Briefs.png

DOMESTIC VIOLENCE III - WHAT DO WE KNOW ABOUT SOUTH AFRICAN CONDITIONS?

The preceding briefs in this series discussed definitional issues and set out international comparative information on domestic violence. This concluding brief will assess information on domestic violence in South Africa.

 

 

Introduction

 

Information on domestic violence in South Africa comes from three sources: (a) indirectly, from police statistics, and directly from (b) Statistics South Africa’s Victims of Crime Surveys and (c) specialised surveys.


 

Police statistics

 

All crimes

 

Figure 1 displays the rate per 100 000 people of violent crimes reported to the police from 2006 to 2015.

 


 

There was a marked drop off between 2006 and 2012, but the rate has remain roughly constant since then.

 

Table 1 sets out the average prevalence of types of violent crime against the person over the period 2010 to 2015.


 

Table 1 – Average prevalence of crimes against the person, 2010-2015

Rate per 100 000

 

Murder

   32

Attempted murder

   31

Assault with the intent to cause grievous bodily harm

 352

Common assault

 323

Common robbery

 101

Robbery with aggravating circumstances

 216

Sexual offences

 110

All crimes against the person

1165


 

Table 1 portrays a very violent society.  The rates for murder, assault, robbery and sexual offences are all above the 90th percentile of the distributions across countries reported in the preceding brief.

 

The Domestic Violence Act

Section 4 of The Domestic Violence Act 116 of 1998 allows complainants to apply to a magistrate’s of family court for a protection order against domestic violence.  If satisfied that the application provides prima facie evidence of abuse, the court can then issue a protection order, which may be contested by the alleged perpetrator.  Failure to comply with the conditions of an order leads to the issue of a warrant of arrest.

New applications rose from 222 919 in 2008/09 to 275 536 in 2015/16 [1].  Over the period, the orders finalised were 36.4% of new applications, with no trend.  Warrants issues for breach of orders rose from 17 936 in 2008/09 to 39 550 in 2015/16.   The prevalence of applications for protection from domestic violence was 498 per 100 000 population in 2015/16.

Of the 246 police stations visited between April and September 2016, none were fully compliant with the expected standards for dealing with domestic violence.  29% were rated as substantially compliant, 56% as partially compliant, and 15% as non-compliant [2].   

 

Victims of crime survey

Since 2011, Statistics South Africa has been conducting an annual victims of crime survey (VOCS).  The most recent statistical release contains information from the 2015/16 survey and the most recent publicly available data set is from the 2014/15 survey.

The VOCS distinguishes between crimes against the household, which include murder and home robbery, and crimes against individuals, including vehicle hijacking, other forms of robbery and assault.  Data on crimes against individuals are collected from people age 16 or older, so the VOCS tells us nothing about actual abuse of children under 16.

It should be borne in mind that there is a difference between crimes in a particular year and victims of crime, since some victims may have experienced a crime more than once.  In 2015/16, 6.9% of households reporting home robbery, and 7.8% of individuals reporting robbery had suffered it more than once.

Table 2 sets out the number of victims of violent crime per 100 000 people in 2015/2016, along with the extent of reporting to the police.

 

Table 2 – Victims of crime per 100 000 people, 2015/16

Crime

Prevalence

Per cent reported to police

Household

   

Murder

  96

95

Home robbery

1095

66

Individual

   

Robbery

 788

44

Assault

 729

49

 

It should be noted that the precision of these estimates is quite low because of limited sample size and of prevalence rates of just over 1% and below [3].  Taking the reporting rate into account, the VOCS estimates are consistent with police statistics for murder and assault, but they are considerably higher for robbery.  Crime rates reported over the preceding five years are only a little higher than for the preceding year, suggesting either a high degree of forgetfulness or the possibility that a part of the population is much more prone to become victims of violent crime than others.

In the case of murder, 43% of the motive behind the murder was attempted rape, 28% long term or sudden anger or jealousy, 21% financial or other reasons and 7% gang violence.  On the other hand, anger or jealousy was behind 58% of assaults, while other motives included financial reasons (13%), robbery (8%), and gang related reasons (7%).  20% of assailants were spouses or lovers, 5% other household members, and a further 51% were known to their victims.  28% of assaults occurred at home.   58% of robberies occurred at home.

As far as perception of violent crime is concerned, 41% of households in 2015/16 felt that street robbery was common, 39% home robbery, 19% assault, 18% business robbery, 16% murder, 10% vehicle hijacking, 6% mob justice/vigilantism, 5% child abuse and 2% political violence.  84% of households felt that walking alone in their areas of residence was very safe or fairly safe during the day, compared with 31% at night.  A third of households avoided open spaces and public parks for fear of crime.  62% of households felt that the perpetrators of violent crime are from the local area, 32% from elsewhere in South Africa and 6% from other countries.  42% of households felt that violent crime had increased in their areas of residence, compared with 30% who felt that it had remained the same and 28% who felt that it had decreased.

Using the 2014/15 VOCS data set, it is possible to disaggregate the prevalence of crime by geotype: (a) metros (b) towns and farms and (c) tribal areas.  The prevalence rates are set out in Table 3.

 

Table 3 – Victims of crime, by geotype, 2014/15

Rates per 100 000 in the previous year

 

Geotype

Vehicle hijack

Other robbery

Sexual assault

Physical assault

Metro

344

1184

137

 679

Towns and farms

111

 520

105

1064

Tribal

 33

 439

125

 790

All

184

 772

125

 814

 

Vehicle hijack and other forms of robbery are more prevalent in metros than elsewhere, not surprisingly, because that’s where the money is.  There is no significant variation across geotypes in the prevalence of sexual assault, while physical assault seems more common in towns and farms than elsewhere.

 

Other sources of information

The South African Demographic and Health Survey 2016 [4]

The DHS asked ever partnered women age 18 and over whether they had experienced physical violence or sexual violence by any partner.  It found that 20.5% had experienced physical violence over their lifetimes, and 9.1% had experienced it in the last year.  The corresponding figures for sexual violence were 6.2% and 2.9%.  While the relative prevalence of physical violence and sexual violence is much the same in South Africa as it is globally, the reported prevalence of physical violence is below the global average reported by the 2013 World Health Organization survey of evidence, a surprising result given the generally high levels of violence crime in South Africa.

 

The Optimus Study on Child Abuse, Violence and Neglect in South Africa 2015 [5]

The Optimus Study surveyed 5 365 old South African adolescents between 15 and 17 years old and asked about their life time experience of child abuse.  Information was gathered both in a household survey and a school survey: in each setting, young people were interviewed about their experiences by trained enumerators, and were also given the opportunity to respond to a small set of questions on a more confidential questionnaire which they completed themselves. The highest reporting rates were obtained from these self-administered questionnaires, particularly in schools.  In what follows, data from the self-administered questionnaires is reported.

The study found that:

  • One in five young people have experienced sexual abuse [6].  More males than females disclosed sexual abuse.  This rate is considerably higher than the average global rate of 12.7% found in a review of studies from around the world [7]. It is, however, very similar to the highest rates identified in the global review: 21.5% for girls in Australia, and 19.3% for boys in Africa.

  • 18.5% of girls and 15.5% of boys suffered from neglect [8], compared with an average of 16.3% in a study covering North American and European countries [9].

  • Just over one-third of young South Africans have suffered being hit, beaten, kicked or physically hurt by an adult caregiver.  The global average cited for comparison is 22.6% [10].  Yet the extent of corporal punishment in the Optimus Survey is not clear, rendering the comparison insecure.

  • 16.1% of respondents reported emotional abuse [11], more common among girls than among boys.  No global comparative estimates are provided.

  • 23.1% of respondents reported exposure to some sort of violence by others in the home.  

  • One in five young people reported being bullied, compared to the one in three found by the UNICEF study cited in the preceding brief.


 

Assessment

 

The available evidence suggests that abuse of women and children in domestic settings may, on the whole, not be much different from global averages, with the exception of sexual abuse of children and, possibly, physical violence from an adult caregiver.  However, traumatisation as a result of household robberies does seem to be markedly higher than the global average.  Combined with the very high levels of violent crime as a whole by international standards, it may well be a distinguishing characteristic of South African violence is that its occurrence outside the home is extraordinarily high by international standards.  Policing is an issue: more than a third of households have to travel more than 30 minutes to get to their nearest police station, and only a third of households report seeing the police every day in their area of residence, with a fifth never seeing them at all.  


 

Charles Simkins

Head of Research

charles@hsf.org.za

 


 

Notes:

[1] Lisa Vetten, Police accountability and the Domestic Violence Act: a review of sixteen years of implantation, Presentation to the Parliamentary Police Committee, 21 June 2017

[2] Civilian Secretariat for the Police Service, Domestic Violence Act: Report for April to September 2016, Presentation to the Parliamentary Police Committee, 21 June 2017

[3] The 95% confidence interval for the prevalence of murder is between 60 and 132 and for the prevalence of assault is between 626 and 832.

[4] Statistics South Africa, South African Demographic and Health Survey, Report 03-00-09, 2016

[5] Centre for Justice and Child Crime Prevention, Research Bulletin: The Optimus Study on Child Abuse, Violence and Neglect in South Africa, 2015
 
[6] Sexual abuse includes  abuse by known or unknown adult(s), abuse by a child(ren) or teen(s), forced sexual intercourse (actual or attempted); sexual-exposure abuse, written or verbal sexual harassment, sexual experience with adult(s), sexual abuse related to gang initiation and sexual abuse related to animal(s)
 
[7] World Health Organization, Preventing child maltreatment: a guide to taking action and generating evidence. 2006, Geneva, Switzerland: World Health Organization and International Society for Prevention of Child Abuse and Neglect
 
[8] Neglect includes living in a home which is broken down, unsafe or unhealthy, and fearing people living in the home or frequently visiting it.
 
[9] J T Manly et al, Dimensions of child maltreatment and children’s adjustment: Contributions of developmental timing and subtype. Development and psychopathology, 13(04), 2001
 
[10] M Stoltenborgh, et al, Cultural-geographic differences in the occurrence of child physical abuse? A meta-analysis of global prevalence. International Journal of Psychology. 48(2), 2013
 
[11] Emotional abuse includes name-calling, scaring and expressions of unwantedness