I used to suffer more, I had lots of debts behind my back, I couldnt even afford to pay everyday expenses with the money I used to have in my pocket (Respondent 45.) Labor brokerage, therefore, can give employees the impression of being not only factors of production but sovereign customers who actually purchase from an employment agency a job placement tailored to their needs. As with the marketization of municipal utilities, the end of stable employment produces certain images of individual empowerment that perform the governmental task of displacing, defusing, and disguising the underlying social processes. At the same time, to maintain their customer relations with labor brokers, workers become more disciplined producers. By tying employees to a specific company, which can call them at any time, brokerage eliminates the indeterminacy of labor mobility as workers can no longer use it to control their work effort. As the companys workforce becomes increasingly divided, regular UCW employees deploy the necessities of their own families to build a boundary discourse

295

and an ethics of responsibility and commitment (Lamont 2002) that juxtapose the genuine concerns of full-time workers to the selfish instrumentalism of the ones hired through the broker: I am permanent and I care for my house, my children, everything. These guys [FEDMO workers] come only to get the money and they are told theyre going to stay for two months, after which they must go. But I am permanent, I am working here more than two months, and those guys dont care about what is going on here in the factory, theyre only here to take their pay and they cannot look three-four years from now, they just say ek weet nie I dont know but here I have to do my best for the future of the factory, because I worry about my children (Interview 27.) But, as much as they subdue and divide workers, corporate strategies for employment flexibility face a singular contradiction as well. To reassert managerial control they must seek a degree of legitimacy by taking workers needs to some extent into account. They become therefore susceptible to unpredictable, potentially unstable expectations and claims outside wage labor. The rules of flexible production are not self-fulfilling but remain the object of contested interpretations (Clegg 1994.) Even when the decline of organized resistance allows the employers to outflank workers meanings and cajole them into fatalistic acceptance, work ers do not lose the capacity to signify the conditions under which they produce. The precariousness management utilizes to reinforce its control can, therefore, come back to haunt it as a condition generating social demands. There are therefore limitations, often not clearly discernible, to the extent threats of redundancy can govern the subjectivity of the employees. Johannesburg municipal workers were particularly apt in mobilizing social needs and community justice to boost claims to workplace fairness. A road worker commented: 296

We are not self-centered in terms of thinking about our benefits in the workplace. We are actually saying that we are workers and we are also part of the community, because for eight hours we work here but at the end of the day we will be faced with a similar monster which is iGoli 2002 also as communities (Respondent 101.) By rejecting the managerial rationale for casualizing and intensifying work, employees enacted the iGoli 2002 saga as a battle between contesting modalities of knowledge. Waste workers argued that the hiring freeze led to chronic labor shortages, which they had to cover by performing multiple tasks, resulting in poorly done, unfinished jobs. They blamed managerial incompetence for impeding optimal performance, therefore failing the community: If we go out at 9:00 we cant finish the streets and we have to make up the following days. And all those bloody jacarandas! When I go to sleep I am dead. . . . Now the rubbish coming on Monday stays on the streets until Friday so the area stays dirty and the residents complain, and its residents who pay me and they should be happy (Respondent 69.) SAMWU members pointed at the possibility of being replaced by inexperienced casual and contract workers as a sign that the city council no longer cared for servicing the poor and therefore betrayed the transformative obligations dictated by the new democracy. A road worker commented: Contracts are doing their jobs in halves and then they leave, we as maintenance people who are running Soweto are supposed to finish the job they started, repair the damage they have done, and those people just go with the money, money of ours as Sowetans, money of ours as workers. . . . At the same time we have people 297

with experience who can do construction jobs and we cant do that for lack of funds and instead they have to finish up the job of contractors that are moreover wasting councils money (Respondent 108.) The precarization of work is not only a productive process; it also involves a struggle to signify human activity in relation to ideas of community and the body politic. It interrogates the imaginative project of the working class as a collective political agency while it reorganizes the connections between production and society as a space inhabited by ambivalent subjectivities combining survival and adaptation with unpredictable needs, strategies, and movements. By harnessing the social as a force of production, the flexible workplace also becomes vulnerable to the disruptions of the social. Johannesburg waste workers are a case in point: they have always been used to the large scale recruitment of contingent labor; yet growing casualization in the wake of iGoli 2002 has proven more contentious for them than for road workers. In some waste depots, even physical clashes between full-time and casual workers have been reported. A workplace-centered perspective would probably find such discrepancies puzzling if it neglects to look at the recruitment of casuals as a socially negotiated and contested process. Before iGoli 2002, customary agreements and implicit norms between the union and depot managers regulated how fixed-term workers were hired; lotteries at the depots gate were a common procedure. The unions intended such rules as a fair balance between the employment needs of jobless residents and their members concern that casual labor was not used to undermine their status. Under iGoli 2002, instead, managers have unilaterally reclaimed the authority to deploy the workforce and ferry truckloads of casual workers between depots, which reminded older employees of the same practice under apartheid-age black municipalities:

298

The casuals, I cant say, when we see these people coming to the gate nobody can say how these people are hired and we have to keep quiet because we dont know, and they say these people just want to work. . . . Before people were hired in the depots and it was a, how you call it, zama-zama. Q: A lottery? A: Yes, but that was stopped. Now the council just brings people in but we dont know how they are hired. Last month we confronted them because this is still happening, trucks of people are downloaded in front of the gate and we dont know where they come from. And we told SAMWU that we dont want these people, we want people hired at the same spot and in the right manner. We have brothers who stay at home and arent working and then the council hires these people (Respondent 91.) Municipal workers perceived that their responses had to transcend the boundaries of the workplace in order to contest the employers capacity to reshape the articulation of production and society. To oppose the city councils unilateralism they tried to reconfigure their jobs as affective labor (Hardt 1999,) a sensible activity of care for local communities. The communities workers appealed to are, however, also complex and contradictory realities that cannot be easily idealized. When municipal workers evoke residents needs to signify their workplace predicament, they encounter fresh divisions and conflicts as they realize that community interactions can be way messier than their normative representations. Residents complaining about poor road repairs and refuse collection may target municipal employees rather than the local government. Waste workers constantly fight with pickers mining collection and dumping sites for materials to recycle; in the year before I interviewed them, three respondents 299

were shot or stabbed in confrontations of this nature. The Avalon waste depot covers much of the southern section of Soweto, where a 2007 survey indicated that two thirds of residents earned no regular income (Frye 2007: 22-23.) Part of the area was the sprawling shantytown of Freedom Park, along the Golden Highway between Johannesburg and Soweto. According to SAMWU representatives, residents expected to be hired by local politicians in refuse disposal projects and chased the municipal waste collectors when they showed up. Articulating a shared language for labor and community claims proved an arduous task: It is a bad situation, when we go to Freedom Park we dont know whether well come back, and I blame the councilors there, the council must speak to these people and tell them that is us who are cleaning the place but the council doesnt talk to those residents. The council, when they call those people, they promise them jobs to clean the place, so when they see us they fight with us because they were promised jobs by the councilors. They want the vote so they promised them jobs. They tell them, vote for me and I will do something for you tomorrow, then they vote for them. . . . The councilors, they are gambling with our lives, so we have to run away (Interview 97.) In conclusion, the social dynamics of insecure work can destabilize the plans of employers trying to use insecurity as an asset, but unions and regular workers are not immune to such disturbances. As wage labor and workplace interactions are often no longer conducive to emancipative imagination, increasingly precarious workers elaborate alternative discourse to articulate claims beyond production. The community is, however, an evanescent social space where the collapse of wage labor amplifies continuous conflicts over scarce income sources, which the very politicians who praise the entrepreneurial opportunities of informality foment 300

through networks of favor and patronage. It is no surprise then, as the next section will discuss, that many workers prefer individual escape from waged work to the elusive, difficult quest for new languages of solidarity.

Entrepreneurs of the Self. Individual Strategies and Life after Waged Employment

South African labor scholars have often tended to look at the workplace as a self-contained space, where dynamics of domination, resistance, adaptation, and negotiation are insulated from the meanings workers attach to their jobs or the influences of wider social interactions, needs, and affects. Industrial relations, instead, allegedly provide workers with the template to improve their productive and, by extension, social condition. When the employers unilateral prerogative held sway in the aftermath of apartheid, various sociologists argued that workers cooperation was indeed the way to claim new legal protections and engage corporate strategies (Von Holdt 2000; Hirschsohn, Godfrey, and Maree 2000.) Judith Hayem (2008) produced a voluminous ethnographic survey of the postapartheid workplace, which she depicts as a place of opportunity and a place of national feelings for black workers because their democratic rights as producers give substance to newly found citizenship. Many of the employees she interviewed reported that under apartheid they were treated like slaves, an image that some of my respondents actually applied to their postapartheid predicament as well. With political liberation, however, Hayem continues, black workers are fully recognized as builders of the national economy while precariousness, despotism, and unfairness disappear from their lives. They can therefore replace old-fashioned adversarialism moved by pass class consciousness with a cooperative leap towards 301

reconstruction and a freshly discovered productivist unanimity. The only problem is for her that management often maintains, irrationally and against all odds, an engrained authoritarian mindset that denotes a lack of interest in the benefits of the humane, felicitous, and productive workplace that democracy promises. By Hayems own admission, however, her analysis eschews as irrelevant any investigation of workers meanings of work that emerge from broader social interactions. She cannot therefore realize that her view of productivist cooperation as the culmination of workers democratic citizenship is self-fulfilling because it is premised on the assumption that the workplace is the only context of meaningful worker agency and the only source of entirely transparent and self-explanatory meanings. Analyses centered on production dynamics are always at risk of essentializing and naturalizing the workplace as the obviously primary social locale where workers express and enact desire. Social sciences and social policies converge in this case around a normative view of the employment relation as the site where workers liberate themselves and deliver progress to the whole nation. In the process, few important questions are obscured: is the workplace actually so central to worker subjectivity, after all? Are attempts to engage their employers the pinnacle of workers hopes, or are they just instrumental components in a complex range of life strat egies? What if workers direct their desire not so much towards contesting restructuring, which they may just endure as a matter of economic compulsion, but to building a more meaningful life outside capitalist work discipline? In an increasingly unprotected postapartheid world of production, workers have tried to stay afloat in uncharted waters with a lot of apprehension and a robust dose of disenchanted opportunism, rather than confidently approaching productivist unanimity as Hayem would want it. Managerial unfairness and bad faith are not the only problems; the main reason why wage labor in South Africa cannot be Hayems place of national feelings is that its 302

degraded material conditions are excruciatingly at odds with its persistent idealization in governmental mythology. Binary conceptual oppositions between compliance and engagement, or resistance and adaptation are ultimately inadequate to grasp how workers reinscribe, to address employment decline, their bodies and personhood with multifarious capacities to act. While the next chapter will elaborate on collective activism and the political discourse of labor-community interactions, I want to conclude this one by discussing the imagination of individual escape as a response to the decline of waged employment. The fact that the workplace is a crucial source of income does not exclude that it can become a frail, peripheral foundation of meanings, sense, and prospects. It is, conversely, simplistic to equate cynicism and disaffection towards employment with abdication from citizenship; they can rather point at the need to flee, or at least alleviate, precariousness and subordination by diversifying existential options. Most East Rand respondents who expressed substantial resentment at their employment conditions and their underlying instability also perceived workplace-based struggles as increasingly isolated, fragmented, and ultimately self-defeating. They conversely fantasized of strategies to reduce dependence on wage labor based on personal initiative. Such feelings cannot be clearly likened to political or social criticism; entrepreneurialism plays, after all, a crucial role in the ANCs anti-welfare discourse and may facilitate workers apathy as the insecurities of restructuring can be perceived as gateways to alternative opportunities. Individual escape departs, nonetheless, from the image of wage labor as a necessary condition of solidarity and progress and rather reflects deep social trends that, in colonial and postcolonial Africa, posit economic independence and regular employment as antagonists. In governmental discourse entrepreneurship and waged employment are equally valuable antidotes to perverse entitlement 303

mentalities; workers ideas of individual self-valorization, however, understand economic initiative as a critique to working for wages. East Rand factory workers were much readier than Johannesburg council employees to contemplate the possibility of a business on their own to replace or complement their waged occupation: 19.3 percent of respondents in metal-engineering plants, 9.7 percent in glass and paper ones, but only 4 percent of municipal workers had a second job, almost always on an informal, self-employed basis. The particularly high figure for metalworkers reflected the collapse of Kelvinator and the high turnover at UCW. Second occupations included small mechanical or electrical repair works, street sales of food and clothes, and part-time construction. Their economic contribution was often vital; metalworkers earned from double jobs an average monthly R 588, or almost one third of their regular wage. Many East Rand workers without a parallel business considered starting one under favorable conditions, but they cited lack of capital, crime, and inadequate training as major impediments.38 Sometimes groups of laid-off workers, as at Kelvinator, talked about pooling their retirement payouts to start some new enterprise only to realize that their families needed the money to repay debts with retail shops and avoid poor credit, blacklisting, and the clutches of loan sharks. The prevailing view, in any case, did not idealize self-entrepreneurship, which respondents tended to consider a mode of coping with extreme duress, rather than an avenue to prosperity: The income for the workers is far too little to allow them to go out and buy what they may sell. On Friday they get their money, do grocery shopping for the family and on Sunday they are left with only five or ten Rand, only enough to pay to come back to work on Monday, and on Monday they are borrowing money from other people. Thats the life we are living in this company. 304

Q: And what about you? A: I dont know, its as good as I can get. Theres this Tata ma Chance lottery, and if I can get maybe a million, that would be good business in Lesotho. Q: And if you dont win? A: I will stay here because in Lesotho there are no jobs. Vosloorus too is a problem when you dont have money, ek se [I say,] so my intention is to have enough money to run a business, because around here, ek se, you are taking a risk if you open a business because you dont sleep when you have got a business; too much crime, especially in the township (Respondent 4.) Sometimes the need to leave a meaningless factory life is, nonetheless, stronger than risk, hardship, and possible dangers: If it wasnt for financial constraints I should have left this company long ago. As an alternative, I have got my industrial relations diploma, I am intending to take another course, there are possibilities in government posts but I must get my degree. But I dont think I will stay here. Q: Do you think your job is becoming insecure? A: No, no, my job is secure, its just a question of it being a shit job (Respondent 181.) In other cases, moonlighting is more clearly articulated as a claim for autonomy that subverts wage labor discipline. A wirer at UCW was also doing electrical repairs around Alberton during weekends, an activity he considered worth the risk of deserting and sabotaging his formal job: Sometimes I can take a day off here at work. Before there was no pay for sick leave, now its paid so I can just take one day and bring back a fake doctors note 305

to cover up. Many people do it, even if one must be careful, they notice when someone brings doctors notices three times a month. Q: So working at UCW for you is mainly something like A: To improve my life, but the main thing is that I can really stand alone. I can work alone, the electrician job is what I really care about (Respondent 39.) For other UCW workers, however, double employment could underlie much gloomier realities of cheap, unprotected labor with no semblance of social advancement. Many had seasonal jobs in the annual cleanup of the giant SASOL oil refinery in Secunda, south of Johannesburg. For a couple of weeks at the end of the year, local employees took leaves and SASOL hired temporary workers with little training and no benefits to scrape toxic and flammable chemicals off, a practice the unions have blamed for several deadly accidents.39 The shakiness of individual survivalism often contributes to corporate-driven informalization, but is not merely synonymous with hopelessness. Strategies of self-employment require the activation of social networks, life experiences, and cognitive abilities that are mostly invisible and marginalized in factory routine. A machine operator at MB Glass, for example, used to make clay pottery after working hours as he nurtured that art through the struggle (Respondent 211) while being trained in Zambia as an underground operative for Umkhonto we Sizwe (MK,) the old armed organization of the ANC. Another worker, a shop-steward and electrical fitter at UCW, grew disenchanted with the company after being repeatedly laid off and rehired on fixed-term contracts too short even to qualify for unemployment insurance. To compensate for the frustrations of wage earning he invested in the parallel, entirely self-taught career of unlicensed architect, as he called himself. For three hours a day after work he drafted, with the help of two hired assistants in a rental office in downtown Nigel, buildings, multi-storey 306

residences, extensions, and churches, sending them afterwards to the local building inspector for approval. Rather than a formal qualification he had just a talent for drawing, which dawned on him while making additions to the house of his mother and family of twelve (Respondent 21.) The UCW wage still provided most of his income, but in his mind it mattered less than being recognized by the community to which he proudly provided an affordable service. An important corollary here is that self-employed escape from waged work can in fact be framed in cooperative terms. It does not necessarily fit the construct of the individual shared, somehow paradoxically, by both neoliberals and left critics of the erosion of collective identities based on occupation and class. Axel Honneths (2004) picture, for example, of the atomized, purposeless emptiness that replaces the warmth and solidarity of the Fordist social contract and the values it supposedly enables is, in the end, as effective in silencing workers life strategies and signifying practices as the economic liberalization it deprecates. For South African workers, instead, images of the enterprising self inform potential connections to social collectivities as much as they reflect market-driven social disaggregation. With all their ambiguities, risks, and counterproductive effects, fantasies of selfemployment grasp the precariousness of wage labor both as an unstable material condition and as a repository of non-economic meanings. To infer political sense from individualized strategies is, however, far more problematic. My respondents critical commentaries on the wage relation do not necessarily underscore a progressive imagination, especially among manufacturing workers with weaker expectations of community solidarity than municipal ones. Idealizations of individual success can be remarkably conservative, especially when religious images of salvation discursively bankroll entrepreneurial options. Religion, as Michele Lamont (2002: 40) remarked offers an alternative to social status and is often defined in opposition to it and can explain the 307

loss of employment in terms other than personal failure. It may also configure, to fill wage labors significational void, self-employment as mundane transcendence enabled by personals virtues of perseverance, faith, and integrity. Spirituality is a central and still largely unstudied motivator of the social justice views of South African workers (Buhlungu 2004.) According to a survey of Gauteng employees conducted at the time of my research, 80 percent regularly participated in church activities, twice the figure for political associations and four times for social and cultural groups (Trres 2000: 415-19.) Most workers I interviewed belonged to established denominations Roman Catholic, Methodist, Anglican, Lutheran and African independent congregations like the Zionist Christian Church (ZCC.) They mostly emphasized reciprocal help and solidarity as remedies to social adversities, but followers of rapidly growing evangelical and apostolic churches almost a quarter in the survey of Gauteng workers (Trres 2000: 358-59) tended to embrace a more competitive ethics of success. As a discourse woven in strong union identities, this latter was not necessarily a version of the prosperity gospel proselytizing in many African countries, but rather reflected a homegrown charismatic Christian tradition grounding social justice into self-reliance (Thompson 1997.) To a self-defined born again Christian worker who had just lost his job at Kelvinator the fate of the company delivered a moral tale on the virtues of assertiveness and the evils of giving up: We have to believe we can do what we want with our lives. For example, in the parable of Lazarus, when Jesus raised Lazarus from the dead, he said before that: Take away the stone from the tomb of Lazarus. What does the stone mean? The stone means disbelief, that people dont believe they can do whatever they want with their lives, they dont believe they can have their own houses, their own car 308

and thats what we should do instead of relying on these whites to give us jobs (Interview 56.) Precisely because workers narratives of self-employment are bendable, multi-layered, and ambivalent they powerfully rebuff the self-validating certainties and the axiomatic assumptions underlying the nexus of work and social citizenship in state ideology and union discourse. Moreover, as a manifestation of labors crisis, self-employment belied the dreams of mass entrepreneurship in more recent versions of governmental imagination. In the mid-2000s, almost two thirds of self-employed workers lived on less than R 500 per month, and half were street vendors. As Neva Seidman-Makgetla argued, they did not resemble the craftsman in his workshop, who knows his place in society and is blessed by the middle-class virtues of hard work, sobriety and savings.40 While high-income freelancers are mostly white, for black workers self-employment is a short-term, high risk, low reward activity, a perpetually unfinished endeavor to escape precarious work where the community is an unstable locale of production and exchange. As Weiss (2004: 10) writes, African postfordism is more about out of luck than just-in-time as the future uncertainties of informal entrepreneurship are grounded in present assets that depend on social networks, chiefly the family, undermined by the same global dynamics that make waged work redundant in the first place. The discursive ambiguities of micro-entrepreneurship connect therefore to their material ones as they respond to the fragility of work only by reproducing and amplifying it. Even if it is not a mere capitulation to neoliberal market discipline, self-employment poses real challenges for trade unions already severely weakened by workplace restructuring and policy shifts within the ANC. Workers may volunteer for layoffs to collect retirement benefits with the plan to invest them in a combi (collective taxi) and start a transportation business rather 309

than listen to the shop stewards exhortations to stay in jobs perceived as worthless. The more the imagination of the unions is folded onto the workplace and isolated from members broader concerns, the stronger its alignment with managerial discipline becomes. Many shop stewards were indeed concerned that rank-and-file disenchantment could subvert established factory procedures and obligations on which the union relied. A NUMSA representative told me with great disquiet that workers can become intentionally undisciplined and disrespectful to be scheduled for layoffs: I discovered that people who dont want to work and have problems, account problems, problems with cash loans, and so on, they volunteer when they hear the company wants to retrench, they know that under packages theyll get lots of money to pay for their accounts. What the company is doing now is to say, We must work as a team and we have to explain people to learn not to come late, because people who come late enjoy coming late, sleeping at work and that is not the right thing. Even when we as shop stewards told them, No, you mustnt do this, lets wait what the company is going to do, and they say, No, the company told us they want to retrench, now you have to remind the company about that, and I say, No thats not my job, to remind the company to retrench people (Respondent 1.) As practices through which workers signify the interactions of work and life diverge from the normative meanings of organized labor, old identities forged in struggle no longer hold and the union comes, therefore, to mean different things to different employees. For some it is still a representative of collective claims to rights and power. Others, however, demand from it a type of support and advice tailored to choices not necessarily revolving around wage labor. For a 310

NUMSA organizer the new worker asks what do you have to offer?. There is a shift in ideology and a search for status. They no longer want to associate with toyi-toyi and strikes (Bonner 2004: 33.) As the next chapter will, however, show, this may not necessarily be the last word on matters of activism and political agency, even if their connections to emancipative narratives of work are not as transparent as they used to be.

Conclusion

In a burgeoning South African literature on employment change precariousness mostly characterizes an insecure occupational status. A recent comparative survey of the transformations of work in South Africa, Australia, and South Korea concludes that the three countries are in a neoliberal second great transformation, to borrow Karl Polanyis terminology, where corporations consciously manufacture insecurity (Webster, Lambert, and Bezuidenhout 2008: vii. Own emphasis.) For the authors insecurity essentially means the dilution of a stable, unionized core workforce into casual jobs or unemployment. It ultimately is a manifestation of domination and market despotism (Webster, Lambert, and Bezuidenhout 2008: 77.) Stable jobs allegedly enable, instead, happiness, fulfillment, and wellbeing. Job satisfaction still stands as a central category in the sociology of work, in ways that often obscure how workers practices and discourse outside production shape the meanings of jobs. To the extent it insulates the workplace from broader signifying practices, job satisfaction can be indistinguishable from attempts to make sense of utterly meaningless activities into which, nonetheless, employees are forced by economic necessity. As this chapter has shown, however, not only is the sense and value of employment lost to workers when occupations 311

become unstable, but precarious employment is itself a manifestation of the inability of work to signify social existence. For South African black workers, in particular, the postapartheid guarantees of the law and the governmental celebration of the citizen as producer did not salvage waged work as a vehicle of social emancipation and fulfillment. Unionized black workers are not among the winners of the postapartheid transition, nor are they mostly on the way to become a prosperous middle class as some sociologists envision. Even many of the benefits they gained in the workplace have proven uneven and fragile. Democratization and racial redress have not largely reversed the injuries of unilateral corporate power. In some ways, managerial prerogatives have indeed been disguised, and therefore reinforced, by company patriotism as a response to global market forces. The demonization of conflict in the ANCs governance model and scholarly fantasies of a post-adversarial workplace have shored up managerial images, resonating with echoes of colonial paternalism, of production as a mutually beneficial terrain. The stories this chapter examined disappearing workplaces, corporate authoritarianism, workers made dependable, despite their cooperative attitudes, as their collective identities and militancy slowly decline starkly contrast, however, with expert views according to which South Africas employment difficulties are mostly due to continuous politicization and confrontational mindsets in the shopfloor (Dickinson 2003.) By sticking to images of workplace-based identities as sites of liberation independent of the broader material and discursive practices shaping them, left intellectual discourse assumes that seriously degraded jobs can still provide a foundation to progressive social compacts, and avoids the task of critical scrutiny. Social conflict has indeed not lost its relevance to the extent it problematizes the impacts of employment precariousness on workers lives and develops alternative practices beyond what union organizations committed to compromise and moderation 312

can, or want to, articulate. If the unions want to bring to the bargaining table the claims of a lifeworld injured by the collapse of wage labor, they will probably need, rather than postadversarialism, a re-energized radicalism and a ruthlessly political imagination. The overwhelming impression my research conveyed is not of a reassured, purposeful, and enthusiastic productivist unanimity, to recall Judith Hayems expression. The employees I talked to were rather fraught with worries, uncertainty, and confusion. Their dominant tones were not necessarily of despair; but anger, disillusioned cynicism, or detached sarcasm surfaced abundantly. Workers defining themselves as permanent could be relieved by not staring at layoffs and poverty directly in the face, and could derive pleasure from the tasks they performed, but very few saw a secure job in terms of, as one respondent put it, respect, honor, and trust. Many more had an instrumental relationship to their occupation, while bestowing more purposeful meanings upon the prospect of a future outside existing jobs, or waged employment altogether. Many respondents did not indeed regard the possibility of being laid off as the unmitigated catastrophe the left and the unions usually imagine. Their resentment at restructuring was not exclusively determined by fears of joblessness, which surely play an important role, but also targeted wage labor as an economic imperative in the name of which employers can perpetrate all kinds of unfairness with impunity. Economic adversity, by contrast, did not stifle workers refusal to automatically associate their waged condition to progress, civilization, and citizenship. The next chapter will look at how workers translate the crisis of wage labor into community discourse, social interactions outside the workplace, and political action. I will further elaborate on some contrasts this chapter raised, particularly the greater propensity of 313

municipal workers to don the moral attributes of the community compared to the industrial employees perceptions of individual escape from waged employment. Both see the workplace as marred by remnants of the old order, yet they consider the social as a realm where democratic South Africa can still fulfill its developmental promise. There are, therefore, echoes of the emerging governmental narratives of self-entrepreneurship and personal initiative in such patterns of workers discourse. Can such parallels hint at a post-wage social compact that revives the ANCs insistence on self-reliance as the alternative to redistribution? Or do workers representations of life after wage labor question established ideological relations between citizenship and economic activity? Or can, conversely, collective identities based on waged employment survive and thrive in some renewed modality of working-class politics? A warning on the future prospects of organized labor comes from the heavy pressures diversified meanings and practices of work already place on the unions. Various writers have focused on how democratization and institutionalization separated COSATU from its grassroots by encouraging bureaucratization, professionalization, and subordination to the ANC in power (Rachleff 2001; Buhlungu 2001.) What happens in the rank and file is, however, equally if not more important. In a precarious world of work trade unions have to deal with complex, often contradictory claims, which old coordinates of struggle based on the blanket glorification of jobs can no longer contain. CEPPWAWUs Andile Nyambezi41 poignantly characterized the dilemma: With all the defeats they have suffered, militancy in that sense has declined. Not that it isnt there, but it has changed. The demands have not changed: wage increases, hours of work remain the main ones. The demand for job security is there but if the wages are settled that is a demand on which they compromise. I 314

dont know, maybe that is informed by the notion that it is an impossible request that you can make. Albeit it is important for the union, job security is seen more as a long-term issue, it is watered down to something that is not tangible. . . . Workers dont demand job security, they dont have faith in the union dealing with the issue. The managerial discourse of corporate patriotism has been, conversely, quick to penetrate the crevices created by workers insecurity, abetted by the work-centered policy discourse of a government still supported by the countrys main union federation. Employers and the state have systematically invited workers to sacrifice for the sake of a national project where corporate competitive success and the health of the body politic are coextensive and interchangeable. Mounting confusion and disorientation characterize therefore rank-and-file conversations that have less and less use for time-honored labor rhetoric. According to NUMSAs George Magaseng:42 Previously it was politics more than bread and butter issues. That has changed, we now deal with real economic issues, we do participate with companies in productivity negotiations, we do also give inputs. Workers take for granted that once there has been a process of democratizing the country everything else will follow, and that's the mistake they are doing. But until they realise it, it's gonna be difficult for them to identify themselves with the union. They can, however, also identify themselves with something else, which requires reconceptualizing employment precariousness from a grammar of suffering and trauma to one of desire and contestation.

315

6 Like a Branch on a Rotten Tree. Recovering Agency after Wage Labor

Hence, too, modern democracys specific aporia: it wants to put the freedom and happiness of men into play in the very place bare life that marked their subjection. Giorgio Agamben (1998: 13)

Introduction

When asked about the challenges and changes his household faced in the five years before my interview, a worker at Baldwins Steel replied: When I was still at school, most of the things we never worried about, which now people are worrying about. For example, Id go to school without any shoes on my feet, I couldnt worry, but now our children have to go out just from the bed to the kitchen and there must be shoes in their feet. You see, its different, I could walk kilometers without shoes. Money makes the world go round. . . . We were walking most of the time but now you must pay for taxi, locations are bigger now and to go from point A to point B you wont be able to walk (Respondent 3.) The social life of workers amplifies the grievances of unstable employment as erratic remunerations and uneven benefits provide a jaded safety net against uncertainties and expenses. Moreover, workers livelihoods confront a deepening commodification as their dependence on the market increases and public services are downscaled and corporatized. Healthcare, education, 316

transportation, and municipal utilities absorb a growing share of the wage, especially where employers revoke the benefits past union struggles gained. To cope with daily expenditures, many wage earners borrow from predatory loan sharks, a category that covers several unregistered moneylenders (mashonisa) operating in the townships. Financial liberalization and the relaxation of usury regulations for microlending allowed lenders and retailers to take advantage, at high interest rates, of a burgeoning demand for credit from previously excluded groups, which can now access new educational and consumption opportunities. In working-class households, spiraling debt mirrors the insecurities of wages, dreams of homeownership, and the rising costs of healthcare, schooling, and utilities. A survey of indebtedness among families depending on full-time employment revealed an average 106.2 percent debt-to-income ratio among respondents up to age 35 and 74.3 percent for those older than 35 (Hurwitz and Luiz 2007: 120.) In 2004, eight workers out of ten took home less than 40 percent of their wages, as the rest was mostly pocketed by microlenders charging an average interest of 175 percent a year.1 All income groups below the wealthy bracket of R 150,000 a year or more needed cash loans to cover basic necessities (Human Sciences Research Council 2003b.) Household debts interrogate the praise by academics and government officials of individual financial responsibility as the solution to poverty. The argument sanitized poverty from prickly political connotations by associating it to vulnerability, which suggests a condition against which it is up to the poor to mobilize their own assets (Ardington 2004.) The Financial Diaries research project, conducted with abundant corporate funding at the University of Cape Town, for example, aimed to humanize the poor instead of representing them in conventionally statistical terms. In line with recent reorientations in the parlance of 317

international financial organizations, it therefore used biographical narratives to show how liberalization converts social precariousness into economic opportunity. With no further inquiry into the structural dynamics of poverty, the poor emerged as proactive investors of portfolios and financial instruments, a seductive terminology of empowerment and choice that embellished meager means of daily survival.2 It would be reductive to dismiss such efforts as cynically manipulative attempts by financial corporations to exploit low-income credit markets with the help of a complacent academia. Their intellectual implications are, in fact, profound as they idealize poverty as a status informed by mutualism, reciprocity, and trust, disguising the quotidian exploitation and socioeconomic violations of predatory lending and casual employment (Bhre 2007b.) To access formal financing or insurance and escape the loan sharks, in fact, regular jobs remain necessary; indeed, more than one fifth of regular workers cannot access banking services (Ardington 2004: 5.) In such cases, the discourse of uplift through financial responsibility is less proof of the progressive mission of entrepreneurship than an instantiation of Bourdieus (1977: 183-97) symbolic violence as the misrecognition of socioeconomic duress. Sometimes violence is also real: some of my respondents reported being threatened, harassed, or physically assaulted, even inside work premises, by loan sharks to whom they owed money. The devastation that precarious employment wrought in many respondents lives operates through a searing gap between wages, as monetary rewards in economic transactions, and incomes, as the socially and culturally defined set of goods and services required for a decent human existence. Workers address the wage-income discrepancy by moonlighting, borrowing from extended families, or participating in community-based pooling schemes. The anxietyridden recourses of creativity and adaptation have often to compensate for the erosion of 318

collective solidarity. The ways workers and the poor eke out their living exemplify, in short, the success of the postapartheid governmental aim to avoid, to recall a previously quoted official, a granny state. While the imagination of the welfare state contained a certain glorified idea of work as creating citizens by protecting them from the uncertainties of the market, South Africas precarious liberation, instead, emphasizes work as the tool individuals wield in their daily battle against a naturalized vulnerability. The experiment is, however, constantly unstable: precarious employment and commodified reproduction reinforce each other in making jobs intolerable by exposing them not just to the violence of the workplace, but to the urgency of social necessities.

Commodification and the Reconfiguration of Workers Lives

The scholarship on commodification has often emphasized that the processes through which livelihoods are subject to market discipline are discursively mediated and perceived in contextually specific ways (Appadurai 1986; Canclini 2001.) The same applies to wages: the fact that they exist as commensurable monetary equivalents of productive activity, and mediate as such human consumption and ambition, does not freeze, as the capitalist dream would want it, their meanings (van Binsbergen 2005.) Workers, rather, constantly challenge their employers interpretation of the wage through their own normative ideas and discourses of needs. Their views, conversely, might not necessarily coincide with the governments, for which the mere fact of earning a monetary remuneration is a major step towards inclusion in the first economy. What workers expect from the wage and the effort required to earn it its subjective value as money of sweat, to recall the Nuer expression discussed by Sharon Hutchinson (1996) is not necessarily limited to consumption but also entails symbolic goods and sociopolitical aspirations 319

that exist outside the realm of commodities. Commensurability remains an unresolved problem in waged employment as workers attach value to their earnings not just as abstract measures of objectified labor but also in terms of the specific desires they validate. For similar reasons it is reductive to argue, as Seekings and Nattrass (2005) do, that what unemployed and precarious workers essentially want is a stable job, because jobs are as discursively signified as any other commodity. For employers they are contractual transactions; the state weave into them images of active citizenship and personal virtue. Workers representations may not, however, be congruent with those other meanings and, especially when wage labor no longer underpins a social order in recognizable ways, the meaning of jobs in relation to claims and imagination is far from obvious. The verb my respondents most commonly used to designate the relationships between their waged occupation and social existence was to cope, which for me encapsulated the grim trajectory from labors past heroic emancipative imagery to the current pragmatics of survival. The inability to provide for families basic needs with wage earning was pervasive and did not differentiate between regular and fixed-term employees. In 2004, households of full-time workers spent on average more than 30 percent of their income on food alone; for casuals and informal workers the figure was probably twice as high (Martins 2004: 49-50, 207.) Recurring images of life becoming faster and more expensive defined the experience of commodification across the workplaces I researched. Many workers were actually playing a decommodifying role for their unemployed relatives, to which they sent portions of their wages to alleviate market pressures (see Table 6.1.) Only a small minority (17.3 percent of glass-paper workers, 12.5 percent of municipal, and 9.7 percent of metal ones) had relatives receiving social grants.3At the time of my interviews, 6.4 persons on average depended to various extents on each 320

South African worker; by 2007 the figure had dropped to 3.5, mostly because the purchasing power of wages had fallen to the point that sharing was less possible and the unemployed were forced to seek casual jobs (Altman 2007b: 27.) [TABLE 6.1 HERE] Glass-paper and metal-engineering workers were mainly sharing their money with family members living with them or residing at relatively small distances, especially in the East Rand townships ravaged by layoffs. Municipal workers, conversely, reproduced a traditional migratory pattern as their household income networks stretched to extended families in rural areas. Remittances were in this case also ways to cultivate social connections, sustain the households modest cultivations or cattle, and possibly provide some stability and accommodation for retirement. Johannesburg city workers could use as much as three quarters of their net wage as remittances, and to do so they often forfeited their health insurance or chose not to have an urban dwelling and stay in 10-beds municipal hostels rooms, with monthly rentals as low as R 20-30. A diminished access to company-subsidized benefits was a further source of anxiety. Rising expenditures, especially for healthcare, threatened to reverse decades of labor struggles to decommodify social provisions and alleviate, to use Patrick Joyces (1980: 125) formulation, the permeation of daily life by the values and routines of work, a state of things summarized in the term the tyranny of work over life. Employers decisions, often motivated with the spread of HIV-AIDS, to limit or downsize healthcare benefits made them increasingly dependent on workers payments so that medical contributions seriously eroded respondents resources for daily necessities (see Table 6.2.) [TABLE 6.2 HERE]

321

A Johannesburg city employee with four dependants pointed at a recent pay slip showing a monthly net wage of R 498 (50 US dollars) out of a gross remuneration of R 2490, most of which went into healthcare deductions. I am going to steal, he added with dismay (Respondent 62.) The commodification of healthcare has mirrored insurance costs rising twice faster than inflation. The caps many employers introduced on claimable benefits, moreover, led low-wage employees to refuse treatments for fear of reaching their annual coverage limit. The insecurities of work revolutionized, therefore, workers temporalities as imponderable future events impinged on the sense of the present while daily calculations allocated scarce resources by weighing todays pressing needs against tomorrows fears. The predicament was particularly stark for Johannesburg municipal migrant workers; despite receiving better health benefits and higher employers contributions than in the private sector, many had to sacrifice coverage altogether to boost remittances to impoverished families. More than 40 percent of respondents in this sector lacked medical coverage. East Rand workers did not fare much better: 63 percent of respondents in the metalengineering industry a figure boosted by the collapse of Kelvinator and labor brokerage at UCW and 32 percent in glass and paper companies had no medical insurance. At the time of my interviews, only 8 percent of NUMSA members in engineering, and about 25 percent overall, had health coverage.4 One truck driver and former CEPPWAWU shop-steward at Paperlink a soldier, as a colleague called him to praise his dedication to the company and the union supported a family with five children, two of whom at school in Queenstown (Eastern Cape,) his wifes birthplace. One of his daughters suffered from epilepsy, had been left partially paralyzed by a recent accident, and required regular hospital visits. She depended for all medical assistance on her fathers wage the family was not poor enough to receive social grants two thirds of 322

which were slashed by medical contributions. Apart from medical insurance, for which the company had just introduced a new managed care plan that capped claimable benefits, he also had a mortgage backed by his retirement money as collateral. The family used in fact to live in a backyard room in Orlando East (Soweto) but, as it added new members, it had to relocate and, since the costs of new middle-class homes or privatized council accommodation were out of reach, they bought a small apartment in Johannesburgs congested downtown. A final net wage of R 700 had still utilities, transportation, and remittances to cover. What remained could hardly buy food, which left the family at the mercy of loan sharks. Survival was, in this case, a matter of painful choices and trade-offs between competing claims of household members; the lesson in individual responsibility taught by wage labor and market discipline came at a heavy price and the duress it produced hardly matched the self-fulfilling citizenry the state imagines. Many respondents rationalized the tough decisions they had to make by deploying a cultural imagery that wrested social benefits from the commodity form. A common explanation for relinquishing medical insurance was that it did not cover witch-doctors, traditional healers (inyanga,), herbalists and diviners (sangoma.) The dangers of HIV-AIDS, in particular, shattered the medical professions claims to effectiveness and enabled competing modalities of knowledge. Workers could therefore perceive as an affront the employers denial to help with healing practices in which, despite their controversial status in the official episteme, they were heavily invested. Healthcare was, in the end, a crucial terrain of contestation over what a job meant for, respectively, the commodified imagination of capital and the decommodified desire of its subjects. To deal with the power of corporate medicine, employees resorted to their own networks, such as the stokvel or moholisano credit associations and savings pooling schemes with which as many as 60 percent of Gauteng workers accessed low-cost services in local clinics 323

(Trres 2000: 415.) Precarious wage labor ultimately makes the employees body an embattled zone of governance (Prakash 1999: 144): workers rejection of institutional medical knowledge was neither just a product of cultural claims to local authenticity as in President Mbekis longstanding opposition to Western notions of healthcare nor an attempt to rescue the inyanga or the sangoma through an alternatively modern episteme of curative effectiveness. It is rather the product of a strategic assemblage of meanings with which workers counteract the intrusion of the commodity form in their lives and reclaim the autonomy of popular discourse against malignant market techniques. Underpinning the reevaluation of traditional medicine was a social justice imagery that wanted healthcare universal and free, as the postapartheid Constitution abstractly proclaims and policies enforced in its name practically evade. I found the following conversation revealing: Q: Are you on medical aid? A: No, never been for 23 years. Q: Why? A: Good question. A person like myself, I dont use medicine. If I am ill, I use just water, coffee and tea. My family is not covered either, they are only covered by the church. ZCC [Zionist Christian Church] is like a clinic, ill people are treated in the church. I am also a doctor, and a priest: if one is sick I know what to do. Q: Can you give me some examples of treatment that you have provided recently? A: I mustnt explain everything, because its a secret [laughter.] Even AIDS I can treat. I only help those who come to us, everybody, and its free. I can only say I tell the guy to bring me the coffee and tea, and I make something. Someone came 324

before me, from the headquarters, to teach me that, how to make medicines. The ZCC is against the use of medicines because the church has got to help you, freely. You pay when you buy tea or coffee, thats it (Respondent 157.) Reliance on spiritual affinities and organizations is surely ambivalent, as it can erode other collective identities, especially when the unions are ineffective in responding to the rollback of employment-based benefits.5 Occult forces also provide, however, alternative, if not overtly resistive, discursive practices to address the injuries of the wage relation. As Adam Ashforths (2005) study of witchcraft in Soweto noticed, and my discussion of workplace religiosity in the previous chapter confirmed, spiritual powers are assets, symbolic as well as material and political, that expand imaginative possibilities of ordinary people in conditions of insecurity, especially when their causes appear numinous and concealed from perceptible reality. While healthcare expenditures were the heaviest burden on respondents households, decaying employment conditions and company benefits undermined other social needs. The combined costs of education (see Table 6.2,) housing (see Table 6.3,) utilities, and transportation absorbed 61.9 percent of after deduction wages among glass-paper workers, 77.3 percent in the Johannesburg city council, and 35.9 in metal-engineering companies, for which transportation costs are excluded as no reliable figure is available. If remittances to relatives are also added, most respondents were regularly ending their month with hardly any disposable income. [TABLE 6.3 HERE] The ambition to move to a separate accommodation with their spouses and families is central to the value workers attach to their jobs as exemplifying the very passage into adulthood as a condition of personal independence. Very few of the youngest respondents resided with their

325

parents; even in the harshest apartheid-age workplaces, employees have been able to define working for wages as a tool to reclaim from their elders the control of resources and affects. The democratic constitution defines housing as a right, but its materialization overwhelmingly depends on the labor market. Very few respondents 9.3 percent of manufacturing workers, none in municipal employment had access to the governments housing subsidies. Those who received them found the program generally inadequate for anything but extremely low-quality dwellings. A decent self-built house could take between R 15,000 and R 30,000 only in construction materials; the full subsidy stood at the time at R 16,000 and could only be claimed by households with a monthly income of R 800 or less, way below the earnings of a full-time worker. Employees living in shacks reported having subsidy applications rejected because officials deemed their informal dwellings as income in excess of the means test. Having a job, therefore, usually meant renouncing any public help, facing the uncertainties of high-interest mortgages, and improvising social networks with friends and coworkers to access low-price materials or borrow vehicles. Workers who built their own residence mostly lived in an umkhukhu (chicken coop,) a shanty made of wood, cardboard, and corrugated tin, mostly with no running water, and an erratic, often illegal connection to the power grid. Of all my respondents, almost 30 percent lived in imikhukhu. A NUMSA member at UCW recounted that in his neighborhood, the historically colored township of Geluksdal near Tsakane and Brakpan, even houses built with subsidies were so small and fragile that local residents called them Unos, after the tiny FIAT car. Housing struggles have for long energized social movement politics in Tsakane and neighboring townships. The end of apartheid and the privatization of council housing had spurred dreams of homeownership among black unionized workers, as developers seemed to 326

take advantage of the deracialization of public subsidies to invest in low-income neighborhoods. As protests continued into the 1990s, however, adequate private financing, which in Tsakane was mostly promised by a major black empowerment housing corporation, did not materialize. Hundreds of families, which included many respondents from UCW and Baldwins, were left with high mortgage payments, to which they responded with organized boycotts. The developer reported at the time that half its loans were not being serviced nationwide, which the media blamed on an intractable culture of entitlement among SAs blacks.6 The ANC sided with such views as it stepped up its Masakhane campaign to persuade residents to resume payments. For many wage earners, expectations that employment would build an independent umzi (homestead) turned into bitter disappointment: I understand the previous government was giving you R15,000 if you were a first time homebuyer. Now these people are talking another language, they say [that when we expect subsidies] were talking about a previous government. But we were boycotting the previous government, now we are talking about this government. When you check the previous governments subsidy the money was R15,000, but this one, they are talking about less than that, depending on your income. So we are saying no, we are not paying. I think COSATU must try to sort this thing out, we must get the money from this government (Respondent 1.) In principle public subsidies could have partially decommodified housing, but they had the opposite effect. They were, in fact, mainly designed for the indigent, who could not raise matching finance to build proper accommodations; wage earners were, instead, excluded and left at the mercy of high-interest mortgages. Those with stable jobs could access credit from banks or non-traditional retail lenders, such as NGOs, cooperatives, or housing associations, but to 327

secure loans they often had to use retirement benefits as collaterals, thus jeopardizing old-age income. Fixed-term and younger workers, however, hardly had adequate accrued retirement contributions for this purpose. Some UCW employees had mortgage applications turned down by banks as soon as they named their employer, well known for layoffs and casualization. Wages were, in these cases, too high to qualify for subsidized housing and too low for affordable private financing, plunging earners in a limbo of uncertainty. Not only did wage labor fail to materialize the constitutional promise of housing rights, but actually undermined it. An alternative possibility was to apply for company housing loans, which brought workers expectations for ownership and respectability to directly weigh on the relations of employment. Loan programs were available in many workplaces I visited and managers generally used them, confirming Hayems (2008) observations, as discretionary rewards to gain the employees loyalty, minimize turnover, and extract concessions on overtime and multitasking. Denying borrowing was, conversely, a way to punish absenteeism and poor performance. Housing loans were in conjunction with threats of layoffs, which would jeopardize their repayment extremely powerful disciplinary instruments, often leading to humiliation and abuse. A worker told me that a human resource manager rejected his application for a loan and suggested him to rent a container as an accommodation more suitable for blacks. When loans were approved, employees often criticized their allocation as the result of favoritism and corruption and resented their amount, barely adequate, as one said, to buy few bricks every week. Housing loans were, nonetheless, double-edged swords for the employers: their immediate function as a device to discipline the workplace generally offended the ideational

328

constructs workers associated with homeownership. In the end employees often gathered the impression that companies were duplicitous and insensitive: The company is not looking after its workers. This company is big, but we start from scratch and we struggle to get the houses and they dont give us loans, they send me to the bank and the bank wants too much interest. They dont do nothing for us, to help us to improve our lives. . . . Even if you aim not to toyi-toyi to change things, managers force you to toyi-toyi (Interview 3.) While prospects of property as an imagined corollary to social advancement faded, many respondents had to fall back on downgraded rental accommodations. Almost 40 percent of municipal workers lived in rental housing, mostly compounds inherited from the apartheid age. A similar percentage of metal-engineering employees were also renters, mostly of backyard rooms or shacks. It is common among respondents to use retirement benefits to address contingencies the collateralization of a mortgage, prolonged periods of unemployment determined by inadequate remunerations and insecure jobs. In most companies I visited the shift from defined benefits to defined contribution schemes has, furthermore, jeopardized retirement security. Union representatives did not necessarily oppose defined contributions programs, and indeed often supported them as they tend to provide better benefits for short-term employees. Older workers coming close to retirement were, however, increasingly restless. For municipal workers, retirement issues were more pressing than for manufacturing ones. The comparatively better benefits provided by city councils were one of the main gains of unionization for black workers as they deracialized erstwhile bastions of white privilege. The main precursor of SAMWU, the former white- and colored-only CTMWA, had significantly 329

placed a heavy strategic emphasis on benefits once it joined the antiapartheid trade unions fold. For migrant municipal employees, in particular, decent retirement decisively redressed an experience of work marked by the abuses of the compound system. The promise of a stable old age income funded through the employers contributions was a powerful image of decommodification and social justice, especially for those planning to retire in poverty-stricken rural areas. The corporate rationality of iGoli 2002, however, foresaw a new, citywide defined contribution fund to replace a range of pre-existing pension products, which the Johannesburg municipality deemed too costly. The city council would contribute 15 percent of the new fund, less than the municipal contributions in the apartheid-age programs, which stood at 20 percent for white employees and 16 percent for most Africans.7 No other issue elicited such a powerfully emotional response among municipal workers, or in any other workplace I visited for this book, as the restructuring of retirement provisions in the wake of iGoli 2002. The proposed increased rates of workers contributions meant more commodified benefits on top of their material insecurity. In a context of managerial unilateralism and lack of consultation, rank-and-file rumors amplified desperation and panic as workers saw their future as utterly disposable. SAMWU members accused the union of failing to provide information on the fate of their retirement and allowing them to be robbed of their contributions. Tensions were palpable: when I visited the Benrose roads depot and a shop steward introduced me to the gathered employees as a researcher conducting fieldwork with SAMWUs approval, few workers immediately directed their hostility at me, thinking I was actually working for the union. An older man verbally assaulted me, while bursting into tears, shouting that the union had sold out, while his colleagues tried to calm him down. After I managed to explain that I was not in SAMWUs employment and conducted my interviews, one worker still urged me to go 330

straight to the union and tell them that the people have said that they want their money back. Many, in fact, wanted to collect their accumulated benefits before it was too late and resented the union for not supporting this demand. Individual fears of the impending future could easily override solidarity and resistance: Coming to iGoli 2002, I am shivering like a branch from a rotten tree. . . . The council is harassing us because they dont want to give us our money like pension funds, provident funds. All those money they deducted from our wages must come back to us, because council is changing now and employing contractors, so we want our money before we can enter those utilities, and the Blue Card money, the U[nemployment] I[nsurance] F[und]. . . . Q: The municipality says they wont retrench anyone. A: It can happen. Its like when you make a proposal to a lady, you can promise her anything, that you will buy her this and that, even if you cant, and that is what they are trying to do with us. . . . There is a person appointed city manager, Khetso Gordhan, hes the one who has transformed railways to Metro Rail: how many people lost their jobs there? . . . When they say our jobs will be secured for just three years, that is nonsense to us, we dont want that. If they can say our job is secured for three thousand years, that is alright, or three hundred years, not three years. When they talk about three years they make us confused. That is why we say, before they do anything they must give us our benefits back and we can start thinking where we want to go with them (Respondent 138.) The disciplining of healthcare, housing, and retirement by market forces qualifies the constitutional commitment of postapartheid democracy to universal social rights by defining a 331

specific type of productive subjectivity: rational risk-taking individuals who are empowered to manage their own affairs. Commodification is, however, also open to contestation. Workers may sometimes lack, as discussed in the previous chapter, a discursive repertoire to systematize and respond to adverse changes in the workplace, but they nonetheless invest issues of social reproduction like health, accommodation, and old-age income with normative meanings that exceed the commodity form. They routinely invoke ideas of existential stability and cultural integrity that resist subjugation to the labor market in ways that are ideologically complex and ambivalent. The following section describes how workers translate uneasiness and anxiety into political meanings underwritten by constructions of selfhood, community, and collective solidarity.

A Future Unlike It Used to Be: Visions of the Apocalypse and Labors Politics of Melancholia

Most respondents experienced their first waged job as a step in social advancement. More rarely it was the result of downward mobility from a previous, high-reward independent occupation, as in the case of a former professional soccer player sent by an injury, after ten years of a promising career, to load metal sheets at Baldwins Steel. For him, it was immediately clear that earning a wage would no longer afford what he called an expensive life, albeit made of hardly extravagant luxuries such as taking my family out, I had a car, my kids were attending a multiracial school, and I managed to buy food from the shop (Respondent 18.) Most respondents arrived later to the conclusion that waged work could not predictably provide basic goods and amenities. 332

The hollowness of labor normativity confronts workers with the challenge of discursively articulating responses to socioeconomic duress. In the South African transition the abstraction, in the Marxian sense, of labor power as a social precondition of accumulation regardless to race or gender paralleled the ideational abstraction of work as a universal signifier of dignity, independence, and responsibility. For black workers, however, experiences of waged employment carry expectations and rules that are too deeply entrenched, and memories and traumas that are too fresh, to be smoothly flattened and rearranged by the states normative framework. The world of work is not confined to political economy, but also involves an economy of sense and emotions in which there is no obvious celebration of production as a progressive space of citizenship. To the contrary, disillusioned recollections of past struggles and recent betrayals acquire a tangible physicality amidst haunted spaces of industrial decay (Edensor 2005.) Kelvinators Freedom Square, a bare courtyard surrounded by empty warehouses and silent machines, stood as a powerful symbolic counterpoint, indigestible within the new regimes monumentality of work. The governmental ethics of employment as a Kantian condition of priceless human worth, autonomy, and self-respect (Sayer 2007) inadvertently fosters workers disenchantment with precarious jobs. Sentences like everything is about money or things are more and more expensive recurred in my interviews as if they were part of some invisible script, or an only partially conscious mental procedure to name and characterize a still largely undersignified existential frailty. On a nationwide basis, a shrinking number of households from 72 percent in 1995 to 60 percent in 2003 in the case of Africans have earned employment-related income, which has also fallen on average by almost one third among Africans in families with employed members (Casale and Desmond 2007: 66.) Among my respondents, 54.2 percent reported a 333

decline in their household living standards over the five years before being interviewed and only 21.2 percent, half of whom had started a second occupation in the meanwhile, mentioned improvements. In the case of metalworkers, mostly hit by casualization and layoffs, monthly household incomes were inadequate to cover basic necessities for nineteen respondents out of sixty, while twenty-six of them regularly failed to save any money at the end of the month. Workers tended to discuss their living standards in objective terms, were usually reluctant to elaborate on them, and did not look keenly at prospects of relying on state programs. Selfreliance clearly played an important role in daily narratives of emancipation and was not merely a knee-jerk reflection of the ANCs opposition to handouts. The most common political conclusion workers derived from the precariousness of jobs was that a government devoted to liberalization made it impossible to provide for their families through hard work and commitment. Far from having their character corroded (Sennett 2000) by flexibility, respondents reclaimed decent jobs, often in explicit opposition to the occupations they actually had, as ethical foundations of disciplined and caring selves (Lamont 2002: 28-33.) Their apprehension was often couched in a socially apocalyptic imagery where low wages, poverty, and joblessness herald turmoil, moral corruption, crime, and the disintegration of the community. As workers no longer perceive their jobs as viable foundations of social order, the emphasis on the waged condition in the policymakers pledges sounds empty and insincere. The previous chapter discussed how images of the community are differentially deployed by distinct groups of wage earners in response to changes in production. The combined impact of precariousness at work and the commodification of life, however, forces workers in general to redefine the meaning of community and the position they and their families occupy in it. Even those who appeal to an ideal communal morality to boost labor claims must often recognize, 334

nonetheless, that, under conditions of socioeconomic duress, actual neighborhoods and households can become uncertain, threatening, possibly violent places. It is not guaranteed, moreover, that social networks originating in the workplace can do much to alleviate such tensions. Reduced resources for leisure activities and the pressures of intermittent unemployment in a fragmented world of work can indeed undermine occupational communities and erode the functions of jobs in social interactions. The continuous relevance of the community indicates, nonetheless, a quest to understand the decline of waged employment in systemic terms rather than as individual failure. Workers imagination of a social sublime where decent jobs underwrite justice and morality starkly diverges from, and therefore criticizes, their frightening existence as disposable bare life, in Agambens terms. Caught in the contradiction between the dignity of employment as imagined by the state and its material realities, between work as it was promised and work as it is, most respondents wove the theme of the community into a sense of melancholia recalling Ranjana Khannas (2007) definition: Melancholia is not only a crippling attachment to a past that acts like a drain of energy on the present. . . . Rather, the melancholics critical agency, and its peculiar temporality that drags it back and forth at the same time, acts towards the future. Workers do not just miss decent jobs as part of irretrievably spent past expectations. The loss of the prospect of decent jobs, rather, determines perceptions of the self as a mutilated subjectivity as much as it motivates psychic projections of ideal social orders. The significational emptiness of work as a practical experience flows into myths of work with dignity as a way to avert the culmination of the present crisis into a future catastrophe. Elsewhere Khanna (2003: 21-23) 335

discusses melancholia in Freudian terms, as an inability to assimilate loss so that the subject criticizes itself rather than the object that is lost. Workers narratives of an imagined lost world of work could therefore long for a symbolic centrality of labor in their communities while lamenting the constraints of their actual productive locations. Melancholic resignifications of the community importantly redefine the relations between the urban and the rural, which in South African political discourse have been sharply polarized. When, in the early 1990s, violence dramatically flared in the East Rands townships, hostels, and informal settlements, labors discourses of resistance often represented urban and rural, migrant and resident as contrasting ontologies. Some scholars argued that workers resisted either by embracing urbanized identities and citizenship claims or by going back to deep seated rural values (Mayer 1981.) Marxist critics rejected the dualism of rural and urban identities arguing that South Africas black proletarian condition was hybrid and, therefore, unionization combined forms of consciousness drawn from the city and countryside (Sitas 1983.) Postapartheid perspectives have reclaimed the promiscuity of cultural and class analysis as old class-conscious migrant workers are replaced by the rural poor and the unions urban solidarity is diluted by nationalist politics and market-based black economic empowerment (Breckenridge 2007.) Rural and urban, therefore, overlap in material practices and symbolic networks rather than defining self-contained identities (Hart and Sitas 2004.) James Ferguson (1999) powerfully criticized the ways in which in the narratives of policymakers, social scientists, and labor organizers alike rural and urban are cultural extremes in a progressive modernization continuum underscored by waged employment. He rejected, in particular, the notion of permanently urbanized labor by stressing how workers connections to the rural areas and images of their eventual return there were neither vestiges of a 336

premodern mentality nor surviving atavistic features within industrial identities. They were rather fundamentally urban styles workers cultivated and deployed to signify waged employment in the city and its adversities. Rural values act, therefore, as critical commentaries and political responses to urban problems. Fergusons idea of style is not a reflection of material conditions or inherent consciousness, but defines the competent performances with which malleable modes of language and self-presentation deal with fragile conditions of production. Terms like urban and rural or permanent and casual are themselves elements of style workers self-consciously and opportunistically appropriate to be publicly visible and make claims that never transparently reflect cultural affinities or organizational belongings, but are always potentially ambiguous as they address different audiences (Ferguson 1999: 225-28.) Ferguson (1999: 110) defines as localism an urban proletarian style motivated by skepticism and disenchantment towards prospects of existential stability centered on the wage. My research confirms Fergusons discussion of localists as workers who, regardless of the depth of their urban experience, imagine the return to a rural home as a solution to their social predicament. A powerfully resurgent ruralism shaped the melancholic narratives of many respondents opposing the travails of industrialism and wage labor to an imaginary agrarian condition. In many interviews, going home was a response to labors unfulfilled promises and to the loss of the citys progressive allure as it came to be configured as an expensive, hypercommodified, materialist, amoral place. Rurality, which many have often never materially experienced, is then evoked as an ideal elsewhere of economic opportunity, relaxed life, and morally sound interactions. Ideas of agrarian exodus acquire a mythical aura, which does not, however, necessarily weaken the underlying critique of waged employment and its dislocations.

337

For several respondents for example older workers in the Johannesburg municipality and founder employees at Kelvinator resurgent ruralism is not, however, only an urbanite mannerism but also indicates persistently strong rural ties and living memories of migrant labor. Many city council employees I interviewed were still living in compounds. Some recalled arriving there when the council asked the amaKhosi [chiefs] for people to work for two cents per week, cheap labor, people from the farms, who didnt know about money (Respondent 88.) Several young workers too, however, ended up there after leaving poverty in the former homelands behind to look for casual jobs at the gates of municipal depots. As township accommodations were out of reach, they experienced the squalor of hostel life: The rooms can accommodate 8 or 10 people, but in my room we are 14. They keep promising they are going to change these hostels but until now nothing has changed, there is water leaking and everything is bad. It is still the same as under apartheid, no change (Respondent 96.) For workers who experienced coercive migration regimes the symbolic investment on the land (which some deferentially called the chiefs land) connects commodification to a broader narrative of dispossession (Hart 2002.) Images of the rural metonymically allude, therefore, to an idealized social order before and outside wage labor. The collapse of Kelvinator left as its jobless residue a large group of founders, older employees often from rural origins who had placed their dreams in the companys social compact of employee share-ownership and industrial citizenship. Unemployment turned them into dependents of relatives they used to support, while loan sharks preyed on retirement payments. Their ideas of dignity shifted therefore from insisting on looking for urban waged jobs to plans to return to the land. Myths of ruralism reversed decades-long mental processes that 338

identified labor struggles with the city. A sixty year old man from Thabankulu (former Transkei) had worked in the Johannesburg municipal waste services since 1969, living in the peri-urban Anthea hostel since 1971. When I interviewed him he was sending R 1,500 of his R 1,800 net monthly wage to his family, wife and six children, none of whom were in a formal occupation: Respondent 106: I want to go back to Transkei because I dont have any family here and I cant bring them here [laughs.] As our culture, I think we dont belong with Johannesburg. The graves of our family are in the homeland, we come here just to work, not to sit, not to get some house here. We could, but when I retire I am supposed to sell that and go back to my roots. A female colleague passed by, overhearing our conversation, and laughed. Q: Why are you laughing? Unidentified Employee (UE): Its about this thing of going back I was born here and I must stay here. Respondent 106: They wouldnt have anywhere to go back to, but we were born outside of Johannesburg, in Mpumalanga, Transkei, and we are supposed to go back there. You see, those people left their houses and their brothers and bought houses here and then when they had to go back they no longer knew where they came from UE: Ahhhh! I dont agree, I know where I come from, I know where my grandfathers place is, Pietersburg, but I am not planning to move there! Respondent 106: When their grandfather passes away their grandchildren forget everything about their roots [the other worker laughs,] but we, our family is there and we come here just to work and after few months I travel back there. If I die 339

here in Johannesburg they will take me back to my forefathers to bury me there at home, we must be there to the grave of our forefathers. As Ferguson (1999: 110) argued, however, fantasies of return to the land rest not only on symbolic foundations but also on what workers can concretely bring home, not just in monetary terms, but as proper manners towards moral codes underpinning social respect and local notables with authority in the allocation of land and resources. It is common in African households that mobility and fluxes are strategic assets for production and reproduction as they connect material and ideational networks (Posel 1999.) For male workers, remittances to rural families enable claims to a moral and economic power that adverse workplace conditions erode in the urban context. The survival of rural households can provide wage earners with a grammar of stability to deal with unpredictable employment changes. Rurality thus encounters a paradoxical fate: once seen as a space of household dissolution through migrancy and degraded homelands, it reemerges now as the image of a material and symbolic household reconstitution (Ngwane 2003.) Some respondents clearly emphasized a melancholic connection between masculinity and ruralism as a space of order sheltered from the ravages of economic decline and urban moral confusion: At home [Gazankulu] it is better than in Johannesburg. In Johannesburg you need a lot of money, you cannot stay here without money, you see, so at home I can get food by ploughing, mielies, you know, selling bananas and mangoes, its better. Life is harder here if you dont have good money. If you are not working, to get a divorce is very easy, but now somebody thinks he cannot lose his wife because hes not working, but that is to start thinking something wrong (Respondent 177.)

340

Resurgent ruralism is not limited to older workers but cuts across age and urban residential status. It is, indeed, seductive for younger workers without a direct experience of agrarian poverty. Second- and third-generation city residents waxed lyrical on how bucolic lifestyles would rebuild communities and liberate productive possibilities. A paper worker born in 1940 in Orlando West (Soweto,) where his family was forcibly relocated from the Kaserne suburb of Johannesburg, was planning to leave for a rural area north of GaRankuwa (former Bophuthatswana,) a municipality with a 40 percent unemployment rate and two thirds of households living with less than R 1,600 per month (Bojanala Platinum District Municipality 2005.) His father had, in fact, moved to family-owned land there soon after the 1976 Soweto revolt, but social turmoil did not convince the son to follow him (the riots were bad for the kids who couldnt go to school, but they were good for me.) The costs of city life, combined to the anxiety generated by talks of layoffs, however, were more than this worker could bear. The Johannesburg he experienced, its indifference, dangers, and corruption, was too cutthroat a place to make a life and painfully contrasted with the bustling and upwardly mobile metropolis in the municipalitys glossy self-presentation: Life in Johannesburg is not that good, because it is too fast, everything here is too fast, people kill each other, do what-what, robberies, hijackings. Eisch! Its very terrible, you cant buy, cant have anything nice, when you stay in the location you must carry a gun all the time and I cant live like that. And people dont have respect for each other, they behave very reckless, you cant expect help from somebody and if you have something better than them that will start jealousy and they will try to remove that from you. There is no solidarity (Respondent 157.)

341

Others are more cautious about idyllic associations of rural life with pride and respect, and display a more utilitarian approach, aware of the competition for scarce resources that runs through rural households. The father of a respondent moved to the highly impoverished QwaQwa area looking for a homeland. But for him, born in Soweto, future life strategies are more about choosing the lesser hardship than buying into a grand narrative of social stability: Q: Are you planning to retire to QwaQwa? A: It is not attractive to me, I dont like the place. The way I see that place is the way we are living in Soweto, I dont see it any different. Before, when hom elands were doing things like farming, people had cows, there was action, but now when I look at that place, it is like I am living in Soweto, I dont see anything different. If I had land and some cattle and I could grow my mielie [corn] I would move there. In Soweto I pay rent, pay electricity, everything is money, everything! (Respondent 165) Common to both mythical and instrumental ruralist imagery is a discourse with strongly conservative overtones as a distanced sneer at the malfunctions of urban employment. The values of wage labor are no longer perceived as effective barriers to social upheaval, normative chaos, and behavioral disruption. Perceptions that roles and mechanisms ideally underpinned by work are losing their grip allude to moral repertoires with clearly defined symptoms and villains. Respondents recursively mentioned, therefore, the youths growing restlessness, unruliness, and disrespect as a sign of social apocalypse. Once again, an age category associated to an unproductive condition is a vehicle of alarm shared by state imagination and ordinary discourse. The image of a vicious spiral of unemployment and lawlessness was common to depict the youths lack of socialization to the ethics of wage labor. Crime was, by far, the most 342

important community problem raised in my interviews, relegating issues like HIV-AIDS or service delivery to distant seconds: I have been treated bad in my life many times, but now I have to have guns pulled out at me by young tsotsis, children really, and that is too much for us. Schoolchildren walk around with guns and if I go to the police they are running for cover, but if you beat your wife you go to jail, but for crimes and hijackings the police doesnt come (Respondent 130.) The image of the tsotsi as an object of working-class panic has deep roots in South Africa. It not only evokes a profoundly destabilizing idea of refusal of work, but specifically defines idle nonworkers preying on productive members of society. A child becoming a work-shy rascal is, in urban as well as rural contexts, the nightmarish representation of the inability of precarious work to guarantee transition into adulthood (Ngwane 2004.) Value systems that compensate for the disappointments of waged employment can either reward masculinized emotional and physical toughness in the community or emphasize sexist chauvinism in the household (Hunter 2004.) In the end, vacated of their emancipative meanings, labors social claims survive as a resentful longing for patriarchal values that ironically mimic images of breadwinners probity and responsibility in old authoritarian industrial orders. The respondent quoted above distinguishes beating your wife from crime by making the former a less urgent issue than public violence. Anxieties over the economic fate and social status of masculinity boosted male fears of womens capacity to renegotiate the tasks of household care. The dreaded possibility of greater female assertiveness made gendered domestic violence recede among the concerns of male respondents. Only three interviewees, all women, specifically indicated it as a social problem. A political economy of misogyny (Ferguson 1999: 343

197) underscored domesticity as a terrain of contested resources. The family becomes then an ambiguous point of intersection for the instrumental logic of economic survival and moral aspirations to respectability (Posel 2006.) Male workers represent their inability to take care of the family not only in terms of material constraints, but also as a symbolically subversive process where the importance of the income of female relatives enhances their claims to power: As in our customs, you are responsible for your family, I must make sure everything is in order, that my children and my wife have enough food. Even if shes working I shouldnt rely on her money (Respondent 88.) Similarly to Isak Niehauss (2006b: 68-69) discussion of gender violence in rural Bushbuckridge, the sexual conduct of men during employment crises points to a dislocated selfrepresentation whereas to stay married is to live with your failure. A precarious male productive position undermines established age hierarchies and domestic roles. The melancholic representation of a lost world of work reacts to the material and symbolic tensions of domesticity through a moral tale with clear-cut characters: the respectful youth and modest women of a timeless, depoliticized countryside contrast the unruly tsotsi and the money-obsessed wives of the city. The postapartheid governance project of naturalizing the family as an institution of social stability takes therefore a peculiarly conservative turn in mundane discourse (Morris 2006.) More disquietingly, the alien has also incarnated social disruption as workers target immigration as a chauvinist outlet for their apprehensions. Approximately one in ten respondents blamed unemployment and crime on foreigners, a feeling captured in comments like here in South Africa we didnt know what drugs, cocaine were before, so South Africans must get jobs and the foreigners must go out (Respondent 56.) In some cases, negative opinions on the 344

immigrants morality disguised resentment at their initiative, aptitude, and industriousness, all characteristics many respondents believed South African youth have lost. In a time when stable jobs dwindle, trepidation about alien intrusions reconfigures citizenship as a signifier of belonging and exclusion. In a sort of deformed mirror image of the official work-citizenship nexus, turning immigrants into scapegoats regressively validates the status of the worker as a member of the national community. In the end, the politics of male working-class melancholia discursively transposed disenchantment with precarious employment into revanchist images of order and purity. Demands to restore the dignity of work stood in stark contrast with the democratic, egalitarian themes of past labor struggles. They rather gave voice to wounded, resentful subjectivities perceiving themselves as embattled and besieged by all sorts of imaginary enemies insubordinate youth, womens claims, undocumented migrants standing for all too concrete social processes. The next, and final, section turns to explicitly political modes of discourse and addresses their potentials and limitations in counter-signifying the precariousness of work.

The Fog of Activism. Working Class Agency and the Uncertain Quest for Citizenship Alternatives

When I visited them for my research, KwaThema and Tsakane were engulfed in conflicts over service delivery, municipal rates, and housing mortgages. Rent boycotts and unauthorized connections to the power grid were widespread; demonstrators destroyed the houses of few local councilors and repeatedly clashed with the police. The KwaThema Crisis Committee (KTCC) was particularly active in protests against electricity cut-offs and supported, similarly to 345

movements in other townships, Operation Khanyisa, or light up, which claimed municipal utilities as decommodified provisions rather than goods provided at market rates or through prepaid devices (Barchiesi 1998.) Many members of the KTCC were unemployed, with a strong presence of women on state grants, a common social composition in social movements like the Anti-Privatization Forum, which in the early 2000s established branches in the East Rand. The ANC and COSATU were from the start hostile to mobilizations for service delivery they often perceived as ultra-leftist and anti-governmental; several respondents were also unsympathetic. A UCW employee from KwaThema, for example, dismissed the KTCC by arguing that paying municipal bills was a residents responsibility: Two weeks ago in our location they robbed Eskom [the electricity company] to get electricity but now the police is trying its best, they arrest those people. . . . [They] make a lot of wrong things, smash company cars, so Eskom doesnt dare to come to KwaThema (Respondent 35.) Far from seeing new social movements as fellow travelers against social injustice, most respondents shared COSATUs condemnation of their direct action tactics and opposition to the ANC-led alliance. Some regarded the confrontational propensities of community activists as part of a pattern of youthful anarchy and moral chaos. I was struck that, even if the rising cost of utilities was a common complaint in my interviews, only about one tenth of respondents defined the commodification of services including market-based rates, aggressive cost-recovery measures, and cutoffs for defaulters as community problems. Particularly surprising in this disjuncture between issues perceived as personal and collective was its contrast with how quickly Johannesburg municipal employees appealed to community claims to advance demands for justice inside the workplace. Otherwise, most workers did not translate the connections 346

between decaying social conditions and precarious employment into an alternative language of political activism. Perceptions of the unions role and potential, on the other hand, varied in relation to age, gender, education, migrancy, and seniority. A large majority regarded organized labor as inadequate on its own to stem or alleviate adverse employment changes, even if they relied on it for protection at the point of production. Older black workers remained attached to images of the union as the organization that, making oppression visible within broader social and historical processes, turned wage labor from a reality of vulnerability and exploitation into a promise of liberation and dignity. The story of a shop steward at MB Glass (Respondent 211) well encapsulated that trajectory. He was born in 1960 in Kwazakhele (Port Elizabeth) and first engaged in the resurgent black workers struggles when visiting relatives near Durban when the city was the epicenter of the early 1970s strike wave. After school he found a job at the Ford automotive plant in Struandale (Port Elizabeth,) where the dismissal of the local civic leader Thozamile Botha sparked a long, militant strike, and all hell broke loose. In the aftermath of the strike, MACWUSA, a community union, broke away from a car workers affiliate of FOSATU, which did not support the mobilization, and Botha convinced my respondent to join. In the Eastern Cape at that time conflicts opposed workerist and populist unions as well as nonracialist nationalist movements and the Azanian Peoples Organization (AZAPO,) a descendant of the Black Consciousness tradition. In Port Elizabeth, and Kwazakhele specifically, ANC activists were more effective than elsewhere in maintaining functioning underground structures, which in the 1980s shaped the antiapartheid political opposition (Cherry 2000: 93.) The local nationalist milieu and its organic intellectuals gave black workers new opportunities to problematize wage labor, seeing it both as a vehicle of solidarity and struggle and a barrier 347

holding collective action within the terrain of production and deferring its translation into political mobilization. Caught in the tension between the limits of workplace activism and desires for radical transformation, my respondent chose, after repeated dismissals and arrests made him increasingly skeptical of trade unionism, to join the ANCs armed wing, the Umkhonto we Sizwe (MK,) to which he was recruited by his brother, also active in the local nationalist underground. After training in Zambia, he returned to the Eastern Cape and, amidst intensified repression, he grew disillusioned with the deepening strife among different antiapartheid traditions. In 1984, after a stint of further education in Lesotho thanks to connections in his Catholic church, he moved to the East Rand, in Katlehong, where trade unions allied to community organizations resisted forced evictions and promised to pry open new spaces for revolutionary politics. While still an undercover MK operative, he initially worked at Khotso House, the Johannesburg headquarters of the South African Council of Churches (SACC,) where his tasks included inducting migrant workers from the Eastern Cape into nationalist organizations. In 1986 MB Glass hired him through a relative serving as a shop steward. Coworkers did not know about his MK activities as the ANC deemed recruitment inside the plant too risky. In the end, labor and political activism remained on separate tracks, one public, the other clandestine, even if the latter infused the former with political meanings, rather than just providing a cover. His approach to the union was, in fact, not merely instrumental and his talent for underground organizing helped establish the Chemical Workers Industrial Union (CWIU,) a COSATU affiliate and precursor of CEPPWAWU, in the plant. For him, in the end, workers, guerrillas, and activists were all people working together in terms of pursuing our political aims. Revolutionary sentiments

348

paralleled, therefore, pragmatism on production issues. He, for example, opposed illegal strikes before union recognition as shooting ourselves in the foot. In the 1990s the MB Glass plant was the stage for physical, often violent confrontations between COSATU members and supporters of UWUSA, the union formed by the Inkatha Freedom Party (IFP.) According to COSATUs policy at the time, as my respondent explained, we need not to draw the dagger inside company premises, which could jeopardize shopfloor organization. He eventually chose to return to full-time political activism in the early 1990s, as he lost his job to the layoffs that started to hit the company. As the East Rand plunged into violence with thousands of casualties, the Kwesine hostel in Katlehong, near where he lived, was an IFP stronghold and a major flashpoint of conflict with residents of local townships and informal settlements. Many migrant workers, attached to an earlier generation of workplacebased labor activism, turned to the IFP when COSATU sided with pro-ANC township organizations. My respondents task was at first to bring a kind of conscientizing and politicizing among squatters from the Eastern Cape and then to organize the local self defense unit: It was comprised of people who were mostly illiterate, they were not educated, and our task was to convince them that we were fighting for a just cause and also convince them that they must not believe in ethnic leanings. He was rehired by MB Glass in 1997 and, as democracy dawned, he could reflect on the distinctive rewards of labors alignment with the national liberation movement. Unionization had brought not only better wages, company benefits, and generous in-house agreements, but also pride in struggles that pushed the breadline level through a type of workplace activism fusing political aspirations and a strong sense of autonomy. As he explained: We are not used to rely 349

too much on our union officials because we have what it takes to be self-esteemed. Political militancy carried other satisfactions in the community, such as a two-room government subsidized house as a reward for my contribution to the struggle and a sense of finally regained personal and family stability. Subsequent restructuring and retrenchments, however, threatened to reverse these gains, as managers took advantage of the favorable moment to roll back many of the provisions the union earned in its militant days. He remained, therefore, convinced that the labor movement should keep a critical distance from the nationalist political leadership to which he kept pledging allegiance: I dont think we have to rely on the government when we want to pursue our aim, more especially as labor. For that matter we do have a right to see that the government hasnt done enough for labor as such, we as workers need to be independent. The complex biographical trajectory of this worker encapsulates wage labors emancipative promise, the ways in which it energized national liberation politics, and the disappointments threatening its undoing. Accompanying it all are sharp insights into the ambiguity of the wage relation as a condition that enables collective solidarity but simultaneously needs to be transcended in order to fulfill its liberatory potential. Most respondents, however, and not only younger ones, shared neither a similar activist path nor the history of sacrifices and achievements that made it possible. They experienced the significance of organized labor in the tangible improvements it allowed, but their relations with political and social movements were generally looser. Their memories of struggle were tied to specific episodes where the protagonist was the union and political organizations could alternatively be contingent allies for a common purpose or self-serving, potentially disruptive 350

actors. Many had serious doubts, in any case, on the current relevance of such past connections. Much more decisive on their union lives were the postapartheid insecurities of work and the confusion generated by the policies of the supposedly friendly ANC government. Another MB Glass worker (Respondent 212) also joined a union at a very early age, in the aftermath of the 1970s student mobilization, when I was at school and I was fighting for my rights, when the apartheid regime was forcing us to learn Afrikaans. The power of the union dawned on him when it obtained his companys recognition in 1987. That time was for him a golden age when we always got what we want by flexing muscles in strike actions. With the onset of the transition to democracy, however, management started to turn things around and two consecutive layoffs peeled two-hundred and fifty workers off MB Glasss headcount. A new feeling of powerlessness crept in as post-1994 labor legislation protected workers organizations but also limited, in the interest of productivity and efficiency, their right to strike over restructuring caused by operational requirements: We didnt go on strike because the LRA [Labour Relations Act] was in favor of them having retrenchments. They said it was an operational requirement, then there was a second restructuring which was made in that period and they brought in some subcontractors, like in canteen, and later some bottle manufacturing and molds have been outsourced (Respondent 212.)8 Many workers recognize that democracy improved the workplace climate as human resources managers recruited from the unions and new consultative forums facilitated mutually respectful interactions. They also see, however, a mixed blessing in such developments, which bypassed union structures and pushed them into largely defensive positions:

351

We are no longer fighting like before. We had a Nampak national strike in 1990, especially in the corrugated division and the paper division. That was a big strike, it took us three months and three weeks. . . . Those strikes have taught us many things, they taught the management a lesson, but they taught us also a lesson because many people have lost their jobs after the strikes. As workers we have been victims of that, we have lost in terms of our salaries and we have lost in terms of our comrades being fired during those strikes (Interview 180.) In other companies workers noticed a two-pronged managerial strategy trying to do away with the union while imposing consultative procedures as paternalist and toothless devices. A Consol shop steward told me it was better in the 1980s: then the union could win concessions from the most hostile employers through prolonged confrontation because an illegitimate government not only did not stand in the way, but actually contributed to the resolve of the workers. On the contrary, when in 1995 the company announced plans for restructuring and downsizing, The union was not prepared because we were not sure what was going on, what they were referring to, we didnt know that this restructuring and outsourcing was a monster or what. Before 1995 we had a good relationship with managers and we had job security. Then in 1995 they started talking about this downsizing, whatwhat, they came up with all these names. They said we had to compete globally to enter the world market and for us to compete they had to downsize (Respondent 195.) Workers were used to the employers hostility and repressive legislation, but the pro -business stance adopted by a government they voted into power brusquely reversed comfortable narratives of progress: 352

It is better with the devil you know. The managers in those years were arrogant but were better than the managers of today. In fact even the LRA in those years was much more better [sic,] it was more union-friendly than the LRA that we are having today whereby you cannot strike if the company is restructuring. In the older days I could organize a strike now and take all the workers but today you cant do that, because of the LRA that we are having (Interview 181.) Reminiscences of the 1980s as a time when union militancy was rewarded and the battle lines were clearly drawn abundantly fed the current melancholia of the black working class. Many respondents had fond memories of strikes, sit-ins, and sleep-ins with which they forced managers to capitulate. Their stories of endurance, dedication, and discipline relate that struggle and conflict, not the work ethic peddled by the government of the day, earned work its dignity. In the postapartheid workplace, however, the sense that the union can shift power relations almost by sheer volition, as if by pushing on a visible and tangible wall, is largely gone. Organized labor confronts indeed unprecedented difficulties in defending what it gained. In place of old radicalism, workers melancholia mimics, while trying to negotiate, the work -centered citizenship imagination of the ANC-led government. Sometimes, as the controversies surrounding Kelvinators demise showed, union leaders actually reined in rank-and-file militancy. Many workers resented the bureaucratization of the unions and their alignment with the employers. In 1991 NUMSA was eradicated at Baldwins Brakpan following an unprocedural strike and the dismissal of the whole workforce. According to a shop steward: We restarted the union here in 1997, and NUMSA Wits or national has done nothing to help us reorganize, they dont even tell us what happens in Dorbyls

353

shop steward council. Is Dorbyl paying them? I always call [NUMSAs Dorbyl organizer.] Q: Well, he was telling me that Dorbyls CEO almost sounds like a union guy A: Managers know [NUMSAs organizer] very well, he was a crane driver and now hes a big man. . . . I dont get enough money, even the organizers dont get enough money, but the man has a house in Tsakane, and a house in Katlehong, and in Delpark, somewhere around there: three houses, man! You can ask yourself the question: why are our leaders relaxing too much? (Respondent 1) In the end, workers growing instrumentalism towards the unions confines their expectations to reliance on their own devices. Research on COSATU members confirmed that wages remain by far their most pressing issue. Job security seemed indeed a way more important demand for white than African workers and motivated only ten percent overall to join COSATU unions (Makgetla and Trres 2006: 11.) Faced with the limits of labor organizations, my respondents articulated four main alternatives. First, many reaffirmed their commitment to labor radicalism and independence from the ANC: It was the workers who changed the government, we were toyi-toying for this new government, we were part of it and now what we want from the government really is to address our problems, because we are not earning. And now when we toyi-toyi, when we start to fight, they say we are killing our economy, our economy is going down. So we ask ourselves: cant we fight anymore? (Respondent 166)

354

For some, engaging government policies is central to labor radicalism as an excessive focus on wages and working conditions would separate regular employees from precarious workers and the unemployed: The union has to engage the government, because the government has been voted by us. The previous government was oppressing us but now we are in a democracy so now if we have some complaints those people must hear us. . . . Every night I am always dreaming about whether I will lose my job, where will I go? What will happen in ten years time in our country if people are not working? The union needs to be more militant in engaging the government, because in the present day I cant be happy that I am working. If the union says, hey, lets go and fight the government and I say no, I cant go there because I am working, then I will be killing myself, at the end of the day I will lose my job. The union must fight both for the unemployed and for the people which are working (Respondent 217.) Second, a more prevalent orientation, especially in companies experiencing rapid restructuring, wanted the unions to act as service organizations with adequate technical skills to negotiate industrial change. Kelvinator workers, for example, were overwhelmingly disappointed at NUMSAs inability to counter managerial claims in the legal procedures of the companys liquidation. Union membership was here less concerned with militancy and social change than with alleviating the impact of corporate unilateralism: When management brings its statements here, do we have on our benches qualified chartered accountants? NUMSA said they dont have any, so how can

355

they read the financial statements? . . . The union is like our lawyer in this company (Respondent 52.) A survey of COSATU organizers (Denga, Marie, and Yanta 2002) detected a widespread perception that, in order to claim legal protections, the unions must operate on a complex procedural terrain, which limits the scope of workers action. As a result, members become cynical and disillusioned towards activism and apathetic towards participation in union activities, preferring instead a business unionism focused on concrete deliverables. Third, some workers, overwhelmingly concentrated in private companies, actually resented what they saw as excessive union radicalism as it allegedly deterred investors and hampered job creation. At Baldwins, where the memories of the disastrous 1991 strike still lingered, few employees articulated this view in remarkably conservative terms: When there was the strike then I went to work, and they started toyi-toyiing in front of my house, calling me sellout. The problem in South Africa is that unions are too militant: tomorrow Telkom is going on strike, teachers might also go on strike. If we are realistic, we have a government here who says, Look, we know people are suffering but the cake is too small. Telkom has given jobs to thousands of youth who passed their matric and now they want to go on strike, and when they go on strike maybe they will get 10.5 percent, but in the long run, what is it going to happen? Telkom will retrench people because they cannot cope now. . . . But now radicalism is gone at Baldwins and maybe I can even go back to the union, because its more interested in building the company. Before it was just intimidation, now its a free country (Respondent 15.)

356

Fourth, in the least favored option few workers argued for alliances between trade unions and social movements around community as well as workplace issues. Proponents of this alternative were preoccupied with the commodification of social services, cutoffs of municipal utilities, and access to housing. The position sometimes reclaimed the political legacy of national liberation: When we refer to the previous manifesto of the ruling party, we were actually saying there must be a community-driven RDP [Reconstruction and Development Programme,] that simply means that the community should reconstruct and develop their areas, the local councils should be influenced by the community because they are actually serving humanity, they are serving the communities (Respondent 108.) The limited support for social movement strategies among union members confirmed that growing protests for service delivery proved a thorny issue for COSATU and its affiliates. In 2003 CEPPWAWU suspended the leadership of its Wits region, which was organizing laid off members into a movement that the union considered anti-ANC and too close to the AntiPrivatization Forum. The purged unionists formed a new organization, the General Industries Workers Union of South Africa (GIWUSA,) which, adopting an anti-neoliberal profile, started recruiting workers of glass, paper, and chemical factories in the East Rand. In general, however, associational life was uneven among respondents. Less than one fifth were actively involved in various community organizations (stokvels, burial societies, club accounts, self-help groups, womens groups, civic associations) and 14 percent identified religious organizations as their main form of social activity. Few took part in fees and rents boycotts, which they usually saw as ancillary to demands to the ANC and local councilors, rather than as antigovernmental actions. Rather than a consciously articulated political tactic, boycotts 357

were often self-reproducing inherited practices, ordinary manifestations of recalcitrance to commodification, or just an automatic thing, we just dont pay anymore. By now it has become a second nature (Respondent 8.) Community-orientated imageries, however, are not necessarily transparent as they often combine radical political rhetoric with a moderate penchant for selfhelp and entrepreneurialism. As one respondent commented with regard to his township: People say Thokoza is too radical and they cant go there, they say people there arent good, they are too rude. We are trying to improve that. We must try to be one, we have a Simunye project that tries to build houses without the help of the government, or we repair sewerage so that we dont have to call for the municipality to come because we pay people from here in the location. We also address problems of crime and violence without waiting for the government (Respondent 57.) A survey of workers attitudes towards the ANC indicated a peculiar mix of disillusioned resignation about the present and idealistic expectation for the future. Most black workers seem to believe that the policies of the postapartheid government greatly favored the employers and only a small minority (even among COSATU members) maintains a clearly positive opinion of the ANCs delivery record. Yet there is also a strong conviction that p atience will in the long run be rewarded with tangible socioeconomic improvements (Makgetla and Trres 2006: 45-46, 56.) More than one third of my respondents trusted their union over the government in representing their interests, while many of those who supported the government questioned the motives and the morality of its officials. Disillusionment was particularly widespread among municipal employees, for whom public institutions are not just distant sources of promises and policies, but also their own employers: 358

Its my government who I have voted for that is my problem. Because they said they were going to change the things and we would be going to live in harmony. But now because it is our government sitting there on top, they are making us suffer. I see only fear and the crisis, and it is our government who is creating all these things because these people from the municipality are part and parcel of the government (Respondent 89.) Most respondents remained, nonetheless, loyal to the ANC: 36.4 percent were members, a figure almost identical to survey results (Makgetla and Trres 2006: 54.) A further 12.8 percent defined themselves as ANC supporters. Most members also belonged to ANC-allied organizations, like SANCO and the Communist Party. Precarious employment does not, therefore, merely lead to political apathy. Black workers allegiance to the ANC, coupled with their reluctance to see COSATU leave the alliance, reflects indeed the enduring stature of the national liberation movement, but also contradicts a widespread skepticism of the ANCs policy performance: If I continue to support them maybe I can make them make those changes. You see, in the past we were suffering, I could not go to that robot [traffic light], they would catch me when I didnt have a reference book. Now the ANC has fought for us and we are free because of the ANC, that is why I support the ANC. [Yet] when we were fighting we expected to get something, I didnt expect to be treated like this (Respondent 148.) Approximately 60 percent of respondents 48 percent of manufacturing workers and a whopping 88 percent of municipal ones considered themselves dissatisfied with the ANC government. To explain the persistent strength of their identification with the party, however, it would be reductive to refer to historical legacies or the lack of political alternatives. Through 359

their support for the ruling party workers refused to experience the employment crisis as an individual predicament and tried to remind those in power of obligations towards popular constituencies. The simultaneity of disillusionment and consent in attitudes towards the ANC may seem paradoxical, but nonetheless enacts the ambiguities of working-class representations of labor as the harbinger of future stability despite its precarious present. Most respondents blamed the ANCs economic policies for what they considered a chasm between their employment conditions and a decent life. Yet, they continued to see job creation, with little qualification as to the type of occupations, as the main governmental contribution to avoid social collapse. Jobs were the most important policy priority for 31.8 percent of respondents, while 30 percent focused on education and training. Despite widespread complaints with unsustainably high costs of living, only 14 percent emphasized strengthening social programs (housing subsidies, public healthcare, and income support for the unemployed.) Surveys showed that almost 90 percent of workers preferred job creation policies to the expansion of state grants (Makgetla and Trres 2006: 47.) Workers, in short, expect from the ruling party a renewed commitment to employment as a socially stabilizing intervention even when they desire and plan to escape their actual, everyday jobs. Seekings and Nattrass (2005) suggested that demands for job creation are evidence of the poors patience and aversion to radical claims. Their examination of employment as an economic transaction rather than an object of signification forecloses, however, a more complex insight into potentially contradictory representations. Longing for jobs tackles future uncertainties while envisaging a productive status for potentially threatening subjects: young women using child support grants to assert their independence; young men choosing crime or reliance on old360

age pensions to undermine their elders authority; foreign migrants flocking to South Africas social provisions. By rhetorically summoning imaginary jobs, male workers in particular cling on the vestiges of authority that work, no matter how degraded, still grants them. Their emphasis on job creation is not only about economic activity, but also elaborates and rationalizes visions of families, communities, and societies. In a survey the ANC conducted in its Gauteng region, for example, the importance of employment policies among working-class members was correlated to moderate or conservative political orientations, including hostility towards social movements and a greater trust in government and religious organizations than in trade unions (African National Congress 2006.) Workers skepticism towards radical community organizing questions the salience of social movement politics as a response to the shortcomings of governance under the ANC. Widespread local protests over the lack of water, electricity, and housing or the governments failure to provide basic utilities have punctuated the South African social landscape in the first decade of the twenty-first century. Many scholars hailed the rise of organized social movements like the Anti-Privatization Forum (APF) or, in the Durban area, the Abahlali baseMjondolo (shack dwellers) movement (AbM) as alternatives to the worn-out identities African nationalism, labor unions of the antiapartheid struggle (Ballard, Habib, and Valodia 2006; Gibson 2006.) As an outlet for the agency of the unwaged, radical community politics plausibly represents one of many manifestations of the fading promise of work. Its focus on the commodification of elementary necessities of life also provides a potential terrain of engagement for the plight of precarious workers ravaged by the violence of the market. It does not seem, however, that wage earners, even precarious ones, significantly rest their hopes upon structures like the APF or the AbM. Research is, on the other hand, almost absent on the extent to which 361

residents with jobs have joined the thousands of, in media parlance, service delivery protests, or place-specific, sudden insurgencies of the poor where the involvement of formal organizations is usually tiny. Part of the problem is that the category of social movement, dominant among South Africanists as a characterization of community politics, obscures more than it reveals, at least when employed in the sense of politically and/or socially directed collectives . . . focused on changing one or more elements of the social, political and economic system within which they are located (Ballard, Habib, Valodia, and Zuern 2006: 3.) One may in fact wonder how useful such tropes of traditional left activism the self-conscious subject, its unambiguous intentionality, the explicitness and transparency of its objectives are in trying to politically name quotidian significations, singular practices, partially elaborated resentments, and ambivalent engagements with mainstream organizations and institutions, which express the dramatic, subversive gap between everyday desire and the requirements of capitalist employment. A case in point is provided by the attitudes of wage earners towards the ANC, seen simultaneously as a responsible for employment decline and its imagined conqueror, a harbinger of globalization and a shelter from its disorders. Contradictory as they appear, such attitudes destabilize nonetheless production-centered identities as deeply as labor market changes. The unions in their alliance to the ruling party, in particular, seem to lose direction in a mist of rankand-file significations that no longer linearly connect the emancipation of work to activist narratives or a clear progressive horizon. Most workers consider organized labor as a constrained agent for change or a defensive tool against uncertainty; proponents of the unions as lawyers and accountants are probably more numerous than those hoping for militant collectivities of producers. Opinions on organized labors political tasks are similarly fragmented among 362

members wanting to oppose, engage, advise the government, or variable mixes of these options. The contested signification of precarious employment can ultimately tear the labor movement apart as it struggles to adapt to fragmented workers experiences and connections between redemptive narratives of work and the ideological appeals of class and nation become evanescent and indeterminate.

Conclusion

As the third decade of democratic South Africa approaches, waged work which many for long believed to be a vehicle of freedom and social emancipation reveals its volatility as a commodity in a world of commodities. The constitutional guarantees of inclusion seem no longer adequate to protect the body of the worker from systematic exposure to personal and socioeconomic violation. Workers resignification of employment in relation to citizenship determines, therefore, a peculiar tension. On one hand, the anxieties surrounding wage labor speak to its separation from decent material existence, let alone social emancipation. Workers alternative meanings project therefore scenarios of self-entrepreneurship, support networks, or spiritual transcendence that escape wage earning or at least problematize its centrality. On the other hand, however, harsh critiques of actual experiences of work can resurrect it as a metaphor of order premised on conservative families, hierarchical communities, and a nation rewarding masculinity and age. What kind of working-class politics and social compacts can result from expectations for job creation underpinned by melancholic longings for order? The last two chapters indicate that the postapartheid work-citizenship nexus hardly rests, in its present conditions, on values of 363

equality and solidarity. If anything, it is acquiring in diverse social imageries a distinctly conservative flavor. For many workers disillusioned with their jobs, dreams of stable and decent employment reveal both urgings to desert the workplace and appeals to social discipline. Grounding such narratives is wage labors perceived inadequacy in relation to the four interconnected crises of gendering systems, livelihood systems, signifying systems, and value systems that for Ari Sitas (2007) define the postapartheid social question. The possibility, therefore, exists for the normative glorification of wage labor to abet what Sitas fears as a defensive and resentful politics of cultural and moral purity underwritten by chauvinist, sexist, and xenophobic claims. Central to defining postapartheid sociality is, to borrow Allen Isaacmans (1996: 5) expression, a negotiation over cultural understandings of how work should be defined and valued. The dominant work-centered citizenship discourse provides, in the end, ammunitions to authoritarian yearnings in response to imagined threats from impure and anarchic popular practices of insubordination and impoliteness (Ndjio 2005: 266) from the young tsotsi who avoids work, to women claiming control of the households income, to grant recipients and illegal electricity poachers with their entitlement culture. The social and the nation are thus enacted as disciplinary communities to normalize the violence of socioeconomic processes and govern its underlying disruptions. Postapartheid governance has so far aimed to curtail possibilities for its precarious subjects to articulate their own languages, imaginations, and struggles. It rather turned them, in Spivaks (2005: 476) words, into a position without identity . . . where social lines of mobility, being elsewhere, do not permit the formation of a recognisable basis of action. Despite the continuous fragility of employment, the democratic state has tried to mold its citizens into 364

workers-in-waiting, not only or not so much to make their labor power exploitable but, as Paolo Virno (2004) put it, to deflect their dispositions, initiative, creativity, and inventiveness from practices that do not coincide with capitalist work ethic. Postcolonial nationalism has been quite apt to the task because it offered, in response to proletarian inquietude, not just an elite discourse but also a heartening celebration of the poors industriousness and ambition (Cheah 2003.) Left and labor organizations have contributed their own narrative of wage labor as a point where juridical rights, social stability, and personal qualities reinforce each other. Similarly to progressive productivism in Western sociology exemplified in the work of Zygmunt Bauman, Ulrich Beck, and Richard Sennett South African trade unions criticized precarious jobs as undermining a dignity of work that they managed to reclaim even amidst past insecurity and exploitation. The idealization of jobs motivates their opposition to a neoliberal globalization that separates precariousness from dignity so that labor no longer provides meanings and security (Strangleman 2007.) In the process, trade unions unwittingly represented casual and fixed-term employment not as material conditions with legitimate claims, desires, and strategies but as threatening others, unorganizable constituencies, and socially pathological byproducts of market rationality. Organized labors difficulties in signifying precariousness leave South African wage earners in a beleaguered position, thorn between material insecurity and reverence for an unfulfilled ideal of full-time, decent work. Such hesitations give the ANC space to regain popular deference through appeals that massage workers anxieties and authoritarian melancholia. Something along these lines seemed to materialize at the end of 2007, when COSATUs anti-neoliberal stance led to the replacement at the helm of the ANC of Thabo Mbeki with Jacob Zuma, who then became the countrys new president following the triumphant 365

2009 elections. As an object of working-class desire, Zumas leadership emphasized cultural traditionalism, family values, toughness on crime, and a self-consciously masculine persona. Zuma was, on the other hand, quick to argue that, for South Africa to be a nation that has values, school dropouts must be taught by force, girls becoming pregnant to claim social grants should lose custody of their children, and suspected criminals should be denied the presumption of innocence.9 The idea of productive employment underpins these fantasies both as the grieved image of a loss and as the pivot of future social discipline and authority. Work-related melancholia limits, as Judith Butler (1997: 23) observed with regard to melancholia in general, political possibilities inasmuch it thwarts critical self-reflection on what is no more, in my case the idea that under capitalism employment can provide dignity. The political uses of workers melancholia are, therefore, in the end, highly ambiguous. Wendy Brown (2000:23) has, in particular, castigated melancholia as an inflection of emancipative discourse characterized by a mournful, conservative, backward-looking attachment to a feeling, analysis, or relationship that has been rendered thinglike and frozen. The left, Brown continues, often tends to cherish freedoms and entitlements that confront neither the domination contained in both nor the limited value of those freedoms and entitlements in contemporary configurations of capitalism. The continuous, stolid attachment to production and employment in discourses of social justice would then enable critical powerlessness and renewed subjugation. Chauvinist rancor and regressive identities can fill the symbolic space vacated by the decline of employment unless the incommensurability between the official values of work and its actual experiences enables alternative political imaginations. My research indicated that such possibilities are indeed open as workers are remarkably autonomous in signifying their own precariousness not only as a labor market phenomenon but as a process subverting identity, 366

community, and citizenship. As Jean and John Comaroff (2001: 114) reminded, the capacity to act effectively on the world is disproportionately distributed over time, territory, and social space and, hence, demands to be demonstrated in all its labile, dialectical complexity. The effectiveness of state discourse in shaping workers subjectivities cannot, therefore, be taken for granted. The elaboration of precariousness into a challenge for authoritarian and chauvinist identifications, however, requires discarding employment-based understandings of liberation. It is worth reminding that casual and informal work repeatedly destabilized the pretense by African postcolonial nationalism to be an immunization paradigm (Esposito 2008) protecting citizens by coercively defining the nation as a productive organism geared to development and consent. To resume such potentialities, one would have to place the precariousness of employment, rather than its idealized celebration, at the core a new grammar of politics.

367

Conclusion

In the kingdom of ends, everything has either a price or a dignity. Immanuel Kant (1998: 42)

Workers have no time for the dignity of labor. . . . Today, the working class need only look at itself to understand capital. It need only combat itself in order to destroy capital. It has to recognize itself as political power, deny itself as a productive force. Mario Tronti1

As I was writing this book, dramatic events ominously reminded the problematic intersections of work, citizenship, and social identities in South Africa. In mid-2008, a nationwide explosion of xenophobic violence caused more than sixty casualties and the displacement of thousands of foreigners and poor citizens amidst media-amplified hysteria blaming immigration for the countrys lack of jobs. Most respondents in a survey in the Pretoria area favored the deportation of all immigrants, legal and illegal, and one third recommended that even South Africans without productive employment be stripped of their citizenship and expelled from the country (Neocosmos 2009.) Meanwhile, workers in particularly insecure and exploitative occupations embarked in harsh and prolonged struggles, like the 2006 three-month private security strike, where fifty-seven died in clashes involving the police and strikebreakers. Steadily escalating 368

prices of basic necessities, especially food, fueled the desperation of working-class households, which at the end of the 2000s also confronted the local repercussions of a global recession facilitating new waves of layoffs. Finally, protesters over service delivery routinely blocked streets, burnt barricades, and clashed with the police, in large and small communities, expressing plights that often alternated defiance of local ANC politicians with expectant supplications for the new leader, Jacob Zuma, to intervene. Frustrations and resentment in the world of work have greatly contributed to close the Mbeki era, tarnished by its association with neoliberal globalization and the casualization of jobs. The Zuma administration has yet to articulate a clearly defined policy agenda. Its utterances on work and citizenship, however, suggest a vision that does not depart from established scripts and indeed delivers a renewed emphasis on order, discipline, respect for social hierarchy, obedience to family authority, celebration of national purity, and the demonization of conflict. In the end, the vicissitudes of the work-citizenship nexus still interrogate the nature and significance of South African democracy. A strange fate has met the concept of wage labor throughout South Africas history as a capitalist society. It has been the target of resistance, or at least tacit contestation, from many quarters, from peasant cultures refusing proletarianization to the continuous insubordination of urban youth routinely classified as deviants. For many generations of colonized South Africans forced to depend on capitalist employment, the idea of working for wages stood for alienation, disintegration of communities, violation of personhood, the ultimate denial of meaningful activity. Yet, competing official discourses recursively converged, no matter how ferociously they fought each other in other realms, in celebrating wage earning. Cecil Rhodes, representing the mining imperialists of his time, was among the first to associate wage labor to dignity and 369

civility for indolent natives. South Africanist segregation defined waged employment as a terrain where natives wanting to be regarded as workers could prove their mettle. For apartheid governments, working for wages was not only a foundation of exclusive white citizenship, but also a universal moral imperative, teaching citizens and natives alike that social virtue resided in hard work, not in claiming public social provisions. At the same time, however, African nationalism too made of the right and duty to work a central component, gradually more prominent than cooperative production or peasant agriculture, of its rhetoric of self-reliance. For the post-1994 democratic government, finally, employment became the exalted condition of real social citizenship, the basis of family life, the norm of participation in national development, and the cure for the laxitude and moral corruption generated by redistributive claims and the culture of entitlement. The working-class left of non-nationalist origins has also elevated waged work in its imagery of citizenship. In the tradition of independent unionism of the 1970s and 1980s employment was the condition of possibility for a universal, nonracial proletariat on the march to class-based socialist politics first and nationalist ideological alignment afterwards. South Africas militant trade unions joined, therefore, a path well trodden by marxisant paradigms that transposed the wage from a precondition of capitalist exploitation into the basis and ongoing practice of social emancipation. It was, in many ways, a reversal of Marxs own deeply felt horror at wage labor in memorably poetic images like the mass of living labour power, which, having the option of being dependent on the sale of its labour capacity or on begging, vagabondage, and robbery was eventually forced by gallows, stocks, and whippings onto the narrow path to the labour market (Marx 1973: 507.)

370

Yet, employment was more nuanced in ordinary workers discourse than in official or nationalist imagination. Rank-and-file vernaculars often reflected a fundamental ambiguity between celebration and refusal of work. The forest full of wild fruits and dangers, with which imbongi Alfred Qabula symbolized labor struggles, was intended to provide solace not only to employees but also to the homeless and the frightened and those who refused to work for nothing. Labors path through the forest also conveyed a purposeful sense of direction, an unstoppable, necessary movement towards fairness, dignity, and justice: What we have made moves forward When its wheels wear out, our unity jolts it forward When they block it on its way to Capetown it does not lose its power, it roars ahead it grumbles on, with flames and fumes and anger (Qabula, 1989: 70.) Twenty years after Qabulas verses, one might wonder whether seeking wage labors respectability and dignity was the most effective compass to navigate a forest that proved thicker and more treacherous than the apparently forward march of the black proletariat probably assumed. Black workers struggles were decisive in the democratic recasting of state sovereignty and enabled new languages of claims and expectations for the formerly oppressed and disenfranchised. In exchange, however, the conjunction of political liberation and economic liberalization reasserted the imperative to work as a tool to discipline such claims and expectations. Subaltern demands for decommodification, or dignified lives independent from employment status and market relations, were concomitantly marginalized. The normative universality of work has recursively allowed the state to cast social imageries that questioned the 371

imperatives of economic activity as its monstrous other. Metaphors of dependency, workshyness, and entitlement ominously scapegoated, in Jean and John Comaroffs (2004b: 527) words, a phantom proletariat [that,] after all, offers a ready explanation for the scarcity of work, for the impotence of youth and the relative wealth of their elders, for the destruction of the labor market, conventional patterns of reproduction, and the community. It is also a potent metaphor for the emptiness of democracy sans a substantive right to material well-being. Public concerns with recalcitrance to work reflected, on the other hand, a typical problematic of liberal governance: how to incorporate within representative democracy subjects prone to articulate claims in terms of social justice and redress rather than in the abstract constitutional language of rights and responsibilities. The postapartheid state has hence classified the demands of its poor and precarious citizens as legitimate or problematic, respectable or threatening depending on whether they sought recognition within the normativity of work or questioned the structural distribution of power and resources. For liberal apologists of the new South Africa, work-centered citizenship provided a much needed neutral terrain for political argument (von Lieres 2005: 24) and a remedy to a political ethos of monism (Buur 2007: 109) harbinger of anticapitalist sentiments. The discursive function of exhortations to work in governing the conduct of the poor is, if anything, reinforced by the fact that, in practical terms, state policies have been focused much less on creating decent jobs than promoting entrepreneurial initiative and a black empowerment often consisting of financial buyouts rather than productive investment.

372

The decline of employment faces, nonetheless, South Africas rulers with thorny challenges. As it is not merely coincidental with social marginality, because it affects the majority, it cannot be simply controlled through stigma. By attaching moral attributes to work, the state thus disguises the social impacts of employment insecurity and delegitimizes its possibly antisystemic outcomes. It also moves precarious workers from the margins to the core of policy concerns. Work as a normative universal aims, in the end, to address the loopholes of constitutional discourse by enforcing guidance under the guise of empowerment and freedom. It practically molds a predictable community of citizens out of the abstract, volatile humanity of constitutional rights. As Anna Tsing (2005) powerfully argued, however, universals are never self-fulfilling. Universals are rather aspirations marked by an insoluble quandary: to be able to structure material connections they must be recognizable from the standpoint of specific claims, which undermine them as universals. To become meaningful in specific existential contingencies and negotiations, therefore, the imperative to work has to relinquish the avowed universality of its values. The precariousness of employment allows, as a result, contestations over the meanings of work and highlights the incommensurability between multiple social practices and the use of work as a signifier of unity and order in official discourse. Rhetorics of employment, therefore, constantly re-territorialize workers attempts to define productive lives not centered on the workplace. In Deleuze and Guattaris (1987: 220) terms, the entwined forces of the states workcitizenship nexus and workers melancholia act as a conjugation that blocks alternatives and possibly enables new repressive compacts. To respond to the disappointments and frustrated expectations of wage earners, the ANCled government has recalibrated its policy discourse around an image of production as 373

constitutively fragmented and characterized by various degrees of casualness. There, informality can be an empowering asset in a cumulative progress from the second to the first economy. Will such discursive shifts govern the decline of stable employment while continuing to stave off perceived threats of dependency? What political and policy developments do they herald for the postapartheid experiment? In 2006 the government launched the Accelerated and Shared Growth Initiative for South Africa (ASGISA,) aimed at the elimination of the second economy and the absorption of social grants recipients within the framework of a democratic developmental state. By reviving the language of developmentalism the state reclaimed its democratic legitimacy and technical expertise as objectively superior solutions to South Africas social question. In its staunchly modernist imagery, the ANC proclaimed its ability to master the science and art of crafting long- and short-term common platforms to ensure that all the motive forces pull in the same direction (African National Congress 2007c: 12.) Instead of being a realm of competing agendas, politics thus becomes the immanent revelation of an unquestionable objective social contract among business, the labour movement (in all its formations), elements of civil society and government around a South African growth strategy. In a society it conceived as an organism of synchronically moving parts, the ruling party figured itself as the demiurge bringing different interests together into one stream (African National Congress 2007d: 16.) ASGISA tried, nonetheless, to emphasize participation and inclusiveness as correctives to the technocratic market rationality of GEAR, which had rallied opposition to the ANCs neoliberalism. While it did not depart from globalization, the new policy aimed, nonetheless, to incorporate organized labor in cross-class alliances to make it a more reliable actor of governance (Gelb 2006.) Rather than merely deferring to market self-regulation, the government 374

sought legitimacy as the sole external actor capable to harness the tumultuous forces of the capitalist economy and steer a pragmatic middle road, insulated from ideological conflicts and political passions, between unfettered neoliberalism and state dirigisme. The ANC defines itself, to that effect, as a disciplined force of the left [which] seeks to put in place a combination of the best elements of a developmental state and social democracy (African National Congress 2007c: 11.) The rhetoric of job creation is here stronger than ever. The ANC had, however, to recognize that the most likely jobs to be created are in service sectors construction, tourism, call centers where wages are low and unions are weak (African National Congress 2007b: 2829.) Sure enough, the ANC has repeatedly condemned the use of labor brokers, which during 2009 were likened to slave traders and rapists in parliamentary and media debates. The party is, however, mostly concerned with the economic burden of employment intermediaries on companies margins. In alternative to labor brokerage it would therefore prefer accommodating some flexibility in [South Africas] labour regime (African National Congress 2005: 24) by loosening rules on hiring and firing, waiving collective bargaining wage minima, and relaxing provisions on paid overtime. The generalization of precariousness seems ultimately to be the narrow path to the first economy. The ANCs understanding of statecraft, as the objective manifestation of a progressive movement that fuses moral imperatives and technical competence, starkly recalls the approaches of governments that preceded democratization. Like them, post-1994 governance adopts modalities Ferguson and Gupta (2002) define as verticality the idea that the state stands to the social as a higher cognitive and ordering function and encompassment the production of the national body as the spatial embrace of policies with the aim to make relations between 375

state and society predictable and controllable. Capital and labor are no longer contrasting class forces but tools that only the state can arrange for the greater good. Minister Trevor Manuel (2004) provided telling insights into this cognitive framework. In his view social antagonisms are necessarily superseded once the advanced knowledge of governmental agencies for statistical measurement has collected the objective data that drive policies. As state sovereignty is asserted as the only objectively true viewpoint over and above self-interested ideologies, the relationships between data collection, definition of the common good, and policymaking become linear, straightforward, uncontroversial: It is not sufficient to extrapolate from contested data into policy, although the temptation is always present. Using contested data creates the danger that we talk past each other, either because we use different sets of statistics, or because the different sets dont confirm each other. As policy makers . . . we cannot afford to make large mistakes, although smaller mistakes are inevitable and probably necessary to arrive at the right policies (Manuel 2004: 2.) Once the truth and righteousness of policies is guaranteed by the right set of statistics , Manuel continued, only self-serving agendas in bad faith can possibly oppose data indicating that Poverty is more likely if people dont have jobs, a finding that I suspect is common to many countries. . . . Our large number of unemployed would materially benefit move out of poverty if they were able to take employment (Manuel 2004: 3.) On the other hand there is no place, in Manuels axiomatic, for subjects that are poor and yet earn wages. Andries du Toit (2005: 20) lambasted the vitiating reductionism of policies and scholarly arguments that regard the labor market not as a realm of unequal power relations, but a 376

conveyor of normative meanings. The view that labor market participation is the way out of poverty, he continues, ignores not only that many poor already participate, but also that it is how they participate that makes them poor. Despite its rediscovery of left parlance, the ANC is not, however, softening its virulence against the alternative of decommodified programs like the basic income grant, . . . which would neither have the broad or deep impact on poverty eradication nor the broad mobilisation of resources to address diverse aspects of poverty and the well-being of our people. It is important that we do so in the context of our challenges as a developmental state rather than against the ideological backdrop of a welfare state (African National Congress 2007d: 16.) Developmentalism was the ideological protagonist of the ANCs 2007 congress in Polokwane, which culminated in the overthrow of Thabo Mbeki by Jacob Zuma. The event witnessed COSATUs revived ascendancy on the ruling party, but did not question the linear link between development, job creation, and social progress in the ANCs imagination. COSATU has indeed downplayed its insistence on universal social grants and strengthened its emphasis on the role of the state in boosting labor-intensive sectors (COSATU 2007b.) It is, nonetheless, legitimate to wonder whether the ANCs narrative of progress can allay workers anxieties, which were crucial in Zumas rise. The rulers modernist rationality can indeed open up new, unpredictable political potentialities as it marginalizes practices and life experiences incompatible with the discipline of labor. As Dipesh Chakrabarty (2000: 253) commented, the constantly fragmentary and irreducible plural nature of the now is a problem to a social science that formulates human future as a project in which reason is realized in some form or other (more democracy, liberalism, rights, socialism, and so on.) To remind of South 377

Africas unsettling plurality of the now are, for example, surveys showing that a good many of [South African youth] say they would rather not be employed than accepting a low-paying job.2 In what I called the politics of workers melancholia itself, expectations of an ideal order centered on respectable work question the current social status of employment. The melancholic laboring subject often perceives the meaninglessness of wage labor not as an individual predicament but as a bitter betrayal of emancipatory projects once vested in the labor market. In contrast with official praise of personal responsibility, workers recognize their precariousness as a socially determined, politically reproduced condition calling for systemic change. Melancholic feelings can be uttered in conservative tones, but they are by no means deferential to politicians or labor leaders and continue to speak a language of social redress that sits uncomfortably with official exhortations to patiently defer and downsize expectations. An alternative imagery of radical social mobilization does not, for sure, clearly emerge from my research, and unionized workers remain indeed in a rather problematic relationship with new generations of community activism. But the precariousness of employment questions, nonetheless, established power relations, rather than producing fatalism, apathy, and subordination. Rallying cries for job creation are shot through with contestations over what jobs mean, the desires they should address, and the kind of life they are supposed to enable. The postapartheid experiment with employment as a master signifier of social existence remains critically unresolved and open to unintended outcomes and unforeseen conflicts. One may doubt that at the end of the road there will be the patriotic, patient, respectable, hard working, law abiding, morally prudent, and politically moderate poor that the government desires as its true citizen.

378

At stake are, ultimately, the significance of the South African transition and democratic politics as massive inequalities and widespread insecurity turn multitudes into a surplus of bare life. According to the eminent writer, Njabulo Ndebele, The South African of the future will live comfortably with uncertainty because uncertainty promises opportunity. But to get the full richness of opportunity, he goes on, would require to run away from unidimensional characterisations of ourselves . . . nurturing imaginative thinking [and] the ability to embrace uncertainty from a position of intelligence and imagination.3 Market discipline has provided, so far, the hegemonic idiom in the postapartheid signification of uncertainty. By questioning the unidimensional characterization of the citizen as employee, however, South Africas precarious liberation is a condition for the continuous reinvention of the possible.

379

Appendix on Methodology

This book examines how the relationship between work and citizenship emerged as an object of contested discourse and imagination in South Africas transition to democracy. To that effect, I define discourse not merely in representational or ideological terms, but as a material and political practice, following Frederick Coopers (1996: 15) observation related to the labor question in decolonizing Africa that discourse does not merely refer to speech acts but to the range of acts laden with power which establish meanings within specific historical contexts. My research has gathered diverse materials to document the complex modalities through which workers, policymakers, experts, and labor organizations attach meanings to labor, citizenship, social claims, and collective identities. Ethnographic and qualitative insights into workers experiences of sociopolitical change are combined with interviews with experts, officials, policymakers, and company executives and documentary analysis, backed by archival documentation, of official governmental and trade union discourse. The related imageries, rationalizations, narratives, and strategies define postapartheid South Africa as a context in the sense of a mentally represented structure of those properties of the social situation that are relevant for the production or comprehension of discourse (van Dijk 2003: 356.) Apart from creating a synchronic conversation among different sources, this work diachronically relates contemporary debates to a historical discussion of the meanings of labor for South Africas rulers, their subjects, and opponents. Instead of privileging archival research or ethnographic fieldwork, I embraced a certain eclecticism in order to integrate partial, limited, and incomplete discursive acts within an illustration of how a contested social reality produces 380

meaning and sense. On one hand I examined policy guidelines, experts assessments, politicians pronouncements, and trade unions declarations, which routinely imagine specific subjects workers, citizens, unemployed, social movements, and so on to cognitively organize material relations and conceptualize, predict, and guide the conducts of actors. On the other hand, ethnographic fieldwork highlighted autonomous subjective capacities to appropriate, modify, and criticize official categories. George Steinmetz (1993: 53) forcefully argued that, even if institutional paradigms have discursive autonomy they are not just a reflection of social interests or bureaucratic rationality they are also structurally underdetermined as officials and policymakers struggle to match their definitions of social problems to continuous change and unpredictable challenges. The postcolonial situation presents, as Gayatri Spivak (1992: 795) noticed, the added difficulty that the states attempt to signify its subjects often operates through discursive violence and expropriation, producing, instead of citizens, subaltern subjectivities. Subalternity is a condition characterized by the unavailability of speech or translation as ways to ontologically restore as, for example, in the intentions of nationalist politics, liberal citizenship, or workers movements subjects that owe their very existence to disempowering acts of signification. Interviews can therefore only produce incomplete meanings that allude to the respondents precariousness, the intrusions of the state, or the brutality of market forces rather than systematizing or conclusively situating them in a conscious historical narrative. The incompleteness of qualitative interviews, and even discordances, omissions, and adornments in the respondents subjective selection of relevant information, do not, however, invalidate, and indeed may even reinforce, their capacity to establish meanings in the dialogical interaction between the interviewer and the interviewee. The interviewees ability to signify the 381

social world is not, in fact, premised or conditional upon the production of coherent, fully transparent reports. Interviews are not accounts aimed at objectivity, but social encounters where the researcher and the researched are mutually and reflexively constituted and signified (Rapley 2001.) The situational reliability (Watson 2006) of respondents, which includes the deployment of emotions and the selection of information as specifically motivated by the context of the interview (Hoffmann 2007,) counters official discourse and questions its ambitions to organize mundane experiences. The social critique and deconstruction of established ideology and discourse is indeed part of the reflexivity of the interviewer-interviewee practice (Finlay 2002.) The contrast between two necessarily partial epistemic modes institutional signification and everyday discourse is what ultimately defines the relations of work and citizenship as a critically unresolved social dimension of the South African transition. From July 1999 to April 2000 I have conducted open-ended, semi-structured interviews with 220 workers, asking them to elaborate meanings of work and employment in relation to practical needs, ideas of citizenship, biographical narratives, and emancipative visions. I visited two specific working-class communities. Eighty interviews were conducted with employees of the Waste Management and Roads departments of the Johannesburg municipality while in the industrial area of the East Rand (Ekurhuleni,) adjacent to Johannesburg, I collected sixty interviews with workers in three metal and engineering factories and eighty interviews in four plants producing glass and paper goods. Access to workplaces was granted through the three main trade unions for the respective industries, the South African Municipal Workers Union (SAMWU,) the National Union of Metalworkers of South Africa (NUMSA) for the metal and engineering companies, and the Chemical, Energy, Print, Paper, Wood and Allied Workers Union (CEPPWAWU) for the glass and paper companies. Union officials provided assistance in 382

the selection of interviewees. My sample roughly reflected the composition of the workforce in production according to race, gender, occupation, and employment contracts. Of the 220 respondents 214 were black, 209 of which Africans, and six were white. The near totality of industrial workers are male, while approximately one quarter of municipal employees are female in response to the gendered structure of low-level jobs in the sector. Interviews have usually taken place inside company premises, where I could also observe interactions in the production process and union activities. Approximately 15 percent of respondents were interviewed at home. Workplaces and communities also had their singular stories to tell as respondents memories wove specific rooms, streets, and artifacts into narrative networks that represented production as a determined and contested place, rather than an abstract, objective, and neutral space (Biernacki 2005, Latour 2005, Herzog 2005.) My use of a non-probability, snowball technique to select respondents and locales did not aim at statistical significance. My claims to empirical validity rather reflect the extended case method approach, which Michael Burawoy (1991) defines as a mode of theoretical reconstruction where connections between singular narratives produce, through consistencies and ruptures, their own generalizations rather than data to be encased within a predefined interpretative frame. Respondents representations, perceptions, and states of mind are sources of knowledge that need to be rigorously unpacked in their complex, incongruous, and multilayered meanings. Divergences and anomalies problematize, on the other hand, the claims to objectivity and universality in existing frameworks, providing therefore bases for critical counter-theories.

383

Notes

Note to Opening Quote

1

Cit. in Zizek (2005: 212,) from the French original in Deleuze (1987.)

Notes to Introduction
1

Cit. in Vandervoort (1999: 105.) Elaine Unterhalter (2000) finds traces of this contradiction in the autobiographies of many

nationalist leaders who celebrated wage labor as a conduit of solidarity and struggle for the people, yet consciously presented their own individual political work as exemplary of a manly, heroic, non-alienated, meaningful activism antithetical to the dullness of waged employment.
3

Cit. in Cohen, Tim. 1996. Mandela Says Sacrifice is a Part of New Patriotism. Business

Day, June 21: 3.

4

The official criteria to measure unemployment in South Africa have repeatedly changed after

the end of apartheid. There used to be a restricted and an expanded unemployment rate, with the latter counting discouraged jobseekers, not considered in the former. In 2008 the governments official definition of discouraged jobseekers was persons who did not have work, but wanted to work and were available to work but did not seek work or try to start a business in the reference period because: they had lost hope of finding work; or they did not have the requisite skills or qualifications; or they believed that no jobs were available in the area. (Statistics South Africa 2008a: iv.) Since 2004, the governments statistical agency has, however, ceased publishing the expanded unemployment rate and has counted discouraged jobseekers as not economically active, with the effect of lowering the labor market participation rates of working-age citizens and, possibly, the official rate of unemployment. Meth (2007a: 30-33) 384

discusses the governments statistical gimmicks as possibly concealing the actual extent of unemployment.
5

In 2006, then deputy-president Phumzile Mlambo-Ngcuka admitted that even if official growth

targets were met, by 2015 the country will still fall two million jobs short of the governments target of a 15 percent unemployment rate (see Hamlyn, Michael. 2006. SA Will Fall Short by 2m in 2015 Job Creation Push. Business Report, November 7: 3.) To make things worse, in 2009 the country was experiencing negative growth for the first time since the end of apartheid.
6

The same data reveal that the share of workers compensations dropped below 50 percent in

1994, the year of the first democratic elections, and never recovered. Before the 1990s companies managed to compress workers wages through high levels of fixed capital investment, which made many militant black workers redundant. During the 1990s and 2000s, however, fixed capital investment has stagnated or fallen, indicating that to transfer resources from wages to profits employers increasingly rely on hiring vulnerable, exploitable, and contingent workers.
7

Da Costa, Wendy J. 2007. Blanket Grant for Poor Not Envisaged. The Star, February 12: 1. SAPA. 2007. Grants Must Link with Development to Work. Business Day, January 31: 3. Address by Minister of Labour, Mr. Membathisi Mdladlana on the Occasion of the

Commemoration of Kgosi Mampuru and Kgosi Nyabela Mamone Royal Kraal, Limpopo Province, 27 January 2007. http://www.labour.gov.za/media/speeches.jsp?speechdisplay_id=11580. Retrieved September 6, 2008.
10

Similarly to old colonial administrators, Wood justifies his argument with an idea of Africa as

essentially stuck in a pre-modern condition where markets are imperfect, reliance on family networks holds sway, and the locals peasant mentality (Wood 2004: 64) prevents 385

commodified behavior, money-mediated exchange, and productivity. It would be therefore pointless, in his view, to provide Africans with a fiscally funded social security system relying on regular monetary revenues. Such a simplistic view of Africa as a land without commodification, and therefore no need for decommodifying social policies, can be usefully contrasted to Jane Guyers (2004) analysis of commercial cultures and monetary transactions in West African contexts or Janet Roitmans (2005) study of contestation between social movements and the Cameroonian state over the definition and regulation of taxable wealth.
11

Gay Seidmans analysis of trade union politics in South Africa and Brazil provided a unique

comparative perspective. She viewed, however, the development of independent trade unionism almost entirely as a manifestation of black workers class consciousness shaped by the triad of urbanization, industrialization, and authoritarianism. Seidman explicitly dismissed the role of workers cultural and discursive practices; while they may shape the expression of demands, she writes, they need not determine their content (Seidman 1994: 11.) Indeed, very little can be found in her book about the meanings ordinary workers attached to their jobs and their unions. She supports her view of waged work as the central repository of workers identities with probably oversimplified claims, like that African workers would not lightly leave a job (Seidman 1994: 88.) On the contrary, Webster (1985: 197-98,) for example, was careful to underline significantly different attitudes between township residents for which refusal of factory jobs remained common and migrants more vulnerable, therefore readier to accept such jobs, but not necessarily attached to them. Trade unions can protect workers against low wages, managerial despotism, and everyday abuse, but this does not necessarily make employment the main determinant of workers identities.

386

12

Such findings are contradicted by research focusing on urban areas, where residents often have

broader income opportunities and support networks than citizens in deeply impoverished rural settings, and where it appears that contestation and refusal of waged work are far from vanishing. A research on youth attitudes towards work and entrepreneurship in Gauteng, KwaZulu/Natal, and the Western Cape found that approximately one quarter of respondents preferred unemployment to low-wage jobs (Maas and Herrington 2007: 33.)
13

Steinberg, Jonny. 2006. More Welfare Grants would Cause Some To Stop Working . . . And

so? Business Day, February 27: 6. Notes to Chapter 1

1

White policymakers strongly disagreed nonetheless on how to manage African workers

precariousness. Municipalities often decried the small fiscal resources available to implement the 1923 act, whose enforcement remained patchy and incomplete. Twelve years after its adoption, Johannesburg administrators requested more flexible measures to deal with a large landless wage-earning class within purely European areas of a people whose culture is different to ours, and who experience great difficulties in adjusting themselves to our social and economic conditions. See Annual Report of the Native Affairs Department, City of Johannesburg, 1935: 149. DHP, Parnell papers, A.1.2.
2

Contrary to Seekings (2007: 376) claim that the 1928 Old Age Pensions Act was an unusual

and precocious measure reflecting the exceptionalism of the South African case, noncontributory pensions arrived in South Africa later than in other British settler colonies and many independent developing countries. New Zealand introduced a program of this type in 1898, Australia in 1908, Canada in 1927, Argentina in 1904, Brazil in 1923, and Chile in 1924 (Social Security Administration 2007: 32 and 151; Social Security Administration 2008: 30, 61, and 73.) 387

By 1928, all European countries except Finland, Greece, Norway, and Portugal had old age pensions programs (Schneider 1982: 199.)
3

The Native Problem. Article in The Cape Times by the Rev. Abner Mtimkulu, May 30,

1924. In Karis and Carter (1972: 216-18.) See also Vahed (1998.)
4

Seekingss (2008: 536) view of the Carnegie Commission as a temporary, inconsequential

derailment that did not stop the growth of the welfare state is, however, simplistic. As I later show, even if policy debates would momentarily swing towards welfare state solutions, actual social programs did not substantially depart, before or after 1948, from the Carnegie commissions emphasis on waged work and hostility towards redistribution.
5

Report of a Deputation from the ANC and the Congress of Urban Advisory Boards to the

Minister of Native Affairs, May 15-17, 1939. In Karis and Carter (1973: 138-45.)
6

Statement on the African Worker, by Dr. A.B. Xuma, January 14, 1946. In Karis and Carter

(1973: 262.)
7

Ibid. Seekings and Nattrass (2005: 171) are, however, mistaken in their speculation that official

concern with refusal of work abated in the 1960s as policing improved, Bantu education put idle youth to school, and better paying jobs became available. In fact, as I show later in the chapter, the governments concern with black work avoidance kept growing throughout the 1970s.
9

Programme of Action. Statement of Policy Adopted at the ANC Annual Conference,

December 17, 1949. In Karis and Carter (1973: 337-339.)

10

Resolution of Central Committee, CPSA, 29 December 1938 2 January 1939. In Davidson

et al. (2003: 294.) 388

11

Isisebenzi 2: 1, 1975. DHP, FOSATU papers, C.3.19.1.2.3. Minutes of the Special TUACC Secretariat Meeting, 21 November 1977. DHP, FOSATU

12

papers, B.4.4.
13

History of Wages Commissions, n.d. (but 1979). MAD, NUSAS papers, B1ii/1982. A Brief History of the TUACC Unions Relationship with TUCSA, TUCSA Affiliate Unions,

14

and Unregistered Unions, May 14, 1975. DHP, FOSATU papers, B.12.2.2.1, Secretariat Report to the Second TUACC Council Meeting, 1 December 1974. DHP, FOSATU papers, B.4.2, and TUACC Secretariat Meeting, 31 January 1977. DHP, FOSATU papers, B.4.4.
15

The quote is from The Labour Situation in South Africa, an undated mid-1970s document by

organizers Foszia Fisher and Harold Nxasana.

16

Alec Erwin, one of the white leaders of the nascent trade unions, underlined the strategic

importance of the factory and waged work as conduits for nonracial class identities in alternative to the focus on black opposition to white privilege in the black consciousness movement. In his view a factory worked more than on just whiteness and in the factory black workers can see that theres something bigger than just the whites that were suppressing them (cit. in Frederikse 1990: 138.)
17

An illuminating example is provided in a 1975 anonymous article: A Movement of Unskilled

Workers. Abasebenzi 3: 3-5. Abasebenzi was the newsletter of the University of Cape Town Wages Commission, later of the Western Province Workers Advice Bureau. The article argued that the interests of black and white workers are basically the same, concedin g at most that skilled and unskilled workers may have different demands. In the article unskilled stood for black. While race was a valid social distinction only as long as translated into occupational

389

categories, the author concluded, the racial question . . . is not a fundamental cause (own emphasis) of social divisions.
18

Andrew Nash (2009: 169-72) provides an insightful discussion of Turners influence on the

independent unions view that a focus on shopfloor struggles did not amount to avoiding political engagement, but rather brought the working class to life as an autonomous democratic and socialist political identity, whereas all other workers movements in South Africas history had fallen under the hegemony of other popular forces.
19

In 1979 former whites-only housing loans and land purchase subsidies provided by the

governments National Housing Corporation were made available to Africans (Kruger 1992: 10.)
20

FOSATU Discussion Document on Pensions, n.d. (but 1981). DHP, FOSATU papers,

C.10.8: 6.
21

In 1980 women, mostly African and colored, constituted 24.6 percent of the total

manufacturing workforce (Gibbs 2007.)

22

Katlehong Local Shop Stewards Stand on Demolition of Shacks, n.d. (but 1983). DHP,

FOSATU papers, C.3.13.

23

Ibid. In the 1970s the term umjondolo started to replace umkhukhu to define a shack.

Umjondolo is perhaps derived from umjendevu (old spinster,) denoting the idea that elderly women expire like the building materials used to build shacks. The use of umjondolo may therefore reflect, as a factor associated with the proliferation of informal accommodation, the growing number of African female-headed households illegally residing in the cities. I am grateful to Mark Hunter for this suggestion.
24

Ibid.

390

25

TGWU Report on Special Central Committee Meeting Held at Wilgespruit on 11.8.84. DHP,

Jane Barrett papers, A.7.2.

26

Second National Congress of FOSATU, St.Peters, Hammanskraal, 10th and 11th April 1982.

DHP, FOSATU papers, C1.8.2: 24.

27

Sweet, Food and Allied Workers Union Discussion Paper, n.d. (but 1985). DHP, FOSATU

papers, C.1.13.8.3.4: 4.
28

Naidoo to Foster, September 5, 1984. DHP, Jane Barrett papers, A.6.2. Erwin, Alec. 1983. Trade Unions and the Community in 1982. An Assessment. DHP, Jane

29

Barrett papers, B.4.8.

30

Federation of South African Trade Unions Policy on a Living Wage, 1981. DHP, FOSATU

papers, C.1.6.1.
31

Anonymous. 1990. LRA/Living Wage Conference Decisions and Recommendations.

COSATU News, June: 1-2; Report from Goods and Services to the COSATU Economics Workshop 3-4 April 1991. DHP, SAHA, M.4.1.9.
32

Anonymous. 1987. Jobs for All at a Living Wage. COSATU News, March: 7; Anonymous.

1988. Campaign for Jobs or Living Benefits NUWCC. COSATU News, May: 14. Notes to Chapter 2
1

Naidoo to All Affiliates and COSATU Regions. July 17, 1990. DHP, SAHA, M.4.1.15. Ibid. A Framework for Economic Reconstruction. Proposals for a COSATU Economic Policy.

N.d. (but 1991.) DHP, SAHA, M.4.1.6.

4

COSATU Economic Policy Workshop. Johannesburg, 1-2 November 1990. DHP, SAHA,

M.4.1.15: 3. 391

Ibid.: 5. Hirschs elaborations, based on official statistics, indicate that while from 1995 to 2003

employed workers have risen from 9.5 to 11.5 million, those in formal occupations have declined from 8.3 to 8.2 million. Another study (Ligthelm 2005b: 14-20) shows that while the countrys GDP rose by 68 percent from 1979 to 2004 (at constant 2000 prices,) formal jobs declined by 8 percent (13 percent for the period following the 1994 elections.) As the absorption rate (workingage population employed) in the formal sector has declined from 39.1 percent in 1990 to 27.7 percent in 2005, the share of informal employment among all South African workers has risen from 16.4 to 20.6 percent over the same period (compared to 7.7 percent in 1980.)
7

Marlea Clarke (2006: 386-88) emphasizes similar problems with the official definitions and

measurement of full-time and contingent employees, which are based on weekly working hours (thirty-five hours or more are required to be defined as permanent) and not on the nature of the contract of employment (permanent, fixed-term, or casual.)
8

Altmans figures show that 55.6 percent of manufacturing employees are poor 38 percent of

which earned less than R 1,000 per month as well as, remarkably, almost two thirds of craft and related workers, a poverty share slightly higher than among service and shop workers and almost identical to plant and machine operators. Other data referring only to employees in the formal sector found that, in 2006, 43 percent received wages lower than R 2,500 per month (Elsley 2007: 82.)
9

Figures in Casale, Muller, and Posel (2005: 19) indicate that working poverty has grown much

faster in the informal than the formal economy. In the former, in fact, the number of workers earning less than two dollars a day rose from 18 to 42 percent from 1995 to 2003. The number of

392

employed workers earning less than the minimum wage for domestic work (the lowest statutory minimum wage in South Africa) rose from 17 to 30 percent over the same period.
10

Figures

reported

by

the

South

African

Savings

Institute,

cited

in

http://www.pretorianews.co.za, retrieved on July 5, 2008.

11

The South African labor legislation does not provide a definition of casual employees,

which is left to the sectoral determinations the Department of Labour issues to establish minimum conditions of employment by industry. The sectoral determination for private security is the most stringent in this regard, and defines as casuals all employees working for less than twenty-four hours in a week (Godfrey and Clarke 2002: 6.)
12

Secretary for Labour to General Secretary, National Industrial Council for the Iron, Steel,

Engineering and Metallurgical Industry, December 5, 1973. DHP, NICISEMI, E.6.2.

13

National Industrial Council for the Iron, Steel, Engineering and Metallurgical Industry to

Commission of Inquiry into Labour, n.d. (but 1977.) DHP, NICISEMI, E.6.1.
14

Memorandum Submitted by the National Industrial Council for the Iron, Steel, Engineering

and Metallurgical Industry to the Department of Manpower on the Proposed Regulation of the Activities of Labour Brokers, n.d. (but 1981); De Jager (NICISEMI) to van der Merwe (Department of Manpower,) October 14, 1981. DHP, NICISEMI, E.6.2. The companies case prevailed in the Labour Relations Act, which defined labor brokers as employers. Companies explicitly used labor brokerage as an anti-union strategy even if they recognized the economic inconvenience of becoming dependent, especially for skilled workers, on brokers charging exhorbitant [sic] rates. See Beukes (Transvaal Regional Council, NICISEMI) to General Secretary, NICISEMI, March 7, 1991. DHP, NICISEMI, E.6.4.

393

15

Representations to the National Manpower Commission by the National Industrial Council

for the Iron, Steel, Engineering and Metallurgical Industry Concerning the Consolidation and Amendment of the Labour Relations Act, No. 28 of 1956, 1990: 3; Cheadle, Thompson & Haysom to Koekemoer (NICISEMI), June 12, 1990. DHP, NICISEMI, E.6.4.
16

General Review of the Labour Broking Industry. April 1, 1991. DHP, NICISEMI, E.6.4. Central collective bargaining declined in the 1990s. By the mid 2000s, only one fifth of

17

workers were covered by bargaining institutions and a further 2.9 percent enjoyed extensions of agreements as non-parties. Such figures were above mid-1980s levels, but mainly because legislation now recognized the right to negotiate for previously excluded public employees. Even if almost 60 percent of manufacturing workers can bargain collectively beyond the workplace, 80 percent of them belong to only five bargaining councils mostly dominated by large, capitalintensive companies (Godfrey, Maree, and Theron 2006a.)
18

Beukes (Metal and Engineering Industries Bargaining Council) to Council Representatives,

February 8, 2002. DHP, NICISEMI, E.6.4. The bargaining council agreed to regulate labor brokerage, which included turning workers hired through brokers into regular employees of the contracting firm after three months.
19

In a 1991 survey of COSATU shop stewards, to the question how should social services be

provided? 56 percent answered by increased taxes and free social services (Pityana and Orkin 1992: 68.)
20

Bengeza Mthombeni, NEDLAC labor representative. Authors interview, June 12, 2000. Manuel, Trevor. 2006. A Delicate Balancing Act. Sunday Times, August 13: 6. The quote is from the decision in Soobramoney v Minister of Health (KwaZulu-Natal), CCT

21

22

32/1997, 31, with which the Constitutional Court rejected an appeal against the denial by a state 394

hospital of life-saving medical treatment to a patient on the ground that limited resources imposed preferential access for individuals with higher survival probabilities.
23

To exemplify the parallels between this line of argument and the disciplining of labor in newly

independent Africa it is worth quoting the labor minister of newly independent Guinea, Camara Bengaly (in Cooper 1996: 418): The workers, without renouncing to any of their rights but convinced of the necessity to use them in good earnest, will go through a reconversion to become the precious collaborators of the authentic elected authorities of the people. . . . [The] orientation of our trade union movement must necessarily correspond to the general policies desired by our populations. Any conception of trade unionism contrary to this orientation must be discarded, and courageously fought in order to be eliminated definitively.
24

Comprehensive Labour Market Commission. Minutes of Plenary 10th-12th February 1996,

19. DHP, Jeremy Baskin papers, L.1.1.

25

Cynthia Alvilar, Acting Chief Director for Labour Market Policy, Department of Labour.

Authors interview, June 13, 2000.

26

COSATU Warns Government of Major Battle over Labour Laws. SAPA Newswire, July 27,

2000.
27

Cit. in Bisseker, Claire. 2007. Jobs for Africa. Financial Mail, October 27: 8. The unionization rate in the public sector jumped from 43 percent in 1997 to 60 percent in

28

2003 (Braude 2004: 53,) when public sector workers constituted 41 percent of COSATUs membership, up from 17 percent in 1994 (COSATU 2003: 1.3.) In 2006, COSATU organized 55 percent of South Africas overall union members, and 80 percent of black union members. 395

Approximately 70 percent of COSATU members were Africans and 7 percent whites (NALEDI 2006: 40.)
29

Comprehensive Labour Market Commission. Minutes of Plenary 1st-4th April 1996, 50.

DHP, Jeremy Baskin papers, L.1.1. Notes to Chapter 3

1

Representative of this line are the important works of Jeremy Seekings and Nicoli Nattrass.

Their contention is that low-wage, precarious jobs are solutions to, and not determinants of, poverty in South Africa. They recognize that the system of social protection is uneven and skewed to the advantage of the formally employed (Seekings and Nattrass 2005: 364-66,) but reject universal redistributive provisions as undue fiscal burdens on investment and corporate profits. Curiously, they also allege that workers themselves dislike universal social grants as leeways to higher taxes, an unproven claim, which disregards the fact that two thirds of South African workers are too poor to pay income taxes (Altman 2006.) Seekings and Nattrass (2005: 367-68) recognize that only ten percent of taxpayers provide sixty percent of the fiscal revenue, but argue that the small tax share of workers is due not to their poverty, but to their political power. In line with their overall view of South African workers as privileged free riders of political connections and collective bargaining, Seekings and Nattrass conclude that, as higher taxes on the rich are anathema, necessary sacrifices to uplift the poor must fall on wage earners. They claim that unionization is a major cause of unemployment and poverty as it drives wage costs upwards, encouraging layoffs, and connects union households to vacant jobs, outcompeting the limited social capital of the poorest of the poor (Seekings and Nattrass 2005: 282.) A step they recommend is to remove contractual minimum standards and union protections from labor-

396

intensive sectors, which would facilitate the creation of low-wage occupations (Nattrass and Seekings 1997b.)
2

In 2006, after a considerable expansion in coverage following the 2001 legislation, 421,000

unemployed received UIF benefits out of an officially recognized unemployed population of 4.2 million (Department of Labour 2007: 21.)
3

Shadrack Mkhonto, Unemployment Insurance Fund commissioner. Authors interview,

October 2, 2000.
4

Dinga Sikwebu, NUMSA education officer. Authors interview, April 18, 2000. Ibid. Jan Mahlangu, CEPPWAWU National Benefits Coordinator and coordinator of the Chemical

Industry National Provident Fund. Authors interview, April 12, 2000.

7

Starting in 2008 the government has begun to lower the pensionable age of male recipients in

view of its gradual equalization with female ones to 60 years by 2010.

8

The massive increase in CSG beneficiaries is the main cause for growing social grant

eligibility. CSG recipients went from zero in 1997 to about 800,000 in 2001, and more than seven million in 2006. SOAP recipients remained more or less stable at around two million from 2001 to 2006, while recipients of disability grants also increased substantially from about 600,000 to approximately 1.3 million (South African Institute of Race Relations 2007: 353.)
9

Mtyala, Quinton. 2007. Grants Used to Support Tik Addiction. Cape Times, October 8: 3. Cit. in Sefara, Makhudu. 2004. Bigger Grants would Discourage Job-Seeking. Cape Times,

10

February 20: 4.
11

See Daniels, Linda. 2006. Social Grant System Cant Be Sustained. The Mercury, August

21: 1. 397

12

See Pressly, Donwald. 2007. Pension Means Test Should Go Minister. Business Report,

September 9: 5.
13

Alex van den Heever, Council of Medical Schemes and Department of Health. Authors

interview, November 29, 2000. See also Department of Health (1997b.)

14

Alex van den Heever. Authors interview, November 29, 2000. See African National Congress, Department of Local Government, National Housing Subsidy

15

Scheme. Discussion Document. Draft 4, May 1994. DHP, PLANACT, 25.24.3.5.

16

Hansard, February 28, 1996: 240. Hansard, May 16, 1995: 1290. Ibid.: 1325. Ibid.: 1268. Ibid.: 1271. Hansard, May 25, 1995: 37. See National Housing Forum, Co-ordinating Committee Third Strategic Workshop.

17

18

19

20

21

22

Glenburn Lodge, Muldersdrift, November 3-5, 1993. DHP, PLANACT, 25.25.10.3.

23

National Housing Forum, Submission to the Proposed National Housing Strategy. October

4, 1994. DHP, PLANACT, 25.11.2.

24

Petal Thring, Graduate School in Graduate and Development Management, University of the

Witwatersrand, Johannesburg. Authors interview, May 22, 2000.

25

Hansard, October 19, 1994: 3297. Hansard, March 28, 1995: 368. Hansard, May 16, 1995: 1215. Hansard, March 28, 1996: 931. 398

26

27

28

29

Hansard, June 20, 1995: 3028. Hansard, March 28, 1996: 1232. Hansard, June 20, 1995: 3028. Frances Lund, oral presentation at the panel Do State Pensions and Grants Create

30

31

32

Dependency? Human Sciences Research Council, Pretoria, November 17, 2007.

33

Leila Patel, former Director General, Department of Social Welfare and Population

Development. Authors interview, June 22, 2000.

34

Ibid. Neil Coleman, chair of the COSATU Parliamentary Office. Authors interview, October 24,

35

2000.
36

Ibid. Fiona Tregenna, COSATU Parliamentary Office. Personal communication, March 27, 2002. Baskin, Jeremy. 1997. Pay the Citizens of South Africa. Mail & Guardian, January 24: 7. Neil Coleman. Authors interview, October 24, 2000. Neil Coleman. Authors interview, December 1, 2000. Anusha Makka, coordinator of the NEDLAC Public Finance and Monetary Policy Chamber.

37

38

39

40

41

Authors interview, November 7, 2000.

42

Ravi Naidoo, director of the National Labour and Economic Development Institute

(NALEDI). Authors interview, October 15, 2000.

43

Bengeza Mthombeni, NEDLAC labor representative. Authors interview, June 12, 2000. Michelle de Bruyn, coordinator of the NEDLAC Labour Market Chamber. Authors interview,

44

November 7, 2000.
45

Oupa Bodibe, COSATU secretariat coordinator. Authors interview, December 5, 2001. 399

46

Ibid. Shipman (NEDLAC business representative) to Koestlich, January 29, 2000. Authors

47

collection.
48

Abie Dithlake, South African National NGO Coalition, oral testimony at the hearings of the

Committee of Inquiry into Comprehensive Social Security, Pretoria, October 10, 2000. Recorded by the author.
49

Viviene Taylor, chairperson of the Committee of Inquiry into a Comprehensive System of

Social Security for South Africa. Authors interview, August 8, 2001.

50

Commission of Inquiry into Social Security. Minutes of the June 1 and June 2, 2000 meeting.

DHP, Anthony Asher papers, A.2.

51

Transcripts of the Committee of Inquiry into Comprehensive Social Security. Meeting of 6

October 2000. 17. Authors collection.

52

Transcripts of the Committee of Inquiry into Comprehensive Social Security. Meeting of 19

and 20 April 2001. Vol.2: 230. Authors collection.

53

Transcripts of the Committee of Inquiry into Comprehensive Social Security. Meeting of 19

and 20 April 2001. Vol.1: 93-94. Authors collection.

54

Transcripts of the Committee of Inquiry into Comprehensive Social Security. Meeting of 19

and 20 April 2001. Vol.2: 196. Authors collection.

55

Ibid.: 238-40. Fezile Makiwane, Director for Social Security of the Department of Social Development.

56

Authors interview, October 16, 2000.

57

Transcripts of the Committee of Inquiry into Comprehensive Social Security. Meeting of 19

and 20 April 2001. Vol.2: 248. Authors collection. 400

58

Ibid.: 217. Viviene Taylor, Authors interview, August 8, 2001. Editorial. 2007. The Risks of a Shift to State Welfarism. Financial Mail, February 23: 3.

59

60

The magazine had previously endorsed the basic income grant as a modest cash payment to enhance mentioning school vouchers in the United States individual choice against welfare paternalism (Lee, Robin. 2004. Say Goodbye to Welfarism. Financial Mail, October 29: 16.)
61

Michaels, Jeremy. 2004. Social Grants Getting Too Costly Manuel. Cape Times, October

27: 4.
62

Ensor, Linda. 2006. Skweyiya Calls for Basic Income Grant for Poor. Business Day,

November 10: 3.
63

SAPA. 2007. Grants Must Link with Development to Work. Business Day, January 31: 3. Gallagher, Christina and Sheree Roussow. 2007. Social Grants Given Dependency Shake-

64

Up. The Star, January 27: 6.

65

Da Costa, Wendy J. 2007. Blanket Grant for Poor Not Envisaged. The Star, February 12: 1. Steinberg, Jonny. 2005. Works Programmes are Not the Way Out of Permanent Poverty.

66

Business Day, October 31: 6.

67

Oral presentation by Lundi Rasmeni to the Social Development Portfolio Committee,

Parliament of the Republic of South Africa, Basic Income Grant. Briefing and Economic Feasibility. November 13, 2002. http://www.pmg.org.za/minutes/20021112-basic-incomegrant-briefing-and-economic-feasibility, retrieved on December 16, 2004.
68

Kaffers is a highly derogatory and racist Afrikaans term referred to Africans. Cit. in Social Development Portfolio Committee, Parliament of the Republic of South Africa, Income Grant. Briefing and Economic 401 Feasibility. November 13, 2002.

69

Basic

http://www.pmg.org.za/minutes/20021112-basic-income-grant-briefing-and-economicfeasibility, retrieved on December 16, 2004.

70

Minister Manuel also complained that land was laying fallow in rural areas because people

were getting social grants. Cit. in SAPA. 2008. Poor Must Show Sense of Responsibility. SAPA Newswire, May 22.
71

Office of the President. Media Statement: Memories of the Future. South Africa Scenario December 10, 2003. http://www.10years.gov.za/review/statements/10dec03.htm.

2014.

Retrieved on November 20, 2008.

72

Musgrave, Amy. 2008. COSATU Opposes Wage Aid. Business Day, February 22: 2. Cit. in Benton, Shaun. 2006. Subtle Shift from Welfare to Responsibility. BuaNews,

73

February 16.
74

Diat von Broembsen, Chief Director of Policy Planning, Department of Housing. Authors

interview, August 7, 2001 and Lucky Mphafudi, Chief Town and Regional Planner, Human Settlement Policy Directorate, Department of Housing. Authors interview, May 8, 2001. Notes to Chapter 4
1

FOSATU Annual Report 1982. DHP, FOSATU papers, 1.9.1, 17. Interview with Organiser of Glass & Allied W.U. (GAWU) and of MAWU as well as the

Exec. of the Germiston Local s/s Council, 7/4/82. Handwritten notes. DHP, Jeremy Baskin papers, R.
3

Discussion Draft. Tsakane Discussions, Notes for Proposed Agreement. DHP, PLANACT

papers, 26.8.4.

402

Eastern Regional Services Council, Regional Planning Land Delivery Study. Phase1, Volume

1: Land Delivery Framework in the East Rand, Report No. 30/93. June 1993. DHP, PLANACT papers, 17.7.24.3.
5

Town and Regional Planning Practice III. Project 2 1990. Wattville/East Rand Housing

Project. DHP, PLANACT papers, 17.7.24.2.

6

Divisional Management Committee 14 April 1992. Project Name: Etwatwa Extension 8.

DHP, PLANACT papers, 25.15.2.

7

Non-European Affairs. DHP, Katlehong papers, Reel 9, 1. Katlehong Advisory Board, Shortage of Housing Accommodation. DHP, Katlehong papers,

Reel 4; Katlehong Town Council, Minutes of the Thirteenth Ordinary Meeting of the Council Held on Tuesday 27 October 1981, 7.5.53. DHP, ERAB papers, C.13.1.2.
9

The Proposed New Pass Bill, n.d. DHP, FOSATU papers, C.3.6, 2. MAWU Transvaal Branch Report to MAWU National General Meeting on the 22nd June

10

1985. DHP, MAWU papers, H.5.2, 4.

11

The Proposed New Pass Bill, n.d. DHP, FOSATU papers, C.3.6, 4. Workers in the Community. Paper presented by FOSATU Springs Local Shop Stewards,

12

n.d. (but 1984). DHP, Jane Barrett papers, B.4.7, 1.

13

Ibid., 2-3. Ibid., 6. Figures based on Metal and Engineering Industry Bargaining Council data provided by Elias

14

15

Monage, NUMSA, April 20, 2000.

16

Authors elaboration based on Statistics South Africa, National Census Database, 1996 and

2001. 403

17

Construction statistics confirm a shift towards service activities in the East Rand/Ekurhuleni

economy. From 2001 to 2003 new industrial buildings declined by 9 percent, while commercial ones increased by 107 percent. By square meters, industrial space increased 12 percent compared to 167 percent for commercial space (Ligthelm 2005a.)
18

George Magaseng, NUMSA Springs organizer. Authors interview, November 11, 1999. Ibid. Elias Monage, NUMSA National Organizer for the Engineering Industry. Authors interview,

19

20

April 20, 2000.

21

The ANC government has generally been unsympathetic to labor brokers, seeing them at best

as an understandable response by employers to legal burdens on hiring and firing. The party has rather argued for enhanced labor market flexibility as a way to reduce the need for labor brokers (Benjamin 2006: 45-46.)
22

George Magaseng, NUMSA. Authors interview, November 11, 1999. The membership figure for 2000 is based on the authors elaboration of data for NUMSA

23

locals provided by Meshack Robertson, NUMSA regional organizer for Wits East. For figures from 1996 and 1989 see, respectively, NUMSA (1996) and NUMSA (1989.)
24

Authors elaboration based on data on housing subsidies approved for the East Rand from

1995 to 2001 in Department of Housing, National Housing Subsidy Database.

25

Sanjee Singh, Assistant Director for Policy and Evaluation, Gauteng Department of Housing.

Authors interview, November 1, 2000.

26

39th Annual Report of the Cape Town Municipal Workers Association for the Year Ending

31 December 1981. DHP, CTMWA, A.1939.

27

Trade Union Summit, April 9 & 10, 1983 in Cape Town. DHP, FOSATU, C.3.16.1. 404

28

Trade Unions Meeting: 3 July 1982 Port Elizabeth, 11:00 a.m. DHP, FOSATU, C.3.16.1. Ibid. Makgane Thobejane, labor relations specialist, Greater Johannesburg Metropolitan Council.

29

30

Authors interview, March 16, 2000.

31

Victor Mhlongo, international relations officer, South African Municipal Workers Union

(SAMWU). Authors interview, May 19, 2000.

32

SAMWU Launches Opposition to the Restructuring of Johannesburg. SAMWU Press

Statement and Briefing Document, April 7, 1999. Authors collection.

33

Cit. in Haffajee, Ferial. 2000. Roger Ronnie Flies the Workers Flag. Financial Mail,

December 8: 21.
34

Hlubi Biyana, Johannesburg branch organizer, SAMWU. Authors interview, April 17, 2000. Ibid. Makgane Thobejane, Authors interview, March 16, 2000.

35

36

Notes to Chapter 5
1

Dorbyl Transformation Minutes, Ndlovu to Girodor et al., May 28, 1997. Authors

collection. Fanagolo refers to the language once used by white supervisors to instruct their African subordinates.
2

Dorbyl TRC Companies Reports (SPI, DHE Vereeniging, Roll Work Vanderbijlpark,

Dorbyl Marine, DSP Containeering, Mechanical Products, MSP and Venco, October 11, 1997; Climax, November 28, 1997; McCarthy Mine Machinery, December 2-3, 1997; Baldwins, March 3, 1998; Busaf, n.d.). Authors collection.
3

Osborne Galeni, NUMSA Dorbyl Group National Organizer. Authors interview, August 11,

1999. 405

Ibid. Coopers experiment was not destined to last. During the 2000s Dorbyl faced a

precipitous decline, and a management buyout of Baldwins Steel and another steel merchant created a separate company, Kulungile Metals, with a 49 percent black empowerment shareholding. See Mawson, Nicola. 2005. Big Steel-Sector BEE Deal. Engineering News, March 4: 13 and Khuzwayo, Wiseman. 2004. Pamodzi Gets Grip on Steel Sector. Business Report, February 12: 6.
5

Works Manager, Baldwins Steel Brakpan. Authors interview, October 27, 1999. Commission for Conciliation, Mediation and Arbitration, a statutory body for the consensual

resolution of labor disputes established by the 1995 Labour Relations Act.

7

Works Manager, Baldwins Steel Brakpan. Authors interview, October 27, 1999. Minutes of Dorbyl National Shopstewards Council. April 13, 1996. NUMSA papers. Logistics Manager, Paperlink Germiston. Authors interview, April 5, 2000. Authors elaboration based on membership figures provided by John Appolis, CEPPWAWU

10

Wits regional secretary.

11

Andile Nyambezi, Germiston Local Organizer for the Chemical, Energy, Paper, Printing,

Wood and Allied Workers Union (CEPPAWU). Authors interview, April 7, 2000.
12

In smaller companies managerial control seemed harsher and unionization more difficult. In

2000 I faxed questionnaires to managers and union representatives in some of these workplaces. A CEPPWAWU shop steward wrote that in his company bosses have become stricter on discipline after we joined the union. . . . We cant afford walking around. Among the main reasons for workers to unionize he cited unpaid overtime and the lack of health benefits and retirement provisions. Another company hired half of its workers as casuals, with wages at one third the national minimum for the business and without benefits. A manager there wrote that 406

strike actions imposed him to hire a more insecure workforce and blamed productivity problems on outside influence, once they joined the union.
13

Human Resources Manager, PFG Building Glass. Authors interview, March 14, 2000. Mathews, Charlotte. 2004. Consol to Be Unbundled from AVI and Relisted. Business Day,

14

October 13: 12.

15

Human Resources Manager, Consol Glass. Authors interview, April 4, 2000. Company information available on http://www.wiegand-glas.de/index.php?id=461&L=1,

16

retrieved on January 18, 2009.

17

Andrew Zulu, Nigel Local Organizer for NUMSA. Authors interview, July 13, 1999. Human Resources Manager, Union Carriage and Wagon. Authors interview, November 3,

18

1999.
19

Venter, Irma. 2008. Train Builder Mulls Big Expansion as South Africas Rail Ambitions

Gain Traction. Engineering News, October 10: 23.

20

Director, FEDMO Employment Services. Authors interview, October 21, 1999. Ibid. General Secretarys Report to the MAWU NEC 6/10/84. DHP, FOSATU, C.1.13.3.5.5. MAWU Transvaal Branch Report for the Period July to October 1985. DHP, MAWU,

21

22

23

H.5.2. Greg Ruiters provided the data on MAWUs membership before the 1984 split.
24

McCulloch v Kelvinator Group Services of SA (Pty) Ltd. Industrial Law Journal 1998:

1399-1413.
25

Zweli Livi, Alberton Local Organizer for NUMSA. Authors interview, September 16, 1999. Koch to Ngwenda, August 4, 1997; Koch to Erwin, July 30, 1997. Authors collection.

26

407

27

Minutes of the Retrenchment Consultation Meeting, August 23, 1999; Raphoto to Livi,

August 25, 1999. Authors collection.

28

Financial Director, Kelvinator South Africa. Authors interview, October 20, 1999. Simon Koch, CEO, Kelvinator South Africa. Authors interview, November 17, 2001. Alan Agaienz, Project Manager, Johannesburg Roads Agency. Authors interview, January 27,

29

30

2000.
31

Roelf de Beer, Executive Officer, Waste Management, Western Municipal Local Council.

Authors interview, October 19, 1999.

32

Manager, GJMC Central Roads Depot. Authors interview, February 16, 2000. Manager, GJMC Waterval Waste Depot. Authors interview, February 17, 2000. Roelf de Beer, authors interview, October 19, 1999. Alan Agaienz, authors interview, January 27, 2000. Human Resources Manager, Kimberly Clark South Africa. Authors interview, March 8, 2000;

33

34

35

36

Human Resources Manager, Sappi Enstra. Authors interview, March 7, 2000.

37

Human Resources Manager, Nampak Corrugated Wadeville, April 10, 2000. For example, in 2005, for one of the most popular forms of self-employment small spaza

38

shops, or convenience stores usually attached to dwellings credit from financial institutions, formal or informal, was the main source of start-up capital for only 6 percent of businesses, while 82.6 percent relied on private savings and loans from families or friends, and 17 percent on severance payments (Ligthelm 2005c: 206-07.)
39

Salgado, Ingrid. 2005. Union Names Sasols Seven Deadly Sins. Business Report,

February 16: 3.

408

40

Seidman-Makgetla, Neva. 2005. Self-Employment Usually Means Scraping a Living.

Business Day, August 26: 6.

41

Authors interview, April 7, 2000. Authors interview, November 11, 1999.

42

Notes to Chapter 6
1

Hlangani, Mziwakhe. 2004. Plea to Help Break Credit Bureaus Grip on the Poor. The Star,

May 27: 9. The figures in this article are based on a research commissioned by the Financial Sector Campaign Coalition, which comprised a broad range of trade unions and civil society organizations.
2

The projects director, a former financial manager and strategist at Lehman Brothers, for

example, wrote a glowing portrait of a man who, at age eighty-one, was still working for a paltry R 1,000 per month to feed a family of four. The author was particularly impressed that the assets in the portfolio of this entrepreneur included not only membership in a stokvel and some livestock but also some investment in a unit trust. The report candidly recognized, however, that such financial instruments did not allow its subject to safely get to the end of the month, for which in the end he relied on his sons ability to find casual employment (Collins 2006).
3

Conversely, 64.5 percent of metalworkers and 47.9 percent of glass-paper ones had other

income earners (formal or informal) in their households, with average monthly wages, before deductions, of, respectively, R 1688 and R 1840.
4

Paul Biyase, NUMSA National Benefits Coordinator. Authors interview, April 18, 2000.

409

It should be noticed in this regard that after apartheid the ZCC has developed its entrepreneurial

profile by, for example, cooperating with large insurance companies providing low-premium coverage to the urban poor (Bhre 2007a.)
6

Charney, Craig. 1995. A New Point of Departure on the Bond Boycotts. Business Day,

December 14: 9.
7

See the statements by the City of Johannesburg in Transformation of Pension Funds in the

City, March 2003, http://www.joburg.org.za/content/view/1333/114/, retrieved on February 21, 2009, and The Transformation of Retirement Funding Arrangements for Employees, March 2003, http://www.joburg.org.za/content/view/1331/114/, retrieved on February 21, 2009.
8

A 2002 amendment to Section 189a of the Labour Relations Act recognized the right to strike

over layoffs but limited it to disputes on procedural fairness.

9

Hartley, Aziz. 2008. Educate Them by Force. Pretoria News, November 10: 1.

Notes to Conclusion
1

Cit. in Thoburn (2003: 111.) Paton, Carol. 2008. Rather Jobless than Low-Paid. Financial Mail, June 6: 13. Ndebele, Njabulo. 2008. Going Boldly into the Unknown. City Press, March 1: 7.

410

Bibliography

a. Interviews Cited or Quoted in Text a.1 Workplace interviews a.1.1 Managerial personnel Baldwins Steel, Works Manager. Brakpan, October 27, 1999. Consol Glass, Human Resources Manager. Wadeville (Germiston), April 4, 2000. FEDMO Employment Services, Director. Nigel, October 21, 1999. Greater Johannesburg Metropolitan Council, Executive Officer (Roelf de Beer) for Waste Management, Western Metropolitan Local Council. Roodepoort, October 19. Greater Johannesburg Metropolitan Council, Project Manager (Alan Agaienz), Johannesburg Roads Agency. Randburg (Johannesburg), January 27, 2000. Greater Johannesburg Metropolitan Council, manager, Central Roads Depot. Johannesburg, February 16, 2000. Greater Johannesburg Metropolitan Council, manager, Waterval Waste Depot. Johannesburg, February 17, 2000. Kelvinator South Africa, CEO (Simon Koch). Telephonic interview, November 17, 2001. Kelvinator South Africa, Financial Director. Alrode (Alberton), October 20, 1999. Kimberly Clark South Africa, Human Resources Manager. Springs, March 8, 2000. Nampak Corrugated, Human Resources Manager. Wadeville (Germiston), April 10, 2000. Nampak Group, national headquarters representative. Telephonic interview, April 12, 2000. PFG Building Glass, Human Resources Manager. Springs, March 14, 2000. Sappi Enstra, Human Resources Manager. Springs, March 7, 2000. 411

Union Carriage and Wagon, Human Resources Manager. Nigel, November 3, 1999. a.1.2 Production workers (numbered according to respondents position in the authors data set; n=220) Respondent 1, male, 45, sales representative, Baldwins Steel. Brakpan, July 12, 1999. Respondent 2, male, 50, dispatcher/loader, Baldwins Steel. Brakpan, July 12, 1999. Respondent 3, male, 48, machine operator, Baldwins Steel. Brakpan, July 15, 1999. Respondent 4, male, 39, sales representative, Baldwins Steel. Brakpan, July 19, 1999. Respondent 6, male, 37, inventory clerk, Baldwins Steel. Brakpan, July 19, 1999. Respondent 8, male, 37, production foreman, Baldwins Steel. Brakpan, July 26, 1999. Respondent 15, male, 36, administrative assistant, Baldwins Steel. Brakpan, August 12, 1999. Respondent 18, male, 42, loader/order maker, Baldwins Steel. Brakpan, September 3, 1999. Respondent 21, male, 47, electrician, UCW. Nigel, July 24, 1999. Respondent 27, male, 33, machine operator, UCW. Nigel, August 7, 1999. Respondent 29, male, 31, spot welder, UCW. Nigel, August 14, 1999. Respondent 33, male, 42, carpenter, UCW. Nigel, August 25, 1999. Respondent 35, male, 31, carpenter/upholsterer, UCW. Nigel, September 8, 1999. Respondent 37, male, 49, electrician, UCW. Nigel, September 15, 1999. Respondent 39, male, 39, electrical tester/wirer, UCW. Nigel, September 22, 1999. Respondent 40, male, 39, electrical tester, UCW. Daveyton, September 29, 1999. Respondent 45, male, 46, carpenter, UCW. Nigel, October 20, 1999. Respondent 51, male, 25, machine operator, Kelvinator. Alrode (Alberton), November 4, 1999. Respondent 52, male, 25, assembler, Kelvinator. Alrode (Alberton), November 4, 1999. Respondent 53, male, 44, picker/line feeder, Kelvinator. Alrode (Alberton), November 11, 1999. 412

Respondent 55, male, 59, driver, Kelvinator. Alrode (Alberton), November 18, 1999. Respondent 56, male, 35, machine operator, Kelvinator. Alrode (Alberton), November 18, 1999. Respondent 57, male, 37, line inspector, Kelvinator. Alrode (Alberton), November 23, 1999. Respondent 62, male, 55, driver, GJMC Waste, Norwood Depot. Johannesburg, November 17, 1999. Respondent 69, female, 50, street cleaner, GJMC Waste, Norwood Depot. Johannesburg, December 8, 1999. Respondent 86, female, 48, street cleaner, GJMC Waste, Waterval Depot. Johannesburg, February 17, 2000. Respondent 87, male, 54, petrol attendant, GJMC Waste, Waterval Depot. Johannesburg, February 17, 2000. Respondent 88, male, 29, truck loader, GJMC Waste, Norwood Depot. Johannesburg, February 23, 2000. Respondent 89, female, 35, street cleaner, GJMC Waste, Avalon Depot. Soweto, February 27, 2000. Respondent 91, male, 51, activity supervisor, GJMC Waste, Inner City Depot. Johannesburg, March 5, 2000. Respondent 96, male, 29, loader, GJMC Waste, Avalon Depot. Soweto, March 12, 2000. Respondent 97, female, 43, litter picker, GJMC Waste, Avalon Depot. Soweto, March 12, 2000. Respondent 101, male, 53, team leader, GJMC Roads, Nancefield Depot. Soweto, November 7, 1999. Respondent 103, male, 49, machine operator, GJMC Roads, Avalon Depot. Soweto, November 13, 1999. 413

Respondent 106, male, 60, supervisor, GJMC Roads, Waterval Depot. Johannesburg, December 3, 1999. Respondent 108, male, 42, repairman, GJMC Roads, Avalon Depot. Soweto, December 12, 1999. Respondent 109, female, 59, office worker, GJMC Roads, Ennerdale Depot. Orange Farm, December 13, 1999. Respondent 116, male, 42, team leader, GJMC Roads, Nancefield Depot. Soweto, February 5, 2000. Respondent 130, male, 57, machine operator, GJMC Roads, Ennerdale Depot. Orange Farm, March 29, 2000. Respondent 138, male, 56, team leader, GJMC Roads, Avalon Depot. Soweto, April 17, 2000. Respondent 141, male, 35, puller, Paperlink. Germiston, February 21, 2000. Respondent 147, male, 41, driver, Paperlink. Germiston, March 13, 2000. Respondent 148, male, 44, truck driver, Paperlink. Germiston, March 13, 2000. Respondent 157, male, 60, truck driver, Paperlink. Germiston, April 13, 2000. Respondent 165, male, 43, machine operator, Nampak Corrugated. Wadeville (Germiston), March 4, 2000. Respondent 166, male, 47, die-cutting operator, Nampak Corrugated. Wadeville (Germiston), March 11, 2000. Respondent 170, male, 44, machine operator, Nampak Corrugated. Wadeville (Germiston), March 17, 2000. Respondent 177, male, 49, recycling operator, Nampak Corrugated. Wadeville (Germiston), April 6, 2000. 414

Respondent 179, male, 41, machine operator, Nampak Corrugated. Wadeville (Germiston), April 15, 2000. Respondent 180, male, 36, strapper operator, Nampak Corrugated. Katlehong, April 15, 2000. Respondent 181, male, 44, mechanic, Consol Glass. Wadeville (Germiston), March 4, 2000. Respondent 195, male, 40, machine operator, Consol Glass. Wadeville (Germiston), April 29, 2000. Respondent 197, male, 39, fitter, Consol Glass. Wadeville (Germiston,) May 5, 2000. Respondent 211, male, 40, machine operator, MB Glass. Germiston, April 28, 2000. Respondent 212, male, 39, general worker/cleaner, MB Glass. Germiston, May 6, 2000. Respondent 217, male, 37, general worker, MB Glass. Leondale (Germiston,) May 16, 2000. a.2 Policymakers, experts, and union officials Alvilar, Cynthia (Acting Chief Director for Labour Market Policy, Department of Labour). Johannesburg, June 13, 2000. Biyana, Hlubi (Johannesburg branch organizer, SAMWU). Johannesburg, April 17, 2000. Biyase, Paul (National Benefits Coordinator, NUMSA). Johannesburg, April 18, 2000. Bodibe, Oupa (COSATU secretariat coordinator). Johannesburg, December 5, 2001. Coleman, Neil (COSATU Parliamentary Office). Johannesburg, October 24, 2000 and December 1, 2000. De Bruyn, Michelle (coordinator of the NEDLAC Labour Market Chamber). Johannesburg, November 7, 2000. Galeni, Osborne (NUMSA Dorbyl Group National Organizer). Johannesburg, August 11, 1999. Livi, Zweli (NUMSA Alberton Organizer). Alberton, September 16, 1999. Magaseng, George (NUMSA Springs Organizer). Springs, November 11, 1999. 415

Mahlangu, Jan (CEPPWAWU National Benefits Coordinator). Johannesburg, April 12, 2000. Makiwane, Fezile (Director for Social Security, Department of Social Development). Pretoria, October 16, 2000. Makka, Anusha (coordinator of the NEDLAC Public Finance and Monetary Policy Chamber). Johannesburg, November 7, 2000. Mhlongo, Victor (international relations officer, SAMWU). Johannesburg, May 19, 2000. Mkhonto, Shadrack (Unemployment Insurance Fund commissioner). Pretoria, October 2, 2000. Monage, Elias (NUMSA National Organizer for the Engineering Industry). Johannesburg, April 20, 2000. Mphafudi, Lucky (Chief Town and Regional Planner, Human Settlement Policy Directorate, Department of Housing). Pretoria, May 8, 2001. Mthombeni, Bengeza (CEPPWAWU deputy secretary general and NEDLAC labor representative). Johannesburg, June 12, 2000. Naidoo, Ravi (director of the National Labour and Economic Development Institute). Johannesburg, October 15, 2000. Nyambezi, Andile (CEPPWAWU Germiston Organizer). Germiston, April 7, 2000. Patel, Leila (former Director General, Department of Social Welfare and Population Development). Johannesburg, June 22, 2000. Sikwebu, Dinga (NUMSA education officer). Johannesburg, April 18, 2000. Singh, Sanjee (Assistant Director for Policy and Evaluation, Gauteng Department of Housing). Johannesburg, November 1, 2000. Taylor, Viviene (chairperson of the Committee of Inquiry into a Comprehensive System of Social Security for South Africa). Pretoria, August 8, 2001. 416

Thobejane, Makgane (labor relations specialist, Greater Johannesburg Metropolitan Council). Johannesburg, March 16, 2000. Thring, Petal (Graduate School of Public and Development Management, University of the Witwatersrand). Johannesburg, May 22, 2000. Van den Heever, Alex (Council of Medical Schemes and Department of Health). Johannesburg, November 29, 2000. Von Broembsen, Diat (Chief Director of Policy Planning, Department of Housing). Pretoria, August 7, 2001. Zulu, Andrew (NUMSA Nigel Organizer). Nigel, July 13, 1999. b. Archival Collections DHP. Department of Historical Papers, University of the Witwatersrand, Johannesburg Asher, Anthony papers, A2940. Barrett, Jane papers, A2168. Baskin, Jeremy papers, AH2920. CTMWA (Cape Town Municipal Workers Association), A1939 ERAB (East Rand Administration Board) papers, AR1934 FOSATU (Federation of South African Trade Unions) papers, AH1999. Katlehong (Mfm) papers, A1880 MAWU (Metal and Allied Workers Union) papers, AH1077. NICISEMI (National Industrial Council for the Iron, Steel, Engineering and Metallurgical Industry) papers, AH3072. Parnell, Susan papers, A3023. PLANACT papers, AL2566. 417

SAHA (South African History Archives), AL2457. MAD. Manuscripts and Archives Department, University of Cape Town NUSAS (National Union of South African Students) papers, BC586. c. Unsorted and Uncategorized Collections (Copies of all documents are in possession of the author) South African Municipal Workers Union (SAMWU), Gauteng provincial office, Johannesburg. National Union of Metalworkers of South Africa (NUMSA), national head office, Johannesburg. d. Newspapers, Newsletters, and Magazines Abasebenzi (Cape Town), 1973-1975 BuaNews (Johannesburg) Business Day (Johannesburg) Business Report (Johannesburg) Cape Times (Cape Town) COSATU News (Johannesburg), 1986-1991 Engineering News (Johannesburg) Financial Mail (Johannesburg) FOSATU Worker News (Johannesburg), 1979-1985 Isisebenzi (Durban), 1973-76 Mail & Guardian (Johannesburg) Pretoria News (Pretoria) SAPA (South African Press Association) Newswire (Johannesburg) Sunday Times (Johannesburg) The Mercury (Durban) 418

The Star (Johannesburg) e. Government Publications Bojanala Platinum District Municipality. 2005. Bojanala Platinum District Growth and Development Strategy. Rustenburg: Bojanala Platinum District Municipality. City of Johannesburg. 2001. Johannesburg. An African City in Change. Cape Town: Zebra Press. City of Johannesburg. 2004. Integrated Transport Plan 2003-08. Updated. Johannesburg: City of Johannesburg Metropolitan Municipality. City of Johannesburg. 2006. Growth and Development Strategy. Johannesburg: City of Johannesburg Metropolitan Municipality. Department of Health. 1997a. Submission to the Retirement Policy Consultative Forum. Pretoria: Department of Health. Department of Health. 1997b. A Social Health Insurance Scheme for South Africa: Policy Document. Pretoria: Department of Health. Department of Housing. 1994. A New Housing Policy and Strategy for South Africa. Pretoria: Department of Housing. Department of Housing. 2004. Breaking New Ground. A Comprehensive Plan for the Development of Sustainable Human Settlements. Pretoria: Department of Housing. Department of Labour. 1996. Restructuring the South African Labour Market. Report of the Presidential Commission to Investigate Labour Market Policy. Pretoria: Department of Labour. Department of Labour. 1999. Labour Market Policy in an Era of Transformation. Discussion Document. Pretoria: Department of Labour.

419

Department of Labour. 2007. Annual Report of the Unemployment Insurance Fund. Pretoria: Department of Labour. Department of Social Development. 2000. Mobilising for a Caring Society. People First For Sustainable Development. Statement by the Minister for Welfare and Population

Development, Dr. Zola Skweyiya, at the launch of a 10-Point Programme of Action for the Welfare and Development Portfolio, January. Department of Social Development. 2002. Committee Report No.5. Social Protection: Employment and Unemployment. Unpublished report for the Committee of Inquiry into a Comprehensive System of Social Security for South Africa. Department of Social Development. 2006a. Report on Incentive Structures of Social Assistance Grants in South Africa. Johannesburg: Kesho Consulting and Business Solutions. Department of Social Development. 2006b. Linking Social Grants Beneficiaries to Poverty Alleviation and Economic Activity. Discussion Document. Pretoria: Department of Social Development. Department of Welfare. 1996a. Report of the Committee for the Restructuring of Social Security (Chikane Committee). Pretoria: Department of Welfare and Population Development. Department of Welfare. 1996b. Report of the Lund Committee on Child and Family Support. Pretoria: Department of Welfare and Population Development. Department of Welfare. 1999. Financing Policy. Developmental Social Welfare Services. Government Gazette 19888. Pretoria: Government Printer. Ekurhuleni Metropolitan Municipality. 2005. Ekurhuleni Metropolitan Spatial Development Framework. Ekurhuleni: Ekurhuleni Metropolitan Municipality.

420

Ekurhuleni Metropolitan Municipality. 2007. Ekurhuleni Southern Spatial Development Framework Review. Ekurhuleni: Ekurhuleni Metropolitan Municipality. Greater Johannesburg Metropolitan Council. 1998. Organisation Review. Final Case for Change Report. Group Executive Report, Group 1. Johannesburg: Greater Johannesburg Metropolitan Council. Greater Johannesburg Metropolitan Council. 1999a. iGoli 2002: Transformation and Implementation Plan. Johannesburg: Greater Johannesburg Metropolitan Council. Greater Johannesburg Metropolitan Council. 1999b. Establishment of a Waste Management Utility. Business Plan. Johannesburg: Greater Johannesburg Metropolitan Council. Greater Johannesburg Metropolitan Council. 1999c. Johannesburg Roads Agency

Implementation Business Plan. Johannesburg: Greater Johannesburg Metropolitan Council. Greater Johannesburg Metropolitan Council. 2000. Proposed Establishment of the Johannesburg Roads Agency: Legal Entity to be Established and Supporting Documentation. Transformation Lekgotla, October 9. Greater Johannesburg Metropolitan Council. 2001. Review of Service Delivery Agreement and Sales of Business Agreement: Johannesburg Roads Agency. Johannesburg: Greater Johannesburg Metropolitan Council, Mayoral Committee. National Treasury. 2007a. Employment, Wages and Social Security. Unpublished report. National Treasury. 2007b. Social Security and Retirement Reform Second Discussion Paper. Pretoria: Government Printer. Office of the President. 1995. Towards a National Growth and Development Strategy. Unpublished report. 421

Office of the President. 2003. Towards a Ten Year Review. Synthesis Report on Implementation of Government Programmes. Pretoria: Government Printer. RSA - Republic of South Africa. 1966. Report of the Committee of Enquiry into Pension Fund Matters (Cilliers Committee). Pretoria: Government Printer. RSA Republic of South Africa. 1979. Report of the Commission of Enquiry into the Legislation Affecting the Utilisation of Manpower (Excluding the Legislation Administered by the Departments of Labour and Mines) (Riekert Commission). Pretoria: Government Printer. RSA - Republic of South Africa. 1992. Report of the Committee of Investigation into a Retirement Provision System for South Africa (Mouton Committee). Pretoria: Government Printer. RSA - Republic of South Africa. 1994a. The Reconstruction and Development Programme. A Policy Framework. Pretoria: Government Printer. RSA - Republic of South Africa. 1994b. White Paper on Reconstruction and Development. Government Gazette 16085. Pretoria: Government Printer. RSA - Republic of South Africa. 1994c. Report of the Commission of Inquiry into the Manner of Providing for Medical Expenses (Melamet Commission). Pretoria: Government Printer. RSA - Republic of South Africa. 1995a. Report of the Committee on Strategy and Policy Review of Retirement Provision in South Africa (Smith Committee). Pretoria: Government Printer. RSA - Republic of South Africa. 1995b. Restructuring the National Health System for Universal Primary Health Care. Report of the Committee of Inquiry into a National Health Insurance System (Broomberg-Shisana Committee). Pretoria: Government Printer.

422

RSA Republic of South Africa. 1995c. Urban Development Strategy of the Government of National Unity. Pretoria: Government Printer. RSA - Republic of South Africa. 1996. Constitution of the Republic of South Africa Act, No.108. Pretoria: Government Printer. RSA - Republic of South Africa. 1997. White Paper for Social Welfare. Principles, Guidelines, Proposed Policies and Programmes for Developmental Social Welfare in South Africa. Pretoria: Government Printer. RSA - Republic of South Africa. 1998a. Government Notice R418, Regulation 3(2). Regulations Regarding the Phasing Out of Maintenance Grants in Terms of the Social Assistance Act, 1992 (No.59 of 1992). Government Gazette 18771. Pretoria: Government Printer. RSA - Republic of South Africa. 1998b. Government Working Document for Use in the Supervisory Committee Meeting of NEDLAC on 11 August 1998. Unpublished document. RSA - Republic of South Africa. 1999. Social Security in South Africa: A Proposal for Reform. Unpublished report of the Interdepartmental Task Team Investigation into an Integrated, Comprehensive Social Security System, July 30. RSA Republic of South Africa. 2001a. A Nation at Work for a Better Life for All. Human Resource Development Strategy for South Africa. Pretoria: Government Printer. RSA Republic of South Africa. 2001b. Committee of Inquiry into a Comprehensive System of Social Security Draft Report into the Principles Underlying the Financing of the South African Social Security System. Report 2: Evaluation of the Budget Process as It Affects the System of Social Security within the Public Sector. Unpublished report. 423

RSA - Republic of South Africa. 2002. Transforming the Present Protecting the Future. Report of the Committee of Inquiry into a Comprehensive System of Social Security for South Africa (Taylor Committee). Pretoria: Government Printer. Union of SA. 1929. Third Report of the Commission on Old Age Pensions and National Insurance, UG 26-1929 (Pienaar Commission). Pretoria: Government Printer. Union of SA. 1936. Departmental Committee of Enquiry: National Health Insurance, U.G. 411936 (Collie Committee). Pretoria: Government Printer. Union of SA. 1944a. Report of the Social Security Committee and Report No.2 of the Social and Economic Planning Council: Social Security, Social Services and the National Income, U.G. 14-1944 (van Eck Committee). Pretoria: Government Printer. Union of SA. 1944b. Report of the National Health Services Commission on the Provision of an Organized National Health Service for all Sections of the People of the Union of South Africa, 1942-1944, U.G. 30-1944 (Gluckman Commission). Pretoria: Government Printer. Union of SA. 1944c. Report of the Witwatersrand Native Mine Wages Commission on Remunerations and Conditions of Employment of Natives on Witwatersrand Gold Mines, U.G. 21-1944 (Lansdowne Commission). Pretoria: Government Printer. f. Books, Articles, and Unpublished Works Adler, Glenn and Eddie Webster. 1999. Towards a Class Compromise in South Africa's 'Double Transition': Bargained Liberalization and the Consolidation of Democracy. Politics and Society 27 (3): 347-85. African National Congress. 1991a. Constitutional Principles for a Democratic South Africa. Johannesburg: African National Congress.

424

African National Congress. 1991b. Draft Resolution on ANC Economic Policy for National Conference. Johannesburg: African National Congress. African National Congress. 1992. Ready to Govern. ANC Policy Guidelines for a Democratic South Africa. Johannesburg: African National Congress. African National Congress. 1996. The State and Social Transformation. A Discussion Document. Johannesburg: African National Congress. African National Congress. 1997. The Core Values of the RDP. Discussion Document for the 50th National Conference, Mafikeng, December 16-20. African National Congress. 1998. The State, Property Relations and Social Transformation. A Discussion Paper towards the Alliance Summit. Johannesburg: African National Congress. African National Congress. 2002. Draft Resolutions of the National Policy Conference. Umrabulo 17. African National Congress. 2004. Understanding Gautengs Changing Economic, Social and Political Landscape. ANC Gauteng Province Base Document towards the 9th Provincial Conference. August 1. African National Congress. 2005. National General Council Discussion Documents. Presented at the National General Council. Tshwane, June 29 July 3. African National Congress. 2006. Gauteng Membership Survey. Johannesburg: African National Congress. African National Congress. 2007a. Policy Discussion Document. Social Transformation. Discussion Document for the National Policy Conference, Johannesburg, June 27-30.

425

African National Congress. 2007b. Economic Transformation for a National Democratic Society. Discussion Document for the National Policy Conference, Johannesburg, June 27-30. African National Congress. 2007c. Building a National Democratic Society. Strategy and Tactics of the ANC. Revised Draft, August. African National Congress. 2007d. Policy Discussion Document. Social Transformation. Discussion Document for the 52nd National Conference. Polokwane, December 16-20. Agamben, Giorgio. 1998. Homo Sacer. Sovereign Power and Bare Life. Stanford, CA: Stanford University Press. Alexander, Neville. 2002. An Ordinary Country. Issues in the Transition from Apartheid to Democracy in South Africa. Durban: University of Natal Press. Alexander, Peter. 2000. Workers, War & the Origins of Apartheid. Oxford: James Currey. Allen, Michael H. 2006. Globalization, Negotiation, and the Failure of Transformation in South Africa. New York: Palgrave MacMillan. Altman, Miriam. 2006. Low Wage Work in South Africa. Paper presented to the IZA/World Bank Conference on Employment and Development, Berlin, May 25-27. Altman, Miriam. 2007a. What Are the Policy Implications of the Informal Sector Beco ming the Informal Economy? Concept paper prepared for the IZA/World Bank Conference on Employment and Development, Bonn, June 8-9. Altman, Miriam. 2007b. Employment Scenarios to 2024. Pretoria: Human Sciences Research Council.

426

Altman, Miriam and David Hemson. 2007. The Role of Expanded Public Works Programmes in Halving Unemployment. Paper for the Employment Growth and Development Initiative, Human Sciences Research Council, Pretoria. Alverson, Hoyt. 1978. Mind in the Heart of Darkness. Value and Self-Identity among the Tswana of Southern Africa. New Haven, CT: Yale University Press. Andrew, Jeremy P. 2004. The Conversion of Members Rights in South Africa Retirement Funds from Defined Benefits to Defined Contributions and the Statutory Apportionment of the Resulting Actuarial Surplus. South African Actuarial Journal 4: 1-62. Anonymous. 1981. Striking at Pensions. Work in Progress 18: 27-38. Appadurai, Arjun. 1986. Introduction: Commodities and the Politics of Value. In The Social Life of Things: Commodities in Cultural Perspectives, edited by Arjun Appadurai, 3-63. Cambridge: Cambridge University Press. Appolis, John and Dinga Sikwebu. 2003. Pressing Challenges Facing the South African Labour Movement: An Interview with John Appolis and Dinga Sikwebu. In Rethinking the Labour Movement in the New South Africa, edited by Franco Barchiesi and Tom Bramble, 205-25. London: Ashgate. Ardington, Elizabeth. 2004. Savings, Insurance and Debt over the Post-Apartheid Period: A Review of Recent Research. Working Paper 65, Centre for Social Science Research, University of Cape Town. Asher, Anthony. 2001. The Fight against Poverty: Social Security, Job Creation and Responsibility. In Defining a New Citizenship for South Africa and the Fundamental Values that will Shape It. Seminar Proceedings, 63-70. Johannesburg: Konrad Adenauer Stiftung. 427

Ashforth, Adam. 1990. The Politics of Official Discourse in Twentieth-Century South Africa. Oxford: Clarendon Press. Ashforth, Adam. 2005. Witchcraft, Violence, and Democracy in South Africa. Chicago: University of Chicago Press. Atkins, Keletso. 1994. The Moon is Dead! Give Us Our Money!: The Cultural Origins of an African Work Ethic, Natal, South Africa, 1843-1900. Portsmouth, NH: Heinemann. Bhre, Erik. 2007a. The Janus Head of Insurances in South Africa: Networks of Inclusion and Bureaucracies of Exclusion. Paper presented at the European Conference on African Studies, Leiden, July 12. Bhre, Erik. 2007b. Reluctant Solidarity. Death, Urban Poverty and Neighbourly Assistance in South Africa. Ethnography 8 (1): 33-59. Balandier, Georges. 1967. Anthropologie Politique. Paris: Presses Universitaires de France. Balibar, Etienne. 1994. Subjection and Subjectivation. In Supposing the Subject, edited by Joan Copjec, 1-16. London: Verso. Ballard, Richard, Adam Habib, and Imraan Valodia, eds. 2006. Voices of Protest. Social Movements in Post-Apartheid South Africa. Durban: University of KwaZulu-Natal Press. Ballard, Richard, Adam Habib, Imraan Valodia, and Elke Zuern. 2006. Introduction. From Anti-Apartheid to Post-Apartheid Social Movements. In Voices of Protest. Social Movements in Post-Apartheid South Africa, edited by Richard Ballard, Imraan Valodia, and Elke Zuern, 1-22. Durban: University of KwaZulu-Natal Press. Barchiesi, Franco. 1998. Delivery from Below, Resistance from Above. Electricity and the Politics of Struggle over Social Needs in Tembisa. Debate 2 (1): 12-27.

428

Barchiesi, Franco. 1999. Kelvinator: Restructuring, Collapse and Struggle. South African Labour Bulletin 23 (6): 65-70. Barchiesi, Franco. 2001. Fiscal Discipline and Worker Response: The Restructuring of Johannesburgs Solid Waste Management. In The Commercialisation of Waste Management in South Africa, by Franco Barchiesi, Msokoli Qotole, and Mthetho Xali, 21-46. Occasional Paper 3, Municipal Services Project, Queens University, Kingston, ON. Barchiesi, Franco. 2007a. Labor and Social Citizenship in Colonial and Postcolonial Modernity. South African Perspectives in a Continental Context. Review 30 (1): 17-43. Barchiesi, Franco. 2007b. South African Debates on the Basic Income Grant: Wage Labour and the Post-Apartheid Social Policy. Journal of Southern African Studies 33 (3): 561-75. Barchiesi, Franco. 2007c. Privatization and the Historical Trajectory of Social Movement Unionism: A Case Study of Municipal Workers in Johannesburg, South Africa. International Labor and Working Class History 71: 50-69. Barchiesi, Franco and Bridget Kenny. 2002. From Workshop to Wasteland: Deindustrialisation and Fragmentation of the Black Working Class on the East Rand (South Africa), 199099. International Review of Social History 47 (Supplement): 35-63. Barchiesi, Franco and Bridget Kenny. 2008. Precarious Collaborations. Working-Class Subjectivities, Community Activism, and the Problem with Social Movement Unionism in Late-Apartheid East Rand (South Africa.) Paper presented at the 8th North Eastern Workshop on Southern Africa (NEWSA), Burlington, VT, October 17-19.

429

Barrett, Jane. 1989. [Untitled.] In Towards a Democratic Welfare System: Options and Strategies. Proceedings of the National Social Welfare Policy Conference, edited by Leila Patel, 16-20. Johannesburg: University of the Witwatersrand. Baskin, Jeremy. 1982. Growth of a New Worker Organ The Germiston Shop Stewards Council. South African Labour Bulletin 7 (8): 42-53. Baskin, Jeremy. 1991. Striking Back: A History of COSATU. Johannesburg: Ravan Press. Baudrillard, Jean. 1983. In the Shadow of the Silent Majorities. New York: Semiotext(e). Bauman, Zygmunt. 1982. Memories of Class. The Pre-History and After-Life of Class. London: Routledge & Kegan Paul. Bauman, Zygmunt. 1998. Globalization: The Human Consequences. New York: Columbia University Press. Bayart, Jean-Franois. 1989. The State in Africa. The Politics of the Belly. London: Longman. Beall, Jo, Owen Crankshaw, and Susan Parnell. 2000. Local Government, Poverty Reduction and Inequality in Johannesburg. Environment & Urbanization 12 (1): 107-22. Beauregard, Robert A. and Richard Tomlinson. 2007. The Discourse of Governance in PostApartheid Johannesburg. In The Making of Global City Regions. Johannesburg, Mumbai/Bombay, So Paulo, and Shanghai, edited by Klaus Segbers, 237-57. Baltimore: Johns Hopkins University Press. Beavon, Keith. 2004. Johannesburg. The Making and Shaping of the City. Pretoria: UNISA Press. Beck, Ulrich. 2000. The Brave New World of Work. Cambridge: Polity Press. Beinart, William. 2001. Twentieth Century South Africa. Oxford: Oxford University Press.

430

Benjamin, Paul. 2006. Beyond Lean Social Democracy: Labour Law and the Challenge of Social Protection. Transformation 60: 32-57. Berger, Iris. 1992. Threads of Solidarity. Women in South African Industry, 1900-1980. Bloomington: Indiana University Press. Bester, Hennie, Simon Hendrie, Timothy Hobden, Christine Hougaard, Richard Ketley, and Omoneka Musa. 2008. Old-age Saving by Low Income South Africans. Report prepared for FinMark Trust and the South African Savings Institute. Johannesburg: Genesis Analytics. Bezuidenhout, Andries. 2004a. Postcolonial Workplace Regimes in the Engineering Industry in South Africa. In Beyond the Apartheid Workplace. Studies in Transition, edited by Edward Webster and Karl von Holdt, 73-96. Durban: University of KwaZulu-Natal Press. Bezuidenhout, Andries. 2004b What Happened to Kelvinator: The Road from Alrode to Ezakheni and Matshapa. Paper presented at the Wits-Ekurhuleni Symposium on Sustainable Manufacturing, June 10-11. Bezuidenhout, Andries and Bridget Kenny. 2000. The Language of Flexibility and the Flexibility of Language: Post-Apartheid South African Labour Market Debates. Unpublished paper, Sociology of Work Unit, University of the Witwatersrand, Johannesburg. Bhabha, Homi K. 1995. In a Spirit of Calm Violence. In After Colonialism. Imperial Histories and Postcolonial Displacements, edited by Gyan Prakash, 326-43. Princeton: Princeton University Press.

431

Bhorat, Haroon and Morne' Oosthuizen. 2005. The Post-Apartheid South African Labour Market. Working Paper 05/93, Cape Town: University of Cape Town, Development Policy Research Unit. Bhorat, Haroon, Murray Leibbrandt, Muzi Maziya, Servaas van der Berg, and Ingrid Woolard. 2001. Fighting Poverty. Labour Markets and Inequality in South Africa. Cape Town: University of Cape Town Press. Biernacki, Richard. 2005. The Event of Place in Historical Sociology. In Looking Backward and Looking Forward. Perspectives on Social Science History, edited by Harvey J. Graff, Leslie P. Moch, and Philip McMichael, 179-211. Madison: University of Wisconsin Press. BIG Financing Reference Group. 2004. Breaking the Poverty Trap: Financing a Basic Income Grant in South Africa. Unpublished report. Bird, Adrienne and Geoff Schreiner. 1992. COSATU at the Crossroads: Towards Tripartite Corporatism or Democratic Socialism? South African Labour Bulletin 16 (6): 22-32. Bollyky, Thomas J. 2002. Balancing Private Rights and Public Obligations: Constitutionally Mandated Compulsory Licensing of HIV/AIDS Related Treatment in South Africa. South African Journal on Human Rights 18 (4): 530-69. Bond, Patrick. 1999. Cities of Gold, Townships of Coal. Essays on South Africas New Urban Crisis. Trenton, NJ: Africa World Press. Bond, Patrick. 2005. Elite Transition. From Apartheid to Neoliberalism in South Africa. Second edition. Durban: University of KwaZulu-Natal Press. Bonner, Chris. 2004. Organisational Renewal in COSATU and Affiliates: An Overview. Johannesburg: NALEDI. 432

Bonner, Philip L. 1990. Desirable or Undesirable Basotho Women? Liquor, Prostitution and the Migration of Basotho Women to the Rand, 1920-1945. In Women and Gender in Southern Africa to 1945, edited by Cherryl Walker, 221-50. Cape Town: David Philip; London: James Currey. Bonner, Philip. 1993. The Russians on the Reef, 1947-57. Urbanisation, Gang Warfare and Ethnic Mobilisation In Apartheids Genesis, 1935-1962, edited by Philip Bonner, Peter Delius, and Deborah Posel, 160-94. Johannesburg: Ravan Press. Bonner, Philip and Vusi Ndima. 1999. The Roots of Violence on the East Rand, 1980-1990. Seminar Paper 450, Institute for Advanced Social Research, University of the Witwatersrand, Johannesburg. Bonner, Philip and Noor Nieftagodien. 2001. Kathorus. A History. Johannesburg: Maskew Miller Longman. Bonner, Philip, Peter Delius, and Deborah Posel. 1993. The Shaping of Apartheid. Contradiction, Continuity and Popular Struggle. In Apartheids Genesis, 1935-1962, edited by Philip Bonner, Peter Delius, and Deborah Posel, 1-41. Johannesburg: Ravan Press. Bonnin, Deborah R. 1987. Class, Consciousness and Conflict in the Natal Midlands, 1940-1987: The Case of the BTR Sarmcol Workers. M.Sc. diss. University of Natal, Durban. Bonnin, Deborah. 1994. We Went to Arm Ourselves at the Field of Suffering: Traditions, Experiences and Grassroots Intellectuals in the Making of Class. Unpublished paper, Centre for Industrial and Labour Studies, University of Natal, Durban. Bonnin, Deborah, Georgina Hamilton, Robert Morrell, and Ari Sitas. 1996. The Struggle for Natal and Kwazulu: Workers, Township Dwellers and Inkatha, 1972-1985. In Political 433

Economies and Identities in KwaZulu-Natal. Historical and Social Perspectives, edited by Robert Morrell, 141-78. Durban: Indicator Project South Africa. Botha, Claire. 2008. Introduction. In Financing South Africas National Health System through National Health Insurance. Possibilities and Challenges, edited by Claire Botha and Michael Hendricks, ix-xiv. Pretoria: HSRC Press. Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press. Bozzoli, Belinda. 1981. The Political Nature of a Ruling Class. Capital and Ideology in South Africa, 1890-1933. London: Routledge and Kegan Paul. Bozzoli, Belinda. 1991. Women of Phokeng. Consciousness, Life Strategy and Migrancy in South Africa, 1900-1983. Johannesburg: Ravan Press. Braude, Wolfe. 2004. South Africa Country Analysis. Global Policy Network, Workforce Development Study. Johannesburg: NALEDI. Breckenridge, Keith. 2005. The Biometric State: The Promise and Peril of Digital Government in the New South Africa. Journal of Southern African Studies 31 (2): 267-82. Breckenridge, Keith. 2007. Promiscuous Method: The Historiographical Effects of the Search for the Rural Origins of the Urban Working Class in South Africa. International labor and Working Class History 65: 26-49. Brookes, Edgar H. 1927. Economic Aspects of the Native Problem. Journal of the Economic Society of South Africa 1 (2): 39-50. Brown, Carolyn. 2003. We Were All Slaves: African Miners, Culture and Resistance at the Enugu Government Colliery. Portsmouth, NH: Heinemann. Brown, Wendy. 1995. States of Injury. Princeton: Princeton University Press. 434

Brown, Wendy. 2000. Resisting Left Melancholia. In Without Guarantees: In Honor of Stuart Hall, edited by Paul Gilroy, Lawrence Grossberg, and Angela McRobbie, 21-29. London: Verso. Brown, Wendy. 2002. Suffering the Paradoxes of Rights. In Left Legalism/Left Critique, edited by Wendy Brown and Janet Halley, 420-34. Durham, NC: Duke University Press. Buhlungu, Sakhela. 2001. Democracy and Modernisation in the Making of the South African Trade Union Movement: The Dilemma of Leadership, 1973-2000. PhD diss. University of the Witwatersrand, Johannesburg. Buhlungu, Sakhela. 2002. From Madiba Magic to Mbeki Logic. Mbeki and the ANCs Trade Union Allies. In Thabo Mbekis World. The Politics and Ideology of the South African President, edited by Sean Jacobs and Richard Calland, 179-200.

Pietermaritzburg: University of Natal Press. Buhlungu, Sakhela. 2004. The Building of the Democratic Tradition in South Africas Trade Unions after 1973. Democratization 11 (3): 133-58. Buhlungu, Sakhela. 2005. Union-Party Alliances in the Era of Market Regulation: The Case of South Africa. Journal of Southern African Studies 31 (4): 701-17. Buhlungu, Sakhela. 2006. Rebels without a Cause of Their Own? The Contradictory Location of White Officials in Black Unions in South Africa, 1973-94. Current Sociology 54 (3): 427-51. Bundy, Colin. 1988. The Rise and Fall of South African Peasantry. London: Heinemann. Burawoy, Michael. 1991. The Extended Case Method. In Ethnography Unbound. Power and Resistance in the Modern Metropolis, 271-87. Berkeley: University of California Press.

435

Burton, Andrew. 2005. African Underclass. Urbanization, Crime and Colonial Order in Dar es Salaam, 1919-1961. Athens, OH: Ohio University Press. Business South Africa. 2003. Memorandum to Portfolio Committee on Social Development on the Report of the Committee of Inquiry into a Comprehensive Social Security System for South Africa. Unpublished document. Butler, Judith. 1997. The Psychic Life of Power. Theories in Subjection. Stanford: Stanford University Press. Buur, Lars. 2007. The Intertwined History of Security and Development. The Case of Developmental Struggles in South Africas Townships. In The Security-Development Nexus. Expressions of Sovereignty and Securitization in Southern Africa, edited by Lars Buur, Steffen Jensen, and Finn Stepputat, 109-31. Cape Town: HSRC Press; Uppsala: Nordiska Afrikainstitutet. Calland, Richard. 2006. Anatomy of South Africa. Who Holds the Power? Cape Town: Zebra Press. Calland, Richard and Mandy Taylor. 1997. Parliament and the Socio-Economic Imperative What is the Role of the National Legislature? Law, Democracy and Development 1: 193-220. Canclini, Nestor G. 2001. Consumers and Citizens. Globalization and Multicultural Conflicts. Minneapolis: University of Minnesota Press. Carmody, Padraig. 2002. The Liberalization of Underdevelopment or the Criminalization of the State? Contrasting Explanations of Africas Politico-Economic Crisis under

Globalization. In Globalization, the Third World State and Poverty-Alleviation in the Twenty-First Century, edited by Bernard I. Logan, 47-62. Aldershot: Ashgate. 436

Casale, Daniela and Chris Desmond. 2007. The Economic Well-being of the Family: Households Access to Resources in South Africa, 1995-2003. In Families and Households in Post-Apartheid South Africa: Socio-demographic Perspectives, edited by Acheampong Yaw Amoateng and Tim B. Heaton, 61-88. Cape Town: HSRC Press. Casale, Daniela, Colette Muller, and Dorrit Posel. 2005. Two Million Net New Jobs: A Reconsideration of the Rise in Employment in South Africa, 1995-2003. Working Paper 05/97. Cape Town: University of Cape Town, Development Policy Research Unit. Chakrabarty, Dipesh. 1988. Rethinking Working Class History. Bengal 1890-1940. Princeton : Princeton University Press. Chakrabarty, Dipesh. 2000. Provincializing Europe. Postcolonial Thought and Historical Difference. Princeton: Princeton University Press. Chamoiseau, Patrick. 1997. Texaco. New York: Pantheon. Champion, Allison W.G. 1927. The Truth about the ICU. Durban: African Workers Club. Chandra, Vandana and Junaid Ahmad. 2000. Applying the Comprehensive Development Framework at the City Level: The Experience of Johannesburg. Panel presentation for the World Bank Urban Forum on Urban Poverty Reduction in the 21st Century. Chantilly, VA, April 3-5. Chanock, Martin. 2001. The Making of South African Legal Culture, 1902-1936. Fear, Favour and Prejudice. Cambridge: Cambridge University Press. Charney, Craig. 1995. Voices of a New Democracy: African Expectations in the New South Africa. Research Report 38. Johannesburg: Centre for Policy Studies.

437

Chaskalson, Matthew, Karen Jochelson, and Jeremy Seekings. 1987. Rent Boycotts, the State, and the Transformation of Urban Political Economy in South Africa. Review of African Political Economy 40: 47-64. Chatterjee, Partha. 1986. Nationalist Thought and the Colonial World. A Derivative Discourse. London: Zed Books. Chatterjee, Partha. 1994. The Nation and Its Fragments. Colonial and Postcolonial Histories. Princeton: Princeton University Press. Chatterjee, Partha. 2004. The Politics of the Governed. New York: Columbia University Press. Cheadle, Halton. 2006. Regulated Flexibility: Revising the LRA and the BCEA. Industrial Law Journal 27 (4): 663-703. Cheah, Pheng. 2003. Spectral Nationality: Passages of Freedom from Kant to Postcolonial Literatures of Liberation. New York: Columbia University Press. Cherry, Janet. 2000. Hegemony, Democracy and Civil Society: Political Participation in Kwazakele Township, 1980-93. In From Comrades to Citizens. The South African Civics Movement and the Transition to Democracy, edited by Glenn Adler and Jonny Steinberg, 86-113. Basingstoke: MacMillan. Chipkin, Ivor. 2003. Functional and Dysfunctional Communities: The Making of National Citizens. Journal of Southern African Studies 29 (1): 63-82. Chipkin, Ivor. 2007. Do South Africans Exist? Nationalism, Democracy and the Identity of the People. Johannesburg: Wits University Press. Clarke, Marlea. 2004. Ten Years of Labour Market Reform in South Africa: Real Gains for Workers? Canadian Journal of African Studies 38 (3): 558-74.

438

Clarke, Marlea. 2006. All the Workers?Labour Market Reform and Precarious Workers in Post-Apartheid South Africa, 1994-2004. Ph.D. diss., York University, Toronto. Clegg, Stewart. 1994. Power Relations and the Constitution of the Resistant Subject. In Resistance and Power in Organizations, edited by John M. Jermier, David Knights, and Walter R. Nord, 274-325. London: Routledge. Coalition for a Basic Income Grant. 2003. Submission to the Portfolio Committee on Social Development on the Consolidated Report of the Committee of Inquiry into a Comprehensive Social Security System for South Africa. June 4. Cobley, Alan G. 1990. Class and Consciousness. The Black Petty Bourgeoisie in South Africa, 1924 to 1950. Westport, CT: Greenwood Press. Cobley, Alan G. 1997. The Rules of the Game. Struggles in Black Recreation and Social Welfare Policy in South Africa. Westport, CT: Greenwood Press. Cockhead, Peter J. 1970. The East Rand. A Geographical Analysis of the Transition of the Economic Base of the Region from Gold Mining to Manufacturing, and its Effects upon Future Economic and Spatial Development. Ph.D. diss. University of the Witwatersrand, Johannesburg. Coleman, Neil. 1999. The Relevance of Workfare to South Africa. South African Labour Bulletin 23 (2): 41-44. Coleman, Neil. 2003. Current Debates around BIG: The Political and Socio-economic Context. In A Basic Income Grant for South Africa, edited by Guy Standing and Michael Samson, 120-42. Cape Town: University of Cape Town Press.

439

Collins, Daryl. 2006. Focus Note: Styles and Profiles. The Connection between Livelihoods and Financial Portfolios of the Poor. Financial Diaries Project, Centre for Social Science Research, University of Cape Town. Comaroff, Jean and John L. Comaroff. 1987. The Madman and the Migrant: Work and Labor i n the Historical Consciousness of a South African People. American Ethnologist 14 (2): 191-209. Comaroff, Jean and John L. Comaroff. 2000. Millennial Capitalism: First Thoughts on a Second Coming. Public Culture 12 (2): 291-343. Comaroff Jean and John L. Comaroff. 2001. Revelations upon Revelation. After Shocks, Afterthoughts. Interventions 3 (1): 100-26. Comaroff, John L. and Jean Comaroff. 2004a. Criminal Justice, Cultural Justice: The Limits of Liberalism and the Pragmatics of Difference in the New South Africa. American Ethnologist 31 (2): 188-204. Comaroff, John L. and Jean Comaroff. 2004b. Policing Culture, Cultural Policing: Law and Social Order in Postcolonial South Africa. Law and Social Inquiry 29 (3): 513-45. Community Agency for Social Enquiry. 1998. Monitoring Socio-Economic Rights in South Africa: Public Perceptions. Johannesburg: Community Agency for Social Enquiry. Cooper, Carol. 2004. The Boundaries of Equality in Labour Law. Industrial Law Journal 25 (5): 813-52. Cooper, Frederick. 1987. On the African Waterfront. Urban Disorder and the Transformation of Work in Colonial Mombasa. New Haven, CT: Yale University Press. Cooper, Frederick. 1989. From Free Labor to Family Allowances: Labor and African Society in Colonial Discourse. American Ethnologist 16 (4): 745-65. 440

Cooper, Frederick. 1993. Africa in the World Economy. In Confronting Historical Paradigms. Peasants, Labor and the Capitalist World System in Africa and Latin America by Frederick Cooper, Florencia E. Mallon, Steve J. Stern, Allen F. Isaacman and William Roseberry, 84-201. Madison: University of Wisconsin Press. Cooper, Frederick. 1996. Decolonization and African Society. The Labour Question in French and British Africa. Cambridge: Cambridge University Press. Cooper, Frederick. 2000. Conditions Analogous to Slavery. Imperialism and Free Labor Ideology in Africa. In Beyond Slavery. Explorations of Race, Labor, and Citizenship in Postemancipation Societies, edited by Frederick Cooper, Thomas C. Holt, and Rebecca J. Scott, 107-49. Chapel Hill, NC: University of North Carolina Press. Cooper, Frederick and Randall Packard, eds. 1997. International Development and the Social Sciences. Essays on the History and Politics of Knowledge. Berkeley: University of California Press. Cooper, Frederick, Thomas C. Holt, and Rebecca J. Scott. 2000. Introduction. In Beyond Slavery. Explorations of Race, Labor, and Citizenship in Postemancipation Societies, edited by Frederick Cooper, Thomas C. Holt, and Rebecca J. Scott, 1-32. Chapel Hill, NC: University of North Carolina Press. COSATU. 1996. Submission to Parliament on the Social Welfare White Paper, November 4. COSATU. 1997a. Report of the September Commission on the Future of the Unions. Johannesburg: COSATU. COSATU. 1997b. Oral Submission to the Portfolio Committee on Welfare Regarding Proposed Changes to the System of Child Support Benefits Arising from the Report of the Lund Committee on Child and Family Support. April 21. 441

COSATU. 1997c. Report on the Alliance Summit, 31 August 1 September 1997. Johannesburg: COSATU. COSATU. 1998. Creating Jobs in South Africa. Key Issues and Strategies. Preliminary draft, February. Johannesburg: COSATU. COSATU. 1999a. Submission on Labour Market Policy in an Era of Transformation, presented to the Department of Labour, June 7. COSATU. 1999b. Response to the Ministerial Task Team Report on Unemployment Insurance and Related Coverage Issues, submitted to the Unemployment Insurance Commissioner, March 31. COSATU. 1999c. Audit of COSATUs Positions on Social Security. Johannesburg: COSATU. COSATU. 1999d. COSATU Submission on the Rental Housing Bill [B29-99]. Presented to the Housing Portfolio Committee, September 8. COSATU. 2000a. Audit of COSATUs Positions on Social Security. Johannesburg: COSATU. COSATU. 2000b. Advancing Social Transformation in the Era of Globalisation. Political Discussion Paper. Johannesburg: COSATU. COSATU. 2000c. Accelerating Transformation. COSATU's Engagement with Policy and Legislative Processes during South Africa's First Term of Democratic Governance. Johannesburg: COSATU. COSATU. 2000d. Submission on Comprehensive Social Security. Submitted to the Taylor Task Team on Social Security. December 5. COSATU. 2002. COSATU Central Executive Committee: Final Political Report. November 57.

442

COSATU. 2003. Secretariat Organization Report to 8th National Congress. Johannesburg, September 15-18. COSATU. 2004. Central Executive Committee Political Discussion Paper. May 24-26. Johannesburg: COSATU. COSATU. 2007a. COSATU Input on Economic Challenges. A Response to the ANC Economic Transformation Discussion Document. Johannesburg: COSATU. COSATU. 2007b. COSATUs Response to the ANCs Draft Strategy and Tactics Document. Johannesburg: COSATU. COSATU, FEDSAL, and NACTU. 1996. Social Equity and Job Creation. The Key to a Stable Future. Johannesburg: Labour Caucus at NEDLAC. COSATU, FEDSAL, and NACTU. 1998. Labour Audit for the Jobs Summit. Johannesburg: Labour Caucus at NEDLAC. Council for Medical Schemes. 2007. Annual Report 2006-07. Pretoria: Council for Medical Schemes. Crankshaw, Owen. 1997. Race, Class and the Changing Division of Labour under Apartheid. London: Routledge. Crichlow, Michaeline. 2000. Under the Shadows of Capital. In Informalization. Process and Structure, edited by Faruk Tabak and Michaeline Crichlow, 166-86. Baltimore: Johns Hopkins University Press. Cruikshank, Barbara. 1999. The Will to Empower. Democratic Citizens and Other Subjects. Ithaca, NY: Cornell University Press. Crush, Jonathan. 1995. Introduction. Imagining Development. In Power of Development, edited by Jonathan Crush, 1-23. London: Routledge. 443

Crush, Jonathan, Alan Jeeves, and David Yudelman. 1992. South Africas Labor Empire: A History of Black Migrancy to the Gold Mines. Boulder, CO: Westview Press. Cunliffe, John and Guido Erreygers. 2004. Introduction. In The Origins of Universal Grants. An Anthology of Historical Writings on Basic Capital and Basic Income, edited by John Cunliffe and Guido Erreygers, x-xxix. New York: Palgrave MacMillan. Das, Veena. 1989. Subaltern as Perspective. Subaltern Studies 6: 310-24. Davidson, Apollon, Irina Filatova, Valentin Gorodnov, and Sheridan Johns, eds. 2003. South Africa and the Communist International: A Documentary History. Volume II. Bolshevik Footsoldiers to Victims of Bolshevisation 1931-1939. London: Frank Cass. Davie, Grace. 2005. Poverty Knowledge in South Africa: The Everyday Life of Social Science Expertise in the Twentieth Century. Ph.D. diss., University of Michigan, Ann Arbor. Davie, Grace. 2007. Strength in Numbers. The Durban Student Wages Commission, Dockworkers and the Poverty Datum Line, 1971-1973. Journal of Southern African Studies 33 (2): 401-20. Davis, Dennis. 2001. Deconstructing and Reconstructing the Argument for a Bill of Rights within the Context of South African Nationalism. In The Post-Apartheid Constitutions. Perspectives on South Africas Basic Law, edited by Penelope Andrews and Stephen Ellman, 194-223. Johannesburg: Wits University Press; Athens, OH: Ohio University Press. Davis, J. Merle. 1933. Modern Industry and the African. An enquiry into the Effect of the Copper Mines of Central Africa upon Native Society. London: MacMillan. Dean, Mitchell. 1999. Governmentality. Power and Rule in Modern Society. London: Sage.

444

De Certeau, Michel. 1984. The Practice of Everyday Life. Berkeley: University of California Press. Deleuze, Gilles. 1987. Quest-ce que lacte de creation? Confrence donne dans le cadre des mardis de la fondation Femis, May 17. http://www.webdeleuze.com/php/texte.php?cle=134&groupe=Conf%E9rences&langue= 1. Retrieved July 23, 2008. Deleuze, Gilles and Felix Guattari. 1984. Anti-Oedipus. Capitalism and Schizophrenia. London: Athlone Press. Deleuze, Gilles and Felix Guattari. 1987. A Thousand Plateaus. Capitalism and Schizophrenia. Minneapolis: University of Minnesota Press. Denga, Samuel, Bobbie Marie, and Thobile Yanta. 2002. Challenges Facing Organizing. Johannesburg: NALEDI. Development Bank of Southern Africa. 2005. Overcoming Underdevelopment in South Africas Second Economy. Development Report 2005. Johannesburg: Development Bank of Southern Africa. Devey, Richard, Caroline Skinner, and Imraan Valodia. 2006. Definitions, Data and the Informal Economy in South Africa: A Critical Analysis. In The Development Decade? Economic and Social Change in South Africa, 1994-2004, edited by Vishnu Padayachee, 302-23. Pretoria: HSRC Press. Dhupelia, Uma. 1982. African Labour in Natal. Attempts at Coercion and Control, 1893-1903. Journal of Natal and Zulu History 5: 36-48.

445

Dickinson, David. 2003. Workplace Change in the New South Africa. Paper presented at the 13th World Congress of the International Industrial Relations Association, Berlin, September 8-12. Di Lollo, Adrian. 2006. A Critical Examination of the Concept of Welfare Dependency: Its Assumptions, Underlying Values and Manifestation in Social Policy, Internationally and in South Africa. MA diss. University of Cape Town. Donzelot, Jacques. 1979. The Policing of Families. Welfare versus the State, London: Hutchinson. Dubow, Saul. 2005. Introduction. South Africas 1940s. In South Africas 1940s. Worlds of Possibilities, edited by Saul Dubow and Alan Jeeves, 1-19. Cape Town: Double Storey Books. Dubow, Saul. 2006. A Commonwealth of Knowledge. Science, Sensibility, and White South Africa, 1820-2000. Oxford: Oxford University Press. Duncan, David. 1995. The Mills of God. The State and African Labour in South Africa, 19181948. Johannesburg: Wits University Press. Du Toit, Andries. 2004. Social Exclusion Discourse and Chronic Poverty: A South African Case Study. Development and Change 35 (5): 987-1010. Du Toit, Andries. 2005. Chronic and Structural Poverty in South Africa: Challenges for Action and Research. Chronic Poverty and Development Policy Paper 6, Programme for Land and Agrarian Studies, School of Government, University of the Western Cape, Bellville. Eckert, Andreas. 2004. Regulating the Social: Social Security, Social Welfare and the State in Late Colonial Tanzania. Journal of African History 45 (3): 467-89.

446

Edensor, Tim. 2005. The Ghosts of Industrial Ruins: Ordering and Disordering Memory in Excessive Space. Environment and Planning D: Society and Space 23 (6): 829-49. Elsley, Trenton. 2007. Outcomes of Collective Bargaining in South Africa: The Quality of Low Wage Employment in the Formal Economy in South Africa. In Bargaining Indicators 2007. A Collective Bargaining Omnibus, edited by Trenton Elsley, 79-97. Cape Town: Labour Research Service. Ernstzen, John. 1991. SAMWU: Three Years on the March. South African Labour Bulletin 15 (8): 30-35. Esping-Andersen, Gsta. 1990. The Three Worlds of Welfare Capitalism. Cambridge: Polity Press. Esposito, Roberto. 2008. Bios. Biopolitics and Philosophy. Minneapolis: University of Minnesota Press. Everatt, David, Graeme Gotz, and Ross Jennings. 2005. Living for the Sake of Living: Partnerships between the Poor and Local Government in Johannesburg. Democracy, Governance and Human Rights Paper 16. Geneva: United Nations Research Institute for Social Development. Falkena, Hans and Christo Luus. 1999. Personal Wealth and Financial Independence, Sandton: Southern Books Publishers. Feinstein, Charles H. 2005. An Economic History of South Africa. Conquest, Discrimination and Development. Cambridge: Cambridge University Press. Ferguson, James. 1999. Expectations of Modernity. Myths and Meanings of Urban Life on the Zambian Copperbelt. Berkeley: University of California Press.

447

Ferguson, James. 2007. Formalities of Poverty: Thinking about Social Assistance in Neoliberal South Africa. African Studies Review 50 (2): 71-86. Ferguson, James and Akhil Gupta. 2002. Spatializing States: Towards an Ethnography of Neoliberal Governmentality. American Ethnologist 29 (4): 981-1002. Fine, Ben and Zavareh Rustomjee. 1996. The Political Economy of South Africa. Johannesburg: Witwatersrand University Press. Finlay, Linda. 2002. Negotiating the Swamp: The Opportunity and Challenge of Reflexivity in Research Practice. Qualitative Research 2 (2): 209-30. Fisher, Foszia. 1978. Class Consciousness among Colonised Workers in South Africa. In Change, Reform and Economic Growth in South Africa, edited by Lawrence Schlemmer and Eddie Webster, 197-223. Johannesburg: Ravan Press. Forrest, Karen A. 2005. Power, Independence and Worker Democracy in the Development of the National Union of Metalworkers of South Africa (NUMSA) and Its Predecessors, 19801995. Ph.D. diss. University of the Witwatersrand, Johannesburg. FOSATU Federation of South African Trade Unions. n.d (but 1981). Federation of South African Trade Unions Report November 1980 to November 1981. Johannesburg: FOSATU. Foster, Joe. 1982. The Workers Struggle: Where Does FOSATU Stand? Review of African Political Economy 24: 99-114. Foucault, Michel. 1997. The Birth of Biopolitics. In Ethics. Subjectivity and Truth. The Essential Works of Foucault, 1954-1984. Vol.1, edited by Paul Rabinow, 73-79. New York: New Press.

448

Foucault, Michel. 2001. Governmentality. In Power. The Essential Works of Foucault, 19541984. Vol.3, edited by James D. Faubion, 201-22. New York: New Press. Foucault, Michel. 2008. The Birth of Biopolitics. Lectures at the College de France, 1978-79. New York: Palgrave MacMillan. Fox-Piven, Frances and Richard Cloward. 1993. Regulating the Poor. The Functions of Public Welfare. New York: Vintage. Frankel, Philip H. 1988. Urbanisation and Informal Settlement in the PWV Complex. Unpublished report. Department of Political Studies, University of the Witwatersrand, Johannesburg. Fraser, Craig. 2003. Shack Chic. Art and Innovation in South African Shack-Lands. Cape Town: Quivertree Publications. Frederikse, Julie. 1990. The Unbreakable Thread. Non-Racialism in South Africa. Bloomington, IN: Indiana University Press. Freund, Bill. 1988. The African Worker. Cambridge: Cambridge University Press. Friedman, Steven. 1987. Building Tomorrow Today. African Workers in Trade Unions, 1970-1984. Johannesburg: Ravan Press. Friedman, Steven and Ivor Chipkin. 2001. A Poor Voice?: The Politics of Inequality in South Africa. Research Report 87. Johannesburg: Centre for Policy Studies. Frye, Isobel. 2007. What is Poverty? A Qualitative Reflection of Peoples Experiences of Poverty. Johannesburg: National Economic Development and Labour Institute. Frye, Isobel and Karen Kallmann. 2003. The BIG Coalition in South Africa: Making it Happen. In A Basic Income Grant for South Africa, edited by Guy Standing and Michael Samson, 102-19. Cape Town: University of Cape Town Press. 449

Gelb, Stephen. 2006. The RDP, GEAR and All That: Reflections Ten Years Later. Transformation 62: 1-8. Gevisser, Mark. 2007. Thabo Mbeki. The Dream Deferred. Johannesburg: Jonathan Ball. Gibbs, Pat. 2007. Women, Labour and Resistance: Case Studies from the Port Elizabeth/Uitenhage Area, 1972-94. In Women in South African History. Basusiimbokodo, Bawelimilambo/ They Remove Boulders and Cross Rivers, edited by Nomboniso Gasa, 315-43. Pretoria: HSRC Press. Gibson, Nigel, ed. 2006. Challenging Hegemony. Social Movements and the Quest for a New Humanism in Post-Apartheid South Africa. Trenton, NJ: Africa World Press. Giliomee, Hermann. 1992. Wretched Folk, Ready for Any Mischief: The South African States Battle to Incorporate Poor Whites and Militant Workers, 1890-1939. Historia 47 (2): 601-653. Gilroy, Paul. 1991. There Aint No Black in the Union Jack. The Cultural Politics of Race and Nation. Chicago: University of Chicago Press. Ginsburg, David, Eddie Webster, Roger Southall, Geoff Wood, Sakhela Buhlungu, Johann Maree, Janet Cherry, Richard Haines, and Gilton Klerck. 1995. Taking Democracy Seriously. Worker Expectations of Parliamentary Democracy in South Africa, Durban: Indicator Project South Africa. Godfrey, Shane. 2007. Collective Bargaining Update. Forms of Collective Bargaining in South Africa and Prospects. In Bargaining Indicators 2007. A Collective Bargaining Omnibus, edited by Trenton Elsley, 27-43. Cape Town: Labour Research Service. Godfrey, Shane and Marlea Clarke. 2002. The Basic Conditions of Employment Act Amendments: More Questions than Answers. Law, Democracy and Development 6 (1): 1-26. 450

Godfrey, Shane, Johann Maree, and Jan Theron. 2006a. Regulating the Labour Market: The Role of Bargaining Councils. Industrial Law Journal 27 (4): 731-52. Godfrey, Shane, Johann Maree, and Jan Theron. 2006b. Flexibility in Bargaining Councils: The Role of Exemptions. Industrial Law Journal 27 (7): 1368-86. Goetz, Graeme. 2000. Shoot Anything that Flies, Claim Anything that Falls: Labour and the Changing Definition of the Reconstruction and Development Programme. in Trade Unions and Democratization in South Africa, 1985-1997, edited by Glenn Adler and Edward Webster, 159-89. Basingstoke: MacMillan. Goodhew, David. 2000. Working-Class Respectability: The Example of the Western Areas of Johannesburg, 1930-55. Journal of African History 41 (2): 241-66. Gorz, Andre. 1999. Reclaiming Work. Beyond the Wage-Based Society. Cambridge: Polity Press. Govender, Prem and James Aiello. 1999. Johannesburgs Strategic Plan for Municipal Service Partnerships. Development Southern Africa 16 (4): 649-67. Grest, Jeremy. 2001. Urban Management, Urban Citizenship and the Informal Economy in the New South Africa: A Case Study of Central Durban. Working Paper No.01-09d, Princeton: Princeton University, Center for Migration and Development. Grosfoguel, Ramon. 2002. Colonial Difference, Geopolitics of Knowledge, and Global Coloniality in the Modern/Colonial Capitalist World System. Review 25 (3): 203-24. Grossman, Jonathan. 1997. The Right to Strike and Worker Freedom in and Beyond Apartheid. In Free and Unfree Labour. The Debate Continues, edited by Tom Brass and Marcel van der Linden, 145-70. Bern: Peter Lang.

451

Gumede, William M. 2005. Thabo Mbeki and the Battle for the Soul of the ANC. Cape Town: Zebra. Guy, Jeff and Motlatsi Thabane. 1987. The Ma-Rashea. A Participants Perspective. In Class, Community and Conflict. South African Perspectives, edited by Belinda Bozzoli, 436-56. Johannesburg: Ravan Press. Guyer, Jane I. 2004. Marginal Gains. Monetary Transactions in Atlantic Africa. Chicago: University of Chicago Press. Hajdu, Flora. 2005. Relying on Jobs Instead of the Environment? Patterns of Local Securities in Rural Eastern Cape, South Africa. Social Dynamics 31 (1): 235-60. Handler, Joel F. 2004. Social Citizenship and Workfare in the United States and Western Europe: The Paradox of Inclusion. New York, Cambridge University Press. Hardt, Michael. 1999. Affective Labor. Boundary 2 26 (2): 89-100. Hardt, Michael and Antonio Negri. 1996. Labor of Dyonisus. A Critique of the State Form. Minneapolis: University of Minnesota Press. Hardt, Michael and Antonio Negri. 2000. Empire. Cambridge, MA: Harvard University Press. Harries, Patrick. 1994. Work, Culture and Identity. Migrant Laborers in Mozambique and South Africa, c.1860-1910. Portsmouth, NH: Heinemann. Hart, Gillian. 2002. Disabling Globalisation. Places of Power in Post-Apartheid South Africa. Durban: University of Natal Press. Hart, Gillian. 2006. Post-apartheid Developments in Historical and Comparative Perspective. In The Development Decade? Economic and Social Change in South Africa, 1994-2004, edited by Vishnu Padayachee, 13-32. Pretoria: HSRC Press.

452

Hart, Gillian and Ari Sitas. 2004. Beyond the Urban-Rural Divide: Linking Land, Labour, and Livelihoods. Transformation 56: 31-38. Hayem, Judith. 2008. La figure ouvrire en Afrique du Sud. Paris: Karthala. Hemson, David. 1978. Trade Unionism and the Struggle for Liberation in South Africa. Capital & Class 6: 1-41. Hendricks, Fred. 2006. Public Pensions in South Africa. Debt, Development and Corporatisation. Unpublished paper for the Project on Public Pensions in a Development Context. Geneva: United Nations Research Institute for Social Development. Herzenberg, Collette and Masibonge Mzwakali. 2003. The Growth and Development Summit: A Terrain to Settle Ideological Battles or a Platform to Map Out A National Growth and Employment Strategy? Cape Town: IDASA. Herzog, Hanna. 2005. On Home Turf. Interview Location and Its Social Meaning. Qualitative Sociology 28 (1): 25-47. Higginson, John. 1989. A Working Class in the Making. Belgian Colonial Labor Policy, Private Enterprise, and the African Mineworker, 1907-1951. Madison, WI: University of Wisconsin Press. Hindess, Barry. 2004. Liberalism. Whats in a Name? In Global Governmentality. Governing International Spaces, edited by Wendy Larner and William Walters, 23-39. London: Routledge. Hindson, Doug. 1987. Pass Controls and the Urban African Proletariat in South Africa. Johannesburg: Ravan Press. Hindson, Doug and Marian Lacey. 1983. Influx Control and Labour Allocation: Policy and Practice since the Riekert Commission. South African Review 1: 97-113. 453

Hirsch, Alan. 2004. Governments Plan to Combat Poverty. Focus. Helen Suzman Foundation 35: 6-12. Hirsch, Alan. 2005. Season of Hope. Economic Reform under Mandela and Mbeki. Durban: University of KwaZulu-Natal Press. Hirschman, Albert O. 1991. The Rhetoric of Reaction. Perversity, Futility, Jeopardy, Cambridge, MA: Belknap Press. Hirschsohn, Philip, Shane Godfrey, and Johann Maree. 2000. Industrial Policy-Making in the Automobile and the Textile and Clothing Sectors. Labours Strategic Ambivalence. In Engaging the State and Business. The Labour Movement and Co-determination in Contemporary South Africa, edited by Glenn Adler, 101-32. Johannesburg: Witwatersrand University Press. Hoffmann, Elizabeth A. 2007. Open-Ended Interviews, Power, and Emotional Labor. Journal of Contemporary Ethnography 36 (3): 318-46. Hofmeyr, Jan H. 1936. The Approach to the Native Problem. Race Relations 3 (2/3): 27-38. Holmstrom, Mark. 1984. Industry and Inequality. The Social Anthropology of Indian Labour. Cambridge: Cambridge University Press. Hong, Yong-Sun. 2005. Neither Singular Nor Alternative: Narratives of Modernity and Welfare in Germany, 1870-1945. Social History 30 (2): 133-53. Honneth, Axel. 2004. Organized Self-Realization. Some Paradoxes of Individualization. European Journal of Social Theory 7 (4): 463-78. Hosegood, Victoria and Ian M. Timaeus. 2006. Household Composition and Dynamics in KwaZulu Natal, South Africa. In African Households: Censuses and Surveys, edited by Etienne van de Walle, 5877. Armonk, NY: M.E. Sharpe. 454

Human Sciences Research Council. 2003a. Synthesis Report on Social and Economic Impacts of Government Programmes since 1994. Prepared for the Ten Year Review Research Programme of the Policy Coordination and Advisory Services, Office of the Presidency. Pretoria: Human Sciences Research Council. Human Sciences Research Council. 2003b. Household Indebtedness in South Africa in 1995 and 2000. Report Prepared by the Employment and Economic Policy Programme of the HSRC for the Micro-Finance Regulatory Council. Pretoria: Human Sciences Research Council. Hunter, Mark. 2004. Fathers without Amandla? Gender and Fatherhood among isiZulu Speakers. Journal of Natal and Zulu History 22: 149-60. Hurwitz, Ingrid and John Luiz. 2007. Urban Working Class Credit Usage and Over Indebtedness in South Africa. Journal of Southern African Studies 33 (1): 107-31. Hutchinson, Sharon. 1996. Nuer Dilemmas. Coping with Money, War, and the State. Berkeley: University of California Press. Iliffe, John. 1987. The African Poor. A History. Cambridge: Cambridge University Press. Isaacman, Allen. 1996. Cotton Is the Mother of Poverty: Peasants, Work, and Rural Struggle in Colonial Mozambique, 1938-1961. Portsmouth, NH: Heinemann. James, Deborah. 2007. Gaining Ground? Rights and Property in South African Land Reform. Abingdon: Routledge-Cavendish. Jansen, Martin. 2001. The Millennium Labour Council. South African Labour Bulletin 25 (2): 54-60. Jensen, Eileen. 1928. Poor Relief in Johannesburg. Journal of the Economic Society of South Africa 2 (1): 26-36. 455

Joffe, Avril, David Kaplan, Raphael Kaplinsky, and David Lewis. 1994. An Industrial Strategy for a Post-Apartheid South Africa. IDS Bulletin 25 (1): 17-23. Johnson, Richard W. 1999. Not So Close to Their Hearts. An Investigation into the Non-Payment of Rents, Rates and Service Charges in South Africas Towns and Cities. Johannesburg: Helen Suzman Foundation. Jones, Chris and Tony Novak. 1999. Poverty, Welfare and the Disciplinary State. London: Routledge. Joyce, Patrick. 1980. Work, Society & Politics. The Culture of the Factory in Late Victorian England. Brighton: Harvester Press. Kalati, Noushin and James Manor. 1999. Elite Perceptions of Poverty: South Africa. IDS Bulletin 30 (2): 117-26. Kant, Immanuel. 1998. Groundwork of the Metaphysics of Morals. Translated and edited by Mary Gregor. New York: Cambridge University Press. Karis, Thomas and Gwendoline M. Carter, eds. 1972. From Protest to Challenge. A Documentary History of African Politics in South Africa 1882-1964. Vol.I: Protest and Hope 1882-1934. Stanford, CA: Hoover Institution Press. Karis, Thomas and Gwendoline M. Carter. 1973. From Protest to Challenge. A Documentary History of African Politics in South Africa 1882-1964. Vol.II: Hope and Challenge 19351952. Stanford, CA: Hoover Institution Press. Keenan, Jeremy. 1981. Migrants Awake. The 1980 Johannesburg Municipal Workers Strike. South African Labour Bulletin 6 (7): 4-60.

456

Kenny, Bridget. 2003. From Insurrectionary Worker to Contingent Citizen: Restructuring Labor Markets and Repositioning East Rand (South Africa) Retail Sector Workers. City and Society 15 (1): 31-57. Kenny, Bridget. 2004. Selling Selves: East Rand Retail Sector Workers Fragmented and Reconfigured. Journal of Southern African Studies 30 (3): 477-98. Kenny, Bridget. 2005. Divisions of Labor, Experiences of Class: Changing Collective Identities of East Rand Food Retail Sector Workers through South Africas Democratic Transition. Ph.D. diss. University of Wisconsin, Madison. Kenny, Bridget and Eddie Webster. 1999. Eroding the Core: Flexibility and the Resegmentation of the South African Labour Market. Critical Sociology 24 (3): 216-43. Khanna, Ranjana. 2003. Dark Continents. Psychoanalysis and Colonialism. Durham, NC: Duke University Press. Khanna, Ranjana. 2007. Indignity. Ethnic and Racial Studies 30 (2): 257-80. Kiernan, J.P. 1977. Poor and Puritan: An Attempt to View Zionism as a Collective Response to Urban Poverty. African Studies 36 (1): 31-42. Klein, Naomi. 2007. The Shock Doctrine. The Rise of Disaster Capitalism. New York: Metropolitan Books. Koch, Eddie. 1983. Without Visible Means of Subsistence: Slumyard Culture in Johannesburg, 1918-1940. In Town and Countryside in the Transvaal. Capitalist Penetration and Popular Response, edited by Belinda Bozzoli, 151-75. Johannesburg: Ravan Press. Kok, Pieter. 1998. South Africas Magnifying Glass: A Profile of Gauteng. Pretoria: Human Sciences Research Council. 457

Kraak, Andre. 1987. "Uneven Capitalist Development. A Case Study of Deskilling and Reskilling in South Africa's Metal Industry." Social Dynamics 13 (2): 14-31. Kraak, Andre. 1992. Human Resources Development and Organised Labour. South African Review 6: 403-23. Kraak, Gerald. 1993. Breaking the Chains. Labour in South Africa in the 1970s and 1980s. London: Pluto Press. Kruger, John. 1992. An Overview of the South African Social Security System. Paper presented at the workshop of the Economic Project on Social Safety Nets, University of Stellenbosch, June 1-2. Kymlicka, Will. 1998. Multicultural Citizenship. In The Citizenship Debates. A Reader, edited by Gershon Shafir, 167-88. Minneapolis: University of Minnesota Press. Lacan, Jacques. 1977. The Freudian Thing, or the Meaning of the Return to Freud in Psychoanalysis. In Ecrits. A Selection, 114-45, New York: Tavistock. Lacan, Jacques. 1998. Seminars. Book 11: The Four Fundamental Concepts of Psychoanalysis. New York: Norton. La Hausse, Paul. 1993. So Who Was Elias Kuzwayo? Nationalism and the Picaresque in Natal. In Apartheids Genesis, 1935-1962, edited by Philip Bonner, Peter Delius, and Deborah Posel, 195-228. Johannesburg: Ravan Press. Lambert, Rob and Eddie Webster. 1988. The Re-emergence of Political Unionism in Contemporary South Africa? In Popular Struggles in South Africa, edited by William Cobbett and Robin Cohen, 20-41.Trenton, NJ: Africa World Press. Lamont, Michelle. 2000. The Dignity of Working Men. Morality and the Boundary of Race, Class and Immigration. New York: Russell Sage Foundation. 458

Lange, Lis. 2003. White, Poor and Angry. White Working Class Families in Johannesburg. Aldershot: Ashgate. Latour, Bruno. 2005. Reassembling the Social. An Introduction to Actor-Network Theory. Oxford: Oxford University Press. Lazarus, Neil. 2004. The South African Ideology: The Myth of Exceptionalism, the Idea of Renaissance. South Atlantic Quarterly 103 (4): 607-28. Lefebvre, Henri. 1977. De lEtat. Vol.3: Le Mode de Production Etatique. Paris: Union Gnrale dEditions. Legassick, Martin. 1974a. South Africa: Capital Accumulation and Violence. Economy & Society 3 (3): 253-91. Legassick, Martin. 1974b. Legislation, Ideology and Economy in Post-1948 South Africa. Journal of Southern African Studies 1 (1): 5-35. Lekota, Mosiuoa. 1986. In Search of the Charter. In 30 Years of the Freedom Charter, edited by Raymond Suttner and Jeremy Cronin, 195-203. Johannesburg: Ravan Press. Lemke, Stefanie, Fanie Jansen van Rensburg, Hester Vorster, and Joachim Ziche. 2002. Food Security, Social and Gender Relations in Black South African Households: Ways of Coping and Survival. In Challenges for Anthropology in the African Renaissance: A Southern African Contribution, edited by Debie LeBeau and Robert J. Gordon, 15062. Windhoek: University of Namibia Press. Lentin, Alana. 2004. Racial States, Anti-Racist Responses. Picking Holes in Culture and Human Rights. European Journal of Social Theory 7 (4): 427-443. Leon, Natalie and Ray Mabope. 2005. The Private Health Sector. In South African Health Review 2005, 32-43. Durban: Health Systems Trust. 459

Le Roux, Pieter. 2001. The Case for a Basic Income Grant in South Africa. Research paper for the Taylor Committee of Inquiry into Comprehensive Social Security, 4th Draft. Institute for Social Development, University of the Western Cape, Bellville. Levy, Norman. 1982. The Foundations of the South African Cheap Labour System. London: Routledge and Kegan Paul. Lewis, Joanna. 2000. Empire State-Building. War & Welfare in Kenya 1925-1952. Oxford: James Currey. Li, Tanya M. 2007. The Will to Improve. Governmentality, Development, and the Practice of Politics. Durham, NC: Duke University Press. Liebenberg, Sandra. 2001. The Right to Social Assistance: The Implications of Grootboom for Policy Reform in South Africa. South African Journal on Human Rights 17 (2): 232-57. Liebenberg, Sandra and Karrisha Pillay. 1998. Economic and Social Rights Report. Vol. 5: SANGOCOs Report on Poverty and Human Rights. Johannesburg: South African Human Rights Commission. Ligthelm, A.A. 2005a. Regional Growth Patterns in South Africa: Evidence from Private Sector Building Activities. Research Report 346, Bureau of Market Research, University of South Africa, Pretoria. Ligthelm, A.A. 2005b. Measuring the Size of the Informal Sector in South Africa, 2004/05. Research Report 349, Bureau of Market Research, University of South Africa, Pretoria. Ligthelm, A.A. 2005c. Informal Retailing through Home-Based Micro-Enterprises: The Role of Spaza Shops. Development Southern Africa 22 (2): 199-214. Lindsay, Lisa. 2003. Working with Gender. Wage Labor and Social Change in Southwestern Nigeria. Portsmouth, NH: Heinemann. 460

Lodge, Tom. 1983. Black Politics in South Africa since 1945. London: Longman. Lodge, Tom. 1999. Policy Processes within the African National Congress and the Tripartite Alliance. Politikon 26 (1): 5-32. Lubeck, Paul. 1985. Islamic Protest under Semi-Industrial Capitalism: Yan Tatsine Explained. International Journal of African Historical Studies 55 (4): 369-89. Lund, Frances. 2008. Changing Social Policy. The Child Support Grant in South Africa. Pretoria: HSRC Press. Maas, Gideon and Mark Herrington. 2007. Global Entrepreneurship Monitor. South African Report 2007. Cape Town: Graduate School of Business, University of Cape Town. Mabin, Alan. 2007. Johannesburg. (South) Africas Aspirant Global City. In The Making of Global City Regions. Johannesburg, Mumbai/Bombay, So Paulo, and Shanghai, edited by Klaus Segbers, 32-63. Baltimore: Johns Hopkins University Press. Mabin, Alan and Susan Parnell. 1984. The Question of Working-class Home Ownership. Second Carnegie Inquiry into Poverty and Development in Southern Africa, Conference Paper 159. Cape Town: University of Cape Town, SALDRU. MacDonald, Michael. 2006. Why Race Matters in South Africa. Cambridge, MA: Harvard University Press. Machaka, Johannes and Simon Roberts. 2006. Addressing the Apartheid Industrial Legacy: Local Economic Development and Industrial Policy in South Africa. The Case of Ekurhuleni. In Sustainable Manufacturing? The Case of South Africa & Ekurhuleni, edited by Simon Roberts, 121-40. Cape Town: Juta. Madonsela, Vusi. 2008. Overview of the Social Security Programmes in South Africa. Paper presented at the International Social Security Association conference on Comprehensive 461

Social Security for Poverty Reduction and Sustainable Livelihoods in Africa: Challenges and Achievements. Cape Town, March 10-14. Magasela, Wiseman, Gemma Wright, Andrew Ratcliffe, and Michael Noble. 2006. Findings from the Indicators of Poverty and Social Exclusion Project: Housing. Key Report 3. Pretoria: Department of Social Development. Mager, Anne and Gary Minkley. 1993. Reaping the Whirlwind. The East London Riots of 1952. In Apartheids Genesis, 1935-1962, edited by Philip Bonner, Peter Delius, and Deborah Posel, 229-51. Johannesburg: Ravan Press. Mahlangu, Qedani D. 2007. Speech by Gauteng MEC for Local Government Qedani Dorothy Mahlangu at the Gauteng Global City Region (GCR) Models Conference. April 20, 2007. http://www.polity.org.za/article.php?a_id=107792. Retrieved on December 2, 2008. Mair, Lucy P. 1944. Welfare in the British Colonies. London: Royal Institute of International Affairs. Makgetla, Neva and Liv Trres. 2006. The Workers Survey for COSATU. Johannesburg: NALEDI. Makhoba, Mandlenkosi. 1984. The Sun Shall Rise for the Workers/Ilanga Lizophumela Abasebenzi. Johannesburg: Ravan Press. Makino, Kumiko. 2003. Social Security Reform in Post-Apartheid South Africa. A Focus on the Basic Income Grant. Paper presented at the 19th World Congress of the International Political Sciences Association, Durban, June 29 July 5.

462

Malinga, Simangaliso S. 2000. The Development of Informal Settlements in South Africa, with Particular Reference to Informal Settlements Around Daveyton on the East Rand, 19701999. Ph.D. diss. Johannesburg, Rand Afrikaans University. Mamdani, Mahmood. 1996. Citizen and Subject. Contemporary Africa and the Legacy of Late Colonialism. Princeton: Princeton University Press. Mangcu, Xolela. 2002. Toward a New Approach to Welfare Policy in South Africa: Building Assets in Poor Communities. In Democratic Governance and Social Inequality, edited by Joseph S. Tulchin and Amelia Brown, 91-102. Boulder, CO: Lynne Rienner. Manuel, Trevor. 2004. Data for Poverty Measurement. Opening Address at the Joint National Treasury-World Bank Workshop. Pretoria, June 28. Marais, Hein. 1993. Squaring the Magic Circle. Work in Progress 90: 13-14. Marais, Hein. 2001. South Africa: Limits to Change. The Political Economy of Transformation. Second edition, London: Zed Books; Cape Town: University of Cape Town Press. Marais, Hein. 2002. The Logic of Expediency. Post-Apartheid Shifts in Macroeconomic Policy. In Thabo Mbekis World. The Politics and Ideology of the South African President, edited by Sean Jacobs and Richard Calland, 83-103. Pietermaritzburg: University of Natal Press. Maree, Johann. 1982. Democracy and Oligarchy in Trade Unions: The Independent Trade Unions in the Transvaal and the Western Province General Workers Union in the 1970s. Social Dynamics 8 (1): 41-52. Marie, Bobby. 1992. COSATU Faces Crisis. Quick Fix Methods and Organizational Contradictions. South African Labour Bulletin 16 (5): 20-26. Markham, Coletane. 1987. Organising Local Authority Workers. South African Labour Bulletin 12 (4): 89-107. 463

Marshall, Thomas H. 1964 (1950). Citizenship and Social Class. In Thomas H. Marshall, Class, Citizenship and Social Development, 71-134. Chicago: University of Chicago Press. Martinot, Steve. 2003. The Rule of Racialization. Class, Identity, Governance. Philadelphia: Temple University Press. Martins, J.H. 2004. Total Household Expenditures in South Africa by Income Group, Life Plane, Life Stage, and Product, 2004. Research Report 326, Bureau of Market Research, University of South Africa, Pretoria. Marx, Anthony W. 1998. Making Race and Nation. A Comparison of South Africa, the United States and Brazil. Cambridge: Cambridge University Press. Marx, Karl. 1973. Grundrisse. Foundations of the Critique of Political Economy. Translated by Martin Nicolaus. Harmondsworth: Penguin. Mashinini, Emma. 1991. Strikes Have Followed Me All My Life. A South African Autobiography. New York: Routledge. May, Julian. 2000. Growth, Development, Poverty and Inequality. In Poverty and Inequality in South Africa: Meeting the Challenge, edited by Julian May, 1-16. London: Zed Books. May, Julian and Nina Hunter. 2005. Poverty, Inequality and the Social Wage: Balancing the Books in Postapartheid South Africa? In Democratising Development. The Politics of Socio-Economic Rights in South Africa, edited by Peris Jones and Kristian Stokke, 10129. Leiden: Martinus Nijhoff. Mayer, Philip. 1981. The Origins and Decline of Two Rural Resistance Ideologies. In Black Villagers in an Industrial Society. Anthropological Perspectives on Labour Migration in South Africa, edited by Philip Mayer, 1-80. Oxford: Oxford University Press. 464

Mbeki, Thabo. 1998. Statement of the President of the African National Congress, Thabo Mbeki, at the 10th Congress of the South African Communist Party. Johannesburg, July 2. Mbeki, Thabo. 1999. Speech at the Annual National Conference of the Black Management Forum. Kempton Park, November 20. Mbeki, Thabo. 2003a. State of the Nation Address of the President of South Africa, Thabo Mbeki, to the Houses of Parliament. February 14. http://www.thepresidency.gov.za/show.asp?type=son&include=president/son/sp021403.htm. Retrieved on January 5, 2008. Mbeki, Thabo. 2003b. Response of the President of the Republic, Thabo Mbeki, to the Parliamentary Debate on the State of the Nation Address. February 18. http://www.info.gov.za/speeches/2003/03021815461003.htm. Retrieved on November 1, 2008. Mbeki, Thabo. 2003c. Address to the National Council of Provinces. November 11. http://www.info.gov.za/speeches/2003/03111115461004.htm. Retrieved on December 1, 2008. Mbeki, Thabo. 2006. Nelson Mandela Memorial Lecture. University of the Witwatersrand, Johannesburg, July 29. Mbembe, Achille. 2001. On the Postcolony. Berkeley: University of California Press. Mbembe, Achille and Sarah Nuttall. 2004. Writing the World from an African Metropolis. Public Culture 16 (3); 347-72. Mboweni, Tito. 1992. The Role of the Trade Union Movement in the Future South Africa. South African Labour Bulletin 16 (8): 24-29. 465

McDonald, David. 2002. The Theory and Practice of Cost Recovery in South Africa. In Cost Recovery and the Crisis of Service Delivery in South Africa, edited by David McDonald and John Pape, 17-41. Pretoria: HSRC Press. Meintjes, Helen. 2001. Washing Machines Make Lazy Women. Domestic Appliances and the Negotiation of Womens Propriety in Soweto. Journal of Material Culture 6 (3): 34563. Meth, Charles. 2003a. Mass Poverty, Slow Economic Growth and Redistribution. Social Dynamics 29 (1): 99-130. Meth, Charles. 2004. Ideology and Social Policy: Handouts and the Spectre of Dependency. Transformation 56: 1-30. Meth, Charles. 2006. Half-measures Revisited: The ANCs Unemployment and Poverty Reduction Goals. In Poverty and Policy in Post-Apartheid South Africa, edited by Haroon Bhorat and Ravi Kanbur, 366-458. Pretoria: HSRC Press. Meth, Charles. 2007a. Sticking to the Facts: Official and Unofficial Stories about Poverty and Unemployment in South Africa. Working Paper 07/123, Development Policy Research Unit, University of Cape Town. Meth, Charles. 2007b. Rational Ignorance and South African Poverty Statistics. Africanus 37 (2): 81-110. Meth, Charles. 2008. Whats the BIG Deal? A Bigger Slice for the Poor! Paper presented at the International Social Security Association conference on Comprehensive Social Security for Poverty Reduction and Sustainable Livelihoods in Africa: Challenges and Achievements. Cape Town, March 10-14.

466

Meth, Charles and Rosa Dias. 2004. Increases in Poverty in South Africa, 1999-2002. Development Southern Africa 21 (1): 59-85. Meth, Charles and Solveig Piper. 1984. Social Security in Historical Perspective. Second Carnegie Inquiry into Poverty and Development in Southern Africa, Conference Paper 250. Cape Town: University of Cape Town, SALDRU. Meth, Charles, Ravi Naidoo, and Barry Shipman. 1996. Unemployment Insurance and Related Coverage Issues. Report of the Ministerial Task Team. Pretoria: Department of Labour. Midgley, James. 1997. Social Welfare in a Global Context. Thousand Oaks, CA: Sage. Mitchell, Timothy. 2002. Rule of Experts. Egypt, Techno-Politics, Modernity. Berkeley: University of California Press. Modsell, Tim. 2006. Free Basic Services: The Evolution and Impact of Free Basic Water Policy in South Africa. In Democracy and Delivery. Urban Policy in South Africa, edited by Udesh Pillay, Richard Tomlinson, and Jacques du Toit, 283-301. Cape Town: HSRC Press. Mohammed, Seeraj. 2008. Macroeconomic Indicators for Collective Bargaining. In Bargaining Indicators, Vol. 12, edited by Trenton Elsley, 105-22. Cape Town: Labour Research Service. Molamu, Louis. 2002. Black Exclusivism and Trade Unionism in South Africa: The Case of the Black Allied Workers Union, 1973-1984. In Participation, Globalisation and Culture. International and South African Perspectives, edited by Gyrgy Szll, Dasarath Chetty, and Alain Chouraqui, 228-45. Frankfurt-am-Main: Peter Lang.

467

Momoh, Abubakar. 2000. Youth Culture and Area Boys in Lagos. In Identity Transformation and Identity Politics under Structural Adjustment in Nigeria, edited by Attahiru Jega, 181-203. Uppsala: Nordic Africa Institute. Mongale, Moathlodi and Pauline Sekaledi. 2005. An Unfinished Strike. South African Labour Bulletin 29 (3): 23-25. Monnier, Jean-Marie and Carlo Vercellone. 2007. Fondements et faisabilit du revenu social garanti. Multitudes 27: 73-84. Moodie, T. Dunbar. 1994. Going for Gold. Men, Mines and Migration. Berkeley: University of California Press. Morapedi, Hosea. 1996. The State of the Union at Dorbyl. Unpublished report. Johannesburg: National Union of Metalworkers of South Africa. Morris, Alan. 1998. Continuity or Rupture. The City, Post-Apartheid. Social Research 65 (4): 759-75. Morris, Alan, Belinda Bozzoli, Jackie Cock, Owen Crankshaw, Leah Gilbert, Lebo LehutsoPhooko, Deborah Posel, Zwelakhe Tshandu, and Elsa van Huyssteen. 1999. Change and Continuity: A Survey of Soweto in the Late 1990s. Department of Sociology, University of the Witwatersrand, Johannesburg. Morris, Rosalind C. 2006. The Mute and the Unspeakable: Political Subjectivity, Violent Crime and the Sexual Thing in a South African Mining Community. In Order and Disorder in the Postcolony, edited by Jean Comaroff and John L. Comaroff, 57-101. Chicago: University of Chicago Press. Moulier-Boutang, Yann. 1998. De lesclavage au salariat. Economie historique du salariat brid. Paris : Presses Universitaires de France. 468

Murray, Martin. 2008. Taming the Disorderly City. The Spatial Landscape of Johannesburg after Apartheid. Ithaca, NY: Cornell University Press. Nakano-Glenn, Evelyn. 2002. Unequal Freedom. How Race and Gender Shaped American Citizenship and Labor. Cambridge, MA: Harvard University Press. NALEDI (National Labour and Economic Development Institute). 2006. The State of COSATU. Phase One Report. Johannesburg: NALEDI. Nash, Andrew. 2009. The Dialectical Tradition in South Africa. New York: Routledge. National Institute of Economic Policy. 2000. Anti-Poverty Policies in South Africa. Unpublished report, National Institute of Economic Policy, Johannesburg. Nattrass, Nicoli. 2005. Economic Growth and Transformation in the 1940s. In South Africas 1940S: Worlds of Possibilities, edited by Saul Dubow and Alan Jeeves, 20-43. Cape Town: Double Storey Books. Nattrass, Nicoli. 2006. Trading Off Income and Health?: AIDS and the Disability Grant in South Africa. Journal of Social Policy 35 (1): 3-19. Nattrass, Nicoli. 2007. Disability and Welfare in South Africas Era of Unemployment and AIDS. In State of the Nation. South Africa 2007, edited by Sakhela Buhlungu, John Daniel, Roger Southall, and Jessica Luchtman, 179-200. Pretoria: HSRC Press. Nattrass, Nicoli and Jeremy Seekings. 1997a. Citizenship and Welfare in South Africa: Deracialisation and Inequality in a Labour-Surplus Economy. Canadian Journal of African Studies 31 (3): 452-81. Nattrass, Nicoli and Jeremy Seekings. 1997b. Constitutional and Legislative Provisions Governing Citizenship, Class and the Labour Relations Act. In Democracy, Human

469

Rights and Economic Development in Southern Africa, edited by Nico Steytler, 29-45. Johannesburg: Lex Patria Publishers. Nattrass, Nicoli and Jeremy Seekings. 2000. Apartheid Revisited: Analysing Apartheid as a Distributional Regime. Paper presented at the African Studies Seminar Series, University of Cape Town, October 11. Ndjio, Basile. 2005. Carrefour de la Joie: Popular Deconstruction of the African Postcolonial Public Sphere. Africa 75 (3): 265-94. NEDLAC (National Economic Development and Labour Council). 1995. Discussion Document on a Framework for Social Partnership and Agreement-Making at NEDLAC. October 16. NEDLAC (National Economic Development and Labour Council). 1999a. Draft Report of the Working Group on Social Security, Social Wage and the Social Plan. April 6. NEDLAC (National Economic Development and Labour Council). 1999b. Social Security Task Team. Draft Conceptual Framework for Developing a Comprehensive Social Security System in South Africa. July 22. NEDLAC (National Economic Development and Labour Council). 1999c. Report of a NEDLAC Workshop on the Commission for Social Security. NEDLAC (National Economic Development and Labour Council). 2000. Social Security Task Team. Draft Report of the Social Security Task Team Meeting. March 1. Negri, Antonio. 2005 (1977). Domination and Sabotage. On the Marxist Method of Social Transformation. In Books for Burning. Between Civil War and Democracy in 1970s Italy, 231-90. London: Verso.

470

Neocosmos, Michael. 2009. Explaining Xenophobia in Post-apartheid South Africa and the Limits of Sociological Analysis: Some Provocative Thoughts. Paper presented at the Annual Congress of the South African Sociological Association, University of the Witwatersrand, Johannesburg, June 28 July 2. Ngoasheng, Moses. 1989. We Organize and Educate. . . Cultural Intellectuals within the Labour Movement. Staffrider 8 (3/4): 29-38. Ngwane, Zolani. 2003. Christmas Time and the Struggles for the Household in the Countryside: Rethinking the Cultural Geography of Migrant Labour in South Africa. Journal of Southern African Studies 29 (3): 681-99. Ngwane, Zolani. 2004. Real Men Reawaken their Fathers Homesteads, the Educated Leave Them in Ruins: The Politics of Domestic Reproduction in Post-apartheid Rural South Africa. In Producing African Futures. Ritual and Reproduction in a Neoliberal Age, edited by Brad Weiss, 167-92. Leyden: Brill. Nieftagodien, Noor. 2006. Economy and Society on the East Rand in the 20th Century. In Sustainable Manufacturing? The Case of South Africa & Ekurhuleni, edited by Simon Roberts, 93-111. Cape Town: Juta. Niehaus, Isak. 2006a. Doing Politics in Bushbuckridge: Work, Welfare and the South African Elections of 2004. Africa 76 (4): 526-48. Niehaus, Isak. 2006b. Biographical Lessons: Life Stories, Sex, and Culture in Bushbuckridge, South Africa. Cahiers dEtudes Africaines 181: 51-73. Njikelana, Sisa. 1984. Unions and the UDF. Work in Progress 32: 30-33. Noble, Michael, Phakama Ntshongwana, and Rebecca Surender. 2008. Attitudes to Work and Social Security in South Africa. Pretoria: HSRC Press. 471

NUMSA (National Union of Metalworkers of South Africa). 1989. NUMSA Wits Regional Secretarys Report to the Wits Regional Congress. Johannesburg, August 13. NUMSA (National Union of Metalworkers of South Africa). 1996. Report of the General Secretary to the 5th National Congress. Kempton Park, September 25-29. NUMSA (National Union of Metalworkers of South Africa). 1999. Bargaining Proposals for the Engineering Industry, Johannesburg: NUMSA. Nussbaum, Martha C. 2000. Women and Human Development. The Capabilities Approach. New York: Cambridge University Press. Nussbaum, Martha C. 2003. Capabilities as Fundamental Entitlements: Sen and Social Justice. Feminist Economics 9 (2/3): 33-59. Offe, Claus. 1997. Towards a New Equilibrium of Citizens Rights and Economic Resources? In Societal Cohesion and the Globalising Economy: What Does the Future Hold?, edited by Wolfgang Michalski, Riel Miller, and Barrie Stevens, 81108. Paris: OECD. Old Mutual. 2005. The Old Mutual 2005 Healthcare Survey. Towards Social Health Insurance. Johannesburg: Old Mutual. Oldfield, Sophie and Kristian Stokke. 2007. Political Polemics and Local Practices of Community Organizing and Neoliberal Politics in South Africa. In Contesting Neoliberalism. Urban Frontiers, edited by Helga Leitner, Jamie Peck, and Eric S. Sheppard, 139-56. New York: Guildford. Olivier, Marius. 2000a. Toward a Coherent Approach to Social Protection, Labour Market Policies and Financing of Social Security: A New Paradigm for South Africa. South African Journal of Labour Relations 24 (3/4): 23-55.

472

Olivier, Marius. 2000b. Towards a Coherent Approach to Social Protection, Labour Market Policies and Financing of Social Security: A New Paradigm for South Africa? Paper presented at the Year 2000 International Research Conference on Social Security of the International Social Security Association, Helsinki, September 25-27. Ong, Aihwa. 2006. Neoliberalism as Exception. Mutations in Citizenship and Sovereignty. Durham, NC: Duke University Press. Palmer Development Group. 2004. Developing a Profile of Urban Poverty in the City of Johannesburg. Cape Town: Palmer Development Group. Parnell, Susan. 2004a. Constructing a Developmental Nation. The Challenge of Including the Poor in the Post Apartheid City. Paper presented to the conference on Overcoming Underdevelopment in South Africas Second Economy. Midrand, October 28-29. Parnell, Susan. 2004b. Building Developmental Local Government to Fight Poverty. International Development Planning Review 26 (4): 377-99. Parnell, Susan. 2007. Politics of Transformation. Defining the City Strategy in Johannesburg. In The Making of Global City Regions. Johannesburg, Mumbai/Bombay, So Paulo, and Shanghai, edited by Klaus Segbers, 139-67. Baltimore: Johns Hopkins University Press. Patel, Ebrahim. 1994. The Role of Organised Labour in a Democratic South Africa. In Worker Rights from Apartheid to Democracy. What Role for Organised Labour, edited by Ebrahim Patel, 1-29. Kenwyn: Juta. Patel, Leila. 1991. Social Welfare Options for South Africa, Vol.1. Ph.D. diss. University of the Witwatersrand, Johannesburg. Patel, Leila. 1992. Social Welfare Options for South Africa. Johannesburg: Ravan Press.

473

Patel, Leila. 2001. Redesigning Social Welfare Services: A South African Case Study. Unpublished paper. Patel, Leila. 2005. Social Welfare and Social Development in South Africa. Cape Town: Oxford University Press. Penvenne, Jean Marie. 1995. African Workers and Colonial Racism. Mozambican Strategies and Struggles in Loureno Marques, 1877-1962. Portsmouth, NH: Heinemann. Pillay, Pundy. 1984. Poverty in the Pretoria-Witwatersrand-Vereeniging Area: A Survey of Research. Second Carnegie Inquiry into Poverty and Development in Southern Africa, Conference Paper 18. Cape Town: University of Cape Town, SALDRU. Pillay, Yogan. 1995. The Policy of Exclusion: Health Policy in South Africa, 1910-1990. Ph.D. diss. Johns Hopkins University, Baltimore. Pillay, Udesh, Richard Tomlinson, and Jacques du Toit. 2006. Introduction. In Democracy and Delivery. Urban Policy in South Africa, edited by Udesh Pillay, Richard Tomlinson, and Jacques du Toit, 1-19. Pretoria: HSRC Press. Piper, Laurence. 2006. Beyond Neo-Liberalism, Beyond the ANC. South African Politics as Progressive Governance. In Shifting Boundaries of Knowledge. A View on Social Sciences, Law and Humanities in South Africa, edited by Tessa Marcus and Alexandra Hofmaenner, 141-59. Durban: University of KwaZulu-Natal Press. Pityana, Sipho and Mark Orkin. 1992. Beyond the Factory Floor. A Survey of COSATU ShopStewards. Johannesburg: Ravan Press. Pollin, Robert, Gerald Epstein, James Heintz, and Leonce Ndikumana. 2006. An EmploymentTargeted Economic Program for South Africa. Amherst, MA: Political Economy Research Institute; New York: United Nations Development Programme. 474

Posel, Deborah. 1991. The Making of Apartheid, 1948-1961. Conflict and Compromise. Oxford: Clarendon Press. Posel, Deborah. 1993. Influx Control and Urban Labour Markets. In Apartheids Genesis, 19351962, edited by Philip Bonner, Peter Delius, and Deborah Posel, 411-30. Johannesburg: Ravan Press. Posel, Deborah. 2005. The Case for a Welfare State: Poverty and the Politics of the Urban African Family in the 1930s and 1940s. In South Africas 1940s. Worlds of Possibilities, edited by Saul Dubow and Alan Jeeves, 64-86. Cape Town: Double Storey Books. Posel, Deborah. 2006. Marriage at the Drop of a Hat: Housing and Partnership in South Africas Urban African Townships, 1920s-1960s. History of Workshop Journal 61: 57-76. Posel, Dorrit R. 1999. Intra-Family Transfers and the Household Division of Labor: A Case Study of Migration and Remittance Behavior in South Africa, Ph.D. diss. University of Massachusetts, Amherst. Posel, Dorrit. 2003. Have Migration Patterns in Post-Apartheid South Africa Changed? Paper presented to the Conference on African Migration in Comparative Perspective, Johannesburg, June 4-7. Pottie, David. 1999. The South African Political Transition and Low-Cost Housing, 1994-1996. Ph.D. diss. York University, Toronto. Pottie, David. 2003. Housing the Nation. The Politics of Low-Cost Housing Policy in South Africa since 1994. Politeia 22 (1): 119-43. PPWAWU (Print, Paper, Wood and Allied Workers Union). 1993. Bi -Annual Report, Fifth National Congress. Johannesburg, July 23-25.

475

Prakash, Gyan. 1995. After Colonialism. In After Colonialism. Imperial Histories and Postcolonial Displacements, edited by Gyan Prakash, 3-17. Princeton: Princeton University Press. Prakash, Gyan. 1999. Another Reason. Science and the Imagination of Modern India. Princeton: Princeton University Press. Price, Max and Phakamile Tshazibane. 1989. Medical Aids: Questions for Unions. South African Labour Bulletin 13 (8): 100-09. Procacci, Giovanna. 1993. Gouverner la misre. La question sociale en France, 1789-1848. Paris: Editions du Seuil. Public Service Commission. 2003. Report on the Evaluation of the National Housing Subsidy Scheme. Pretoria: Public Service Commission. Qabula, Alfred T. 1989. A Working Life, Cruel Beyond Belief. Johannesburg: National Union of Metalworkers of South Africa. Quijano, Anibal. 2000. Coloniality of Power, Eurocentrism, and Latin America. Nepantla. Views from South 1 (3): 533-80. Rabinow, Paul. 1989. French Modern. Norms and Forms of the Social Environment. Cambridge, MA: MIT Press. Rachleff, Peter. 2001. The Current Crisis of the South African Labour Movement. Labour/ Le Travail 47: 151-70. Rapley, Timothy J. 2001. The Art(fullness) of Open-Ended Interviewing: Some Considerations on Analysing Interviews. Qualitative Research 1 (3): 303-23.

476

Rogerson, Chris. 2006. From National Industrial Workshop to Rustbelt? Restructuring the Manufacturing Economy in Ekurhuleni, 1980-1999. In Sustainable Manufacturing? The Case of South Africa & Ekurhuleni, edited by Simon Roberts, 112-20, Cape Town: Juta. Roitman, Janet. 2005. Fiscal Disobedience. An Anthropology of Economic Regulation in Central Africa. Princeton: Princeton University Press. Rojas, Cristina. 2004. Governing through the Social. Representations of Poverty and Global Governmentality. In Global Governmentality. Governing International Spaces, edited by Wendy Larner and William Walters, 97-115. New York: Routledge. Roos, Neil. 2005. Ordinary Springboks. White Servicemen and Social Justice in South Africa, 1939-1961. Aldershot: Ashgate. Roux, Renee. 1981. Striking for Pensions. South African Labour Bulletin 6 (8): 34-54. Royston, Lauren. 2002. Security of Urban Tenure in South Africa: Overview of Policy and Practice. in Holding their Ground. Secure Land Tenure for the Urban Poor in Developing Countries, edited by Alain Durand-Lasserve and Lauren Royston, 165-81. London: Earthscan. Ruiters, Greg. 2000. Urban Struggles and Defeats in 1980s South Africa. Reflections on the Politics of Scale. Urban Forum 11 (2): 289-310. Rust, Kecia. 2006. Analysis of South Africas Housing Sector Performance. Johannesburg: FinMark Trust. Sagner, Andreas and Raymond Z. Mtati. 1999. Politics of Pension Sharing in Urban South Africa. Ageing and Society 19 (4): 393-416.

477

Samson, Melanie. 2004. Reprivatising Services, Regendering the Labour Market: A Feminist Analysis of the Formation of the Pikitup Waste Management Utility by the Contracting Local State in Johannesburg. MA diss., University of the Witwatersrand, Johannesburg. Samson, Melanie. 2005. The Commodification of Street Cleaning in Johannesburg. Refracturing Citizenship and Resegmenting Labour Markets. Paper presented at the annual conference of the Canadian Association of African Studies, Montreal, April 27 May 1. Samson, Michael. 2007. An Options Assessment with Respect to Making the State Old Age Pension Universal. In Reform of Retirement Provisions. Feasibility Studies, 1-17. Pretoria: Department of Social Development. Samson, Michael, Una Lee, Asanda Ndlebe, Kenneth Mac Quene, Ingrid van Niekerk, Viral Gandhi, Tomoko Harigaya, and Celeste Abrahams. 2004. The Social and Economic Impact of South Africas Social Security System. Cape Town: Economic Policy Research Institute. Samson, Michael, Kenneth Mac Quene, Ingrid van Niekerk, Sheshi Kaniki, Karen Kallmann, and Martin Williams. 2007. Review of Targeting Mechanisms, Means Tests and Values for South Africas Social Grants. Final Report. Department of Social Development Project SD 13/2005. Cape Town: Economic Policy Research Institute. SAMWU (South African Municipal Workers Union). 1997. Secretariat Report to the 5th National Congress. Johannesburg, October 29-31. SAMWU (South African Municipal Workers Union). 1998. Memorandum to Minister of Provincial Affairs and Constitutional Development. July 23. SAMWU (South African Municipal Workers Union). 2003. Secretariat Report to the 7th National Congress. Johannesburg, August 19-22. 478

SANLAM. 2004. Twelfth Biennal Survey on Retirement Benefits. Johannesburg: SANLAM. Sapire, Hilary. 1987. The Stay-away of the Brakpan Location, 1944. In Class, Community and Conflict. South African Perspectives, edited by Belinda Bozzoli, 358-99. Johannesburg: Ravan Press. Sapire, Hilary. 1993. African Political Organisation in Brakpan in the 1950s. In Apartheids Genesis 1935-1962, edited by Philip Bonner, Peter Delius, and Deborah Posel, 252-74. Johannesburg: Witwatersrand University Press. Saul, John S. 2002. Cry for the Beloved Country. The Post-Apartheid Denouement. In Thabo Mbekis World. The Politics and Ideology of the South African President, edited by Sean Jacobs and Richard Calland, 27-51. Pietermaritzburg: University of Natal Press. Sayer, Andrew. 2007. Dignity at Work: Broadening the Agenda. Organization 14 (4): 565-81. Schneider, Saundra K. 1982. The Sequential Development of Social Programs in Eighteen Welfare States. Comparative Social Research 5: 195-219. Schram, Sanford. 2000. After Welfare. The Culture of Postindustrial Social Policy. New York: NYU Press. Scott, James C. 1990. Domination and the Arts of Resistance. Hidden Transcripts. New Haven, CT: Yale University Press. Seekings, Jeremy. 1997. Civic Organisations during South Africas Transition to Democracy, 1990-1996. In The Post-Colonial Condition. Contemporary Politics in Africa, edited by D. Pal Ahluwalia and Paul Nursey-Bray, 138-57. Commack, NY: Nova Science Publishers.

479

Seekings, Jeremy. 2005. Visions, Hopes and Views about the Future. The Radical Moment of South African Welfare Reform. In South Africas 1940s. Worlds of Possibilities, edited by Saul Dubow and Alan Jeeves, 44-63. Cape Town: Double Storey Books. Seekings, Jeremy. 2006. Beyond Heroes and Villains: The Rediscovery of the Ordinary in the Study of Childhood and Adolescence in South Africa. Social Dynamics 32 (1): 1-20. Seekings, Jeremy. 2007. Not a Single White Person Should be Allowed to Go Under: Swartgevaar and the Origins of South Africas Welfare State, 1924-29. Journal of African History 48 (3): 375-94. Seekings, Jeremy. 2008. The Carnegie Commission and the Backlash against Welfare StateBuilding in South Africa, 1931-37. Journal of Southern African Studies 34 (3): 515-37. Seekings, Jeremy and Nicoli Nattrass. 2005. Class, Race, and Inequality in South Africa. New Haven, CT: Yale University Press. Seidman, Gay. 1994. Manufacturing Militance. Workers Movements in Brazil and South Africa, 1970-1985. Berkeley: University of California Press. Sen, Amartya. 1985. Commodities and Capabilities. New York: Oxford University Press. Sennett, Richard. 2000. The Corrosion of Character. The Personal Consequences of Work in the New Capitalism. New York: Norton. Sevenhuijsen, Selma, Vivienne Bozalek, Amanda Gouws, and Marie Minnaar-McDonald. 2003. South African Social Welfare Policy: An Analysis Using the Ethics of Care. Critical Social Policy 23 (3): 299-321. Sewpaul, Vishanthie. 2005. A Structural Social Justice Approach to Family Policy: A Critique of the Draft South African Family Policy. Social Work/Maatskaplike Werk 41 (4): 31022. 480

Shanafelt, Robert. 1988. The Heroic Ideal in Three Sotho Novellas of Labour Migration. African Studies 47 (1): 47-61. Simone, Abdoumaliq. 2004a. For the City Yet to Come. Changing African Life in Four Cities. Durham, NC: Duke University Press. Simone, AbdouMaliq. 2004b. People as Infrastructure: Intersecting Fragments in

Johannesburg. Public Culture 16 (3): 407-29. Sitas, Ari. 1983. African Worker Responses on the East Rand to Changes in the Metal Industry, 1960-1980. Ph.D. diss., University of the Witwatersrand, Johannesburg. Sitas, Ari. 1985. From Grassroots Control to Democracy: A Case Study of the Impact of Trade Unionism on Migrant Workers Cultural Formations on the East Rand. Social Dynamics 11 (1): 32-43. Sitas, Ari. 1996. The New Tribalism: Hostels and Violence. Journal of Southern African Studies 22 (2): 235-48. Sitas, Ari. 1998. The New Poor. Organisational Challenges. South African Labour Bulletin 22 (5): 16-22. Sitas, Ari. 2004. Thirty Years Since the Durban Strikes. Black Working-Class Leadership and the South African Transition. Current Sociology 52 (5): 830-49. Sitas, Ari. 2007. Disabled Systems and Disabled Social Actions in South Africa. African Identities 5 (1): 33-38. Smith, Laila and Nolene Morris. 2008. Municipal Entities: A Panacea for Service Delivery Constraints? In Consolidating Developmental Local Government. Lessons from the South African Experience, edited by Mirjam van Donk, Mark Swilling, Edgar Pieterse, and Susan Parnell, 423-44. Cape Town: UCT Press. 481

Social Security Administration. 2007. Social Security Programs Throughout the World: Asia and the Pacific. Washington, DC: Social Security Administration. Social Security Administration. 2008. Social Security Programs Throughout the World: The Americas. Washington, DC: Social Security Administration. Sociology of Work Programme. 1994. Restructuring at Nampak: A Strategy for Worker Involvement. Unpublished report, Sociology of Work Programme, University of the Witwatersrand, Johannesburg. Somers, Margaret and Fred Block. 2005. From Poverty to Perversity: Ideas, Markets, and Institutions over 200 Years of Welfare Debate. American Sociological Review 70 (2): 260-87. South African Cities Network. 2006. State of the Cities Report, 2006. Johannesburg: South African Cities Network. South African Human Rights Commission. 2004. The Right to Social Security. 5th Economic and Social Rights Series. 2002/2003 Financial Year. Johannesburg: South African Human Rights Commission. South African Institute of Race Relations. 2007. South Africa Survey 2006/07. Johannesburg: South African Institute of Race Relations. Southall, Roger. 1985. Monopoly Capital and Industrial Unionism in the South African Motor Industry, Labour, Capital and Society 18 (2): 304-42. Spivak, Gayatri C. 1992. Acting Bits/Identity Talk. Critical Inquiry 18 (4): 770-803. Spivak, Gayatri C. 2000. Translation as Culture. Parallax 6 (1): 13-24. Spivak, Gayatri C. 2005. Scattered Speculations on the Subaltern and the Popular. Postcolonial Studies 8 (4): 475-86. 482

Standing, Guy. 2002. Beyond the New Paternalism. Basic Security as Equality, London: Verso. Standing, Guy. 2003. The South African Solidarity Grant. In A Basic Income Grant for South Africa, edited by Guy Standing and Michael Samson, 7-18. Cape Town, University of Cape Town Press. Statistics South Africa. 2000. Survey of Employment and Earnings. June. Pretoria: Statistics South Africa. Statistics South Africa. 2008a. Quarterly Labour Force Survey. Quarter 1 and Quarter 2 2008. Pretoria: Statistics South Africa. Statistics South Africa. 2008b. Quarterly Labour Force Survey. Quarter 3 2008. Pretoria: Statistics South Africa. Stavrakakis, Yannis. 1999. Lacan and the Political. London: Routledge. Steinberg, Carol. 2006. Can Reasonableness Protect the Poor? A Review of South Africas Socio-economic Rights Jurisprudence. South African Law Journal 123 (2): 264-84. Steinmetz, George. 1993. Regulating the Social: The Welfare State and Local Politics in Imperial Germany. Princeton: Princeton University Press. Stivale, Charles J. 1998. The Two-Fold Thought of Deleuze and Guattari: Intersections and Animations. New York: Guildford. Strangleman, Tim. 2007. The Nostalgia for Permanence at Work? The End of Work and Its Commentators. Sociological Review 55 (1): 81-103. Sunde, Jackie and Vivienne Bozalek. 1995. (Re)presenting The Family. Familist Discourses, Welfare and the State. Transformation 26: 63-77. Surender, Rebecca, Phakama Ntshongwana, Michael Noble, and Gemma Wright. 2007. Employment and Social Security: A Qualitative Study of Attitudes towards the Labour 483

Market and Social Grants. Report for the Department of Social Development, Employment and Social Security Project. Oxford: Centre for the Analysis of South African Social Policy, Oxford University. Suttner, Raymond. 2005. Legacies and Meanings of the United Democratic Front (UDF) Period for Contemporary South Africa. In From National Liberation to Democratic Renaissance in Southern Africa, edited by Cheryl Hendricks and Lwazi Lushaba, 59-81. Dakar: CODESRIA. Swart, Sandra. 2000. Desperate Men: The 1914 Rebellion and the Politics of Poverty. South African Historical Journal 42: 161-75. Swilling, Mark. 1984. Workers Divided: A Critical Assessment of the Split in MAWU on the East Rand. South African Labour Bulletin 10 (1): 99-123. Taylor, Scott D. 2007. Business and the State in Southern Africa. The Politics of Economic Reform. Boulder, CO: Lynne Rienner. Taylor, Viviene. 2007. A Choice of Change: Social Safety Net or Comprehensive Social Protection. New Agenda 27: 12-20. Terreblanche, Sampie. 2002. A History of Inequality in South Africa, 1652-2000. Pietermaritzburg: University of Natal Press. Theron, Jan. 2004. Employment Is Not what it Used To Be: The Nature and Impact of the Restructuring of Work in South Africa. In Beyond the Apartheid Workplace. Studies in Transition, edited by Edward Webster and Karl von Holdt, 302-23. Durban: University of KwaZulu-Natal Press.

484

Theron, Jan. 2007. No Room for Complacency: Strategies toward Informalization in South Africa. Paper presented at the conference Informalising Economies and New Organising Strategies in Africa. Uppsala, Nordiska Afrikainstitutet, April 20-22. Theron, Jan, Shane Godfrey, and Peter Lewis. 2005. The Rise of Labour Broking and Its Policy Implications. Development and Labour Monograph Series 1/2005. Cape Town: University of Cape Town, Institute of Development and Labour Law. Thobejane, Makgane. 2000. iGoli 2002, Transformation and Implementation. Chronology of Events, Labour Relations Section. Unpublished report. December 20. Thoburn, Nicholas. 2003. Deleuze, Marx and Politics. London: Routledge. Thompson, Glen. 1997. Ministering to the Oppressed? Change and the Spiritual Economy of the Faith Movement in Durban During the Late Twentieth Century. Journal of Natal and Zulu History 17: 60-98. Tomaselli, Keyan G. and Arnold Shepperson. 2001. Re-Semiotizing the South African Democratic Project: The African Renaissance. Social Semiotics 11 (1): 91-106. Tomlinson, Richard, Robert Beauregard, Lindsay Bremner, and Xolela Mangcu. 2003. The Postapartheid Struggle for an Integrated Johannesburg. In Emerging Johannesburg. Perspectives on the Postapartheid City, edited by Richard Tomlinson, Robert Beauregard, Lindsay Bremner, and Xolela Mangcu, 3-20. London: Routledge. Trres, Liv. 2000. Amandla. Ngawethu? The Trade Union Movement in South Africa and Political Change. Oslo: FAFO. Trapido, Stanley. 1971. South Africa in a Comparative Study of Industrialization. Journal of Development Studies 7 (3): 309-20.

485

Tsing, Anna L. 2005. Friction. An Ethnography of Global Connection. Princeton: Princeton University Press. Turner, Richard. 1978. The Eye of the Needle. Towards Participatory Democracy in South Africa. Maryknoll, NY: Orbis. Twala, Chitja. 2004. The Letsema/Ilima Campaign: A Smokescreen or an Essential Strategy to Deal with the Unemployment Crisis in South Africa? Journal for Contemporary History 29 (1) 184-98. United Nations Development Programme. 2003. South Africa Human Development Report 2003. The Challenge of Sustainable Development in South Africa: Unlocking Peoples Creativity. Oxford: Oxford University Press. Unterhalter, Elaine. 2000. The Work of the Nation: Heroic Masculinity in South African Autobiographical Writing of the Anti-Apartheid Struggle. European Journal of Development Research 12 (2): 157-78. Urban-Econ. 1999. Eastern Gauteng Services Council. Volume 1: Perspective. Germiston: Urban-Econ Development Economists. Vahed, Goolam. 1998. Control of African Leisure Time in Durban in the 1930s. Journal of Natal and Zulu History 18: 67-123. Van Binsbergen, Wim. 2005. Commodification: Things, Agency, and Identities: Introduction. In Commodification. Things, Agency, and Identities (The Social Life of Things Revisited), edited by Wim van Binsbergen and Peter Geschiere, 9-51. Mnster: LIT Verlag. Vandervoort, Bruce. 1999. Wars of Imperial Conquest in Africa, 1830-1914. Bloomington: Indiana University Press. Economic Development

486

Van der Berg, Servaas. 1996. Fiscal Implications of the White Paper on Social Welfare, Social Work/Maatskaplike Werk 32 (1): 8-19. Van der Berg, Servaas. 1997. South African Social Security under Apartheid and Beyond. Development Southern Africa 14 (4): 481-503. Van der Horst, Sheila T. 1942. Native Labour in South Africa. Oxford: Oxford University Press. Van der Westhuizen, Christi. 2007. White Power and the Rise and Fall of the National Party. Cape Town: Zebra Press. Van Dijk, Teun A. 2003. Critical Discourse Analysis. In The Handbook of Discourse Analysis, edited by Deborah Schiffrin, Deborah Tannen and Heidi E. Hamilton, 352-71. Oxford: Blackwell. Van Ginneken, Wouter. 2003. Extending Social Security: Policies for Developing Countries. International Labour Review 142 (3): 277-94. Van Niekerk, Robert. 2003. The Evolution of Health and Welfare Policies in South Africa: Inherited Institutions, Fiscal Restraint, and the Deracialization of Social Policy in the Post-Apartheid Era. Journal of African American History 88 (4): 361-76. Van Onselen, Charles. 2001 (1982). New Babylon, New Nineveh. Everyday Life on the Witwatersrand, 1886-1914. Johannesburg: Jonathan Ball. Van Parijs, Philippe. 1995. Real Freedom for All. What (If Anything) Can Justify Capitalism? Oxford: Clarendon Press. Vavi, Zwelinzima. 2005. Twenty Years of Historic Struggle for a Better Life for All. Umrabulo 24. http://www.anc.org.za/ancdocs/pubs/umrabulo/umrabulo24. Retrieved on November 2, 2008.

487

Verhoef, Grietjie. 2006. From Friendly Society to Compulsory Medical Aid Association. The History of Medical Aid Provision in South Africas Public Sector, 1905-1970. Social Science History 30 (4): 601-27. Virno, Paolo. 2004. A Grammar of the Multitude. For an Analysis of Contemporary Forms of Life. New York: Autonomedia. Von Holdt, Karl. 1987. Trade Unions, Community Organisations and Politics. A Local Case Study on the East Rand. Labour Studies Research Report 3, Sociology of Work Programme, University of the Witwatersrand, Johannesburg. Von Holdt, Karl. 1994. COSATU Congress. Challenging Elite Anxieties, South African Labour Bulletin 18 (5): 40-46. Von Holdt, Karl. 2000. From the Politics of Resistance to the Politics of Reconstruction? The Unions and Ungovernability in the Workplace. In Trade Unions and Democratization in South Africa, 1985-1997, edited by Glenn Adler and Eddie Webster, 100-28. Basingstoke: MacMillan. Von Holdt, Karl. 2002. Social Movement Unionism: The Case of South Africa. Work, Employment and Society 16 (2): 283-304. Von Holdt, Karl. 2003. Transition from Below. Forging Trade Unionism and Workplace Change in South Africa. Pietermaritzburg: University of Natal Press. Von Holdt, Karl and Edward Webster. 2005. Work Restructuring and the Crisis of Social Reproduction. A Southern Perspective. In Beyond the Apartheid Workplace. Studies in Transition, edited by Edward Webster and Karl von Holdt, 3-40. Durban: University of KwaZulu-Natal Press.

488

Von Lieres, Bettina. 2005. Marginalisation and Citizenship in Post-Apartheid South Africa. In Limits to Liberation after Apartheid. Citizenship, Governance, and Culture, edited by Steven L. Robins, 22-32. Oxford: James Currey. Von Schnitzler, Antina. 2008. Citizenship Prepaid. Water, Calculability, and Techno-politics in South Africa. Journal of Southern African Studies 34 (4): 899-917. Vorster, Jan, Hester Rossouw, and Gerald Muller. 2000. Phasing Out the State Maintenance Grant within the Context of Developmental Social Welfare. Report for the Department of Welfare and Population Development. Datadesk and Department of Sociology, University of Stellenbosch. Walters, William. 2004. Some Critical Notes on Governance. Studies in Political Economy 73: 27-46. Watson, Cate. 2006. Unreliable Narrators? Inconsistency (and Some Inconstancy) in Interviews. Qualitative Research 6 (3): 367-84. Watts, Michael. 2005. Baudelaire over Berea, Simmel over Sandton? Public Culture 17 (1): 181-92. Weber, Max. 1994 (1895). The Nation State and Economic Policy (Inaugural Lecture). In Political Writings, edited by Peter Lassman and Ronald Speirs, 1-28 (Cambridge: Cambridge University Press). Webster, David. 1984. The Reproduction of Labour Power and the Struggle for Survival in Soweto. Second Carnegie Inquiry into Poverty and Development in Southern Africa, Conference Paper 20. Cape Town: University of Cape Town, SALDRU. Webster, Edward. 1979. A Profile of Unregistered Union Members in Durban. South African Labour Bulletin 4, 8: 43-74. 489

Webster, Edward. 1985. Cast in a Racial Mould. Labour Process and Trade Unionism in the Foundries. Johannesburg: Ravan Press. Webster, Edward. 1988. The Rise of Social Movement Unionism: The Two Faces of the Black Trade Union Movement in South Africa. In State, Resistance and Change in South Africa, edited by Philip Frenkel, Noam Pines, and Mark Swilling, 174-96. London: Croom Helm. Webster, Edward. 1997. Democratic Transition: South African Sociology. Contemporary Sociology 26 (3): 279-282. Webster, Edward. 2005. Making a Living, Earning a Living: Work and Employment in Southern Africa. International Political Science Review 26 (1): 55-71. Webster, Edward. 2006. Trade Unions and the Challenge of the Informalisation of Work. In Trade Unions and Democracy. COSATU Workers Political Attitudes in South Africa, edited by Sakhela Buhlungu, 21-43. Pretoria: HSRC Press. Webster, Eddie and Judson Kuzwayo. 1975. Consciousness and the Problem of Collective Action. A Preliminary Case-Study of a Random Sample of African Workers in Durban. Unpublished paper. Durban: Institute of Industrial Education. Webster, Edward, Rob Lambert, and Andries Bezuidenhout. 2008. Grounding Globalization. Labour in the Age of Insecurity. Oxford: Blackwell. Weiss, Brad. 2004. Contentious Futures: Past and Present. In Producing African Futures. Ritual and Reproduction in a Neoliberal Age, edited by Brad Weiss, 1-20. Leyden: Brill. White, Luise. 2000. Speaking with Vampires. Rumor and History in Colonial Africa. Berkeley: University of California Press.

490

Weyland, Kurt. 2006. Bounded Rationality and Policy Diffusion. Social Sector Reform in Latin America. Princeton: Princeton University Press. Wilder, Gary. 2005. The French Imperial Nation-State. Negritude and Colonial Humanism between the Two World Wars. Chicago: University of Chicago Press. Wilson, Francis and Mamphela Ramphele. 1989. Uprooting Poverty. The South African Challenge. Cape Town: David Philip. Wolpe, Harold. 1972. Capitalism and Cheap Labour-Power in South Africa: From Segregation to Apartheid. Economy and Society 1 (4): 425-56. Wood, Geoffrey. 2004. Informal Security Regimes. The Strength of Relationships In Insecurity and Welfare Regimes in Asia, Africa and Latin America. Social Policy in Development Contexts, edited by Ian Gough and Geoffrey Wood, 49-87. Cambridge: Cambridge University Press. World Bank. 2001. World Development Report 2000/2001. Attacking Poverty. Washington, DC; World Bank. Wylie, Diana. 2001. Starving on a Full Stomach: Hunger and the Triumph of Cultural Racism in Modern South Africa. Charlottesville, VA: University Press of Virginia. Wynter, Sylvia. 2003. Unsettling the Coloniality of Being/Power/Truth/Freedom. Towards the Human, After Man, Its Overrepresentation An Argument. New Centennial Review 3 (3): 257-337. Young, Robert J.C. 2004 (1990). White Mythologies. Writing History and the West. London: Routledge.

491

Zack, Tanya and Sarah Charlton. 2003. Better Off, But Beneficiaries Perceptions of the Governments Housing Subsidy Scheme. Occasional Paper 12. Johannesburg: Housing Finance Resource Programme. Zain, Pia. 2000. A Small Process of Dying: The Impact of the Cancellation of the State Maintenance Grant. Research Report 13. Johannesburg: National Association of Democratic Lawyers. Zizek, Slavoj. 1993. Tarrying with the Negative. Kant, Hegel, and the Critique of Ideology. Durham, NC: Duke University Press. Zizek, Slavoj. 2000. The Ticklish Subject. The Absent Centre of Political Ontology. London: Verso.

492