Punitive Welfare on the Margins of the State: Narratives of Punishment and (In)Justice in Masiphumelele

Abstract

While there is an established literature on the relationship between political economy and state punishment, there is less work on how punishment is constituted from below in contexts of inequality. This article analyses the discourse around incidents of lethal collective violence that occurred in 2015 in a former black township in South Africa. I use this as a lens for examining how punitive forms of popular justice interact with state punishment. Whether via the slow violence of structural inequality or the viscerally corporeal high rates of interpersonal violence, my interviewees were intimately acquainted with violence. Although they supported long-term imprisonment, none of them came across as stereotypical right-wing populists. Instead, they adopted complex positions, calling for a type of punitive welfarism, which combined harsh solutions to crime with explicit recognition of the importance of dealing with ‘root causes’. I argue that when the state is perceived to be failing to both impose punishment and provide welfare, violence becomes a technology of exchange, which simultaneously seeks both more punishment and more welfare. The result is an assemblage of exclusionary penal forms.

Keywords

The procession started at Khoza square, it went past the church, all the way around, back to the clinic. They dragged them like that, one person put tires on him, they told him to walk while he had a burning tire on his body – around his neck and shoulders – like a sling.…There were lots of people in the procession, lots of singing, toyi toying, shouting. You do feel like that – a sense of connection, as if you are part of a community when you are all standing being against the baddies. It hasn’t been like that since then, it hasn’t completely stopped though, every now and again it still happens

Thembeka (19 December 2018)

Introduction

In the early hours of 15 September 2015, Amani Pula, a 13-year-old boy, was brutally murdered in his own home, by three intruders who were allegedly high on tik (a highly addictive methamphetamine). That evening, at an emotion-laden mass meeting, held outside in Masiphumelele’s Khoza Square, and attended by at least 500 people, there was a ‘public denunciation’ (Garfinkel, 1956: 421) of drug dealers and a collective decision was made to search for Amani’s murderers. A delegation of young men, aged between 20 and 35, was mandated to carry out this task. They were specifically looking for someone who had been arrested earlier during the day as a suspect in a rape that took place just prior to Amani’s murder. This person was released, because the victim could not identify him, but a rumour was circulating that he had been seen with Amani’s murderer and was somehow complicit. He was found while sleeping in a car, dragged to the main square and beaten to death while a massive crowd looked on.

About 5 days after the first Khoza Square meeting, the Masiphumelele Youth Forum emerged as the ‘formation’¹ at the forefront of a mass mobilization against crime. Consisting of people between the ages of 20 and 25 it embarked on a ‘#DrugsMustFall’ campaign, aimed at publicly denouncing and acting against general criminality, drug use, and drug dealing. Its members forced skollies (young boys who were known drug users and who allegedly robbed in order to feed their drug habits) to reveal where they were obtaining their supplies; they conducted drug searches; and meted out violent punishments (ranging from purported expulsions to executions). At least 8 suspected drug dealers were killed, although some claimed that the true figure was 22.

Eventually, on 22 October, the police arrested and detained Lubabalo Vellem. Described in court as an ‘influential community leader’ (Dolley, 2015) and a chief instigator of the violence, he was charged with multiple offences, including the murder of a 32-year-old man in a ‘mob vigilante attack’ (Phaliso, 2015). Vellem’s arrest precipitated a series of mass protests resulting in 37 people being arrested for public violence. After a highly charged hearing, where more than 1000 protestors singing struggle songs and toyi toying,² disrupted the proceedings and demanded the release of their leader, Vellem was released on bail, subject to stringent conditions. He was prohibited from returning to Masi until the charges against him were finalized and he had to stay with his sister in Khayelitsha (more than 30 km away), where he was to report to the police station three times a day. After his release, and the withdrawal of the public violence charges against all but eight people, the protests died down. In June 2019, more than 3 years later, the state withdrew the murder charge against him, but the public violence, attempted murder, and arson cases were yet to be tried. Vellem continued to visit Masi regularly, although he no longer lived there.

Conceptual Puzzle and Main Arguments

As Foucault (cited in Arellano, 2012: 134) notes, violence is always ‘legitimated and constituted through narrative practices’ and ‘Truth’, with a capital T, is an elusive concept. Instead of seeking the ‘Truth’ about what happened during this period, I use the discourse on vigilantism as a window onto everyday practices of punishment and justice in a former black township in South Africa. My focus is on incidents of collective (as opposed to individual) vigilante violence.

In their classic 1974 article, Rosenbaum and Sederberg (1974: 559) argue that vigilantism is a ‘conservative phenomenon’ because vigilantes identify with the ‘established order’. However, they acknowledge that, in certain circumstances, dissident (right-wing) and revolutionary (left-wing) violence shade into vigilante (anti-crime) violence. It then becomes difficult to distinguish between vigilantes, counterrevolutionaries and revolutionaries (Huggins, 1991; Rosenbaum and Sederberg, 1974; Super, 2017).

My case study highlights the overlaps between vigilantism and popular justice. I use the latter term both broadly, to denote its location ‘on the boundary between state law and [“traditional”] indigenous law’ (Merry, 1993: 35) and also in a more specifically socialist (or revolutionary) sense. Both popular justice and vigilantism are ‘fleeting’ (Merry, 1988: 49) forms of local ordering that exist in the interstices between law and non-law; they are capable of ‘sliding in one direction or another’(Abrahams, 1998: 7), and as I have argued elsewhere (Super, 2016, 2017), there are, in a South African context, historical overlaps between them.

The standard argument about vigilantism in South Africa’s former black townships is that it is caused by the failure of the criminal justice system – due to an absence of police, poor prosecution and/or lenient court sentences.³ In a counterargument, Rush Smith (2019: 5) argues that it is not the failure of the system that causes vigilantism, but its ‘technical success’. According to him, the introduction of a procedural rights regime enables vigilantism because vigilantes attribute criminality to the modernizing, rights-based state and blame it for granting ‘criminals…too many [constitutional] rights’ (Rush Smith, 2019: 203).

Yet, as my case study demonstrates (see also Super, 2016, 2020), denial of rights for those accused of crime is desired only in relation to some and not all forms of illegality. The fact that Masi residents demanded the release of Lubabalo Vellem implied that in his case, they actually supported the presumption of innocence and right to bail. As was the case with Goldstein’s (2003) work in Bolivia, the state’s authority to arrest and punish was both claimed and rejected. Given that the law and punishment are ‘social institutions’, which combine conflicting objectives and ‘organize the relations of often antagonistic interest groups’ (Garland, 1990: 282), this ambivalence and contestation, over who should be punished and who should be protected, is not surprising (see also Hornberger, 2013; Orock, 2014).

I argue that the collective violence that occurred in Masi in 2015 was not merely a response to a failing criminal justice system (even though this was an important contributing factor) but was also rooted in the ideology of ‘revolutionary justice’. What might on the face of it have appeared as a simple (albeit) violent campaign against drugs had clear echoes of current, and past, popular justice initiatives. As was characteristic of other ‘community protests’ (Paret, 2015), which since 2004 have become a central facet of the South African political landscape (Kynoch, 2016; Makhulu, 2015; Paret, 2015), the Masi protestors drew on the technologies and discourse of the liberation struggle. The name of the ‘#Drugsmustfall campaign’ was a clear reference to the ‘#RhodesMustFall’ and ‘#FeesMustFall’ student protests, which were occurring at universities throughout South Africa at roughly the same time. Whereas ‘the Fallists’, as they were known, shut down universities, occupied buildings and organized mass marches to force the state and university managements to implement their demands for a free and decolonized higher education (Gillespie and Naidoo, 2019), the Masi ‘Fallists’ campaigned against drugs and drug dealers. Far from being conservative supporters of ‘establishment violence’ and/or challengers of the human rights enshrined in the South African constitution, the vigilantes in Masi did in fact have an alternative social vision, one which drew on radical, redistributive and traditionalist-conservative ideology.

My interlocutors viewed the provision of state welfare as a means to prevent crime and simultaneously viewed punishment as a baseline necessity for welfare. In this article, I use the term ‘welfare’ loosely to include state welfare programmes (such as social grants), state services (such as the provision of water, sanitation and social services) and also state redress programmes (such as land redistribution and the provision of housing to compensate for the historic injustices resulting from apartheid and colonialism). The form of security that Masi residents desired encompassed not only protection from crime but also security more broadly construed, and they coupled their demands for a harsher state with a demand for more welfare from this very same state (Caldeira and Holsten, 1999; Goldstein, 2003; Koch, 2019; Mnisi Weeks, 2017). I call this ‘punitive welfarism’. As an ‘ideology’, it lies at the intersection of popular justice and vigilantism and the technology of violence, specifically corporal punishment, is central to it.

Like Koch (2019: 160) who argues that popular support for more law and order on a disadvantaged council estate in England ‘cannot be divorced from how democracy is experienced in the first place’, I argue that populist punitivism in informal settlements in South Africa is more than a simple rejection of ‘processes of democratic state formation’ (Rush Smith, 2019: 5). It is not that residents in poor black townships reject the Rule of Law, but, to the contrary, they are demanding a ‘thick’ version of it – one which involves much more than just law (Krygier, 2015). Without a ‘broader set of institutions’ (Kelly, 2018: 18) which are specifically geared towards distributive, or socio-economic justice, the criminal justice system, and indeed the Rule of Law, will fail to attain legitimacy, particularly among the poor (Kelly, 2018: 171). Thus, the call by South Africa’s poor for the state to act punitively against ‘criminality’ is integrally tied to demands for more welfare, hence the term ‘punitive welfare’.

By calling on the state to act harshly against criminals, and simultaneously claiming their own right to decide who the criminals are and how to punish them, my case study is an example of how vigilantism and/or popular justice both limit and uphold law’s ‘transcendent being’ (Fitzpatrick, 1992: 10). The result is support for an assemblage of exclusionary penal forms, ranging from formal (state) punishments to informal and illegal ‘non-state’ punishments. Both penal forms are based on a ‘hostile solidarity’ (Carvalho and Chamberlen, 2018: 217) against ‘criminals’ (Carvalho and Chamberlen, 2018; Garland, 2005; Mead, 1918) and, as I argue, are rooted in projects of ‘othering’.

In what follows, I tease out the concept of ‘punitive welfarism’ and the relationship between punitive justice, violence and the demand for welfare. I start by arguing that support for retribution and distrust of the criminal justice system emanates from a deep sense of social injustice. I then discuss the central role that corporeal violence plays in informal punishment and redistribution and, finally, the moral ambivalence and contradictions inherent in the process of ‘othering’. This ambivalence, coupled with the structural precarity that I describe next, promotes instability. Thus, as I demonstrate in the last section, outbursts of collective vigilantism, like popular justice, are inevitably short-lived.

Methodology

This article is based on research I conducted between 2015 and 2019. Although it is a case study of a series of events that unfolded in one former black township at a specific point in time, it is substantially informed by previous research I have carried out in other townships in Cape Town. Thus, while the events that I describe played out in a specific locality, my findings are informed by, and relevant to, a broader South African context.

Masiphumelele (meaning ‘we will succeed’) is less than 2 km wide, covers approximately 0.45 km² and is one of the most densely populated former black townships in the Western Cape. Approximately 40,000 people live in the original (formal) township and 15,000–21,000 in the Wetlands informal settlement. The latter is a flood-prone area of uneven and damp land, which is owned by the City of Cape Town (Development Action Group, 1996; Freedom House, 2017; Masicorp, n.d.). Unlike Khayelitsha, a suburb situated some 35 km outside Cape Town and created by apartheid planners in terms of the Native Urban Areas Act, Masi was established after a black ‘squatter community’ won the right to land in a white group area in the late 1980s (Development Action Group, 1996).⁴ As such, unlike Khayelitsha, it is surrounded by middle-class, affluent and largely white neighbourhoods.

Like other former black townships and informal settlements, Masi is a vivid example of how, 25 years after the first democratic elections in 1994, poverty still has a ‘strong spatial (and racialized) dimension’ (World Bank, 2018: xxii–xxiii). 85% of residents in Masi live in shacks in the backyards of privately owned houses or in the adjacent ‘Wetlands’ informal settlement (Masicorp, n.d.). It is not uncommon for the backyard of a formal home to house 22 people in one or two roomed shacks, with everyone sharing an outside toilet, contributing to electricity and paying rental to the owner in the range of 600–800 ZAR per month (between US$41 and US$55). Given its proximity to affluence, Masi is a graphic reminder that South Africa is the most unequal country in the world, with obscene wealth and poverty rubbing up against each other (World Bank, 2018). In the Wetlands (as in other informal settlements) forced, often violent, removals of illegal shacks, at the behest of the City of Cape Town, are common.

My research design consisted predominantly of 60 selective, semi-structured, in-depth interviews supplemented with observations via site visits and attendance at court hearings. Most interviews were conducted with residents living in Masi, although I also interviewed seven people living in neighbouring ‘white’ suburbs. Interviews averaged an hour and were conducted in English, apart from with three people where I used the services of a research assistant to translate from IsiXhosa to English. Most of my interviewees were either unemployed, subsisting on meagre state grants,⁵ or employed precariously, in piecemeal, part-time work in the building or service industry. Many referred to themselves as unpaid volunteers in various community ‘upliftment projects’. As such, they are part of the ‘chronic poor’ deriving most of their income from government social grants (World Bank, 2018: 36).

I followed an inductive approach which I started by interviewing people who I met when I attended the bail hearing for Mr. Vellem in October 2015. Thereafter, I interviewed community leaders, ordinary residents, a defence lawyer and a victim of vigilantism. I sought to interview as many people as possible who had personal knowledge of the events that played out between September and November 2015, with my initial analysis determining where to go and what to look for in my next round of interviews (Massoud, 2015). My research assistant, a respected community organizer, assisted in arranging some of the interviews. He also arranged for me to share my research findings with about 40 residents in July 2019. As I proceeded with my interviews, it became clear to me that some of my interviewees had been more directly involved in the violence than others: some of them had been at the front of the crowd(s), others claimed to have been at the back and still others claimed to have not been involved at all. For reasons of confidentiality, I use pseudonyms for all my interviewees and have refrained from identifying my interlocutors through their activities as office bearers, or activists, in various local organizations and/or social movements.

I paid interviewees who lived in Masi a small stipend, in recognition of their time. Interviews were conducted inside shacks in the Wetlands, in backyard shacks in the formal settlement, at cafes, in my car, at places of employment and inside one of the classrooms at Ukhanyo Primary School. In all instances, I was guided by my interlocutors as to where they preferred to meet me. I recorded most interviews, and also took notes, which I read through and typed up on a daily basis. In certain instances, where the information being shared with me was of a particularly sensitive nature and/or my interviewee requested that I switch off the recorder, I only took notes. During the process of typing up my notes, I coded the responses by reading and rereading what I had written and listening to the recordings. I underlined words and phrases that consistently appeared and highlighted recurrent issues. Thus, I sought a continuous interplay between data collection and analysis. I identified representative quotes from the interviews as a means of highlighting my main findings and to describe a particular situation in the most graphic way possible (Beckett and Evans, 2015).

With the assistance of various interviewees, I went on multiple site visits to the Wetlands and to locations where instances of collective violence had occurred. Since informal settlements have either non-existent or sporadic and inadequate access to the services that ratepayers in more affluent areas pay for, conditions in the Wetlands are appalling. Approximately 3000 people live in each of its seven sections; there is only one waterpoint and the ratio of outside, often malfunctioning, toilets to residents is approximately 1 to 100 (Bulelani, 21 June 2018; field notes, 21 June 2018). The canals that run between the sections are often clogged with rubbish, including sewage; the area is prone to fires and flooding and massive mounds of garbage lie heaped throughout the narrow footpaths that traverse the area (field notes, 21 June 2018). During one site visit, I witnessed the aftermath of a City of Cape Town authorized shack demolition and a shack fire, which residents doused themselves by forming a human chain, passing buckets from one person to the next. Despite the fact that someone had called the fire department, a fire engine did not arrive. This is typical of the way in which residents in informal settlements have been forced (out of necessity) to assume responsibility for ‘self-governance’ (Steinberg, 2016: 11), including the investigation of crime and punishing of alleged criminals (Steinberg, 2016).

A Sense of Injustice

Although in the post-1994 era, South Africans are (theoretically at least) ‘equal before the law’, with punishment no longer permitted to be ‘cruel and unusual’ and/or to violate dignity, the social meanings attached to it are still very much underpinned by a palpable sense of racial discrimination. As was the case during apartheid poor, ‘non-whites’ are still disproportionately imprisoned and victimized by violent crime. Thus, what Forman (2017: 12) refers to as ‘racism’s enduring role’ in the United States is hyper-salient in South Africa. There are countless examples of police failing to keep crime victims in poor black townships informed about their cases, disappearing dockets or not responding at all (O’Regan and Pikoli, 2014). Despite the fact that almost 65,000 people (possibly more) live in Masi, the closest police station is in Oceanview (a former coloured township),⁶ some 5 km away.

The desire to know ‘what is the justice’s work’ (Nobuntu, 18 June 2018) is not just an empty rhetorical question but a deeply important issue which is intimately related to the empathy that is intrinsic to ‘doing justice’ (Dubber, 2006). It is also a painful reminder of the ‘periphractic marginalization’ (Razack, 2014: 68) of residents in former black townships and informal settlements. They are ‘in the city, but not of it, [and] are fenced off invisibly’ (Razack, 2014: 55), not only because of their poverty and lack of access to the corridors of power (both literally and metaphorically) but also because ‘black lives and deaths’ (Gillespie, 2015: 204) are largely invisible in the greater context of the city (Gillespie, 2015). Thus, not only is criminal punishment intertwined with racial oppression (Garland, 2005), but the victims of crime (particularly violent crime) are also disproportionately racialized.

There was an overwhelming perception among my interviewees that constitutional rights were used selectively, in favour of ‘criminals’ and against the poor (see also Gillespie, 2015; O’Regan and Pikoli, 2014; Rush Smith, 2019). This sense of injustice was compounded by the glaring inequality between ‘white’ middle-class suburbs and informal settlements. As one person put it, ‘ the only thing we want is for the police to be visible – like in town or Seapoint [a mostly affluent, formerly white area on the Atlantic Seaboard]’ (Joseph, 27 June 2017). My interviewees questioned why ‘rapists, murderers and drug-dealers’ were released ‘immediately’, but ‘our people’ (the 37 protestors who were arrested in 2015) were detained for a few days and had to obtain the services of an expensive lawyer to secure bail. Similarly, there was widespread agreement that Vellem’s bail conditions were unfair and too harsh. As Lunga put it, ‘You even see the drug-dealers in the next hour (but) our people they kept them for a couple of days’ (21 June 2018).

It is particularly traumatic for the families of crime victims when suspects are suddenly released back into the community, with no forewarning. This happened to Thembeka, a young woman who lived in a two-room backyard shack, with her husband and two children, sharing a toilet with 15 people. Thembeka described to me how, in 2010, her young son and sister witnessed her mother’s murder, by their mentally unstable neighbour. For 2 years after the suspect was arrested and charged with murder, she repeatedly sought information about the investigation. This entailed travelling back and forth to the Simonstown and Wynberg Courts to speak to criminal justice officials, each time having to pay for a taxi to travel 13 km to Simonstown and 25 to Wynberg. One day she saw the murderer back in the same building. This traumatized her so much that she moved away. As she said, ‘Even today I’m asking myself why did my mom die because it will be much better if I know the reason’ (19 December 2018).

These types of cases (and everyone I interviewed in Masi had a personal story to relate about crime) contribute to deeply held perceptions about the unfairness (and thus illegitimacy) of the administration of criminal justice in South Africa. It was not that my interlocutors denied the granting of rights to those accused of crimes but that rights were being granted to the ‘wrong people’ (Tokozane, 16 June 2018). Thus, Tokozane (16 June 2018) stated that he did not agree with ‘rights for bail, rights for this, and this and this’ because they are ‘not in favour of poor people’ and that this is why ‘we want to change the Constitution so it can favour the poor people, so it can deal with unnecessary rights given to wrong people’. Clearly, his comments were not only about criminality but about racially skewed land ownership patterns in South Africa. Tokozane was alluding to the political debate about changing the contested ‘property clause’ in the Constitution so as to provide for state expropriation of privately owned land without compensation. Thus, the narrative that ‘rights are given to the wrong people’ justified both popular struggles for socio-economic justice (particularly insofar as land and housing was concerned) and vigilantism. A different, yet related instance of the intertwining of popular justice and vigilantism, and another example of punitive welfarism, was evidenced by the Masi Youth Forum’s decision to destroy the shacks of drug dealers, but in those cases where a dealer lived in a brick-and-mortar house, they would ‘only’ be evicted and the house given to someone in need. Thus, in terms of a violent form of redistributive justice, the vigilantes forcefully evicted a female cannabis dealer and her family and gave her home to an ‘old lady’ (Bulelani, 18 June 2018).

In a type of inversion of the ‘less eligibility’ principle, most interviewees believed that prisoners enjoyed better state benefits than law-abiding citizens. According to Tokozane (18 June 2018), ‘to be in prison is like you are in the restaurant or hotel’, and Sipho was angry that ‘they destroy you-they destroy somebody’s life outside, and they go inside, they promote their lives’ (22 June 2018). He complained about the fact that there were only four social workers in Masi, with no one to counsel rape victims, whereas prisoners received counselling in prison (22 June 2018). Yet, despite this support for harsher conditions and longer prison sentences, my interviewees recognized that overcrowded townships and poor living conditions were key contributors to crime (Nomthetho, 9 July 2018; Sipho, 22 June 2018; Tokozane, 18 June 2018). Despite calling for harsh measures against ‘criminals’, they also supported a type of left realist approach, which explicitly recognized the importance of dealing with the ‘root causes’ (Forman, 2017: 29) of crime. These included the need for ‘job creation to fight poverty’ and recognition that ‘people end up doing funny things because they have nothing to do’ (Andile, 19 June 2018). For this reason, the Masi Youth Forum members embraced a type of coercive developmentalism. They would confront skollies, ask them what ‘their talents were’ and attempt to channel them into a youth group to ‘keep [them] off the street’ and apply their skills to honest activities (Joseph, 11 December 2018).

The demand for the punitive treatment of ‘criminals’ was always coupled with demands for houses, land and general services (including sanitation, water and electricity). This desire to see a stronger, more protective and more distributive state is central to punitive welfarism. My interlocutors were certainly not denying procedural rights to all criminally accused, and in those instances, when they did, it was always imbricated in the state’s perceived failure to respect the dignity of poor black crime victims. In the next section, I discuss how violence, a central technology of punitive welfarism, is intertwined with this struggle for dignity and justice.

Violent ‘Justice’

Traumatic experiences, such as witnessing or having a close family member or friend murdered, have a ‘rupturing effect’ (Gobodo-Madikizela, 2008), which violates a person’s sense of self, leaving the individual ‘with only anger and resentment…in place of the loved one’ (Gobodo-Madikizela, 2003: 96). Violence is also a ‘numbing mechanism’ to avoid confronting trauma (Gobodo-Madikizela, 1999: 86), and indeed, for my interviewees, the idea of ‘justice’ (ubulungiswa) often blurred with retribution and revenge. Their support for retribution and distrust of the criminal justice system emanated from a deep sense of social injustice and the ‘moral (in)appropriateness’ (Kelly, 2018: 162) of the state’s perceived leniency towards criminality added to this. Rather than claiming that the state lacked a right to blame (because of the unjust circumstances within which it did so (Kelly, 2018)), my interlocutors called on the state to blame better, and more harshly.

According to Joseph (18 June 2019), who acknowledged having participated in the search for Amani’s murderer, ‘justice is if you got the right sentence [20 years to life imprisonment without early release in the case of rape or murder] for the crime you did’. Although he referred to justice in ‘our tradition’ as meaning ‘when you get punishment (isohlwayo)’, his reference to the ‘right sentence’ was clearly a gesture towards formal law, which provides for minimum sentences of 5, 7, 10, 15, 20, 25 years and life, to be imposed for a range of offences.⁷ He equated ‘justice for Amani’ with beating the suspected perpetrator because since people ‘go to prison and come back out so it’s better to beat him up’. Thus, justice was better served by assault than by ‘revolving-door’ imprisonment.

Although my interviewees also referred to ostensibly non-violent punishments, such as the payment of compensation by a wrongdoer’s family, and in the case of a serious offence, expulsion from an area (Bulelani, 25 July 2019; see also Jacob, 18 June 2018; Mzu, 13 July 2018; Tokozane, 18 June 2018; Mnisi Weeks, 2017; Super, 2020), these were often secured through the threat (whether express or implicit) of the violence that would ensue in the case of non-compliance. Like the long-term prison sentences that most of my interviewees demanded, these expulsions served to purge local communities of ‘criminals’ and skollies. In the case of the latter, they were, like imprisonment, couched in a reformative discourse. According to Joseph (11 December 2018) due to a lack of state support, the only solution was for ‘us black people…to take the [“problem”] kid out of Cape Town to the Eastern Cape’ (see also Super, 2020). Nomthetho (9 July 2018) also thought that sending someone ‘back’ to the Eastern Cape, where the dagga (cannabis) was ‘not strong’, was a good idea because, being a rural area, there would be less bad influences and negative peer pressure. Similarly, Tokozane (16 June 2018) justified what was essentially an expulsion as a person being taken ‘home’ to promote an ‘improvement’ in behaviour. Thus, the act of banishing was justified as a way to avoid prison and to ‘rehabilitate’ in a context of great scarcity and lack of access to state run social services. However, as Joseph (11 December 2018) acknowledged, this did not solve the problem of drug addiction and was experienced as a punishment by those on the receiving end.

Thus, whether by threat or otherwise, the meting out of violence onto the body of the ‘criminal’ is critical to local perceptions of justice. As I discuss below, it also functions as a technology of redress which, unlike the official state and/or private insurance, is always available for the poor. It is also used as a mechanism of instruction and is central to punitive welfarism.

It’s Okay to Beat but Not to Kill

Corporal punishment, whether official (in the form of judicial and penal punishment) or unofficial (in the form of the random and arbitrary violence imposed by white settlers on native subjects), was a central pillar of colonial (and apartheid) rule (Alexander and Kynoch, 2011; Hansen and Stepputat, 2005; Mbembe, 1992). It was an instrument of racialized and paternalistic control (Alexander and Kynoch, 2011; Glaser, 2018; Mbembe, 1992; Ocobock, 2012; Van Onselen, 1985) and was an integral mechanism of the ‘civilizing mission’ in terms of an ethos which framed Africans as ‘child-like’ and only able to comprehend the language of bodily violence (Anderson in Alexander and Kynoch, 2011: 400; Super, 2013). By the early 1990s, just before the practice was ruled unconstitutional,⁸ the South African state was carrying out more than 32,000 juvenile whippings per year (Skelton, 2005). These were primarily inflicted on the bodies of young black males, as a means to divert them from prison. Yet, the practice itself was far more widespread than revealed by official statistics. It was not only a central technology of discipline and identity formation among the subjects of apartheid rule (Crais, 1998) but liberation movements in exile also relied heavily on flogging as a disciplinary technique. Likewise, during the 1980s ungovernability campaigns in former black townships, beatings were inflicted by some people’s courts as a mode of social control for those deemed counter revolutionary (Buur, 2003; Crais, 1998; Kynoch and Alexander, 2011; Super, 2017).

Whereas residents in middle class, formerly white neighbourhoods, rely on insurance policies when their material possessions are stolen, and contract out the use of violence against suspected criminals to private security companies (Cooper-Knock, 2016; Diphoorn, 2016), in marginalized areas, it is through violence that people secure material and symbolic compensation (Buur, 2003; Mnisi Weeks, 2017; Super, 2016).

All of my interviewees regarded ‘beating a little’ in pursuit of information about stolen property and/or to obtain the truth about some other crime as acceptable. As Buur (2003: 34) notes in his study of New Brighton township in Port Elizabeth, ‘there is always someone who has seen something’. According to Nobunto (and indeed many others), ‘we have no choice, it’s the only way to get our TVs back’ (18 June 2018). Sometimes, after stolen property is retrieved, the person is handed over to the police, ‘so that the law can take its course’(Bulelani, 11 December 2018; see also Super, 2016). When this occurs, the police usually look the other way, often arresting a suspect for theft after the person has been severely beaten.

Public corporal punishment also plays a central role in ‘community’ censure and in assuaging collective trauma, even if the person being punished did not directly harm those meting out the punishment. As Joseph explained:

Sometimes when a thief is being caught, if you lose something – even if he did not steal it, you beat him because you blame him, even those other people who lost their stuff…when people are passing you always follow the mob because, in our community you want to be a witness. You want to see what’s going to happen with that guy. You want to see him being punished. (4 July 2018)

Despite approving of corporal punishment, most of my interviewees, including those like Tokozane (16 August 2018), who supported the reintroduction of the ‘hangpaal’ (death penalty), drew a line between killing and beating. Whereas beating was justifiable if it was carefully calibrated so as to discipline and punish, killing was an unacceptable, although not entirely unanticipated, outcome of collective punishment. Even Mr. Mfeketo, who was still traumatized by his nephew’s death at the hands of vigilantes in 2015, supported corporal punishment to obtain the truth and/or to punish. As he stated:

I believe in corporal punishment, because I’m growing up that time. My mother and my father, when I’m do something wrong, they hit me. But not to kill. It’s a good thing to hit, it was helpful for me but don’t kill. You see? (11 December 2018)

According to him, the crucial thing was not to lose control and not to ‘hit on the head’. It was however acceptable to ‘hit on the body, just like a children, when your children is doing something’ (11 December 2018).⁹

Similarly, Mandla distinguished between killing and beating:

Even if South African law doesn’t agree with us, sometimes it (beating) is a winning formula. Us people coming from the rural areas we know that there’s not much crime there because they can punish you physically there but they can’t kill you. They can punish you physically and then you are back on track. Criminal behaviour can be changed by beating. Us as Africans have experienced that in many ways – because people who were thieves have changed because they don’t want to be punished any more but they are still alive. If you kill a person you don’t give them a second chance to change. (26 June 2018)

Max, like Mandla, also drew selectively on ‘cultural markers’ (Beinart, 1992) in his justification of corporal punishment. As he put it:

A black person is not like a white child, a white child you can sit in a chair and say as parents: No John, this what you are doing is not right, can you stop this what you are doing. He/she will understand…but a black child before he stop that he must feel the pain. That’s why we believe when you catch a skollie you must beat the skollie to get him to stop or to show us where he sell the stuff (18 June 2018).

In differentiating between ‘black’ and ‘white’ children, my interlocutor was making a clear distinction between poor black township residents, who had nothing, and the white, middle classes with their ready access to privatized social services, which they could draw on instead of corporal punishment.

Violence (specifically beating) is thus a crucial technology of exchange, a way to teach someone a lesson and/or a mechanism to retrieve stolen property, without having to pay. It is a stark reminder of the inequality of lived experiences between the privileged civility of whiteness and being black and poor in South Africa. On the other hand, as I discuss next, there was considerable moral ambivalence and skirting around the issue of lethal collective violence.

Moral Ambivalence and Construction of the ‘Other’

Punishment and Vengeance

Despite justifying corporal punishment as playing an educational role and being ‘central to the production of order’ (Buur, 2003: 35), my interviewees acknowledged that what started out as being a ‘simple beating’ or a banishment had the potential to collapse into uncontrollable lethal violence (see also Super, 2016, 2017, 2020). This was attributed to leaders losing control over the crowd.

When I first interviewed Joseph, he told me that when the ‘community [is] at large’, you can’t stand in front and say ‘please don’t do this’ (24 October 2016). However, he subsequently claimed that when community censure resulted in death, this was a form of justice because ‘if you rape and kill, you supposed to die…to feel pain’ (18 June 2019). Similarly, Bulelani initially told me that ‘more than 15 people were killed by us between September 2015–2016’ because

it is easy for people to act without control: when they catch him someone will stab him behind his back, someone will put a light and someone will do something else, without any negotiation. Even if he says no – people will take a stick and beat him, someone will take a knife and stab him, a tire, someone will pour petrol and burn it (18/06/18).

However, he also subsequently stated that someone who raped or killed deserved to be beaten to death, ‘to be brought to book’ (25 July 2019), whereas someone who stole a cell phone should only be beaten, but not to death. As he put it, ‘if you rape or kill they’ll beat you to death [and]when you rape they cut out your testicles – you won’t survive’ (25 July 2019). He acknowledged that this kind of punishment was rooted in anger and a loss of control.

The majority of my interviewees, however, claimed not to support lethal collective violence at all. Despite this, they still thought that Masi was ‘safe’ during ‘this period of Asiphelelange’ (Nobuntu, 18 June 2018). Roughly translated from IsiXhosa, Asiphelelange means ‘we are not enough’ (because ‘one of us is missing’) and is the title of a song sung by protestors during the 1980s when the apartheid state exercised its draconian powers of detention without trial over township activists. In 2015, in its reinvented form, Asiphelelange was both a song sung by protestors to lament Vellem’s detention and the informal name for the anti-drugs campaign.

Everyone nostalgically referred to a time when they were able to walk about with their cell phones at night and in the early mornings without the fear of being robbed by skollies (see, e.g. Ncediswa, 29 June 2018; Nobuntu, 18 June 2018; Nomaindia, 14 December 2018; Lunga, 21 June 2018; Thembeka, 17 July 2018). This highlights the ambivalent relationship that my interlocutors had with violence. It is possible that they did not want to admit – to me – that they had participated in the violence, instead they blamed ‘the community’ for it. Another explanation is that even if they did not overtly participate in and/or support lethal vigilantism, they nevertheless enjoyed being able to walk around without fear of being robbed at knifepoint. Even those bystanders, who either did not participate in the violence or drew a line between beating and killing regarded ridding the community of a threat as a ‘socially acceptable’ (Kelly, 2018: 168) result.

Clearly, not all crimes posed the same risk: those who killed as ‘self-help’ (Black, 1983) were regarded as engaging in ‘moral’ forms of crime as ‘social control’ (Black, 1983: 352). As Innes notes in relation to signal crimes, ‘some murders matter(ed) more than others in shaping collective risk perceptions’ (Innes, 2004: 352). This observation is particularly relevant in instances of ‘generalized vulnerability’ (Shotland and Goodstein cited in Johnston, 1996: 234). Thus, although Vellem was charged with murder, he was not perceived as morally culpable – partly because all but one of my interlocutors in Masi questioned why he was singled out for arrest when the ‘whole community’ had participated in the violence and also because he was regarded as a hero who was trying to make Masi safe. Thus, when he was arrested, it seemed like the ultimate act of discrimination or symbol of a state that did not care.

According to Gobodo-Madikizela (1999: 89), ‘deindividuation’ occurs when ordinarily rational individuals ‘become carried away by the intensity of a violent crowd or group’ and lose their capacity (or desire) for ‘restraint’. Thus, what starts out as an assault can, when carried out in front of a crowd, collapse into lethal violence. This is particularly the case where the subject of the violence fulfils some stereotyped image of the ‘other’ and is, as such, dehumanized (Gobodo-Madikizela, 1999). Although the ideal typical identity of the ‘enemy’ has shifted – from the impimpi of the 1980s to the ‘criminal’ of the ‘post-apartheid’ era, dehumanization is not a new method of community building. Nor are the punitive technologies that accompany it. Clearly, the rape of Amani acted as a catalyst that precipitated an outpouring of anger, at a range of social problems, onto the bodies of suspected offenders (Gillespie, 2014; Godoy, 2006; Goldstein, 2003). This brings to mind Mary Douglas’s notion of ‘abjection’ in terms of which a human being is perceived (at least for that moment) as ‘deserving the pain inflicted on them’ (Thomas, 2012: 16). Even if the person being beaten is not the actual suspect but has been seen with the suspect, then she or he will be assaulted, partly to obtain information but also because of the mere association. As Nomaindia explained, ‘you can’t be a friend with someone who’s a tsotsi and you aren’t also a tsotsi’ (14 December 2018). Thus, in a Nietzchean¹⁰ type of scenario, vengeance and punishment collapse into each other and the criminal becomes the archetypal ‘oppressor’, the ‘common enemy’ or ‘traitor’ against whom ‘a formidable right to punish is established’ (Foucault, 1995: 90) and who must be expelled from the community.

Yet, and this is a crucial part of my argument, the discourse and practices around imprisonment are also, like the vengeance referred to above, based on the removal of the offender from their community (Dubber, 2006). Even when justified by the ideology of rehabilitation, the offender’s character is, to a greater or lesser extent, presented as ‘deficient’ and in need of correction. In this sense, imprisonment – whether for the purposes of rehabilitation or incapacitation – is linked to the vengeance from which it distances itself (Dubber, 2006: 122). The violence that is meted out on the body of the offender by vigilantes refracts and amplifies the ‘othering’ that underpins long-term imprisonment: in both instances, there is a lack of identification with the offender’s ‘personhood’ (Dubber, 2006). The degree might differ, but the essence is the same. In both cases, the goal is to remove the offender from society, for a stipulated period of time (as in the case of court-imposed imprisonment); indefinitely, in the case of unofficial banishments and/or permanently in the case of lethal collective violence.

The Fracturing of Self-Help

Although my interviewees wanted criminals, particularly drug dealers, to be expelled from the community, the ‘#DrugsMustFall’ campaign was inconsistently implemented. Firstly, it was not entirely clear what constituted a dangerous ‘drug’, and secondly, rumours that questioned the honesty of the moral crusaders themselves started to circulate. This fractured the narrative of moral ‘self-help’ (Black, 1983) against a clear enemy, revealing contradictions in the discourses of punitive welfare and popular justice.

While there was unanimity about the dangers of tik, the position on dagga (cannabis) was less clear-cut. At the time of the campaign, possession and dealing in cannabis was criminalized in terms of the 1965 Medicine and Related Substances Control Act. However, on 18 September 2018, the Constitutional Court decriminalized the possession and cultivation of a limited amount of cannabis for personal private consumption (Minister of Justice and Constitutional Development and Others v. Prince). According to Sipho, selling cannabis was an ‘in between offence’ because ‘dagga smokers aren’t violent’ and traditional healers used cannabis for spiritual purposes (21 June 2018). However, he supported the decision to expel dealers who were selling cannabis because ‘when you clean up, you must clean up everything’. Similarly, Joseph told me in July 2018 that ‘after getting criminals kicked out people went for drug dealers’ and that ‘dagga was also a drug so they were also kicked out’ (4 June 2018). When I interviewed him again in December, after the Constitutional Court ruling, he said that he ‘never had a problem with dagga even before’ and that he only had a problem with chemicals like ‘mandrax and tik’. He claimed that because cannabis was a facade for selling ‘harder’ drugs, the #DrugsMustFall campaign was aimed at expelling all dealers (11 December 2018). Thus, his justification had shifted: expulsion of cannabis dealers was no longer because cannabis was a drug, just like all other drugs, but because it was a ‘facade’. This brings to mind Fitzpatrick’s (1992: 180) idea of ‘mythic mutuality’: although ‘popular justice takes its identity in opposition to formal law, it is based on a core of formal, juridical power’. In 2015, all drugs had to fall because they were all illegal, but in 2018, when cannabis was decriminalized, the rationale had subtly shifted.

In addition to this vacillation on what constituted a drug, the expulsions were inconsistently implemented. Sometimes the discussions about which house or shack to search were leaked in advance of the search. As Nomthetho explained:

If my brother is a drug dealer and I knew the community is going to search his house (we have a meeting and decide) I will call my brother to tell him ‘ok tomorrow we gonna search’, because what happens they go quietly straight to your house, you can’t hear nothing, only when they are knocking on the door. So, if I heard that I would warn my brother and I think other people they are also doing that.…So someone was warning people around the corner, hiding people and telling them tomorrow we gonna do this and this. (9 July 2018)

According to Bulelani, one of the key influencers in the decision to ‘go against people selling weed’ was herself found to be selling it (18 June 2018). He blamed the return of the dealers on a ‘lack of unity’ and fear of arrest (see also Joseph, 10 December 2018; Max, 18 June 2018; Thembinkosi, 22 June 2018). Tokozane (6 July 2018) told me that one of the leaders of the campaign was himself either dealing in drugs and/or protecting his drug-dealing friends and the target of this rumour claimed, in turn, that ‘people that were with you in this struggle – they are the ones selling drugs. Now they make us look dirty because we were the ones cleaning’ (11 December 2018). Others referred to allegations that drug dealers were asking certain vigilantes to sell drugs for them (Sizwe, 22 June 2018).

Thus, the hegemony of those at the forefront of the drug purge eventually fractured. The fact that Vellem’s bail conditions resulted in him being ‘banished’ to Khayelitsha played a role in this; there were also internal leadership struggles within the campaign itself and the narrative of a united ‘Masi community’ began to falter. Newspaper articles referred to ‘angry residents’ blaming ‘individuals acting on their own for the violence, saying it was not the work of vigilantes’ because no community meetings had been held to plan ‘what action to take’ (Nkalane, 2015). Despite the ‘prison is a five-star hotel’ (Thembinkosi, 22 June 2018) narrative, when the prospect of imprisonment was immanent, it did not seem quite as welcoming an institution.

As Hansen and Stepputat (2005: 2) note, ‘sovereign power’ is always an unstable and tentative project. In situations of socio-economic inequality and historic marginality, this is even more the case. Thus, the space for violent expulsions closed down and some of the dealers returned. When I was conducting interviews in 2018, Nobuntu told me that ‘Drugs are on. The Nigerians are doing everything they want to do. Those who were gone at the time of Vellem and Asiphelelange they are back now’ (18 June 2018). According to Tokozane, ‘those that were chased out are back again in full force and heavily armed. It’s a matter of pressing the button then they will react’ (16 June 2018). Sipho explained that although ‘the DrugsMustFall campaign took us somewhere the campaign is finished now, the dealers have come back – more than before’ (22 June 2018).

Conclusion

This article has asked how punishment and justice are viewed by residents living in South Africa’s informal settlements. I have argued that popular punitivism in these spaces cannot be considered apart from the historical context of state-imposed violence (both physical and structural), precarity, and material deprivation. Here, where the sense of failure to realize the expected fruits of democratization is acutely felt, inequality has ‘drastically narrowed the options’ (Forman, 2017: 29) for both victims and perpetrators of crime. It is therefore unsurprising that there is overwhelming support for long-term imprisonment and a ‘hostile solidarity’ against ‘wrongdoers’.

As residents in South Africa’s informal settlements seek instant “solutions” to complex social problems the physical exclusion of those labelled “criminal” is both a means to “efface” pain, in a context where there is a severe lack of social (and other “psy”) services; a “distancing device” (Gobodo-Madikizela, 2003) rooted in anger at the invisibility of poor black crime victims; and a protective mechanism deployed as a means to deter further crimes. In this context, Amani Pula’s murder became a ‘prism for long standing complaints’ (Seeds, 2018: 193) about police absence, complex social problems (including high rates of violent crimes and drug abuse) and overcrowded informal settlements.

The violent responses and widespread protests triggered by Amani’s death were a way to propel Masi’s problems into ‘wider visibility’ (Gillespie, 2015: 204) to force the state to take the problems of poor black township dwellers seriously. They were also a cleansing technique in terms of which expulsion served the ostensible purpose of repairing a severely fractured social fabric. Yet, despite the toyi toying, struggle songs and protests (which clearly provided an occasion for people to come together against a common ‘enemy’), the sense of ‘community’ was fragile and fractured by leadership struggles and micro-dynamics between participants. Thus, to quote from Mead (1918: 559): the ‘solidarity’ that emerged out of the creation of a common enemy in the shape of the criminal was only a ‘temporary relief from social frictions’. It soon collapsed.

In what I have referred to as ‘punitive welfarism’, I found a combination of a left-wing approach to crime (blaming crime on poverty and inequality) and a simultaneous desire to expel ‘criminals’ from society. This support for an assemblage of welfare-based strategies and punitive individualizing ones differs from the standard punitive narratives, which ignore poverty and structural circumstances. Indeed, the security that residents in South Africa’s informal settlements desire is far broader than ‘only’ security from crime (Caldeira and Holston, 1999; Mnisi Weeks, 2017). My interviewees were not just calling for punitive action against ‘criminals’ but also for ‘root cause solutions’ (Forman, 2017: 29) such as more welfare, jobs, activities for youth, land, houses and sanitation. In other words, like many other poor black South Africans, they desired the ability to lead dignified lives, a promise that has not been realized in ‘post-apartheid’ South Africa. As such, the sense of injustice and indignity that was vocalized by everyone I spoke to was deeply entwined with and rooted in the historical legacy of racism and inequality, which is still a dominant fault line in South Africa. Thus, in South Africa’s former black townships, vigilante violence is not only linked to conservative (or ‘traditional’) values which promote violence as an integral part of social control but also arises from an alternative vision for a better society with less inequality (i.e. ‘revolutionary’ or ‘popular’ justice). I have referred to this assemblage as punitive welfarism.

Although, on the one hand, it could be argued that vigilantes view human rights as promoting crime (Rush Smith, 2019), I found that in fact, my interviewees had an ambiguous relationship with human rights and, indeed, with official law. They were deeply ambivalent about the ‘post-apartheid’ state, calling on it to provide security and to mete out justice (in the broadest sense of both terms) while simultaneously claiming the right to impose violent and punitive punishments. The latter both drew on legal discourse and also challenged the state’s authority and its’ claim to monopolize legitimate violence. As I have argued, both official (state) and unofficial (popular) punishment are grounded in a mutually constitutive continuum of ‘othering’, which is targeted at ‘criminals’. Instead of being based on recognition of the humanity of both victim and offender (Dubber, 2006; Forman, 2017), justice is conflated with vengeance (whether soft or hard), in terms of which the aim is to remove certain criminals from society, via long-term imprisonment, banishment, or lethal collective violence. Yet, the perception among my interviewees was that South African prisons (and indeed, the rest of the criminal justice system) had failed – both to incarcerate and/or to rehabilitate. Thus, it was not so much that the law was successful but that it was failing dismally, even on its own terms.

Although this case study is specific to a particular place and time, the underlying tensions and overlaps between ‘justice’, punishment and vengeance are broadly applicable to other contexts (not only in South Africa). Indeed, the conflation of ‘justice’ for crime victims with punishment, and the elective affinity between punishment and vengeance, which are central tenets of populist punitivism are also, albeit not in quite the same way, reflected in punitive welfarism. These are clearly not only South African problems.

It is precisely by looking at seemingly extreme cases that we gain a sense of the underside of liberal penality. It is here, in spaces where ‘law’s universal pretensions’ (Fitzpatrick, 1992: 179) fragment, that we gain insight into the failure of the narrative myth of liberal punishment as representing a shift from vengeance to rational, non-violent and more humane forms of ‘justice’ (Foucault, 1995; Garland, 1990).

Footnotes

Acknowledgements

I would like to thank Dr. Marie Fox and the anonymous reviewers for their insightful comments. Thanks also to my colleagues from the Socio-legal Writing Group at the University of Toronto: Ellen Berrey, Catherine Evans, Phil Goodman, Bea Jaregui, Filiz Kahraman, Anna Korteweg, Sida Liu, Max Mishler, Ayobami Laniyonu and Chris Smith. Further thanks to Mark Hunter and Ashley Rubin for reading the earlier drafts. This research would not have been possible without the incredible generosity and kindness demonstrated towards me by the people who I interviewed in Masiphumelele and beyond. All remaining errors are mine.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Connaught Programme at the University of Toronto and a Government of Canada Social Sciences and Humanities Research Council Insight Development Grant.

ORCID iD

Gail Super

Notes

References

Minister of Justice and Constitutional Development and Others v Prince [2018] ZACC 30.

S v T Makwanyane and M Mchunu Case No. CCT/3/94 [1995] (6 June 1995).

S v Williams CCT20/94 [1995] (9 June 1995).

Abrahams

(1998) Vigilant Citizens: Vigilantism and the State. Cambridge: Polity Press.

Alexander

Kynoch

(2011) Introduction: Histories and legacies of punishment in Southern Africa. Journal of Southern African Studies 37(3): 395–413.

Arellano

(2012) Vigilantes and Lynch Mobs, Narratives of Community and Nation. Philadelphia: Temple University Press.

Ballard

(2014) The discourse and policies around sentencing and punishment in South Africa, 1992-2014. Available at: http://www.nylslawreview.com/wpcontent/uploads/sites/16/2014/11/Ballard.pdf (accessed on 21 May 2018).

Beckett

Evans

(2015) Crimmigration at the local level: Criminal case processing in the shadow of deportation. Law and Society Review 49: 241–277.

Beinart

(1992) Introduction: Political and collective violence in Southern African historiography. Journal of Southern African Studies 18(3): 455–486.

10.

Black

(1983) Crime as social control. American Sociological Review 48(1): 34–45.

11.

Buur

(2003) Crime and punishment on the margins of the post-apartheid state. Anthropology and Humanism 28(1): 23–42.

12.

Buur

Jensen

(2004) Introduction: Vigilantism and the policing of everyday life in South Africa. African Studies 63(2): 139–152.

13.

Caldeira

Holston

(1999) Democracy and violence in Brazil. Society for Comparative Study of Society and History 41: 691–729.

14.

Carvalho

Chamberlen

(2018) Why punishment pleases: Punitive feelings in a world of hostile solidarity. Punishment and Society 20(2): 217–234.

15.

Cooper-Knock

(2016) Behind closed gates: Everyday policing in Durban, South Africa. Africa 86(1): 98–121.

16.

Crais

(1998) Of men, magic, and the law: Popular justice and the political imagination in South Africa. Journal of Social History 32: 49–72.

17.

Development Action Group (1996) Masiphumelele: A case study of the role of the development action group in the informal community of Noordhoek. unpublished.

18.

Diphoorn

(2016) Twilight policing: Private security practices in South Africa. British Journal of Criminology 56(2): 313–331.

19.

Dolley

(2015) ‘Marchers Sow Fear in South Peninsula’. Weekend Argus. Available at: https://www.iol.co.za/weekend-argus/marchers-sow-fear-in-south-peninsula-1938564.

20.

Dubber

(2006) The Sense of Justice: Empathy in Law and Punishment. New York: New York University Press.

21.

Fitzpatrick

(1992) The mythology of modern law. London and New York: Routledge.

22.

Forman

(2017) Locking up Our Own: Crime and Punishment in Black America. New York: Farrar, Straus and Giroux.

23.

Foucault

1995 Discipline and Punish, The Birth of the Prison (transl. Alan

Sheridan

, 1977). New York: Vintage Books.

24.

Freedom House (2017) Masiphumelele Social Cohesion Profile. Available at: https://freedomhouse.org/sites/default/files/Masiphumelele_ZA_Community_Case_Study_FINAL.pdf (accessed on 1 March 2019).

25.

Garfinkel

(1956) Conditions of successful degradation ceremonies. American Journal of Sociology 61(5): 420–424.

26.

Garland

(1990) Punishment and Modern Society, A Study in Social Theory. Chicago: University of Chicago Press.

27.

Garland

(2005) Penal excess and surplus meaning: Public torture lynchings in twentieth-century America. Law and Society Review 39(4):793–834.

28.

Gillespie

(2014) The context and meaning of mob justice in Khayelitsha, report prepared for the Khayelitsha Commission: Commission of inquiry into allegations of police inefficiency and a breakdown in relations between SAPS and the Community of Khayelitsha. Available at: http://www.khayelitshacommission.org.za/images/towards_khaye_docs/Khayelitsha_Commission_Report_WEB_FULL_TEXT_C.pdf (accessed on 12 January 2018).

29.

Gillespie

(2015) Murder and the whole city. Anthropology Southern Africa 37(3–4): 203–212.

30.

Gillespie

Naidoo

(2019) #Mustfall: The South African student movement and the politics of time. The South Atlantic Quarterly 118(1): 190–194.

31.

Glaser

(2018) Nostalgia for a beating: Discipline, generational authority and corporal punishment at a Soweto High School, c. 1960-2000. History of Education 48(3): 395–409.

32.

Gobodo-Madikizela

(1999) Legacies of violence: An in depth analysis of two case studies based on interviews with perpetrators of necklace murder and with Eugene de Kock. Unpublished PhD Thesis, University of Cape Town, Cape Town, South Africa.

33.

Gobodo-Madikizela

(2003) A Human Being Died That Night: A Story of Forgiveness. Cape Town: David Philip.

34.

Gobodo-Madikizela

(2008) Psychological repair: The intersubjective dialogue of remorse and forgiveness in the aftermath of gross human rights violations. Journal of American Psychoanalytical Association 63(6): 1085–10123.

35.

Godoy

(2006) Popular Injustice, Violence, Community and Law in Latin America. Stanford: Stanford University Press.

36.

Goldstein

(2003) In our own hands: Lynching, justice, and the law in Bolivia. American Ethnologist 30(1): 22–43.

37.

Hansen

Stepputat

(2005) Introduction: On empire and sovereignty. In: Hansen

Stepputat

(eds) Sovereign Bodies: Citizens, Migrants, and States in the Postcolonial World. Princeton: Princeton University Press, pp. 9–37.

38.

Hornberger

(2013) From general to commissioner to general – On the popular state of policing in South Africa. Law & Social Inquiry 38(3): 598–614.

39.

Huggins

(1991) Introduction: Vigilantism and the state – A look south and north. In: Huggins (ed.) Vigilantism and the State in Modern Latin America. Essays on Extralegal Violence. Westport, New York, Connecticut, London: Praeger.

40.

Innes

(2004) Signal crimes and signal disorders: Notes on deviance as communicative action. British Journal of Sociology 55(3): 336–355.

41.

Johnston

(1996) What is vigilantism? British Journal of Criminology 36(2): 236–2220.

42.

Kelly

(2018) The Limits of Blame, Rethinking Punishment and Responsibility. Cambridge, London: Harvard University Press.

43.

Koch

(2019) Personalizing the State: An Anthropology of Law, Politics, and Welfare in Austerity. Oxford: Britain Oxford University Press.

44.

Krygier

(2015) Rule of law (and Rechtsstaat). In: Wright

(ed) International Encyclopedia of the Social & Behavioral Sciences. 2nd ed. Oxford: Elsevier, pp. 780–787.

45.

Kynoch

(2016) Apartheids afterlives: Violence, policing and the South African state. Journal of Southern African Studies 42(1): 65–78.

46.

Makhulu

(2015) Making Freedom, Apartheid, Squatter Politics and the Struggle for Home. Durham and London: Duke University Press.

47.

Masicorp (n.d.) ‘About Masiphumelele’. Available at: http://www.masicorp.org/aboutmasiphumelele.html (accessed on 17 June 2018).

48.

Massoud

(2015) Work rules: How international NGOS build law in war-torn societies. Law and Society Review 49(2): 333–364.

49.

Mbembe

(1992) Provisional notes on the postcolony. Africa 62(1): 3–36.

50.

Mead

(1918) The psychology of punitive justice. American Journal of Sociology 23(1918): 577–602.

51.

Merry

(1993) Sorting out popular justice. In: Merry

Milner

(eds) The Possibility of Popular Justice: A Case Study of Community Mediation in the United States. Ann Arbor: University of Michigan Press.

52.

Merry

(1988) ‘Legal Pluralism’. Law and Society Review, pp. 869–896.

53.

Mnisi Weeks

(2017) Access to Justice and Human Security: Cultural Contradictions in Rural South Africa. New York: Routledge.

54.

Nkalane

(2015) Man burnt to death in Masiphumelele. Cape Times 10 November.

55.

O’Regan

Pikoli

(2014) Towards a safer Khayelitsha, report of the commission of inquiry into allegations of police inefficiency and a breakdown in relations between SAPS and the community of Khayelitsha. Available at: http://www.khayelitshacommission.org.za/images/towards_khaye_docs/Khayelitsha_Commission_Report_WEB_FULL_TEXT_C.pdf (accessed 15 September 2014).

56.

Ocobock

(2012) Spare the rod, spoil the colony: Corporal punishment, colonial violence, and generational authority in Kenya, 1897-1952. The International Journal of African Historical Studies 45(1): 29–56.

57.

Orock

(2014) Crime, in/security and mob justice: The micropolitics of sovereignty in Cameroon. Social Dynamics 42(2): 408–428.

58.

Paret

(2015) Violence and democracy in South Africa’s community protests. Review of African Political Economy 42(143): 107–123.

59.

Phaliso

(2015) Masiphumelele calm after residents rampage. Cape Times 4 November. Available at: https://www.pressreader.com/south-africa/cape-times/20151104/281625304177148.

60.

Razack

(2014) It happened more than once’: Freezing deaths in Saskatchewan. Canadian Journal of Woman and the Law 26(1): 51–80.

61.

Rosenbaum

Sederberg

(1974) “Vigilantism: An Analysis of Establishment Violence”. Comparative Politics 6: 541–571.

62.

Rush Smith

(2019) Contradictions of Democracy: Vigilantism and Rights in Post-Apartheid South Africa. Oxford: Oxford University Press.

63.

Seeds

(2018) Disaggregating LWOP: Life without parole, capital punishment, and mass incarceration in Florida, 1972–1995. Law and Society Review 52(1): 172–205.

64.

Skelton

(2005) The influence of the theory and practice of restorative justice in South Africa with special reference to child justice. LL.D. Dissertation, Faculty of Law University of Pretoria.

65.

Steinberg

(2016) How well does theory travel? David Garland in the global south. The Howard Journal of Crime and Justice 55(4): 514–531.

66.

Super

(2011) Punishment and the body in the ‘old’ and ‘new’ South Africa, a story of punitive humanism. Theoretical Criminology 15(4): 427–443.

67.

Super

(2013) Governing Through Crime in South Africa: The Politics of Punishment and Race in Neo-Liberalizing Regimes. Surrey, Burlington: Ashgate, Advances in Criminology series.

68.

Super

(2016) Volatile sovereignty: Governing crime through the community in Khayelitsha. Law and Society Review 50 (2): 450–483.

69.

Super

(2017) What’s in a name and why it matters: A historical analysis of the relationship between state authority, vigilantism and penal power in South Africa. Theoretical Criminology 21(4): 512–531.

70.

Super

(2020) Three warnings and you’re out: Banishment and precarious penality in South Africa’s informal settlements. Punishment and Society 22(1): 48–69.

71.

Thomas

(2012) The Power of Naming, ‘Senseless Violence’ and Violent Law in Post-Apartheid South Africa. Cape Town: Centre for the Study of Violence and Reconciliation and Centre for Humanities Research University of the Western Cape.

72.

Van Onselen

(1985) Crime and total institutions in the making of modern South Africa the life of Nongoloza Mathebula 1867-1948. History Workshop Journal 19(1): 62–81.

73.

World Bank (2018) Overcoming poverty and inequality in South Africa, an assessment of drivers, constraints and opportunities. Available at: http://documents.worldbank.org/curated/en/530481521735906534/pdf/124521-REV-OUO-South-Africa-Poverty-and-Inequality-Assessment-Report-2018-FINAL-WEB.pdf (accessed on 5 April 2019).