Racial Politics and Campaign Strategy in South Africa's 2009 Election

Abstract

South Africa's 2009 election featured competition between the ruling African National Congress (ANC), the steadily improving Democratic Alliance (DA), and the newly formed ANC splinter party, the Congress of the People (COPE). In 2009, national-level parties faced the same strategic imperative as in previous elections: to gain support among the Black majority. But the emergence of COPE had potentially important implications for election strategy. Competition was no longer only between the liberation party (the ANC) and the successor of an apartheid-era party (the DA). COPE had its own set of liberation-linked leaders that presented a new challenge to both the ANC and the DA. This paper examines the main parties' use of candidate lists, voter outreach, and campaign rhetoric to target the Black majority in 2009. It updates and expands previous studies of racial politics in South Africa, providing novel empirical data on the campaign tactics of the ANC, DA, and COPE.

Notes

1. The three parties' lists totaled over 2000 names. In order to conduct the racial coding, each name was evaluated as Black, Indian, White, or Coloured. Racially distinctive names in South Africa make this method highly accurate. However, since White and Coloured names are often difficult to distinguish, additional internet and facebook searches were conducted to confirm the person's race. A native South African and political science graduate student was employed as a research assistant for this task, which follows the method used by Ferree.

2. Ferree's racial coding of lists combined Whites and Coloureds, while this project separated the two. For comparative convenience, Coloureds and Whites are combined for 2009.

3. An additional note of caution is necessary for interpreting these data. The vast majority of voting-age South Africans reported no direct contact with any campaign, with 92%, 90.9%, and 80.1% of those surveyed reported no direct contact from the DA, COPE, or the ANC, respectively. Table 2 indicates the percentages of those who were contacted by a political party or candidate, not of the whole voting-age population.

4. COPE did not have a ‘traditional' constituency because 2009 was its first election.

5. Some additional clarifications are necessary. COPE's invocations of the aforementioned service delivery issues were coded as bonding. COPE needed to poach traditional ANC voters in order to gain power and, therefore, its use of ‘Black’ issues was considered strategic. Crime and security is a non-Black issue, but different coding logics were used for each party. DA references to the issue were considering bonding and COPE references were considered bridging, given their respective traditional constituencies. ANC references to crime and security were considered neutral. Even though this is a non-Black issue according to surveys, the ruling ANC is compelled to address all issues of service delivery. Even so, its references to crime and security were minimal and changing the coding to bridging would have had little effect. References to service delivery writ large, education, economic management, and corruption, regardless of which party said it, were coded as neutral. Blacks and non-Blacks were equally concerned about these issues.

6. Business Day, Cape Argus, Cape Times, Citizen, City Press, Daily Dispatch, Daily News, Diamond Fields Advertiser, Leadership SA, Mail & Guardian, Star, Sowetan, Sunday Independent, Sunday Times, The Herald, and Witness.