Back to homepage...


Chapter 9

The Labour Movement, Radical Reform and
the Transition to Democracy in South Africa

Glenn Adler and Eddie Webster 

Introduction

The truly remarkable nature of South Africa's transition was captured in the election and the inauguration of Nelson Mandela as the new president during the last week of April and the first week of May 1994. Contrary to widespread expectations of violence descending into chaos, the three voting days were among the most peaceful in the country's recent history. Whites and Blacks queued for hours - indeed days in some cases - to cast their votes. For South Africans, it became a deeply emotional experience, with the overwhelming majority of eligible voters participating in this first-ever democratic election. The atmosphere of euphoria carried through into the following week of national celebrations, culminating in an all-day gathering outside the former seat of White political domination, the Union Buildings in Pretoria. Music and dancing provided by South Africa's best entertainers continued well into the night, and similar scenes of jubilation were repeated throughout the country.

However, the ambiguous character of this astonishing climax to South Africa's transition to democracy was clearest when Mandela stepped from his automobile to be ushered to the inaugural platform by none other than the White general officers of the security forces, the same men who had prosecuted the apartheid regime's brutal war against its opponents, especially the African National Congress (ANC), within the country's borders and throughout southern Africa. The most enthusiastic response of the day came immediately after Mandela's inaugural speech, when the Air Force staged a dramatic fly-over, concluding with a squadron of six jet fighters releasing smoke trails in the colors of the new national flag. As the commander-in-chief, Mandela took his place at the head of a Government of National Unity, which, although dominated by the ANC, included the most important leaders of the old regime.

From all appearances, South Africa seems to be a textbook case of democratization, as understood by leading contributors to transition theory. These theorists argue that the transition from authoritarianism to democracy can be brought about only as a result of negotiations and of pacts between the reformers in the ruling regime and the moderates in the opposition. The political implication of this alliance between the reformers and the moderates is that prodemocratic forces must be prudent; they must be prepared to offer concessions in exchange for democracy. The corollary is that the democracy that results from this process is inevitably conservative, economically and socially.

While this literature is suggestive and in many senses quite appropriate in any credible effort to grasp the dynamics of South Africa's transition, its emphasis on elites leads it to misunderstand the role of popular movements and struggle in the origin, development, and outcome of actual transitions. In particular, transition writers neglect the role of labour movements as important actors in transitions (but see Keck, 1992; Payne, 1991 and Valenzuela, 1989). Popular and radical movements are given scant attention and are understood as `maximalists' who, if allowed free reign, risk prompting a right-wing reaction, thereby scuttling the entire transition process. Attention falls instead on an alliance between `reformers' in the authoritarian regime and `moderates' among the pro-democratic forces in civil society, both of whom corral the `extremists' in their respective camps on the way to brokering a centrist pact.

Our argument is that this model distorts the role of the left; between unwise maximalism and co-optation lies a range of strategic possibilities for influencing the course of the transition. By undervaluing the capacity for innovation and the strategic use of power by social movements, the literature overlooks actors with extremely important influence over the transition. The crucial consequence of this approach is an inappropriate narrowing of possible outcomes. Through a strategy we label radical reform, disciplined and sophisticated social movements may be able to inject more progressive content into the democratization process and wrest important concessions from reformers and moderates alike. In other words, a conservative outcome is in no way given in advance.

The South African case thus has important implications for debates about transitions to democracy. Rather than being a force to be restrained by the alliance between reformers and moderates, a mobilized civil society and powerful social movements - especially the labour movement - played a central and constructive role in creating the conditions for the transition, in shaping its character, and indeed in legitimizing the transition process itself. Whether the labour movement in particular can continue to play this role is an open question. With the advent of parliamentary democracy and the creation of corporatist-type policy making fora and institutions, as well as the emphasis on national reconciliation, pressures toward incorporation and demobilization have never been stronger. However, countervailing trends point toward the persistence of these movements and their democratic culture and toward a postapartheid politics of continued contestation within the new democratic institutions. These contradictory possibilities are clearest in the emerging politics over the ANC's new policy framework, the Reconstruction and Development Programme (RDP). The RDP embodies the longstanding central demands of the prodemocracy movement for a more radical social and economic program of transformation.

The South African case is not simply one of parochial interest only. It maps out possibilities for more progressive potential in transitions than is admitted in the literature. Between the sterile polarity of revolutionary rupture on the one hand and co-optation into conservative pacts on the other lies the potential for a society that begins to meet people's basic needs while developing practices of participatory democracy. The content of these policies is subject to challenge, but already a program has emerged whose vision goes beyond the narrow confines of transition theory. The democratization process has created certain conditions that favor this possibility: important labour activists and intellectuals have been inserted into key state structures. In combination with this involvement in state structures, a strong civil society and labour movement capable of mobilizing and restraining its members opens up possibilities of deepening and extending democracy during this phase of the transition process.
 
 

1. The Emergence of the Democratic Trade Union Movement

Trade unions were first founded in South Africa during the 1880s by immigrant mineworkers, including Cornish tin miners who flocked to the Witwatersrand gold fields (For accounts of trade unionism in South Africa see Seidman, 1994; Baskin, 1991; Hirson, 1989; Freidman, 1987; Webster, 1985; Luckhardt and Wall, 1980; Simons and Simons, 1969). From craft origins, trade unions developed into a relatively large and sporadically militant movement. Powerful industrial unions emerged with the development of secondary manufacturing during the interwar period, with large memberships and political and economic muscle. However these organizations were deeply fractured on racial lines in their membership, leadership, and indeed in their very organizational structure. Until 1979, South Africa had a dualistic structure of industrial relations: a legalistic formal guarantee of certain industrial rights to White, Colored, and Indian workers through the 1924 Industrial Conciliation Act (from 1981, the Labour Relations Act, or LRA), and a repressive labour regime for African workers resting since 1953 on separate legislation known as the Black Labour Relations Act. African workers were prohibited from joining "registered" (or officially recognized) trade unions and engaging in formal collective bargaining in Industrial Councils, bipartite bodies established under the LRA.

White workers dominated the trade union movement. Their historical privileges, entrenched through their craft unions established at the turn of the century, made them something of a "labour aristocracy" within the workforce. The organization of Black workers occurred through bodies subordinated to White unions or through highly vulnerable independent organizations under left-wing leadership on the periphery of the official labour movement.

South Africa's first efforts to diversify from dependence on mining and agricultural exports came during the l92Os as a result of state-led industrialization policies. Despite industrial growth during World War II, South African industry was still relatively undeveloped through the 1950s. Production was largely labour intensive, was located in small enterprises, and remained deeply dependent on imported raw materials and machinery financed from profits on mineral exports (Innes, 1984, p. 169).

Important structural changes took place in the economy during the 1960s, which prepared the ground for the emergence of new Black trade unions. The South African economy expanded rapidly, bringing important changes in the relations of production in industry. The growing concentration and centralization of capital led to the growth in the absolute size of workplaces and a tremendous advance in their technological sophistication. Most manufacturing industries experienced increasing demand for semiskilled and skilled labour, which could not be met either by local White or immigrant workers. The changes allowed for large numbers of Black workers to fill the void.

The transformation of the division of labour caused by the rise of mass production

assembly line industries is familiar to students of import substitution around the world. But the racial division of labour in South Africa created peculiar conditions, which provided additional sources of grievance for Black workers and possible bases of collective action. Racially discriminatory practices affected every aspect of the employment relationship. The intersection of brutal industrial conditions and direct racial oppression in the workplace served as a source of deep discontent among workers and increased their sense that the political and economic systems were fundamentally unjust.

These conditions created an important base on which workers could build industrial unions. The unofficial erosion of race-based job protection was accompanied by a growing awareness and confidence by Black workers in their economic power, encouraging efforts to revitalize and develop Black trade unions. Gradually, a few older, independent Black trade unions as well as newly formed ones began testing their strength, suffering numerous defeats but also gaining important victories.

The direct controls over African workers had functioned efficiently during the period of strong repression in the 1960s after the major Black political and labour organizations were banned or dispersed. But they became increasingly untenable with the economic changes of the 1960s and were entirely overwhelmed by the huge increase in Black worker militancy, which first surfaced in Durban in 1973. The slow emergence of nonracial industrial unions during the 1970s challenged the comfortable "social contract" between the apartheid state, employers, and White labour based on protectionism and cheap Black labour. By the 1970s, the official (White) labour movement was a spent force, with falling membership and a rapidly declining ability to protect members' material interests. At the same time, Black workers began building the modern labour movement, organizing on a formally nonracial class basis, and soon displaced the racist unions as the institutional center of South African labour.

The budding movements focused their organizational efforts in the workplace to develop decentralized structures that could survive when state repression threw the head office into disarray. The strategy depended on "the development of a cadre of shop stewards integrally linked into the constitution and decision-making structure of the unions." (Cheadle, 1987, p. 7) Shop stewards were directly elected by shop floor workers, usually through a secret ballot, and were directly accountable to their constituents. They operated on the basis of strict mandates from the membership and were subject to recall. Furthermore, elected worker representatives dominated the regional and central executives of these unions, including the president and national officers, who were constitutionally required to be full-time shop floor workers. Thus the imperatives of organizing under authoritarian conditions coincided with the structural changes in the economy which had created a vast new semiskilled workforce with potential leverage over production, leading to the creation of powerful shop floor union organization.

The new unions, especially those in Durban, emphasized building strong structures in individual factories rather than diffusing organizing efforts in a general union of workers drawn from an array of industries in a particular area. The emphasis on organization rather than numbers was reflected in the policy that it was "better to have 100 members in one factory than a thousand in a hundred factories"; the collective strength in one factory "put pressure on their employer in a way that the dispersed membership in 100 factories could not." (Cheadle, 1987, p. 8).

The unions focused on important industries with large workforces. They found particular success at multinational corporations where international pressure could be brought to bear, especially in the metal, textile, and chemical industries. These were soon linked together in the first trade union center since the repression began during the early 196Os. Through mergers and further organizing, in 1979 this center grew into the first national trade union body, the Federation of South African Trade Unions (FOSATU), with 20,000 workers organized in ten industrial unions.

Growing labour militancy and deepening township protests culminating in the Soweto Uprising of 1976 propelled the state into an equivocal and hesitant reform project. Labour law reforms, introduced by the Wiehahn Commission of Enquiry between 1977 and 1979, opened civil society, allowing for the formal recognition of African trade unions within the official industrial relations system for the first time in South African history.

Once African workers were granted statutory trade union rights under the LRA, FOSATU's membership rocketed to 95,000 by 1981. The federation became one of the central advocates of building broad unity among the emerging unions and in 1985 helped engineer the creation of a "super-federation," the Congress of South African Trade Unions (COSATU), the largest trade union center in South African history. In 1990, COSATU claimed more than 1.2 million members organized in fourteen industrial unions (Baskin, 1991, p. 448).

The fledgling unions, in particular those affiliated to FOSATU, had made important strategic innovations, which profoundly affected trade union development as well as the course of political struggle in South Africa. They successfully combined a radical vision with a strategy of reform; we call this strategic use of power radical reform (see Adler, Maller and Webster, 1992; Webster, 1993; Saul, 1991; Saul, 1994). In pursuit of the long-term goals of ending apartheid and creating a socialist economy, the unions emphasized legal means of struggle (see Turner, 1972). They sought inclusion of all workers within the industrial relations system and decided to register their unions under the LRA (On the registration debate see Maree, 1987). Finally, they eschewed involvement in national political issues and refused to align themselves publicly to any political movement.

Reforms were not regarded as ends in themselves but rather as dynamic phases in a progressive struggle to achieve the longer-term goals. With the strong backing of their members, shop stewards had the power to push for concessions from management, which not only created space for further advances but also won concrete improvements in workers' conditions, thereby reassuring them of the efficacy of direct action.

There were two important components to the union's approach to the strategic use of power: (1) democratic processes to win voluntary consent from members for mobilization and for restraint when necessary and (2) tactical flexibility, which included a capacity to distinguish principles from tactics and to choose those tactics most likely to succeed, including negotiation and compromise.

The negotiation of recognition agreements, which set out the rights and duties of shop stewards and trade unions in the workplace, was an important step in establishing the "rule of law" on the shop floor. Trade unions did not win these rights without a fight. During the late 1970s and early 1980s, there were hundreds of strikes in support of the demands for recognition.

Throughout the early 1980s, workers and trade unions used their increasing strength and rights on the shop floor to fight for better wages and working conditions as leave, hours of work, safety, and pensions became matters for negotiation. But workers not only were becoming a formidable force on the shop floor, they also were beginning to engage in centralized industrial bargaining in the Industrial Councils established under the LRA. These joint employer-union bodies provided a forum where unions could engage in industry-wide negotiations, which could potentially be extended to include industrial restructuring issues.

In its emphasis on political independence, tactical flexibility, and compromise with employers and the state, the strategy stood in marked contrast to the political and military struggle then being waged by the ANC and its internal allies. This political strategy aimed at the state's overthrow through a national democratic revolution: a sudden shift in the balance of power in which the old ruling class was destroyed altogether. This strategy of revolutionary rupture stressed abstention from involvement in the apartheid state and all its institutions on the assumption that the leadership of any subject group would be co-opted and the status quo would remain. By contrast, the labour movement developed a strategy of engaging rather than boycotting the state from an independent and disciplined power base resting on strong factory structures, held together through practices of democratic accountability. The differences between these two strategic visions caused innumerable conflicts between the emerging unions on the one hand and the exiled ANC and the developing internal political movement on the other. However, the strategy of radical reform led to neither revolutionary rupture nor co-optation but to the legitimization of the union as an institution and the rapid growth of unions during the l980s.

The growth of union membership was paralleled in the townships and the schools by the development of powerful social movements such as civic organizations and student and youth congresses. From 1979, these movements launched community-linked unions, rejecting the strategy of radical reform in an attempt to bring workers into the national democratic struggle. By the mid-1980s, these new community-based formations in civil society began to challenge the state directly, both at the local and the central levels, creating conditions of near insurrection. This deepening crisis sharpened the differences in strategy between those committed to radical reform and the growing political movement, since 1983 in a national umbrella body named the United Democratic Front. In particular, the unions were forced to reevaluate their policy of political nonalignment or face isolation (see Baskin, 1991). In the end, the industrial unions committed to radical reform forged an alliance with these social movements, actively participating in the series of general strikes and political actions from late 1984. In l985, these two main currents in the labour movement merged to form COSATU, signaling a strategic compromise in which the integrity of the industrial unions was acknowledged while the new federation committed itself to participation in the national democratic struggle under the leadership of the ANC (see Fine and Webster, 1989; Lambert and Webster, 1988 and Eidelberg, 1993).

The new alliance forged a more powerful, unified opposition to apartheid, fueling insurrectionist tendencies while strengthening Hardliners in the state who argued for a strategy of repression. The increasing tempo of mass action during the first half of 1986, symbolized by growing strike activity and township "ungovernability," led to a declaration of a national state of emergency in June.

As part of the repression, in 1988 the government - with the support of employers -passed the Labour Relations Amendment Act, which sought to curb union power by rolling back the rights established since 1979. The threat posed by the new LRA galvanized unions. Extending the logic of radical reform, the unions negotiated with employers while pressuring them through continued engagement in mass action. Over the next two years, the labour movement held worker summits and launched a campaign of protests, stayaways, and overtime bans against the act.

The range of union campaigns, protests, wildcat industrial action, and disruption on the shop floor demonstrated the limited effectiveness of repression in ending mobilization.

Furthermore, the mass action took its toll on employers, many of whom abandoned their support for the repressive labour legislation. Nor was labour the only social force to mobilize during this period. Regrouped under the name of the Mass Democratic Movement, the constituent formations of the now-banned United Democratic Front renewed mass action and mounted a highly effective campaign of civil disobedience. The state was faced with a choice: intensify repression toward the labour movement and other mobilized formations in civil society and risk alienating capital with no guarantee of eliminating mobilization from below, or turn toward more fundamental political reform. During late 1989, the reformist F. W. de Klerk replaced P. W. Botha as state president and signaled his intention to reform apartheid. In October, de Klerk released the veteran leaders of the ANC who had been jailed since the early l960s. In February 1990, he announced the unbanning of the ANC, the Pan Africanist Congress, and the South African Communist Party (SACP), and he freed Mandela. During this period of abertura, COSATU formalized the political relations forged in the Mass Democratic Movement by entering into a formalized pact with the ANC and the SACP, known as the Tripartite Alliance.

There was a noticeable shift in the attitudes of major employers as it became clear that South Africa was moving toward a transition to democracy. In May 1990, a historic accord was struck between the main labour federations, the national employers association and the Department of Manpower. The "Laboria Minute" repealed the objectionable amendments of 1988 and signaled the extension of basic trade union rights to all workers. It was agreed that all future labour laws would be considered by the employers' body and the trade union movement prior to being put before Parliament. The trade unions agreed to participate in a reconstructed National Manpower Commission, a statutory consultative body established in the Wiehahn reforms, which they had always boycotted due to its unrepresentative structure and lack of political power.

In February 1991, the LRA was finally amended along the lines of the Labouria Minute. The new act was hailed as the first piece of postapartheid legislation and consolidated a democratic industrial relations system accepted by labour, employers, and the state. The timing of these reforms was crucial for the entire transition process: the labour movement entered the transition phase not as a wounded giant hobbled by an authoritarian regime but as a movement with unprecedented freedom of action and flushed with the success of the anti-LRA campaign. The anti-LRA campaign marked the victory of the labour movement's struggle against those forces attempting to turn back the clock, demonstrating that real gains could be made through the strategy of radical reform.

The anti-LRA campaign was a high point of the period of resistance, yet it ushered in a new era characterized by the politics of democratic transition and reconstruction. The emergence of a powerful labour movement committed to radical reform opened up the possibility of a distinct role for the trade union movement in the process of transition to a new democratic order. However, according to transition theory, the transition process contains distinct constraints on the role of progressive social movements, leading toward conservative social and economic outcomes. Consideration of labour's role in the transition poses certain problems for transition theory, particularly in its conceptualization of movements in civil society and its claims about the range of possible outcomes. At the same time, transition theory prompts an important question as to whether the unions and the broader democratic movement will be able to take forward the strategy of radical reform into the new period of democracy and parliamentarism.
 
 

2. Transition Theory, Social Movements, and Democratization

There is a great irony in the current political moment. While South Africa stands at the threshold of a historically unprecedented democratic order, the constraints on the transition have never been more apparent. After decades of opposition to apartheid, culminating in the insurrectionary period of the 1980s, pro democracy forces are participating in a Government of National Unity in which their main enemies are now their partners and face a global economy that dictates neoliberal market-based policies as the only acceptable solutions. Radical possibilities seem to have evaporated. While activists bravely accept that half a loaf may be better than none, few started out with this goal in mind. The expansive -indeed, romantic - hopes motivating those activists in the past now appear contained within the rather claustrophobic space of a pact between former adversaries.

This irony is not lost on the key theoreticians of transition in Eastern Europe and Latin America (see O'Donnell, Schmitter and Whitehead, eds, 1986; Przeworski, 1991; Malloy and Seligson, 1987; Di Palma, 1990). Indeed, it is the central insight these writers have to offer to the understanding of transitions to democracy in the modern world. In short, for a democratic transition to succeed, democracy itself must be limited. Radical expectations for democracy as a means to remake society, perhaps to institute socialism, are not only unfeasible but also threatening to the transition process as a whole, as they risk provoking an antidemocratic reaction. The consequence is that prodemocratic forces have to offer concessions on their economic and social programs in exchange for democracy. Thus, successful transitions from authoritarianism can be brought about only as a result of negotiations and of pacts between adversarial elites.

For many transition writers, movements - indeed all protest groups - are grasped in their functional role as "collective organizations that are capable of coercing those whose interests they represent." (Przeworski, 1991, p. 12) Unions are especially important in transitions, but due to their power to restrain members' wage demands and thereby contain inflationary pressures which would otherwise jeopardize economic recovery. In Przeworski's argument, unions' disciplining function contributes to the marginalization of "radicals" on the left, solidifying the leadership of "moderates" in the pro-democratic forces. This process parallels a similar movement whereby "reformers" in the old regime marginalise "hardliners" on the right.

In this respect transition writers reproduce classic assumptions about the relationship of structure and agency common to some social movement and much corporatist theory. A fairly stark divide is drawn between social movements and state institutions, based on the underlying Michelsian assumption that participation in formal institutions leads to a demobilization of movements through the centralization and bureaucratization of decision making. In what Michels described as the process of "goal displacement," movement leaders, committed to pacts with their adversaries, become distanced from their members; they cease to represent rank-and-file demands and rather exercise control over them. While this process is viewed by some political theorists and union democrats as a vice, for many transition writers it is not only a virtue, but the main contribution unions can make to democratization (Pateman, 1971).

At first glance, social movement theory would not appear to be the most promising place to search for an alternative. Indeed, in much of contemporary social movement theory, an overdrawn distinction is usually made between new and old social movements, highlighting the distinctiveness of the former against the materially oriented, class-based (rather than identity-based) old social movements. According to new social movement theory, in becoming part of the established institutional order, labour movements in particular have fallen victim to bureaucratization and demobilization. In fact, new social movements are often seen to be directly challenging these now institutionalized old movements (see Touraine, 1982; Inglehart, 1990; and for a criticism of this view, Olofsson, 1988). We argue, however, for effacing the distinction between new and old movements, positing that the labour movement is not the poor historic forerunner of new social movements but is a legitimate object of contemporary social movement study.

Indeed, writers on South African labour have fruitfully applied the concept "social movement unionism" to account for the particular form taken by the labour movement (see Waterman, 1991; Webster, 1988). This form of unionism blurs the demarcation between unions as formal organizations and social movements as loosely structured networks of action or, more drastically, the distinction falls away. In other words, a form of social movement unionism exists when the formal organizational features characteristic of unions are fused with the mobilization capacity and looser structure of social movements. In South Africa, the twin grievances of capitalist exploitation and apartheid compelled the labour movement to seek both economic and political solutions to workers' problems. Furthermore, the union movement has been characterized by patterns of organization suited to a movement fighting an authoritarian enemy: leadership devolved to local levels, a wide repertoire of mass protest tactics and strategies, and alliances with community and political groups.

Social movement unionism has defied easy categorization as reformist or revolutionary. In advocating socialist-inspired visions of alternative economic models, these movements have prioritized democracy and empowerment both as a societal goal and, in true new social movement fashion, as their internal operating principle. They have demonstrated the pragmatic ability to survive - and retain their progressive vision - in hostile authoritarian political systems. As such, these movements are almost never "labour arms" of programmatic left-wing parties. If they are not doctrinaire adherents to classical notions of revolutionary rupture, neither can their goals be comfortably contained within the neoliberal orthodoxy of contemporary development thought nor within the conservative elite pact advocated in Przeworski's approach to democratic transition.

While our arguments about social movement unionism grew out of empirical evidence from the South African case, the concept did not originate in South Africa and has been applied successfully to other semi-industrialized countries such as Brazil and the Philippines. Social movement unionism therefore could make a contribution to social movement theory, helping to clarify the differences between trade unions and social movements while specifying conditions under which they become more similar. Indeed, the concept may have applicability beyond the cases of authoritarian industrializing capitalist countries. Tarrow's work on social movements in western Europe (Tarrow, 1991) makes a strong criticism of Michels-inspired arguments about institutionalization and bureaucratization, an approach conceptually similar to that used by writers on social movement unionism. He has explored cycles of protest and movement renewal as well as the articulation of formal organization with movement-type activity: movements transform institutions through the invention of new forms of representation and hence create new possibilities for movements to operate through institutions.

In South Africa, and possibly in other Third World transitions such as Brazil, the labour movement has never been a passive participant in social change nor has it been blindly maximalist in its struggles with capital and the state. Instead, as we described in section I, it has combined long-term radical goals of social transformation with reformist strategies and tactics to achieve meaningful victories without provoking a `hardliner' reaction. Despite participation in institutions, it has retained its social movement character, which enabled it to contribute to the crisis that provoked the transition as well as to the pace and character of the transition itself. Whether it is able to retain its social movement character and influence the outcome of the transition process in a more radical direction is explored in section III.
 
 

III. Radical Reform or Demobilization?

Not surprisingly, transition theory has found a receptive audience in South Africa as political leaders and organizations have struggled to find an accommodation between "reluctant reconcilers." In fact, the literature has sparked a local social science growth industry (see Lee and Schlemmer, eds, 1991; van zyl Slabbert, 1992; Adam and Moodley, 1993; Friedman, 1993). Arguably, the local interpreters of transition theory have had a crucial role in shaping the transition itself through numerous workshops and direct consultancies with the central participants as well as through influential media interventions. Curiously, these writers, like the international literature they employ, overlook the labour movement as an important actor in precipitating and securing the transition to a democratic order.

As argued earlier, the labour movement emerged at the height of the state of emergency as the de facto leader of the internal democratic opposition. It not only grew as a movement with the formation and consolidation of COSATU, but it extended its radical reform strategy to contest state policy in its successful campaign to rewrite the LRA. With the emergence of state Reformers, led by de Klerk as dominant within the state, the political terrain changed dramatically with the unbanning of political organizations and the establishment of "normal" politics. The opening created the opportunity for direct negotiations between the reformers' in the state and the moderates' in the ANC (to use Przeworski's categories) and its allied organizations. From 1989, the center of gravity within opposition politics shifted as the ANC began to reassert its hegemony as the political leader of the anti-apartheid movement, a process that was completed in the run-up to the April 1994 elections (see Eidelberg, 1993).

However, the Radicals, in the form of the labour movement, were neither demobilized nor tamed by the Moderates. Instead a triangular relationship developed between Reformers in the state on the one hand and Moderates and Radicals in the prodemocracy movement on the other. Crucially, the labour movement has its own independent power base through its ability to mobilize collectively in the key sectors of the economy. Its capacity to mobilize is not dependent on the Moderates, and thus labour can continually put pressure on Moderates and Reformers alike according to the labour movement's own economic and political agenda.

In addition, through years of practicing radical reform, labour has its own institutionally structured and deep relationship with employers as well as with the state by virtue of its involvement in the campaign against the LRA and its participation in the National Manpower Commission. This relationship is longstanding, ongoing, and - most importantly - neither dependent on nor mediated by its relationship with the ANC. These relationships have given labour leaders a deeper knowledge of their bargaining opponents' strategies, tactics, and real interests. Such knowledge has provided labour leaders with an important ability to weigh options, see through bluffs, and better understand the stakes and limits of their adversaries' positions. It has also given them a strategic understanding of the strengths and weaknesses of their own position. This is what the strategic use of power is all about: an accumulation of knowledge from more than two decades of struggle. Combined with the unions' structure of shop floor representation, accountability, and mandates, labour leaders are in a very strong position to appreciate their members' interests as well as the limits of their willingness to pursue these interests. Finally, Moderates and Reformers alike appreciate that the Radicals, especially those linked to the SACP, were hardly committed to a maximalist program but instead have played a crucial role in cementing unity among pro democracy forces via the strategy of radical reform.

Indeed, the resources the labour movement brings to the transition allowed COSATU to make a distinct mark on the process through the application of its radical reform strategy to the transition. At the outset of negotiations, the federation articulated three main concerns over the process of transition. First, the constitution and the process of drawing it up should be as democratic as possible, allowing space for the widest political participation and consultation by all citizens. Second, the constitution should enshrine worker and trade union rights. Finally, the constitution should facilitate worker participation in economic decision making and the sharing of the fruits of economic activity. In pursuit of these objectives, it is possible to identify five major interventions by COSATU.

First, the unions' tradition of combining mass action with negotiation through the strategic use of lightning general strikes transformed the prodemocracy movement's approach to the transition. This was best illustrated through the highly successful two-day general strike over the government's proposed new Value-Added Tax (VAT) when the unions mobilized nearly 4 million workers to back their demands for fundamental revision of the law. The successful strike established the labour movement as a key participant in negotiations over the transition by virtue of its unique capacity to mobilize a large, disciplined mass base in pursuit of its demands. Labour reasserted its strategy of mass action to back up its negotiating position and confirmed the wisdom of extending the strategy to the political sphere. Since the release of Mandela and the start of the process of political negotiations, this strategy had been downplayed and negotiations had taken place exclusively between leaders of the various parties and organizations. The VAT stayaway ensured that negotiating positions were supported with pressure from mass mobilization.

The combination of negotiation and mass action had an even more profound influence over the negotiations in mid-1992, when the central forum for negotiating the transition, the Conference for a Democratic South Africa (CODESA), stalemated. The governing National Party refused to budge in its demands for a veto over the adoption of a new constitution; in frustration with this position, the ANC alliance, led by COSATU, embarked on a campaign of rolling mass action, including a major work stayaway in August as well as marches and workplace stoppages. The mass action precipitated a number of violent reactions by Hardliners in the state, including a "third force"-directed massacre of nearly 50 squatters at Boipatong, south of Johannesburg. The widening mass action and the vicious reactions from the security forces threatened to derail the negotiation process entirely and raised the possibility of a return to confrontationist politics between insurrectionists and Hardliners. Peering into the abyss, both sides drew back: the Reformers dropped their insistence on entrenched vetos, and the tripartite Alliance put forward compromises known as "sunset clauses" safeguarding the positions of existing civil servants. The negotiations resumed in early 1993 under the guise of the Multiparty Negotiating Forum, and an election date - a key tripartite Alliance demand - was set for April 1994.

Second, through the fight over VAT, it became clear to labour leaders that merely delaying the new tax would not achieve their overall goals of participation in economic decision making. Furthermore, after entering the National Manpower Commission, COSATU withdrew from the body, complaining that the government was dragging its heels on its restructuring and on extending labour legislation to workers not covered previously. These experiences motivated the labour movement to seek a new general forum where economic policy could be negotiated and intensified the importance of the second demand raised during the VAT stayaway: for a macroeconomic negotiating forum where economic policy could be discussed in a more coordinated, global manner. Jay Naidoo, COSATU General Secretary, argued that piecemeal interventions such as tinkering with the terms of VAT were no basis for addressing macroeconomic policy issues during the transition. Socio-economic development, Naidoo asserted, could not be postponed pending the arrival of political democracy. Nor, he could have added, would COSATU trust that Reformers and Moderates would magically arrive at policies that addressed workers' concerns. The anti-VAT campaign brought home to the labour movement the necessity for the negotiating process to construct new institutions in which a wide range of formations from civil society would be able to address qualitatively new demands. These institutions would enable economic and political issues to be addressed simultaneously and immediately.

Not only did labour claim the right to participate in the process of restructuring the economy through the proposed National Economic Forum (NEF), but it also demanded membership rights at CODESA. The demand was rejected, but in compensation COSATU leadership was offered participation through a working group consisting of its alliance partners, and four COSATU representatives sat in the SACP delegation to CODESA.

Labour's initiation of and participation in the NEF was based on its growing appreciation that short-term collective bargaining issues were increasingly bound up with the long-term health of the economy. Labour became acutely aware of the need to address the deep structural distortions and decline caused by apartheid. In this vein, COSATU embarked on a number of research initiatives to explore the possibilities of restructuring work and the economy and producing viable industrial strategies. Its policy formulations demonstrated the federation's preeminent position in the debate on new macroeconomic policy (see Joffe and Lewis, 1992).

Third, labour contributed to the interim constitution devised in the Negotiating Forum through the NMC, where labour and the employers agreed to a proposal that workers' right to strike would be entrenched in the interim constitution, and employers would have the right to lockout. By writing both rights into the proposed Bill of Rights, the COSATU representatives believed that the unions would have something to gain (the right to strike, which they did not possess previously) and nothing to lose, as employers already had the right to lockout (see von Holdt, 1993).

Fourth, thirty COSATU leaders were nominated for the ANC's list of candidates for the Constituent Assembly in the new Government of National Unity and a further forty for the regional parliaments. The former union officials elected to Parliament will no doubt become influential members of the new government. Their expertise cannot be ignored at the socioeconomic level, especially in drafting legislation and formulating policies in key portfolios such as Labour, Public Service, Trade and Industry, and Mining and Energy Affairs.

A more significant intervention in electoral politics has been COSATU's mobilization for the ANC Alliance in the April 1994 elections. Extensive federation resources, both infrastructural and personnel, were devoted to voter education and canvassing. COSATU's intervention was crucial in securing the turnout for the ANC (see Buhlungu, 1994).

Finally, the most important intervention by COSATU was its contribution to the formulation of the Reconstruction and Development Programme (RDP), adopted by the Alliance as its central policy program (see ANC, 1994). The RDP, launched in January 1994, attempts to build a broad social consensus through a multipartite accord. A central idea underlying the program is that social movements such as the trade union movement, the civics movement, women's groups/movements, youth and student organizations, and associations of the unemployed and the aged will be part of an organized pact to reconstruct society. Because many of these sectors are the weakest, poorest, and most marginalized, they will experience the most difficulty in developing their capacity. It is for this reason that the RDP believes that "attention must be given to enhancing the capacity of such formations to adapt to partially changed roles." (African National Congress [ANC], 1994, p. 121) Democracy, the document promises, will not be confined to periodic elections but is viewed rather as an active process enabling formations of civil society to contribute to reconstruction and development.

Although the state will play a "leading and enabling role in guiding the economy and the market toward reconstruction and development," the RDP is not a statist blueprint. (ANC, 1994, pp. 79-80). The emphasis in the document is on the importance of participatory and accountable policy making procedures involving all parties: The democratic government, the trade union movement, business associations and relevant organizations of civil society must cooperate in formulating economic policy." (ANC, 1994, p. 81) A similar commitment is given to the importance of involving forums such as the NEF for "broad consultation on policy matters." (ANC, 1994, p. 130) Although the RDP is an Alliance document, it originated in COSATU and represents a significant attempt to move policy beyond socially and economically conservative goals by taking as its point of departure people's basic needs.

While these five interventions have significantly influenced the transition process, they have not been without problems for COSATU. First, their involvement in a wide range of forums, including the Negotiating Forum, has raised fundamental questions about the capacity of COSATU to participate effectively while retaining its coherence as a labour federation. The clearest example of this problem was an abortive strike called by COSATU's Central Executive Committee (CEC) over the lockout clause in the interim constitution agreed to by its own representative in the NMC. The abortive strike underlines the problem of the relation of the NMC negotiators to the constitutional structures of the federation. According to a prominent labour analyst, "There is seldom time for the NMC delegates to meet with the leadership and discuss NMC issues. In this incident, the NMC delegate had to rely on discussions with COSATU's legal advisers however the COSATU delegate failed to inform the CEC of his support for the compromise in the NMC. If COSATU cannot ensure adequate mandating and report back in the NMC" (von Holdt, 1993, p. 39), which is so directly concerned with labour interests, how can these problems be avoided in the other forums in which it participates?

These capacity problems have been exacerbated by the departure of senior leaders to Parliament, to the civil service, and to the corporate sector. Due to their organizational skills and vast experience, these officials are being recruited to take leading positions in sectors previously closed to cadres from the democratic movement. Not only does the federation lose their skills, but these leaders were in many cases the originators of the radical reform strategy and bearers of the independent union tradition. The impact of this "brain drain" has led a key labour observer to comment:

Several affiliates are experiencing internal instability, with wholesale change of leadership and internal conflict. There is a widening gap between leadership and base, between the "big three" [COSATU's metal, mining, and textile affiliates] and the smaller affiliates, and members of affiliates complain of a lack of service. There is a steady drift of skilled and experienced officials out of the movement. COSATU seems to have lost the capacity to develop strategies and campaigns (von Holdt, 1994, p. 20). These capacity problems have emerged at exactly the moment when the sophistication of issues confronting the movement has increased dramatically. Major changes in state policy, the restructuring of industry, and the creation of corporatist-type bargaining fora require unprecedented reliance on the very intellectuals and strategists who are now leaving the movement.

A serious problem for COSATU is a growing gap between leadership and the base. A leading union organizer has written on the decline of union locals, which are being turned into "the passive recipients of the national directives." More significantly, he points to the decline of the vision that drove union organizers before February 1990 to "make enormous personal sacrifices and push the union movement into achievements well beyond the resources available. (Marie, 1992, p. 21)" A nationwide survey of COSATU members has discovered disturbing evidence of this growing gap. (Ginsburg and Webster, 1995) Of the members surveyed, 80 percent knew nothing about the NMC or the NEF and had never been present when there was a report back on either forum. Even more disturbing, 75 percent did not know what the RDP was and had never participated in any discussion around it.

This growing gap between leadership and base seems to support orthodox interpretations of the demobilizing effects of corporatist-type arrangements as the leaders are drawn into elite tripartite institutions, leaving their members behind. These problems have led to an ongoing debate inside the federation over the wisdom of such participation. As far back as the campaign against the LRA, dissident officials - especially among regional and local organizers-have inveighed against participation and have rejected, in particular, the notion of a social contract or accord between labour, capital, and the state (see Vally, 1992).

The gap is all the more disturbing in a federation such as COSATU, where democratic accountability is founded on a structured link between peak-level leaders, shop stewards, and the rank and file. But more than internal democracy is at stake; radical reform depends on the capacity of leaders to mobilize members around shared goals and to restrain members when necessary. If the link between leaders and the rank and file is ruptured, the strategy breaks down and the union leaders may lose their bargaining power within the Alliance, in the state, and even before employers.

A more fundamental challenge lies within the Alliance itself. The ANC is a broad cross-class nationalist organization attempting to build social consensus among a range of constituencies, especially the oppressed and excluded communities of South African society. It cannot afford to give primacy to the interests of organized labour any more than it could give primacy to women, the unemployed, or youth. In spite of COSATU's demand that a Platform on Worker Rights be adopted along with the RDP, a separate document was abandoned and worker rights clauses were subsumed within the broader RDP. Furthermore, the new government will be constrained less by a threat from Hardliners on the right and more by domestic and international capital and international agencies, such as the IMF and the World Bank, to follow neoliberal economic policies. An ANC-led Government of National Unity will be under tremendous pressure to reduce the social wage, weaken wage-setting arrangements including trade union rights, and enforce laissez-faire market systems. The ANC will be pushed from these quarters to minimize the radical potential within the RDP. A new government will almost certainly expect a reversal of the primary role of unions: to become developmental rather than representational actors, on the grounds that they are a small and relatively privileged sector of the population. This tendency may reinforce the gap between leaders and members identified previously.

Nor will these pressures be easily resisted by the union leaders COSATU released to stand for Parliament. By choosing to enter electoral competition, these new Members of Parliament made a series of definite choices that arise from the particular organization of a parliamentary system and that may undermine their capacity to represent worker interests. These choices are threefold: to seek advancement of labour's interests within the existing institutions of capitalism, to rely on multiclass support, and to seek reforms and partial improvements (see Przeworski, 1985). These leaders had to resign their union positions and run for office as ANC, not COSATU, candidates. They are subject to parliamentary caucus discipline and are no longer formally accountable to COSATU members. Furthermore, a clause in the new constitution forces members of Parliament to resign their seats if they leave their party, which provides the party leadership with powerful sanctions over dissident members.

If, as argued earlier, the labour movement had played a hegemonic role in the alliance during the high point of anti-apartheid struggle in the late 1980s, by the elections of April 1994 the ANC had clearly reestablished its hegemony in the prodemocracy movement. While COSATU had been able to place its stamp on the transition to democracy - developing new institutions, new policies, and new practices - it will be increasingly under pressure to sacrifice its "narrow" interests to the goals of national development, as defined by the new political leaders.

However, the same COSATU membership survey identifies the persistence of a tradition of direct participatory democracy among the rank and file, where leaders elected by workers are expected to be accountable and to report back to their members. Importantly, the majority were committed to the view that parliamentary democracy must be substantially the same and consist of elected members held regularly accountable to the citizens. If the new government fails to deliver, they claim they will resort to ongoing mass action to force the government to live up to its electoral promises. Despite the organizational breakdowns identified earlier, the principles of radical reform remain deeply embedded in the culture of the organization and point toward potential conflict within the Alliance over the implementation of the RDP.
 
 

Conclusion

We have shown how the South African labour movement has played a central role in the origins of the transition process and in the development of the transition itself through a variety of interventions driven by the strategy of radical reform. However, with the creation of a new parliamentary democracy, there are increasing signs of a widening gap between the leadership and the base, echoing the classic features of goal displacement first described by Michels. If the labour movement does not address the problems head on, it indeed runs the risk of bureaucratization and co-optation, with its power - historically based on its capacity for disciplined mobilization - slowly ebbing away.

To pursue the strategy of radical reform in a new democracy - to realize the progressive possibilities of the RDP - priority will have to be given to organizational renewal or to reestablishing the tight relationship between leaders, shop stewards, and the rank and file. As in the past, the labour movement's influence within the state will depend on its organizational strength and its ability to use its collective power strategically. With this power, the labour movement will be able to take advantage of the opportunities that have opened in the democratic transition: friendly ministers in key state departments, the possibilities of gaining access to state resources - research, information, training, and finances - for capacity building, and greater freedom of action under a new constitution and Bill of Rights.

The centrality of the labour movement in the South African transition to democracy alerts us to a serious gap in contemporary analyses of transition. Not only have writers generally ignored the role played by labour movements, but where they have given consideration to labour, its role has been misunderstood as little more than one of mobilizing or restraining its members as dictated by the pace of negotiations over an elite pact. This is not merely an empirical lapse or a failure to pay attention to particular facts but suggests problems in the conceptualization of transitions themselves, which have prescriptive implications. Such arguments may have the practical effect of delegitimizing political projects with the potential to realize a fuller measure of democracy.

In place of an elite-centered approach to democratic transition, we have argued that popular movements - as demonstrated by the labour movement in South Africa - can play central roles in creating the conditions leading to transitions as well as in using their intelligence and other resources in shaping the form and content of the transition to promote their objectives of democracy and equality. It is certainly an open question whether the South African labour movement can continue to play this role during the period of consolidation of democracy. Nonetheless, the profound effect its strategy of radical reform has had on the transition itself shows that the expansive hopes of those who struggled for progressive change in South Africa need not be contained in the claustrophobic space of a conservative negotiated outcome. The limited elite pact is but one possible alternative to chaos. Sophisticated and powerful movements, such as labour, offer possibilities for alternative outcomes that may be able to propel a transition beyond such narrow confines. The South African case encourages advocates of progressive transition not only to reassess completed transitions to account more fully for the role of popular movements but to explore current cases to identify progressive potentials that have been either overlooked or declared impossible by transition theorists.
 
 

Postscript

Since this article was written on the eve of South Africa's first democratic election, labour has continued to play an important but uneven role in the consolidation of democracy. Labour has grown significantly. In the case of COSATU, membership increased between 1994 and 1996 from 1.3 million to 1.9 million, with growth coming in mining, service, and the public sector. Furthermore, the "brain drain" appears to have abated, as unions report fewer staff departures and virtually no vacancies. New leadership is emerging from within the ranks of the movement. Most importantly, though strike statistics are down overall, the unions have prosecuted successful wage strikes in particular sectors, even in those facing strong threats from international competition, such as automobile and clothing and textile. This mobilizational capacity has not been limited to collective bargaining issues, but has also been deployed in pursuit of broader political questions, such as the April 1996 stayaway around the ultimately successful demand to prevent a lockout clause being inserted in the new constitution.

Labour has also expanded its direct influence in the state. Ex-COSATU leaders occupy key positions in the Departments of Labour and Trade and Industry, and lead crucial committees in parliament, such as Mineral and Energy Affairs. In these roles they have contributed to innovative policies, including the construction of a new corporatist/co-determinist labour regime as well a new human resource policy. However the new government has come under strong pressure from domestic and international capital to implement neoliberal economic and social policies. These pressures led to the redefinition of the RDP along more market-driven lines, and ultimately to its marginalisation symbolised by the closure of a separate RDP office. The RDP's development vision has been replaced by the government's new macroeconomic strategy. The "Growth, Employment and Redistribution" strategy (GEAR) retains some of the redistributive goals of the RDP, but within a framework that emphasises fiscal deficit reduction, gradual relaxation of exchange controls, reduction in tariffs, tax reductions to encourage private sector (and especially foreign direct) investment, and the restructuring of state assets (privatisation).

GEAR has created profound stresses within the Tripartite Alliance between the ANC, the South African Communist Party and COSATU, which have yet to be resolved. However the Alliance has also enabled COSATU to make interventions in specific areas. Though it had little purchase on macroeconomic policy-making, it was able to influence privatisation policy through negotiating with the government a National Framework Agreement for the public sector as a whole. The Agreement creates bipartite structures in each subsector which give labour a major role in shaping the pace, extent, and form of restructuring.

In general, labour is struggling to adapt to the new terrain of policy making in the new democracy. This involves more than simply a lack of capacity, although capacity problems exist. It entails the difficulties of shifting from outright resistance to the state to a subtle combination of opposition with engagement and co-responsibility for particular policies. The shift is made more difficult given that little has changed in the workplace, which remains characterised by low trust, racism and the continuation of the "apartheid workplace regime." (von Holdt, in press) Labour is not yet fully committed to these new institutions at the centre of the new democracy, and as a result, has not fully shifted its resources and intellectual orientations towards active engagement. Moreover, the shift requires specific policies and a transformative vision to give coherence to them, and both remain in short supply.

It is easy to underestimate the constraints facing governments in a liberalising world, but these constraints are not an iron cage, and remain subject to contestation and negotiation. They have become more apparent in the two-and-a-half years since the elections, and not only with respect to capital's power. During this period the democratic movement has come up against the intractability of institutions, and has had to develop a longer view of the pace of transformation. In part, this has to do with problems associated with the old guard in the state - the persistence of the past. More importantly it grows out of the difficulties in changing complex institutions while simultaneously attempting to deliver resources to people in the short term through the same institutions.

Whether labour succeeds in achieving its objectives depends in the first instance on itself. It retains its independent base and its capacity to mobilise in pursuit of its goals. Furthermore, it is unlikely that it will be repressed by the government, especially given the new labour regime. But it could face increasing marginalisation if it fails to adapt itself to the new terrain.
 
 

BIBLIOGRAPHY

Adam, H and Moodley, J (1993), Negotiated Revolution: Society and Politics in Post-Apartheid South Africa, Johannesburg : Jonathon Ball.

Adler, G Maller, J and Webster, E (1992) "Unions, Direct Action, and Transition in South Africa," in N Etherington (ed) Peace, Politics, and Violence in Southern Africa, London : Hans Zell.

African National Congress, (1994) The Reconstruction and Development Programme: A Policy Framework, Johannesburg: Umanyano.

Baskin, J (1989) Striking Back: A History of COSATU. Johannesburg: Ravan.

Buhlungu, S (1994) "COSATU and the Elections," South African Labour Bulletin, 18, No. 2.

Cheadle, H (1987) "Recognition at the Workplace: The Formulation of a Strategy," unpublished seminar paper, Yale University, Southern Africa Research Program, February 1987.

Di Palma, G (1990) To Craft Democracies: An essay on Democratic Transitions. Berkeley and Los Angeles: University of California Press.

Eidelberg, P (1993) "The Unions and the African National Congress" South African Historical Journal, 28, May, 270-291.

Fine, A, Webster, E (1989) "Transcending Traditions: Trade Unions and Political Unity," in South African Review 5, (ed) by G Moss and I Obery, Johannesburg : Ravan.

Friedman, S (1987) Building Tomorrow Today: African Workers in Trade Unions, 1970-1984. Johannesburg: Ravan.

Friedman, S, ed. (1993) The Long Journey: South Africa's Quest for a Negotiated Settlement. Johannesburg: Ravan.

Ginsburg, D and Webster, E, et al. (1995) Taking Democracy Seriously: Worker Expectations and Parliamentary Democracy in South Africa. Durban: Indicator Press.

Hirson, B (1989) Yours for the Union: Class and Community Struggles in South Africa. London: Zed.

Inglehart, R (1990) "Values, Ideology, and Cognitive Mobilization in New Social Movements," in Challenging the Political Order: New Social and Political Movements in Western Democracies, edited by RJ Dalton and M Kuechler. Cambridge : Polity Press, 43-66.

Innes, D (1984) Anglo American and the Rise of Modern South Africa, Johannesburg : Ravan, 169.

Joffe, A and Lewis, D. (1992), "A Strategy for South African Manufacturing," South African Labour Bulletin, 16, no 4, 24-31.

Keck, M E (1992), The Workers' Party and Democratization in Brazil. New Haven, CT : Yale University Press.

Lambert, R and Webster, E (1988), "The Re-emergence of Political Unionism in Contemporary South Africa?", in Popular Struggles in South Africa, (ed) W Cobbett and R Cohen, London : James Currey.

Lee, R and Schlemmer, L, eds. (1991) Transition to Democracy: Policy Perspectives 1991. Cape Town: Oxford University Press.

Luckhardt, K and Wall, B (1980), Organize...or Starve! The History of the South African Congress of Trade Unions. New York : International.

Malloy, J M and Seligson, MA, eds. (1987), Authoritarians and Democrats: Regime Transition in Latin America. Pittsburg, PA: University of Pittsburg Press.

Maree, J (1987), The Independent Trade Unions, 1974-1984: Ten Years of the South African Labour Bulletin. Johannesburg: Ravan.

Marie, B (1992) "COSATU Faces Crisis: Quick-Fix' Methods and Organizational Contradictions," South African Labour Bulletin 16, no. 5.

O'Donnell, G, Schmitter, PC and Whitehead, L eds. (1986) Transitions from Authoritarian Rule: Prospects for Democracy. Baltimore, MD: Johns Hopkins University Press.

Olofsson, G (1988) "After the Working-Class Movement? An Essay on What's `New' and What's `Social' in the New Social Movements," Acta Sociologica, 31, no. 1, 15-34.

Pateman, C (1971) Participation and Democratic Theory. Cambridge: Cambridge University Press.

Payne, LA (1991) "Working Class Strategies in the Transition to Democracy in Brazil," Comparative Politics, 23, no. 3, 221-238.

Przeworski, A (1985) Capitalism and Social Democracy. Cambridge: Cambridge University Press. 

Przeworski, A (1991) Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. Cambridge: Cambridge University Press.

Saul, J (1991) "South Africa: Between Barbarism and Structural Reform," New Left Review, 188.

Saul, J. (1994), "Thinking the Thinkable: Globalism, Socialism, and Democracy in the South African Transition," in The Socialist Register, eds, R Miliband and L Panitch, London: Merlin.

Seidman, GW (1994) Manufacturing Militance: Workers' Movements in Brazil and South Africa. Berkeley: University of California Press.

Simons, HJ and Simons, RE (1969) Class and Colour in South Africa, 1850-1950. Baltimore, MD: Penguin Books.

Tarrow, S (1991) "Struggle, Politics, and Reform: Collective Action, Social Movements, and Cycles of Protest," Occasional Paper no. 21 (2nd ed.), Cornell University, Center for International Studies, Western Societies Program.

Touraine, A (1982) The Voice and The Eye. Cambridge: Cambridge University Press.

Turner, R (1972) The Eye of the Needle: Towards Participatory Democracy in South Africa: Johannesburg: Ravan.

Valenzuela, JS (1989) "Labour Movements in Transitions to Democracy," Comparative Politics 21, no. 4, 445-472.

Vally, B (1992) A Social Contract: The Way Forward? Johannesburg: Phambili Books.

van zyl Slabbert, F (1992) The Quest for Democracy. Johannesburg: Penguin.

von Holdt, K (1993) "The Lockout Strike: Confusion in COSATU," South African Labour Bulletin 17, no. 6, 38.

von Holdt, K (1994) "Impressive Gains, Organisational Crisis," South African Labour Bulletin 17, no. 5, 20.

von Holdt, K (in press) "From the Politics of Resistance to the Politics of Reconstruction? The Union and Ungovernability' in the Workplace," in Adler, G and Webster, E, eds.

Democratisation in a Liberalising World: Trade Unions and Transition in South Africa, London: Macmillan, in press.

Waterman, P (1991) "Social-Movement Unionsim: A New Model for a New World," Working Paper Series no. 110, Institute of Social Studies, The Hague, Netherlands.

Webster, E. (1985), Cast in a Racial Mould: Labour Process and Trade Unionsim in the Foundaries, Johannesburg : Ravan.

Webster, E. (1988), "The Rise of Social-Movement Unionsim: The Two Faces of the Black Trade Union Movement in South Africa," in State, Resistance, and Change in South Africa, eds, P Frankel, N Pines, and M Swilling, London : Croom Helm.

Webster, E. (1993), "Moral Decay and Social Reconstruction: Richard Turner and Radical Reform," Theoria 81/82.

Back to the top