Civil society in South Africa is generally celebrated as a space for action to promote social justice, either through organisations that play the role of “watchdog”, or through mobilisation by the poor themselves around their own concerns. However, civil society can reflect and reproduce many of the pathologies and injustices of the wider society. Sometimes it works to benefit a specific ethnic group or political group, and also reflects some unsatisfactory aspects of culture to which the constituents of civil society belong. In this study, both qualitative and quantitative analyses show that the associational activities and social movements in Cape Town reflect some kind of pathologies or injustices of the wider society. The qualitative analysis focuses on the ‘toilet war’, which took place throughout 2010 in Khayelitsha, Cape Town. It surely has an aspect of “service delivery protest”, where the poor and the marginalised undertake protest action to express their grievances for the improvement of service delivery. Nevertheless, the data shows that the “poorest of the poor” are actually excluded and deserted by democracy within the community, the partisan purpose, and the rage against the legacy of apartheid. Ironically, democratic and participatory processes of decision-making in the small community contribute toward ignoring the voices of the poorest of the poor, who are still the minority there. The partisan purpose and the rage against racism restrict the interest of the poorest of the poor, too. Also there is a subtle but critical disjuncture between the “commander” of the ‘toilet war’ and its followers, which makes it more difficult to ensure the civil rights of the poorest of the poor. On the other hand, the quantitative analysis shows that participation in associational activities and protests is not correlated so much to incomes and grievances. Rather, variables such as race, political attitudes, and psychological resources are more correlated with participation. This is particularly true with associational activities like being a member of community-based groups or attending community meetings. Cape Town has a substantial overlap amongst race, income and grievances, but the regression analyses indicate that race still has significant correlation with participation in civil society. Opposed to the general expectation for civil society in South Africa, participation in civil society is not always the channel for the poor to express their grievances. Although further research should be conducted on why black people are more likely to participate in civil society than other races, this study disputes the general romantic notion of civil society in South Africa.