Crime, Community and the Governance of Violence in Post-Apartheid South Africa

Abstract

The South African government has embarked on a programme of encouraging social cohesion in South Africa first to address concerns stemming from high levels of violent crime which characterise the society, and second, to foster positive national identity in a complex, heterogeneous, racialised and stratified nation. Through a discussion of the impact of violent crime on emergent forms of community, this paper argues that the practices of communities evolving in the post-apartheid period show tendencies toward fragmentation rather than unification, undermining efforts of ‘nation-building’.

This paper considers the relationship between violence in South Africa and its impact on citizenship and governance. Violence considered outside of the normative terrain of law, and as structural, physical and psychological phenomenon, has wide ranging implications for the forms of community emerging in post-apartheid South Africa. There have been calls for increased levels of social cohesion as a remedial action to counter the propensity toward violence as a trend which breaks down the social fabric. This paper critically considers the implications of relying on ‘community’ as a way to reduce violence. It also discusses the practices of cohesion and community which violence brings into existence. These actual forms of community in the post-apartheid era may be at odds with unifying nation-building efforts of the government, and would result in balkanised zones of governmental practices mapped onto spatially distinct networks of living, regulated by racialised calculations of risk and mortality.

Based on the official statistics that have been released by the South African Police Service, there has been a consistent decrease in the number of reported crimes committed in South Africa. This reduction continued at a rate of 2.6 per cent in relation to the previous reporting period as of July 2007. Within the overall trends which point toward a decrease, two categories of crime have shown an increase. These two categories fall within what are called ‘contact crimes’. Two of the eight categories of contact crime increased: aggravated robbery and murder increased by 4.6 per cent and 2.4 per cent, respectively. Car hijackings and house robberies, which are sub-categories of aggravated robbery, increased by 6.0 per cent and 25.4 per cent, respectively (SAPS, 2007).

Based on studies of dockets, the geographical locations of crime, and analysis of the time frames within which crimes occur, the South African police argues that contact crimes like murder and assault are likely to occur amongst people who know each other and are friends, relatives or acquaintances. Based on their docket analysis, an alarming 89 per cent of assault to do grievous bodily harm and common assault, 82 per cent of murders and 76 per cent of rape cases involve people who know each other. In the category of attempted murder 59 per cent of cases showed a similar pattern (SAPS, 2007). In addition to these patterns, a number of other trends emerged. The contact crimes committed showed strong patterns involving alcohol and/or other substance abuse, and spatially these were associated with informal settlements in large or ‘mega-townships’. Furthermore, most serious violent crimes committed involved alcohol, and happened over weekends.

Based on these trends the Minister of Safety and Security argued that in relation to serious violent crime, ‘the social conditions people live under have an impact on the incidence of crime in South Africa … Poorer communities experience more violent crime than wealthier ones’ (Nqaqula, 2007). This conclusion is supported by a number of studies, which show that high incidence of violent crime are occurring in areas characterised by high levels of poverty, low levels of formal education, housing, electricity and sanitation (Gentle, 1984; Butchart and Brown, 1991; Nell and Brown, 1991; Demombynes and Ozler, 2006).

It is worth drawing a distinction between the actual incidences of violent crime, as it pertains to race and class, and the perception, or ‘fear of violent crime’. When the actual incidents of violence crime is contrasted to the ‘fear of violent crime’, surveys suggest that the so-called White and Asian respondents feel significantly more prone to the possibility of violent crime than Black African and ‘Coloured’ respondents (HSRC, 2008), with 34 per cent of Black South Africans feeling that they worry about the possibility of being a victim of violent crime, while 51 per cent and 74 per cent of White and Asian respondents, respectively, worry that they would suffer that fate. White and Asian South Africans generally occupy a higher class position than Black South Africans, indicating that the victims of violent crime are divided along race and class lines, and so too are the perceptions. Given that the legacy of the Group Areas Act of the apartheid period still largely survives, this has important spatial implications since race and residence still overlap to significant degrees.

Violent crime is an important indicator of a range of socio-political factors which have long term implications for social cohesion, democratic practices and economic development in South Africa. This paper takes both ‘crime’ and ‘violence’ as categories that need to be understood not simply as legal categories, but as social constructions. The activities that we identify as ‘criminal’ are defined as such because they fall into, or outside the domains and norms of behaviour which are governed by the law. The intentional killing of someone, for example, is considered ‘murder’ when it falls outside of the lawful prescriptions on killing someone. In countries that have the death penalty, for example, the taking of a life by the state is not considered murder, since it is accepted as punishment accorded to an individual, following a judicial process deemed fair.

Violence, in and of itself, may not be criminal, but its relationship to the existing legal framework will decide whether its particular form is criminal or not. This is quite apart from normative views, which might inform an individual or a group's ethical, moral and religious views on violence (Benjamin, 1986).

Outside of the legal categorisation of violence as ‘crime’, such as murder, or assault, we can think of violence in two categories or forms. The first is to think of violence as involving physical assault, the infliction of pain, or death. This notion of violence is consistent with the definition of violence developed by the World Health Organisation (2008):

the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community, that either results in or has a high likelihood of resulting in injury, death, psychological harm, mal development, or deprivation.

The World Health Organisation's definition is widely cited by those who work in conflict prevention, conflict resolution and public health interventions. It does seek to extend an understanding of violence by including psychological trauma and forms of deprivation which we might not immediately consider to have damaging physical somatic effects on the body. It does however require that the actions which have violent effects, whether material or psychological, are done so ‘intentionally’, with the premeditated aim of creating a violent effect.

Another consideration of violence emphasises its social, cultural and political dimensions as a process, which is often less visible. The violence involved in murder and assault is more likely to get attention in the form of newspaper reporting, rumour, gossip and social hysteria (Siegel, 1998). As anthropologist, Nancy Scheper-Hughes, notes, ‘focusing exclusively on the physical aspects [of violence] … misses the point and transforms the project into a clinical, literary, or artistic exercise, which runs the risk of degenerating into a theatre or pornography of violence’ (Scheper-Hughes and Bourgois, 2004, p. 1).

Scheper-Hughes invokes the concept of ‘everyday violence’ to describe those forms of violence which become part of the routine lived existence of citizens, and therefore attract less attention: ‘the everyday violence of infant mortality, slow starvation, disease, despair, and humiliation that destroys a socially marginalised human with even greater frequency are usually invisible’.1 This resonates with the conception of violence which Hannah Arendt (1963) developed in her work on the routine violence which resides in the organisation of bureaucracies, which she described as the ‘banality of evil’.

Similarly, John Galtung has drawn attention to what he calls ‘structural violence’. Galtung argues that violence is not simply a physical act, but should also be understood as a ‘process that can be embedded into wider social structures’, and that can therefore be ‘built into the structure of society, showing up as unequal power and consequently as unequal life chances’ (Galtung, 1969, p. 171).

Everyday, routine, or normalised violence does not often appear within the ambit of research on violence because it does not fall within categories of law that we consider ‘illegal’. ‘Violent crime’ connotes specific categories of violent behaviour, practices and intentions. It is worth bearing in mind however that it does not describe all forms of violent relations in South African society. This paper focuses on the effects of what we might call ‘manifest violence’, but it keeps in mind the ‘everyday’ structural violence which results from the fact that more than half of the South African population lives below the poverty line. Given that the official unemployment figure is above 30 per cent, and much higher in urban Black areas where most of the violence takes place, this has important implications (Bentley and Habib, 2008). One of which is that the lack of regular income, and the insecurity of employment, places poor South Africans in a high degree of vulnerability, which might be considered as a form of structural violence.

Violence, cohesion and governance

Discussing the implications of violence for South Africa's future requires taking into account trends in both forms of violence that permeate relations within the society and affect social cohesion. More recently, social cohesion has been made visible through the work of Robert Putnam (2000). Putnam has drawn the link between social cohesion, and the ability of communities to mobilise social capital, the forms of symbolic and material capital which fosters reciprocal relations of trust between horizontal and vertical groups in a society. Significant concerns have been expressed about the applicability of Putnam's model on a global scale, which are discussed later in the paper (Amin, 2005).

For the purposes of this paper, I am taking what I have described as relations of ‘governance’, to refer to the multiple networks through which social life is facilitated and regulated, informed loosely by the conception of ‘governmentality’ developed by Michel Foucault (Dean, 1999). Second, whilst these links work at vertical and horizontal levels, my discussion here focuses on the latter rather than the former, indicating the fragmenting effects of violence on horizontal cohesion. These do impact on the disarticulations between the formal spheres of governance, indicated by the politicised nature of the discussion on ‘crime’ in South Africa. Low levels of trust are engendered and therefore poor levels of social cohesion follow.

Governance is used here in its broadest sense to describe multiple practices which co-exist and overlap, and which describes practices and relations at both an individual level as well as at a group level. Specifically, there are four senses in which I am using governance in this paper: governance refers to the ways in which the self is managed (Foucault, 1990). This has implications for the multiple ways in which we manage personal conduct, in relation to morality, ethical conduct and the configuration of choice. These are of course choices configured within group sensibilities and are part of the values, norms and practices shaped by socialising agents and institutions.

Second, governance refers to the ways in which we manage the construction, maintenance and transformations of identity. Identities often prescribe or establish ways of being in the world, which entail normative modes of conduct and behaviour. The construction of gendered identities, for example, creates normative expectations of femininity and masculinity which are taken for granted. When these are challenged or threatened by encounters with difference, it produces multiple effects, some of which result in forms of violence. Threats to male honour for example, might occasion violent behaviour to re-assert a masculine sensibility. The ways in which we respond to threats are themselves socially conditioned, but are often inscribed into the body so as to become a reflex reaction, the ‘ways in which from society to society men know how to use their bodies’ (Mauss, 1973, p. 70).

More broadly, we may also refer to governance in the more conventional political sense. This refers to the institutional mechanisms through which we regulate the public aspects of social and community life. The distribution of political power occurs within the structures of governance we establish. In a democratic constitutional dispensation, such as South Africa, the governance of social life is therefore mapped by the division of spheres of authority and competencies horizontally and vertically, for example, local, provincial and national government. The Constitution governs the relations among citizens, and between citizens and the state—relations established by the rights which underpin it. This establishes the legal frameworks of legitimate and illegitimate practices. The government of political life further refers to responsibilities for taking care of different aspects of collective and individual well being—economic well being, education, public and private health, and safety and security, for example.

Fourth, governance also entails the regulatory networks of interdependence that characterise social and collective life which fall outside the direct responsibility of the state and within what we refer to, within a secular state like South Africa, as the private domain. This involves practices within the family, the church, mosque, temple or synagogue, the ‘community’, or the ‘nation’ in its cultural iterations.

Taken together, the character, depths, mediation and articulation among these different spheres of governance, along a continuum from the personal to the political, describe what is referred to as ‘social cohesion’. Whilst social cohesion is a normative aspiration in a fractured society like South Africa, its presence or absence is the effect rather than the cause of practices established by the ways in which governance is practiced in the multiple forms described above. In other words, social cohesion cannot be created by simply insisting on it as a normative disposition towards others within a family, community or nation. To approach it in this way does not take into account the historicity of the existing lived forms of governance which exist in these spheres of social life.

State responses to violent crime

The post-apartheid security services, and the South African Police Service in particular, has been acutely aware that in order to provide effective policing the perception by the public had to be changed (Marks and Fleming, 2004). From serving the interests of an authoritarian and racist regime which protected the welfare of a racial minority, the police had to become an institution which now served the majority of South Africans. It had to become the custodian of security rather than an agent of insecurity. The South African Police Service heralded the renaming of the force, as well as heralding the inauguration of a new ethic of policing that upheld human rights practices, was based on detection and accumulation of evidence and a state objective to serve ‘the community’ (Brogden, 1993; Rauch et al., 1994; Shaw, 2002).

The most comprehensive framework for transforming the role of the police to date was the National Crime Prevention Strategy (NCPS). The NCPS advocated a multi pronged approach which emphasised a range of interventions, which included community partnerships, public education and reshaping social values, and changes in environmental design. As Clifford Shearing and Mark Shaw noted, ‘What was remarkable about the document was that it challenged the conventional wisdom about policing being fundamentally police business. While it did not recognise the risk-focused nature of apartheid policing’, it emphasised that

security within post-apartheid South Africa will require the creation of a new set of apparatuses for governing security that re-established the police as one node in a network of resources … At the nub of the NCPS is the argument that the governance of security is effective when it involves more than simply police resources. It argues that post-apartheid policing strategy should be developed to create new systems for mobilising resources to reduce the risk of crime across the society (Shaw and Shearing, 1998, p. 3).

As Shearing and Shaw note, however, all elements of this plan were not implemented equally. In effect, the plan carried forward the policing aspects, which focused on the more conventional understanding of crime as something to be reduced by a more effective administration of detection, arrest and incarceration. The actual response to crime has therefore been to provide legal empowerment to the administration of justice, and to create mechanisms which improve the efficiency and scope of the criminal justice system through the employment of more police personnel and prosecutors. In effect therefore, ‘the net result of all of this has been that, despite the intention of the NCPS to shift policing reform away from the police, it has become little more than a strategy for police structural reform’ (Shaw and Shearing, 1998, p. 3).

The shift away from the comprehensive approach to policing has been driven by a number of factors, which are likely to continue to exert considerable influence on the shaping of policing policy. A key factor has been the need to respond to public outcry around crime levels, animated by both the actual statistical levels of crime, and growing perception of increased insecurity. This public outcry is mediated through newspapers and television, and initially was substantially driven by those who had a historical sense of entitlement to feeling safe,2 who felt empowered to articulate their views in newspapers in English, and who understood the legal and constitutional obligations of the state towards its citizens. In effect therefore, it has been White citizens who most vocally articulated views around ‘crime’ as a pressing socio-political concern. In addition, an important articulator of the public discourse around crime has been the business community, who have emphasised the need for a safe environment in order to be effective producers and retailers of goods, and to create an enabling environment for investment in long term investment. To that end, business has been proactive in establishing policing partnerships, private security, surveillance cameras, and so on, in urban areas and central business districts.3

The return to community

The observation by the Ministry of Safety and Security that more than 80 per cent of violent contact crime occurs between people who know each other, has once again brought to the fore an element of the NCPS that the police cannot solely prevent violent crime since they cannot be everywhere all the time, more so in social spaces considered the private domain of citizens. This realisation has led to the current emphasis on the need to inculcate and transform the social norms, values and practices of citizens to reduce violent and property related crime. The emphasis is therefore on creating a ‘socially cohesive’ community where levels of trust engender more socially harmonious relations, and where rights and obligations are practiced which are consistent with the laws which govern the country. The importance of social cohesion was highlighted through the development of specific programmes to be undertaken and monitored by the Cabinet of the South African government which noted in its cabinet lekgotla in July 2007 that such ‘programmes would be informed by the need to improve the quality of social relations by promoting positive values, including social solidarity and trust’ (GCIS, 2007, http://www.gcis.gov.za/media/releases/2007/070729.htm).

The mechanisms through which these could be achieved at the level of policing are through community policing forums and sector policing, for example. Key institutional agents of socialisation, like the family and the public education system, could also be important sites of intervention through which to transform relations amongst citizens in order to create a socially cohesive community.

The successful implementation of policies and programmes, based on this realisation would be an important goal in the years ahead. If South Africa is to address violent crime, and increased property crime, then security will have to be understood in broader terms than punishment, deterrence and incarceration. The question of which section of the population's security needs will most effectively be met is of vital importance. However, making the communities in which the majority of South Africans find themselves, safer, and building ‘social cohesion’ through state–civil society partnerships, will have to confront a number of challenge. In the following section, I take up three key challenges.

Crime and the postcolonial condition

The ways in which we conceptualise the relationship between rights and obligations, social capital, and social trust have to take into account the historical influences which shape the actual practices of these ideas. In the South African context, the African experience of citizenship, shaped by colonialism and apartheid is crucial. As noted, social cohesion has been expressed as a normative aspiration of the South African government. And social cohesion requires forms of interaction based on shared conceptions of the legitimacy of juridical authorised norms and values, commonly referred to as the ‘rule of law’. So in the postcolonial context, what is the force of law where cohesion might be hampered by the proliferation of divided loyalties to multiple publics?

Nigerian political scientist, Peter Ekeh (1973) argued in a seminal article that the conventional division between the private and the public sphere, between the state and civil society, cannot simply be transferred analytically from its Western genealogy to an African context. The African colonial experience, in particular the experience of indirect rule, Ekeh argued, creates multiple spaces of rights and obligations and two publics rather than one. Ekeh distinguished between a public realm, a private realm and a realm of ‘societal morality’ (1973, p. 91). Whilst the same individuals may overlap as members of these realms, each of these realms makes a different claim on ethics and morality, and rights and obligations. In particular, Ekeh distinguishes between a primordial public, a civic public and a private realm. One feels an obligation to the primordial public, but in exchange one gets solidarity and identity, not economic benefit. The civic realm on the other hand, which is the colonial public sphere, is an amoral space where economic benefit is derived: ‘duties … are de-emphasized, while rights are squeezed out of the civic public with the amorality of an artful dodger’ (1973, p. 107).

Ekeh was intent on historicising the idea of ‘corruption’ and ‘tribalism’ in Africa, by showing the impact colonialism had in creating different sensibilities about rights and obligations amongst different social groups and ‘communities’. Corruption, as a practice which is labelled as ‘criminal’, Ekeh argued, had to therefore be understood as a practice which could be justified because the law of the postcolonial state lacked legitimacy since it had historically been used to exclude and dominate Africans. In the South African context, a homologous referent could be the sensibility of ‘ukupanda’, of always trying one's luck. This resonates with the ‘artful dodger’ that Ekeh describes. As the South African literary theorist Sam Raditlhalo (2007, http://www.litnet.co.za/cgi-bin/giga.cgi?cmd=cause_dir_news_item&news_id=21262&cause_id=1270) observes,

young Blacks in the ghetto use the notion of hustling, ‘ukupanda’, or to ‘make life’ as their armour. For them this obsession with making it good involves, as for youth anywhere, a desire for recognition, legitimacy, status, wealth, and, ultimately, being allowed membership of mainstream society, and Johannesburg is as good a place as any for the hustlers … Beyond the platitudes of the ruling party, youth out there are going to make it by hook or crook, and yes, some of this makes for the higher crime rates that we experience. Having grown up in a tough neighbourhood that included the likes of ‘Fingers’ Rabotapi, I know not the number of funerals of hustlers I have attended over the years.4

Above all what the practice of ukuphanda points to is the divergence between an ethical conduct condoned by a domain of sociality which is not commensurate with the legal conduct authorised by judicial norms. To this extent, legal conduct that regulates the domain of ‘civil society’ applies effectively to a limited section of the society, as Marx noted in his usage of ‘civil society’. Criminal violence, I would argue, must be ‘thought’ in this liminal space, as neither within the domain of the state, nor the domain of ‘civil society’.5

Second, it is important to note that significant forms of social organisation, which expressed grass roots political demands were, demobilised with the inauguration of a democratic government in 1994 (Jack and Cherry, 2003). The expectation was and continues to be that the new state will deliver, and that communities would be the recipients of development rather than its proactive agents. This view was held by ruling African National Congress and it set about its tasks as a government with this approach in mind.6

Third, it is worth bearing in mind that the existence of the apartheid state, and the mobilisation against it, created multiple socially cohesive communities around the binary of domination and resistance. The rationality of domination encouraged the formation of paramilitary community safety forums, youth brigades and community networks mobilised around support for the army amongst White citizens. The rationality of resistance created communities mobilised not only against the infliction of harm through repression, detentions and torture, but also mobilised against the everyday violence of apartheid, the violence of racialised citizenship, restrictions on movement, poor education and health services. In other words, the socio-economic violence of apartheid as the bureaucratic administration of racist policies which systematically undervalued Black life, which Hannah Arendt (1963) referred to as the ‘banality of evil’. Bringing into focus the apartheid state as the source of benefits and security, and suffering and insecurity provided an animating energy around which ‘communities’ were brought into existence and maintained by ‘traditions’ of civic duty on the one hand, and community activism, community ‘justice’, and ‘struggle’ on the other. These sensibilities, their idioms and institutional forms could not have been erased with the demise of the apartheid state.

Fourth, the effects on these communities on the post-apartheid government's economic policies need to be critically considered. Whilst there have been considerable improvements in political stability, provision of social welfare grants, access to water and electricity, these have been uneven and their means of delivery have had mixed, sometimes unintended, consequences. South Africa is ranked by the UNDP as a ‘middle-income’ country. The macro-economic indicators suggest consistently modest growth of the economy, with a drop in inflation levels, increases in profits, trade and investment (Manual, 2005). For nearly a decade these aggregate macro-economic indicators also contained a less optimistic story of ‘jobless growth’. However, ‘between September 2004 and September 2005, more than 650,000 jobs were created’, according to the official statistical agency's Labour Force Survey (Statistics South Africa, 2005, p. ix). Whilst there is a translation of growth into jobs there is still some concern about the sustainability of these jobs, created in sectors which rely on a flexible labour market, and whether they provide sufficient rewards to lift people out of poverty.

The unemployment crisis presents a serious challenge: wide definitions hovers at around 40 per cent whilst ‘strict’ definitions place it in the low 30 per cent.7 The economy's most dramatic cleavage, the racialised gap between the rich and poor, and which has widening since the end of apartheid, is only marginally deracialising. The Gini-coefficient, which measures income inequality, puts South Africa in competition with Brazil as the two countries with the highest level of income inequality. Some 45 per cent of the population lives below the poverty line. ‘In spite of the pro-poor policies, South Africa still remains one of the highest in the world in terms of income inequality’ (World Bank Report, 2006, p. 13).8

Building social cohesion often relies on the assumption that communities possess a level of material and symbolic resources, what is called ‘social capital’. This new ethic, argues Ash Amin (2005, p. 6) ‘is epitomized by talk of revitalising social capital, community cohesion, civic responsibility, and public spaces’ However, as Amin notes,

why should we expect so much from community in such places? … The down and out—if this is not to caricature the social pathology of deprived areas—are least equipped and least motivated to play at such lofty community expectations, stripped as they are of the material, institutional, experiential and possibly also civic and public life. It simply does not make sense to base regeneration on community in these contexts.

Amin's point, which is worth taking up, is not that local areas do not have any capacity or agency. Rather it is, first, that it might be misplaced to give a normative moral character to the kinds of cohesion projected onto these communities as a ‘lack’, which needs to be revived in social relations. And second, he reminds us that the fate of a ‘community’ is not simply determined by communities themselves, but is also determined by the networks of spatial links, economically and politically, that communities are trapped into, or delinked from.

Fragmented cohesions: ‘community’ after apartheid

Areas where violent crime occur the most are also the areas most affected by everyday or structural violence. They are Black communities in mega-townships, mostly working class, with high levels of unemployment and poverty. They are also communities least articulating crime as the major concern affecting them, but they are increasingly mobilising around the expectations of improvements to their basic conditions of living—housing, land, and ‘service delivery’. By May 2007, according to a study by the Freedom of Expression Institute, ‘6000 protests were officially recorded during the 2004/05 financial year of which almost 1000 were banned … 15 protests are being held per day somewhere in South Africa—besides the number of unrecorded protests’ (Delaney, 2007, http://www.fxi.org.za/content/view/83/). The study went on to note, significantly for our consideration here,

A disturbing pattern emerges from the research: activists who oppose the government's macro-economic strategy and their communities' slide into deeper poverty and misery, are finding themselves isolated and targeted by local municipalities and the law enforcement machinery. In the process, they are denied their constitutional rights to freedom of expression and assembly. Police officers are often ignorant of the Gatherings Act or, more worryingly, abuse the Act to prevent people from protesting and marching in public (Delaney, 2007, http://www.fxi.org.za/content/view/83/).

Ekeh's argument outlined above suggests that the social status of the Law in postcolonial societies cannot be assumed to be normatively prescriptive of the social morality within which it is embedded. In fact, there is a disjuncture between the social morality of legitimate practice and the legal prescriptions which govern the ‘civic sphere’. The conceptualisation of a policing strategy in post-apartheid South Africa took this into account, and the formulation of the NCPS set out to establish the legitimacy of policing by extending the conception to include community partnerships, taking a broad view of ‘safety and security’ as not just the task of a police force. This however was not fully implemented.

‘Community’, could also become an identity which coheres around a notion of marginalisation and social exclusion, as victims of a lack of delivery, and poor local government representation and service. This cohesion, whilst seeming geographically sporadic, and often times spontaneous, is being established against a particular object of derision. If municipalities and local government agencies respond, as indicated above, with disproportionate force and disregard legal restrictions, then the police are less likely to establish legitimacy within communities in the future. The law will therefore be experienced as repressive if experienced as the sharp end of a stick, and likely be seen to be working against poorer communities rather than for them.

More than 50 per cent of Black South Africans are victims of violent crime in their residential areas, while more than 80 per cent of White South Africans are victims of similar crimes away from their residential areas. The growing call for action against criminals has seen an increase in police arrests. However, the increased arrest rate has not seen a commensurate increase in conviction rates. Rather, there has been a decline in conviction rates. The result is a decline in the perception of the efficacy of the South African Police as the ultimate provider of safety and security. These have lead to the growth of various community initiatives, which have taken multiple forms depending on the resources which communities can mobilise.

A decline in the general trust of citizens in the formal juridical processes, and the perceptions of the efficacy of the policing capacity, has encouraged the emergence of vigilante groups across the country, such as the People Against Gangsterism and Drugs (Pagad) in the Western Cape. Similarly in Natal organisations were formed which started out as community neighbourhood watches, some of which have taken on violent and racialised forms. In his ethnographic study of community policing within a mostly Indian neighbourhood in Chatsworth, Thomas Blom Hansen notes that ‘the patrols became involved in a range of intractable and unpredictable activities … arrests at times turned into shootings and savage beatings of young Africans’ (Hansen, 2006, p. 89). ‘Community’ in the case of Pagad and the neighbourhood watches in Chatsworth speak to the recovery of a religious and racially hegemonic social morality fraying at the seams as a result of substance abuse, gangsterism, and poverty (Pillay, 2002).

In less organised forms, there have also been sporadic violent outbursts of community action against criminals or those suspected of having committed crimes. One such story, witnessed by a reporter of the Cape Argus describes how a thief who stole vegetables was chased by a group of residents and stoned to death:

A mob of angry Gugulethu residents chased down a man who was believed to have robbed a fruit and vegetable stall, crushing him with a concrete slab. Police have since arrested two men linked to the vigilante murder, and say more arrests may follow. The man, who is believed to be one of a group of two or three men who targeted the stall on Saturday afternoon, was pronounced dead on the scene. He is believed to have stolen two bags of potatoes and a pumpkin (Hawker, 2007, p. 3).

Black South African townships, suffering the structural violence of poverty and unemployment have also cohered around that which comes from outside and threatens, or perceived to threaten, or impede the life chances of local citizens. Within this category of external threats are those who are seen to be taking prospective jobs, and income generating opportunities, or undercutting local business people by selling staple goods at a cheaper price within townships and informal settlements. The target around which ‘community’ coheres in this particular instance becomes foreign Africans, designated colloquially as Amakwerere, who have become victims of xenophobic violence. A stark case in point has been the Masiphumele informal settlement in Cape Town, where more than 30 Somali traders and their families have been killed and their businesses destroyed. Similar attacks occurred on Somalis in Johannesburg, Northern Limpopo and the Free State provinces as well. One victim, Hadith Haji Adam lamented that ‘Some people in the community like us, but others don't want the competition. I think some local shopowners are behind the violence against us—they organised the attacks on our businesses and now we have nothing’ (IRIN, 2006, http://www.irinnews.org/report.aspx?reportid=60776). These attacks are not confined to Somali's, and the number of migrants to South Africa from the North, including the DRC, Mozambique, Nigeria, Senegal and Zimbabwe, continues to surge.

The insecurity of life, in its biological, economic and social reproduction also further entrenches the widespread belief in occult practices of witchcraft. Where trust is eroded, threats and jealousy circulate invisibly. The use of sangomas to protect health, and reduce susceptibility to the bad intentions of others is a widespread and often unacknowledged practice which disrupts modernist biomedical and legal conceptions of care and social order. In a survey conducted in the Slovo Park informal settlement outside of Pretoria, it was found that almost 70 per cent of residents resorted to the use of muti in order to safeguard them from risk. This included herbal products such as Ubulawu, which assists in communication with ancestors and keeps attackers away from a house; Vimbela, which is said to make attackers burn alive; Isithunzi, which provides a shadow against evil; Ntelezi, which is sprinkled around the home and the body to protect against misfortune, and Iwasho, which is a mixture of soap and holy water used to ward off evil (Emmett and Buchart, 2000, p. 169).

In his study of the relationship between witchcraft and democracy in Soweto after apartheid, Adam Ashforth observed:

Life in a world with witches must be lived in the light of a presumption of malice: one must assume that anyone with the motive to harm has access to the means and that people will cause harm because they can … It is dangerous to assume that an instance of suffering might be accidental or a product of purely impersonal forces devoid of connection with human or spiritual agency … Thus it is wise, when living in a world of witches, to seek protection against suffering being so caused. And it is also wise, when misfortune does occur, to other reason than to afford protection against further mishap (2005, p. 69).9

Those who are seen to succeed, or are upwardly mobile can sometimes be marked as doing so at the expense of others in the community, or through the use of occult powers. Those perceived to be witches are also vulnerable to attack, as has been the case in the North West province in particular, but in other parts of the country as well. In a recent case at a Kwazulu-Natal school, pupils at Mahnlenga High School burnt to death two women whom they accused of being ‘umthakathis’ (witches), who were putting a curse on the school. An investigation brought to light that one of the women accused of being a witch had a granddaughter who was a grade 12 pupil at the school. According to a newspaper report, she was the only one passing, and the girls at the school began to blame her grandmother for their failures, alleging that she was an umthakathi placing a curse on them. They marched over to the grandmother's house, who happened to have a female visitor: ‘The two women in their 60s—Mangubane Msaba Zungu and Qibile Thabitha Thusi—were dragged from a home and taken to a sports field where they were doused with petrol and set alight. Zungu died at the scene and Thusi later died in hospital’ (Nzama and Andrew, 2007, p. 5).

Safety and the production of ‘communities’ of exclusion

With the perception of state inadequacy to provide security, and the perception of rising levels of insecurity there is the unravelling of ‘one of the foundational myths of modern societies: namely that the sovereign state is capable of providing security, law and order, and crime control’ (Garland, 1996, p. 446).

Noting that for most White South Africans, victim surveys suggest that most violent crime occurs away from the place of residence, it is the residential area that has become the zone which has to be fortified. Middle class residents are able to mobilise resources, information, technologies, and organisation. A website of one such organisation, the Community Safety and Information Network, which aims to ‘fight crime through intelligence’, notes that the state cannot be relied on to ensure effective criminal justice, given the low levels of convictions:

statistically, out of an estimated 2,115,946 serious & violent crimes committed in South Africa in 2004/5 there were only 62,717 convictions—a conviction rate of 2.96%. Only 1,100,292 (52%) were reported to and/or recorded by the South African Police Services (S.A.P.S.) This means that at least 2,053,230 criminals, equivalent to the population of a city like Cape Town or Durban, who perpetrated serious and violent crimes in 2005 alone are walking free on our streets, posing a massive threat of crime to the citizens of South Africa as well as all international visitors—even the President of Taiwan, Chen Sui Bien, was held up at gunpoint in South Africa when he visited our country while he was still a legislator! (http://www.csin.org.za)

As the blurb clearly insinuate, if the President of a foreign country is not immune from criminality in South Africa, how could ‘ordinary’ citizens expect to have their safety guaranteed by the state? The ability of wealthy communities to mobilise resources for their safety has given rise to a number of industries geared towards this. A company by the name of Cingulum Security for example, offers subscribers the preventive measure of becoming part of their network of surveillance cameras that monitor 900 households in four suburbs of Johannesburg. The cameras are linked to a central control room which can deploy a rapid reaction security force. Working on a strategy of ‘Tactical Intervention’, company director Geoff Schapiro extolled the virtues of his approach:

To move past response to early detection and prevention of crime, you need field intelligence, and that is what the cameras give us. Video surveillance has great advantages. Technology does not fall asleep on duty or take lunch and tea breaks. It's in action 24/7 (http://www.axis.com/files/success_stories/ss_gov_suburb_anticrime_30776_en_0802_lo.pdf).

The socio-spatial experience of violent crime suggest that violent crime accompanies property related crime in middle class and formerly White areas to a greater degree than in Black urban areas. Two trends are important here. First, the proliferation of ‘gated communities’. Typically these have involved setting up access control around older neighbourhoods, blocking off a street or blocks of streets, organised by residents within an area. Private security guards are employed to regulate the inward and outward movement of people and vehicles within a neighbourhood. There are more than 300 road closures in the Gauteng metropolitan area, most of which have not been approved by the local municipality (Landman, 2002).

A second model, the ‘security village’, is the format taken by newer property developments, based on town house and cluster home models, are often built as a form of gated community, with controlled access points, surrounded by high walls, and electrified razor wire. The safety offered by these developments adds to the retail value of the property, indicating their desirability in the middle to upper class housing market, particularly in Gauteng, Durban and Cape Town.

Second, the lack of confidence in state agencies providing adequate levels of safety has led to the massive growth of the private security industry in South Africa. The increased investment in residential fortification, and private security, in order to both ensure safety, and feel safer, may paradoxically, in the long run, create further social fragmentation and insecurity. In her study of spatial separation in Cape Town, Charlotte Lemansky observes that, based on the Latin American experience,

walls and gates have reinforced a vicious cycle of poverty and exclusion by concentrating the poorest social groups in spaces with minimal economic and political leverage … Furthermore, enclaves do not just respond to difference and fear, but actually deepen segregation and reinforce fear by excluding difference and limiting social mixing, thus increasing paranoia and mistrust between groups (2004, p. 107).

Fear has also tended to consolidate a sense of community within former White areas, which overlaps onto shared sensibilities of siege by ‘the terrorists’ during the apartheid period. With considerably more resources and capacity, active community policing forums have been set up in many former White areas, where short staffed police stations can be personelled with volunteer staff, body-corporates are organised, and there is a collective mobilisation against crime. In a survey of how residents are responding to crime in two residential areas, one predominantly White and middle class, and the other predominantly Black and working class, it was found that the middle class community could much more effectively mobilise around ensuring services which would secure their neighbourhood than the working class one (Emmett and Buchart, 2000, p. 173).

However, these positive formations of community are often mobilised against an ‘other’ that increasingly fosters separation. In the study mentioned above, it became clear that in the middle class White area the view was that ‘crime originated from outside the community’ (2000, p. 173). The community could therefore establish a level of ‘trust’ and social relations which bound them together in relation to an external threat which they could collectively keep a look out for and draw on, to develop practices and technologies in order to do this. Similarly, in her study of gated communities in Sao Paulo, Caldiera noted that the ‘withdrawal of the upper classes from public space into enclaves leads to the emergence of a discourse associating criminality and poverty, and generates stereotypical images of the poor as inherently “dangerous”’ (2000, p. 55).

This imaginary of the poor, as those to be placed outside the spatial zones of safety, means relinquishing the shared social space of citizenship and narrowing conceptions of rights and obligations. In her work on Brazil, Scheper-Hughes noted the mutually determining relationship between the fate of the urban poor and the fears of the middle classes: ‘Meanwhile, the affluent and the powerful, tucked away in gated communities and in homes protected by armed guards and mechanical surveillance reminiscent of medieval fortresses safely imagine themselves as endangered rather than as endangering populations’ (2004, p. 14). Under siege, the middle classes do not see the proportion of wealth at their disposal as directly of consequence to the level of fear they live with in a society with stark material inequalities.

Their residential spaces of safety are accompanied by fortified enclaves where work takes place, as well as consumption and leisure. These zones are connected by road and transport systems that favour the wealthy, creating an interlinked ‘fortified network’, which could eventually ‘disembed’ the city. The trend towards malls, high ways, and transport systems like the Gautrain project, which favours the middle classes, may be indications of this trend in South Africa.

Conclusion

Violence impacts on South African society at physical, psychological and structural levels. Rather than suggest a breakdown, I am arguing that we need to look at the kinds of cohesion this violence brings into existence. The forms of social cohesion it creates show signs of being at odds with the forms of social cohesion envisioned by the national government. Whilst there are positive community formations to manage risk, poor communities are also showing signs of cohering around marginalisation, social exclusion, xenophobia, and susceptibility to gender and sexual violence. Violence, as experienced by middle class communities increasingly pessimistic about the state's capacity to provide safety, shows signs of creating enclave communities with privatised security, which could lead increasingly to cohesion around fear of ‘the poor’.

If social cohesion continues in these fragmented and mutually exclusive spatial zones, social polarisation would continue to create racially separate ‘publics’, with different benefits, rights and obligations, and fragmented experiences of citizenship. Each dependent on the other, yet rendering a conception of the Other which threatens the identity of the self, where the self is considered racially, as an economic agent, nationally, and gendered. The terrain of the encounter between Self and the Other remains therefore a terrain saturated with everyday violence, which may be seen to require an act of violence to secure the self.

Additional information

Notes on contributors

Suren Pillay

∗Democracy and Governance, Human Sciences Research Council of South Africa & Political Studies, University of the Western Cape, South Africa. Email: [email protected]

Notes

Ibid., and also Scheper-Hughes (1992, chapter 20).

It is worth bearing in mind that in 1996 more than 74 per cent of police stations were located in formerly white areas.

An example of this is the Business Against Crime initiative (www.bac.co.za).

See also Peterson (2003).

Partha Chatterjee (2006) has noted this space as the domain of ‘political society’.

For a critical reflection on this by a senior member of the African National Congress and the South African Communist Party, cf. Sheehan (2002).

The ‘strict’ definition does not include people who are unemployed and are no longer seeking a job.

This trend is also noted in the study by Altman (2006) and Trieghaardt (2006).

On occult belief and neo-liberalism, see Comaroff and Comaroff (1999).