The following is a response to a BBC report on a study concerning rape in South Africa. It can be found at http://news.bbc.co.uk/2/hi/africa/8107039.stm
There are a number of things that might be worth saying.
I think one of the keys is what I call the 'flat' structure of South African social relations. (I say 'social relations' since using terms like 'society' or 'social structure' is not quite right in this context.) South Africa is one of the youngest and most mobile societies in the world. No social 'structure' that still functions today has a history of more than a couple hundred years and the same is true for its social identities, ethic categories, racial categories and so on. I think this is even true of its system of gender. Along with this, is an extremely fluid and 'loose' system of kinship, marriage and family property transmission.
There is almost no property passed from generation to generation in families through inheritance (except for a few elite 'oligarchs'--White, Indian, & Black), and very little heritable property of any kind. Outside of the 'canonical' large business enterprises (e.g mining such as Anglo-Ashanti Gold, etc.) and parastatals, there is very little political hierarchy of any kind (that is, hierarchies in which the 'higher' exercise effective control over the real actions of those 'beneath' them, although there are status hierarchies in 'traditional', religious, class and racial terms.) This is true of kinship as well.
This results in almost no control being exercised by the elders over juniors. Although few would accept it (!), in my ethnographic and personal experience of marriage and gender relations in South Africa, there is even very little actual control exercised by men over women. It is the case, however, that there is a universal and powerful ideology that men should and do exercise control over women (This is called 'patriarchy', after early European kinship theorists such as Bachofen, Engels, Freud and many others.) The outcome of this is that men cannot, in fact, exercise the control over women that they believe they should do. Violence results not from excess of control, but rather from an effective lack of it, and the consequent frustration felt by many men who cannot realise their own sense of self.
I know of many cases, too, in which people do not know who their father is. In many cases, they are not even certain who their mother is, since wide-spread fostering, early teen pregnancies, and pervasive mobility often disrupt bonds between mothers and their children. Where mothers die young, or at childbirth, their relation to their children is obviously disrupted, and this is more common that it should be. (SA has one of the worst records of maternal mortality, and this has become steadily worse since 1994.)
The power of ideology also obscure the sociological vision of this society since it is believed by all, including the researcher (Rachel Jewkes of the Medical Research Council, in this case) that only men are implicated in this system, in which 'South African masculinity' and 'patriarchy' become cover categories to explain everything. In my experience, women are also likely to be violent in 'domestic' cases of conflict. They also express the view that men should exercise control and that they should provide; when they don't, or can't, women can become exceptionally abusive and violent. I have seen this is countless court cases in the traditional courts where I work, and in virtually every domestic relationship that I am familiar with in the course of my work. Men always and absolutely deny that they have been attacked violently by girlfriends, wives and lovers, but many men in the town where I work can show scars (usually on their faces) where they have been attacked and deeply injured in domestic 'quarrels'.
Incidentally, my ethnographic work over 10 years in a small South African town in eastern Mpumalanga Province is one of the least violent in the region, and one of the oldest and most established.
It seems to me, even, that the extent and pervasiveness of sexual networks functioning as social structure, that it is almost the case that the South African sexual network amounts to a vast and strangely integrated system of kinship, a sort of national system of 'kinship' in which sex (sexual relations) functions as a primary form of social relation.
All of this has a strong bearing the pervasiveness of rape in southern Africa (NB not just South Africa, although only SA is actually studied in depth in this respect; very little is known about Lesotho, Mozambique, Zimbabwe or Namibia with respect to sex and sexuality; Swaziland and Botswana, however, are better studied.)