- The New South AfricaA Season for Power-Sharing
Constitutional engineers commonly overlook a significant factor in democratization: time. Some institutions that may be workable or even necessary in the short term may undermine the long-term process of democratic consolidation. This is true of several elements of Arend Lijphart’s well-known model of consociational democracy. Lijphart’s consociationalism has four basic components: the sharing of executive power by all significant groups (government by “grand coalition”); such groups’ retention of a high degree of cultural or regional autonomy; proportionality in the distribution of civil-service positions, public funds, and legislative seats (the latter through proportional representation, or PR); and a minority veto on most vital issues. 1
This essay assesses the changing value of power-sharing for South African democracy over time. In particular, it will be shown that while government by grand coalition was necessary to bring about South Africa’s transition to democratic rule, its continuation would hinder the institutionalization of the new regime. An examination of the intellectual debate reveals that both proponents and opponents of consociationalism follow a similar logic. For example, Lijphart and two of his South African critics, Janet Cherry and Steven Friedman, proceed from a specific conception of democracy, focusing on its virtues and understating its costs. They fail to recognize that each version of democracy contains features that other reasonable people would find undemocratic. Furthermore, none of the three sufficiently distinguishes between democracy and democratization. [End Page 67]
South Africa’s 1993 interim constitution—which is currently being revised by a Constituent Assembly, made up of the directly elected National Assembly and the indirectly elected Senate—provides both for PR (with the election of parliament from national and regional party lists) and a government of national unity (GNU). Any party with at least 5 percent of the national vote was guaranteed representation in the executive cabinet. Moreover, any party with at least 20 percent of the vote became entitled to a deputy presidency. In the April 1994 national elections, the National Party (NP) garnered 20.4 percent of the vote, thereby securing a deputy presidency (filled by former president F.W. de Klerk) and six cabinet portfolios. The Inkatha Freedom Party (IFP), with 10.5 percent of the national vote, obtained three cabinet positions.
The new government is clearly a power-sharing system. Whether the South African settlement fully exemplifies Lijphart’s consociationalism, however, is open to debate. Lijphart’s own conclusion was that the “newly founded [South African] democracy . . . embodies all of [consociationalism’s] basic principles” and represents “optimal power-sharing.” 2 Yet at least one component of the model—an explicit minority veto—is missing. More important, the constitution does not contain a single group-based provision, focusing instead on the rights of the individual. Indeed, the general consensus is that the South African constitution is not a consociational document but takes the individual as the basic unit of reference, as do the constitutions of most liberal democracies. 3
It may be too early to say whether the South African experience vindicates consociationalism, or even just some of its power-sharing features. Lijphart celebrated the success of South African consociationalism within little more than two months of the April 1994 elections. Critics of power-sharing in South Africa were almost as quick to condemn it. Any serious assessment will require a much longer view.
An Essential Precondition
There is no doubt that power-sharing was an essential precondition of democratization in South Africa. Indeed, it was the minimum that the NP would have been willing to settle for. For the African National Congress (ANC) it was, at worst, an irritant. In fact, it served as a confidence-building device for all concerned. Broadly, it was the minimum requirement for stability. There were also specific payoffs for the various actors. For the ANC, it guaranteed continued and essential support from whites. For the NP, it provided a means of monitoring and limiting the power of the ANC. For the international community, it met a combination of objectives: it brought political legitimacy through the ANC, as well as the promise of pressure for free-market policies from...