Statistics on violence – particularly towards women and children – are highly politicised in South Africa. Typically, they serve one of two purposes: either they provide government with evidence of a good story to tell; they are winning the war against against crime. Or they highlight a crisis and the government's inability to crack the problem.
However, both approaches poorly reflect reality and result in interventions inadequate to solving the problem.
Official statistics don't tell the full story
While the official police statistics reflect decreases in sexual offences, other sources say something very different.
Figures released by the South African Police Service earlier this year show that crimes against children decreased by more than 3000 between 2012-13 and 2013-14. Conviction rates increased to 75% and more than 1832 life sentences were secured for perpetrators of these crimes. The numbers echoed the 2014 crime statistics, which showed that rape had declined by 6.2% between 2008-09 and 2013-14.
Various media outlets have, in the past, reported that Interpol has crowned South Africa the rape capital of the world at least twice. The BBC alleges that South African women are more likely to be raped than learn to read. Another British newspaper asserts that up to 40% of women suffer this fate in their lifetimes. The "Blow the Whistle" campaign from a local non-government organisation claims that one in three South African women will experience rape.
While these statistics may be alarming, none are derived from credible research. And they do not assist the public to understand the problem around violence against women and children. And neither of these sets of numbers capture the rapes that are never brought to the police's attention.
A more accurate picture of the extent of sexual violence in the country is emerging from credible bodies such as the Medical Research Council, as well as research by not-for-profit organisations.
In 2002, drawing on data from the 1997 national South African Demographic and Health Survey, the Medical Research Council found that only one in seven women who had been raped reported the matter to police.
Research conducted 12 years after the 1997 survey suggests that under-reporting persists. Interviews conducted in the Gauteng province in 2009 showed that one in 12 women in the province had been raped that year, but only one of every 13 women raped by someone who was not their partner reported the matter. The number was even more startling for women who had been raped by their partner. Only one in 25 reported this to police.
The same study was undertaken in KwaZulu-Natal and Limpopo. In KwaZulu-Natal, one in seven women who had experienced rape reported the matter while one of every six women raped by non-partners in Limpopo informed the police.
Creating a barrier to reporting crime
In 2010, the South African cabinet set the police the target of reducing rape and other contact crime by between 4% and 7% annually. Although these crimes need to be reduced, the reality of the reduced target means that if only one in every six women were reporting the crime, these targets are the equivalent of asking the police to ensure that only one in ten women in Limpopo reports in the future.
These targets also appear to be having a perverse effect. Research shows that the performance of many police stations is tied to meeting targets. This has led to situations in which dockets are dumped, victims turned away, or their cases not registered.
It is unsurprising that some members of the police have resorted to such tactics. Many of the factors associated with these crimes are not within their power to alter. These include experiences of childhood neglect, abuse and other difficult life circumstances, a sense of sexual entitlement and pervasive hostility to women.
It is therefore entirely possible that the figures show the police to be very successful in reducing the number of women who report violent crimes – while simultaneously having almost no effect on reducing the number of these crimes.
If so, then the police may just have become an obstacle to the reporting of sexual offences and domestic violence.
A new policy is needed
To address this, South Africa needs a different policy around the reporting of rape and other gendered crimes. It should emphasise ways to reduce the barriers to reporting rape and domestic violence rather than reducing the number of reports.
A new policy could measure stations' performance by their compliance with legislation like the Domestic Violence Act, and the officers' abilities to correctly apply law and policy around domestic violence and sexual offences.
These would include the procedures and competencies they need to help deaf victims of sexual offences, or lesbian and male victims, or those with intellectual disabilities.
These sorts of indicators treat the police's quality of services to victims, rather than a reduction in numbers, as their measure of success.
Making this policy real also requires the police to record their statistics differently. For example, they could record the relationship between victim and perpetrator. Currently this information can only be obtained through costly research carried out by external agencies.
If the police disaggregated their data in this way, the number of murders that occur within the context of a domestic relationship could be tallied annually. This information would provide insight into the dynamics of such murders – and perhaps even their prevention.
If this approach was followed then an increase in the statistics would be for the good. Increases could suggest a greater trust in the police, rather than evidence of a crisis, or the government's inability to win the war on crimes affecting women and children. This would remove the incentive to juggle the statistics.
And when used in this way, police statistics are more likely to be measuring what counts, rather than counting what can't be measured.
Explore further: Domestic violence victims may be hurt by mandatory arrest laws