The ‘Fuck White People’ Phenomenon in South Africa: A Discursive and Statistical Analysis

ABSTRACT

During the recent student protests in South Africa ‘fuck white people’ appeared on T-shirts and walls. When understood only as anti-white sentiments, these words could cause panic and fear. However, we need to determine the extent to which these opinions are shared, by whom, and what should be done to people who propagated these views. To address these questions, an item was inserted in the 2016 nationally representative South African Social Attitudes Survey (n = 2988). An overwhelming majority of the adult population (96%) expressed a clear preference for how this phenomenon should be addressed. Of this, 26% recommended that those wearing/writing this phrase should be informed that they are hampering the prospects of peaceful coexistence between different race groups (the social cohesion argument), while 39% advocated refrain from such rhetoric. 16% adopted a retributive stance and little more than a tenth (13%) non-interventionist. Only a nominal proportion (2%) believed such behaviour ‘should be praised’ for drawing attention to persisting white privilege in South African society. We discuss these findings and their socio-demographic variations discursively, as part of ‘white fragility’ and ‘rainbow nation mythology’, observing the provocative and productive dimensions of this phenomenon.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Benjamin Roberts http://orcid.org/0000-0002-0607-5447

Notes

1 The only subtle difference is among those aged 25–34 years, with punishment marginally less likely to be cited than non-intervention (13% vs. 14%), placed the latter in a higher rank (third) than is the case for other age groups (fourth).

2 Significance was determined through One-way ANOVA analysis [F(3, 2984) = 16.09, p-value = 0.000] based on a dichotomized version of the coded responses, together with post hoc Scheffe tests.

3 One-way ANOVA post hoc Scheffe tests, where F(3, 2984) = 11.91, p-value = 0.000.

4 This difference is statistically significant, based on One-way ANOVA post hoc Scheffe tests [F(3, 2984) = 19.67, p-value = 0.000].

5 Only 18% of those with no formal schooling chose the social cohesion response, compared with 20% of those with a primary level education and 25% of those with an incomplete secondary education. For all three groups of South Africans, this was significantly lower than for those with a tertiary education (33%). Those with a matric-level education were more inclined to select the social cohesion response than those with a primary level education (28% vs. 20%).

6 Those that are ‘just getting by’ stress the refrain option more than the non-poor (43% vs. 33%), while the non-poor favour punishment slightly more than those ‘just getting by’ (19% vs. 14%). All other group-based mean score differences are not statistically significant.

7 Even though the variation on the punishment response does not achieve statistical significance, it is important to observe that the share mentioning this was less preferred among those in informal settlements (10%) than in other environs (16–19%).

8 The share requesting that protestors to refrain from using ‘fuck white people' slogans in KwaZulu-Natal is higher than in all other provinces excepting Northern Cape and Mpumalanga. In the case of the Northern Cape, the share supporting this view is higher than in Eastern Cape, Gauteng, Limpopo and Western Cape (54% vs. 33–35%).

9 The punishment option actually is actually supported by a marginally higher share than the refrain option, meaning that it is the top ranked response in Limpopo, but the difference is at the decimal level (33.9% vs. 33.8%).

10 This figure is significantly lower than in five of the eight other provinces – all except Mpumalanga, Eastern Cape and Mpumalanga.

11 The percentage range refers to the shares selecting this option in five provinces, namely Northern Cape, Gauteng, KwaZulu-Natal, Western Cape and Limpopo. There is no statistical significance in mean share supporting this response between the Eastern Cape and North West, Free State and Mpumalanga (15–18%).

12 Those that were uncertain of their political identification were also more inclined to support behavioural refrain than DA supporters (30% vs. 29%).

13 The politically uncertain are also more partial to the damaging social cohesion argument than ANC supporters (32% vs. 22%).