Religion & politics, Never the twain should meet

Social attitudes survey

South Africans tend to oppose political intervention by religious leaders in relation to swaying voting decisions during times of elections and broader state decision-making. The latest round of the annual South African Social Attitudes Survey (SASAS), conducted nationally in November and early December 2008, included a detailed module on religious beliefs and practice, report Jaré Struwig and Ben Roberts.

As the country approached the 2009 general elections there was increasing debate about the extent to which politicians were using religion as a platform for securing support. Similarly, there have been instances where concern has been raised about the role of religious leaders in the political realm. What is missing from such discussion is the voice of South Africans on the role of religion in politics. Does the public want religion to have an impact on political decisions or not?

The national consensus

The latest round of the SASAS included a detailed module on religious beliefs and practice. The sample is nationally representative and includes 3 173 respondents aged 18 years and older (age eligible to vote in 2009).

Although the religion module in the SASAS covered many diverse issues, two statements in particular relate to the role of religion in politics, say HSRC researchers Jaré Struwig and Benjamin Roberts. Using a five-point scale, respondents were asked to state their level of agreement or disagreement with the statements ‘religious leaders should not try to influence how people vote in elections' and ‘religious leaders should not try to influence government decisions'.

They found that most South Africans (Figure 1) were not in favour of political involvement by religious leaders in relation to electioneering and government decision-making. A considerable majority (72%) stated that religious leaders should not try and influence votes during elections, while two-thirds of South Africans stated that they were opposed to religious leaders influencing government decisions. This shows that the issue of religion is not an especially divisive one in South Africa, with a broad consensus around supporting the idea of separating religion from politics.

It also seems that this orientation among South Africans may have gained some momentum since 2001. The absence of significant diversity in opinion across political and religious affiliation, as well as many other socio-demographic variables, is notable.

Where differences do exist, in most instances the size of the attitudinal gap on this issue is not particularly large. One notable exception is among those with higher living standards, who appear more tolerant towards religious leaders attempting to affect political decisions than those with lower living standards.

Struwig and Roberts say the findings should not be interpreted as a negation of the important societal and community roles that religious institutions and their leaders continue to perform as our democracy matures.

‘South Africa is a deeply religious society, with approximately 85% declaring a religious affiliation. In the 2008 SASAS survey, the vast majority of South Africans (83%) expressed high levels of confidence or trust in religious institutions, especially relative to politicians and political parties, the latter receiving a mere 29%.

‘As such, they would probably expect these institutions and affiliated leaders, as part of their role in the spiritual and moral welfare of the people, to make statements about important social and political concerns and raise their voices in related debates. Despite this, the inescapable fact is that there remained a strong preference for non-intervention of the church directly in relation to electioneering and governance.'

A considerable majority (72%) stated that religious leaders should not try and influence votes during elections, while two-thirds of South Africans stated that they were opposed to religious leaders influencing government decisions

Recent dynamics: A secular or sacred turn?

Since the fourth and fifth waves of the World Values Survey (WVS) (undertaken in 2001 and 2007 in South Africa) included religion and politics, it is of interest to compare these results with the SASAS 2008 findings. Between 2001 and 2007, the percentage who agreed with the statement saying that religious leaders should not influence voting decisions remained virtually unchanged, though there does appear to have been a slight increase between 2007 and 2008 from 64% in 2001 to 72% (Figure 1).

This observed difference is statistically significant, say Struwig and Roberts, but they emphasise that the increasing level of agreement is likely to be partly attributable to methodological differences between the surveys as well as real value change shaped by contextual events.

Nonetheless, the results do seem to indicate that South Africans have remained fairly entrenched in their view that religious leaders should not try to influence electoral outcomes through the pulpit.

A similar trend is evident when considering the pattern of responses to the statement about religious leaders and their influence on government decisions. Between 2001 and 2007 there was a modest rise in support for less involvement by religious leaders in government decisions, with a further consolidation of this viewpoint between 2007 and 2008.

Attributes of support for religion in politics

When analysing the 2008 SASAS data by demographic characteristics, the results showed that no significant differences were found between males and females when support for religion in politics was determined. In terms of age, young people aged 25-29 years were less inclined than the other age groups to want to separate the role of religion in politics. The largest gradient of difference is between 25-29 year-olds and those older than 60 years, with the latter group expressing the most resistance to a religious influence on politics.

Surprisingly, the political party that people support does not have a bearing on
whether they want more or less involvement of religion in politics

When the statements were analysed by socio-economic characteristics, including the Living Standard Measure (LSM) and educational attainment, it emerges that higher LSM groups tend to prefer greater involvement of religious leaders in matters of political interest.

Surprisingly, the political party that people support does not have a bearing on whether they want more or less involvement of religion in politics, as the observed difference in attitude between party supporters is not statistically different.

‘We arrive at a similar conclusion with regard to religious affiliation. Irrespective of whether one belongs to no religion or declares belonging to a specific religious denomination - whether Protestant, Catholic, Zionist/Shembe, Hindu, Muslim or other faith. There was a commonly held viewpoint that there should not be a religious influence on political decisions,' Struwig and Roberts said.

Jarè Struwig is a senior research manager in the Knowledge Systems (KS) research programme and Ben Roberts is a research specialist in the Child, Youth, Family and Social (CYFSD) research programme.