SciELO - Scientific Electronic Library Online

 
 issue55 author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Article

Indicators

Related links

  • On index processCited by Google
  • On index processSimilars in Google

Share


SA Crime Quarterly

On-line version ISSN 2413-3108
Print version ISSN 1991-3877

SA crime q.  n.55 Pretoria Mar. 2016

http://dx.doi.org/10.17159/2413-3108/2016/v0n55a154 

RESEARCH ARTICLES

 

Pervasive, but not politicised. Everyday violence, local rule and party popularity in a Cape Town township

 

 

Laurence PiperI; Joanna WheelerII

IProfessor of Political Studies at the University of the Western Cape (UWC), lpiper@uwc.ac.za
IISenior Research Associate at the Sustainable Livelihoods Foundation (SLF), and a Postdoctoral Fellow in the Political Studies Department, UWC. joanna.wheeler@livelihoods.org.za

 

 


ABSTRACT

Through examining violence in the township of Imizamo Yethu in Cape Town, we show that leadership in this community is not based on violence, despite its pervasiveness in the settlement. Further, rule by local leaders and the state is often weak, and normally not violently enforced. This account challenges three common views in the literature. The first is that, under conditions of weak rule, violence is primarily about contests over political power. The use of violence by a variety of social actors in Imizamo Yethu, but rarely by political leaders or parties, challenges this assumption. The second is that violence is central to maintaining local rule - but in Imizamo Yethu leaders have seldom used coercion. Lastly, our case illustrates that effective local rule is not necessarily a condition of party identification, which is rooted in larger dynamics of state patronage and race politics that may even weaken local rule.


 

 

On the morning of 6 December 2011, we arrived in Imizamo Yethu as a small group of researchers1prepared to start a three-month action-research project on violence, local (dis)order and rule. As we drove into the township, we noticed that large rocks and tyres had been pushed into the road, blocking the way. Smoke drifted across the settlement and we saw broken glass and debris in the road. We started counting the numbers of cars with smashed windows. At the police station at the entrance to the township, a large group of angry people was gathered, shouting and arguing.

Eventually, after speaking with various community leaders and residents, we were able to establish that there had been an outbreak the previous night of 'taxi violence', involving two different factions of the local Imizamo Yethu taxi operators. The dispute centred around the licencing process and access to lucrative taxi routes, with the more established association refusing entrance to others. When a second, less formalised group began to operate taxis in the area, the more formal group retaliated by attacking their cars and stabbing a driver. The situation then spiralled into a series of retaliations between the different taxi groups.

What emerged on our first day in Imizamo Yethu was an indication of the complexities surrounding the pervasiveness of violence in the settlement, and the implications for social and political order: two taxi associations with uncertain links to competing factions of the South African National Civic Organisation (SANCO) local leadership; stories of episodic neighbourhood watches; and allegations of local linkages between all of these and the main political parties in Cape Town. Over time, the frame of violence covered by our fieldwork extended to dealing with issues of crime, xenophobia and service delivery protests, but all of these were threaded through by the dynamics of local political rivalry and weak rule by both state and local leaders, as demonstrated by the taxi violence on our first day in Imizamo Yethu.

By late 2015, while we were writing this article, Imizamo Yethu witnessed a popular mobilisation of unprecedented scale against two drug gangs, the amaXaba and the Bad Boys Company, whose turf battle had led to regular stabbings and even deaths over the preceding year. In August, a large vigilante group confronted and killed the leaders of both gangs. In the days that followed, SANCO leaders met with the remaining gang members to end the conflict and demobilise the gangs. Since that period a nightly community patrol has been in place that, by all accounts, has significantly reduced crime. Once again, local leaders reacted to the use of violence by other actors in Imizamo Yethu, although this time they endorsed the violence and new forms of coercion which united the community, local leaders and even the police.

This article explores the relationship between violence, local rule and political actors in order to contribute to the current debate on social cohesion, inequality and security in cities of the global South. The relationship, we suggest, is a lot less linear than often assumed. We show how violence has not been used to gain or consolidate local leadership in Imizamo Yethu, at least not yet, and has been used only in exceptional cases to enforce local rule. Rather, violence is a pervasive and a constant background presence in many private and some public interactions, and is used to police particular moral views, such as the immorality of drug use -reinforcing the kind of social cohesion that leads to the vigilante mobilisation described above.

 

Rebelocracy and violence

Some influential recent literature on violence explores the linkages between micro-level analysis and meso-and macro-level analyses of conflict, in order to construct new arguments about the implications of violent conflict for wider political, economic and social processes,2 and the extent to which violence is used by particular groups to establish political order.3 Thus there is an emergent literature from conflict resolution studies exploring the relationships between armed actors and specific regimes of governance.4 This study interrogates some of the assumptions reflected in the literature with regard to how violence constitutes the political under conditions of weak state rule, and poses questions about social cohesion by revealing that weak social cohesion need not result in rule through violence.

As part of the attempt to bridge different levels of analysis of violence, Ana Arjona offers a typology of authority regimes within civil and political conflict that entails a 'degree of intervention of armed groups in civilian affairs' and the 'presence of social contract between armed group and local population'.5 Although this includes situations of civil war, in theory it includes any context in which state monopoly of violence is contested or filled by other armed non-state actors. Indeed, as Davis points out, this is a common phenomenon across the global South, particularly in respect of economic control of local areas, for instance with drug gangs or militias.6 The inability of the South African state to address endemic levels of insecurity in poor, urban settlements of South Africa, and the proliferation of actors pursuing violence, from gangs to vigilante organisations to moments of popular mobilisation such as xenophobic attacks, reveal the relevance of armed, non-state actors to local rule in South Africa.

With this in mind, we return to Arjona's typology, where 'rebelocracy' refers to a high degree of intervention by armed groups in civilian affairs, and a sense of a social contract between the armed group and the local population. This would include the provision of services similar to those provided by a state, as well as other symbols of state-like power, such as a flag or nationalist symbols. On the other end of the typology, Arjona identifies 'aliocracy' as a narrow range of interventions by armed groups in civilian affairs, and a social contract between the armed group and the local population, such as militia expelling drug gangs from the favelas of Rio. The lack of a social contract between armed groups and the local population would fall into 'disorder' in Arjona's typology.

We suggest that since its formation in the early 1990s, Imizamo Yethu has been slipping from a form of 'aliocracy' under a relatively strong civic leadership, towards 'disorder' with the weakening of SANCO and local state rule. This is despite the fact that SANCO is closely identified with the African National Congress (ANC) at local level. Notably, however, the ANC's legitimacy remains in place through the influence of national ideas and racialised experiences of life in Hout Bay, rather than through legitimate local political leadership and effective rule in Imizamo Yethu. The possible exception to this trend is the recent antidrug gang mobilisation which has, paradoxically, strengthened both SANCO and its relations with the local police on the back of an implicit anti-crime social contract with the residents of Imizamo Yethu. How long this will last is hard to say.

Key to understanding local politics in Imizamo Yethu is the fact that this is an ANC-aligned community in a city that has been run by the Democratic Alliance (DA) since 2006, and in a province that has been run by the DA since 2009. As argued elsewhere,7 this has placed significant strain on relations between community leaders and the city and province, as SANCO's partisan identity threatens rather than reinforces its claim to legitimate leadership of Imizamo Yethu in the eyes of these two spheres of the state. Conversely, DA governance desires 'non-partisan' community representation, an approach attempted in Imizamo Yethu by the SANCO leadership of 2007 until this undermined their relations with the local ANC, and they were eventually supplanted in 2015 by a clearly more partisan group. These dynamics, we suggest, are mostly driven by the larger logic of race and party politics in South Africa, and thus it seems likely that tensions between the DA province and city and SANCO will continue into the future, potentially undermining local rule in Imizamo Yethu, and leading from the current state of 'aliocracy' back down the path to disorder.

Arjona, Kalyvas et al. and Davis make the case that contexts of civil war and political violence cannot be treated as homogenous political spaces, and this is also consistent with our argument - that the informal nature of governance within the settlement means that political rule is not only about the state, or one form of state/non-state rule, but may vary significantly across place. In this regard the dominant party literature on the South African political system is particularly useful. Thus Butler8suggests that the enduring rule by the ANC in South Africa, arguably reinforced by race politics where the ANC is seen as the leader of black people, leads to a blurring of party and state, with a range of positive benefits for governance but negative consequences for accountability. On the one hand, the ANC's popularity means it can make unpopular but wise long-term policy choices on, for example, land reform, but on the other, it can ignore a public outcry about unpopular choices or corrupt practices. Some scholars have pointed to the role of liberation nationalism and the access to state resources in cementing the idea of a 'party-state', where racial identity (black African), political party (ANC) and state power are seen as both instrumentally and normatively linked.9

At the local level, we can add the notion of 'partysociety' that conjoins racial identity, political party and community leadership.10 This leadership is not exercised by the ANC alone, but also by its allies, often in the form of SANCO. In this context, the claim that the ANC (and its allies) is entitled to rule as the historic champion of oppressed black South Africans is reinforced at local level by the dependency of poor communities on the ANC-run state for development. Consequently, as Benit-Gbaffou notes, the most reliable way of accessing the state for most poor, black communities throughout the country is often through networks in the party rather than state channels.11 It is often these informal networks that mediate state-society relations, more than formal processes or structures are able to.

As already noted, in Imizamo Yethu the link between party and state has been weakened with the advent of the DA to political power in the city and in the province. While this has weakened state patronage to local Imizamo Yethu leaders through the party, it has not necessarily weakened the popularity of the ANC. A key reason for this centres on the politics of development, in particular the long struggle over what to do with the vacant land adjacent to Imizamo Yethu.12 The debate centres on whether to build community-specific facilities like a school, as advocated by leaders of the white community, or build more houses and have Imizamo Yethu children attending schools in other parts of Hout Bay, as advocated by SANCO. The vision of the white community is seen by Imizamo Yethu leaders as an attempt to entrench racial segregation in Hout Bay, rather than challenge it by constructing one set of schools, clinics and other public facilities for all residents. It is offered as evidence of white racism. Thus the politics of race, party and place reinforces a form of political cohesion in Imizamo Yethu despite, or even because of, weak local rule.

In addition, violence in Imizamo Yethu is more commonly practiced by non-political actors, who are not obviously connected to local leadership. The one exception here has been the anti-gang mobilisation, which has brought the community, SANCO and the police together through the use of violence against criminals. This exceptional moment brings us to the question of social cohesion. Social cohesion is fundamentally a normative concept that prescribes a shared sense of morality, purpose and order within a particular context. Thus Forest and Kearns state:

Social cohesion can emphasise the need for a shared sense of morality and common purpose; aspects of social control and social order; the threat to social solidarity of income and wealth inequalities between people, groups and places; the level of social interaction within communities or families; and a sense of belonging to place. By implication, a society lacking cohesion would be one which displayed social disorder and conflict, disparate moral values, extreme social inequality, low levels of social interaction between and within communities and low levels of place attachment.13

Other research on urban neighbourhoods in Chicago shows that more organised and socially cohesive localities may have higher levels of organised violence, as the levels of mutuality and social networks provide a resource for violent actors.14 The authors make the case for 'negotiated co-existence', which could be seen to correspond to Arjona's type of aliocracy. However, as we will show, neither of these categories fully capture the realities of the relationship between violence, regimes of authority and local order in Imizamo Yethu. The key point here is that most of the evidence points to uneven and transitory forms of social cohesion, in which violence is used both to enforce a notion of social cohesion (e.g. by expelling certain foreigners) and to unravel a sense of social cohesion (e.g. high levels of insecurity due to crime and interpersonal violence). The overall levels of violence are high, but not highly politicised. Instead, violence is tied to everyday crime, inter-personal relations and business competition (e.g. taxi violence).

 

Cohesion, violence, insecurity and (dis)order in Imizamo Yethu

The analysis that follows is presented on the basis of evidence gathered over a period of five years in several different research projects. In 2011 we conducted a representative household survey of Imizamo Yethu, interviewing 306 respondents on mostly demographic and livelihoods issues. That same year we also embarked on a participatory action research project exploring insecurity that involved five participatory focus groups in addition to a week-long digital story-telling workshop with 11 residents. From 2012 to 2015 we conducted over two dozen in-depth elite interviews with SANCO, ANC leaders, non-governmental organisation (NGO) staff, community organisers, local government officials and leaders of the migrant community on projects dealing with housing, transport, xenophobia, and waste and water. In addition, we have kept a close eye on the local media, both print and social, in relation to these themes.

 

Social cohesion

Central to the Forest and Kearns account of social cohesion are shared norms, identities, order, equality, solidarity and sense of belonging. While at its formation in 1991, with just 450 families, Imizamo Yethu was sometimes described in these terms, today most long-standing residents explicitly contrast the plurality, mobility, diversity, violence and disorder of the present unfavourably with the social order of the past. Demographically it is clear that Imizamo Yethu has changed tremendously in 20 years, growing at a rate of nearly 1 000 new people a year to about 25 000 people today.15 When asked in 2011 how many people live in Imizamo Yethu, the most common response from our 306 householders was 'too many'. Further, most of the residents are migrants, mostly from the Eastern Cape, but increasingly from the rest of Africa.16

In addition to growing quickly, Imizamo Yethu has also become more diverse, and obviously so, with foreign nationals prevalent in the informal business sector in the township, especially Somalis in spaza shops, Namibians in taverns, and Congolese in hair salons, and with significant numbers of Angolans, Malawians and Zimbabweans resident throughout Imizamo Yethu. Along with the diversity of nationalities come language and cultural differences that have limited the ability of the isiXhosa-speaking majority to unite the community in cultural terms. Hence, language at community meetings has become an issue of contention, as isiXhosa is used rather than English, which is also understood by the coloured and foreign residents of Imizamo Yethu. The inability of SANCO to include foreign residents in its various meetings is clear ('they don't want us, but we see SANCO as a South African thing anyway'.)17 The general decline of SANCO has undermined its ability to manage increasing diversity. Hence, as one former leader put it:

[I]n the 1990s SANCO was strong, every meeting had hundreds of people, and we controlled everything. It was a good place to live. Now people only come to a meeting if they're going to get a house, and people do what they want.18

With this diversity has come greater inequality, as many migrants are poorer than more established families. Since 2000 the upper slopes of the Oranje Kloof mountain have become crammed with shacks that number over 4 300, to the roughly 1 100 formal houses in Imizamo Yethu.19 Further, local leaders complain of having no ability to control migration into Imizamo Yethu: 'People just rent out their backyards, and sometimes the whole house to foreigners, without telling anyone, so we don't even know about it.'20 Our own experience working with community members over a five-year period also suggests a significant turnover of residents, with people moving in and out of the township at a high rate. This is the pattern of 'churning' in poorer urban settlements more widely in South Africa.21 Thus several of the leaders we started working with in 2011 left Imizamo Yethu within a couple of years for other townships in Cape Town.

Perhaps the clearest evidence of the complexity surrounding the relationship between violence and social cohesion in Imizamo Yethu comes from the ubiquitous levels of insecurity and violence we encountered. Indeed, there is a real sense that violence is a key social norm, with many respondents referring to the need to establish order through violence, for example by reducing crime through 'community justice'. Hence one respondent wistfully longed to return to 'the old ways' or 'sorting people out' when they committed a crime. 'We caught a thief near the school', he told us one day, 'and sorted him out over there [pointing], and then over there, and over there.'22

Crime and insecurity: cats and dogs, not the police and SANCO

Overall levels of crime, and interpersonal violence in particular, are significant in Imizamo Yethu, and contribute to a sense of insecurity in the township. Police statistics for Hout Bay indicate an average of 13 deaths per year for the last five years.23 This amounts to a figure of 26 per 100 000, which, while lower than the national average of 33 per 100 000, is still high. Our research also suggests that many crimes, particularly those related to sexual violence, mugging and robbery, are significantly underreported, as revealed by many respondents in our workshops who related intimate crimes they had not reported to the police or even shared with their families. Perhaps more striking has been the rise in drug-related crime, which has seen a notable spike in the last few years. There is more to this than better policing, as many respondents reported the emergence of drug gangs in Imizamo Yethu for the first time in its history.24

In our participatory workshops, discussions highlighted the question of insecurity. Notably, all respondents feared crime, especially at night, and in all parts of the settlement other than in the section where they lived. While respondents felt that 'the police, SANCO/ANC and the community' should be the leading actors in reducing crime, in that order, they reported that those who made the community safe were 'cats and dogs, neighbours and family', in that order. Cats, because they 'kill rats and mice that eat food', and dogs because 'they bark at tsotsis' .25

Despite these general observations, the nature of insecurity experienced by respondents changed depending on their social positioning. Some with resources, like shopkeepers, reported 'always watching for thieves'.26 Certain foreigners (e.g. those from Zimbabwe, Malawi and the Democratic Republic of the Congo) were more vulnerable to violence than others (Angola, Namibia) because 'everyone knows the Namibians don't play. They will fight every time.'27Women, especially young women, felt vulnerable at night in areas of the township outside their immediate neighbourhood, in particular shebeens. There were also important differences in perceptions of safety by type of housing, with those living in shacks feeling more vulnerable than those in formal housing. As one woman put it: 'We live in fear of someone kicking down the shack door and raping us.'28 Lastly, some respondents reported deliberately befriending powerful people in the township to get protection: 'I always make friends with gangsters so no-one messes with me.'29

This work threw into sharp relief the meaning of insecurity and violence for social cohesion and local governance. It was clear that these generalised levels of insecurity led respondents to see the state, and the police and justice system in particular, as inadequate. This lack of trust in the state was not replaced by a faith in local political actors such as SANCO. Indeed, the space of effective security governance, whether by state or society, remained mostly a vacuum in Imizamo Yethu, at least until the anti-gang violence of late 2015 and the nightly community patrols. Although some respondents complained that 'patrols sometimes beat up the wrong people', all agreed that they had made the township much safer.30

Taxi violence: amaphela versus amahoender

Generally in Cape Town, taxi associations are important role players, as they are well armed, organised and relatively wealthy, and have in the past been accused of operating in mafia style.31 In both Hout Bay and Imizamo Yethu there is a history of violent conflict linked to business competition between taxi owners,32 as noted above. In 2011 the conflict we encountered was between two informal taxi associations that did local routes around Hout Bay, respectively known as the amaphela (cockroaches) and amahoender (chickens). After much bargaining, they merged in 2013.

Most recently, a former taxi owner has become the chairperson of SANCO, following community protests against resettlement linked to building a Bus Rapid Transport (BRT) turnaround point on the Wynberg route that runs past Imizamo Yethu. Thus, although the protest invoked the long-standing demand by SANCO for housing rather than other facilities to be built on the vacant land in Imizamo Yethu, it also coincided with the particular interests of a small group of taxi owners yet to be incorporated into the system. Despite this, the overwhelming impression we gained from respondents over many years' fieldwork is that while taxi owners are influential in Imizamo Yethu, they do not control SANCO. As one respondent put it, 'they are much too busy making real money to worry about small politics'.33

Consequently, our workshops identified taxi associations as a source of both insecurity and security. At times, taxi associations have been drawn in to intervene to protect residents, such as in incidents involving street gangs. At other times, taxi associations drive violence and insecurity through internal disputes, such as the conflict over local routes that we stumbled on in 2011. Indeed, this lack of clear alignment with both wider community agendas and community leaders undermines conditions for both social cohesion and effective local governance by contributing to a generalised sense of unpredictability.

Xenophobic attacks and protection rackets

Our 2011 household survey of Imizamo Yethu revealed that the vast majority of respondents (85%) were South African. This finding runs against the received wisdom of many who live in Imizamo Yethu that the proportion of foreign migrants in the settlement is around 40 to 50%. Our survey findings are closer to the 2011 census, which identifies 3.3% of the Hout Bay population as 'other'. Assuming the vast majority of the 'other' live in Imizamo Yethu, this would be about 7% of the settlement - roughly half of what we found. Notably, a 2003 Development Action Group survey found just 5% of Imizamo Yethu were foreign nationals, so our 2011 figure is a threefold increase in eight years.34 Although this uncertainty around the number of African migrants in Imizamo Yethu is unresolvable without more careful research, the available evidence suggests that it is probably closer to the 20% mark (4 000 people) rather than the 40% (8 000 people) often invoked by Imizamo Yethu residents and local leaders in public forums.

While many foreign-born residents experience violence in Imizamo Yethu, there are many reasons to believe that it is not always about national identity. Foreign residents as well as South African nationals experience everyday crime, business competition, personal conflicts and the like. However, there is no doubt that xenophobia is real, as reported by many foreign residents. We also encountered this first hand when a digital storytelling workshop collapsed after conflict between South African and foreign participants.35 In addition, Imizamo Yethu has been the site of several attempts to expel foreign residents by mobs threatening violence; the most substantial of which was during the 2008 xenophobic wave that swept the country.36 Notably, the only time we encountered stories of protection rackets against foreign nationals in Imizamo Yethu was in respect of allegations that certain ANC Youth League members approached foreign shopkeepers in the wake of expulsions in order to extort money for protection -with the implication that they could prevent (or incite) xenophobic violence.

While the ambivalence among South Africans towards foreign nationals clearly undermines a shared sense of belonging in Imizamo Yethu, and thus social cohesion, it also weakens local rule. This is not only because of the ambivalence of local leaders towards foreign nationals but also because many residents know that foreign nationals are less likely to go to the police when robbed, as many feel vulnerable to state persecution due to inadequate documentation. Indeed, this exclusion from rule is also manifest in the reluctance of the vast majority of foreign-born residents to participate in SANCO structures, thus further parsing representation in Imizamo Yethu rule along national lines.

Service delivery protests: houses or buses?

Compared to most townships, Imizamo Yethu has a relatively limited history of service delivery protests,37excluding the xenophobic and vigilante attacks already discussed. However, the most recent one occurred in April 2015 when a community protest led to the destruction of eight aluminium-framed houses built by the City of Cape Town for families who were going to be displaced by a MyCiti Bus Rapid Transport (BRT) turnaround station planned for the northern entrance to Imizamo Yethu. According to a leader of SANCO, 'the city could not explain why these few people had got housing ahead of others who have been waiting for years ... also why was MyCiti going where we had agreed there would be housing? This angered the community and we tore them down.'38

This story is consistent with accounts of service delivery protests that identify both the failure to deliver and poor communication by municipalities as key reasons for protest.39 Indeed, if we return to the taxi conflict we encountered on the first day we entered IY in 2011, the two groups of SANCO leaders involved in that were divided by the issue of housing. Where one group had supported the building of a new school instead of houses, the other saw this as selling out. The issue was not just that they were taking the side of the 'white' DA-run city, but also that the decision to build community facilities specifically for Imizamo Yethu, rather than share existing ones in the white area, would further racially segregate Hout Bay. As one leader put it: 'We are all Hout Baynians, we should share the same facilities.'40 The complaint that the white residents of Hout Bay were racist and did not want black people in 'their area' was one made frequently by one SANCO faction at the time, as well as by local ANC leaders.

At the same time, however, there was more than race politics driving the anti-BRT protests of 2015, as some of the key SANCO leaders involved also have interests in the taxi industry, and are potentially threatened by the extension of the BRT to Wynberg. Thus, while this protest represented a moment in a longer struggle over access to housing and integration into Hout Bay, it also provided an opportunity for some local elites to pursue personal interests. This is a good example of what Von Holdt terms as 'protests within protests',41 and potentially undermines faith in SANCO to reliably champion all the residents of Hout Bay. Indeed, the claim of corruption was publicly wielded by various local leaders in Imizamo Yethu. Paradoxically then, service delivery politics and protest do not necessarily unite Imizamo Yethu, and may even exacerbate rivalry for local leadership.

At the same time, however, the weakening of SANCO and enduring conflicts over housing and other forms of development in Imizamo Yethu have not undermined the ANC at election time. As revealed by an examination of national election results for the two Imizamo Yethu voting districts, ANC popularity has remained constant at just below 90% since 1999 (Table 1) and indeed, voter turnout in Imizamo Yethu has increased with every national election, and is comparable to the white community who live in the valley (Figure 1). Indeed, this is not just a phenomenon of national elections; a similar positive trend is evident in support for the ANC in local government elections in Hout Bay (Table 2). Whatever the travails of SANCO and weak rule, the ANC remains electorally strong in Imizamo Yethu.

 

Conclusion

The pervasive and varied nature of violence in Imizamo Yethu is implicated in complex dynamics of order and rule, such that there is no simple relationship between violence and social cohesion. Most recently, popular mobilisation against drug gangs led to the deaths of the gang leaders at the hands of a mob, and ongoing nightly patrols by the community. These events appear to have re-established a form of social cohesion between the community, SANCO leaders and the police around the violent repression of crime. However, this moment represents an exception in the history of rule in Imizamo Yethu over the last 10 years, where local leadership and local rule have become increasingly weak. Notably, during this period neither local rule nor the contest for leadership were settled through violence.

Key to explaining these paradoxes, we suggest, is the disjuncture between local leaders in Imizamo Yethu and the local state. Thus, unlike much of the rest of South Africa, identification with the ANC in Imizamo Yethu is a disadvantage in accessing the DA-run city and province, and hence the patronage available through the 'party state' is limited. At the same time, though, the exclusion felt by Imizamo Yethu leaders, best represented in the struggles over housing and racial segregation in Hout Bay, reinforces the idea that the ANC is the true champion of the racially oppressed black people of South Africa. Thus weak local rule poses no threat to ANC popularity, and under certain circumstances may even reinforce it. Ironically it is the popular demand for security, manifest in the anti-drug mobilisation of 2015, that has disrupted the slide from 'aliocracy' to 'disorder' in Imizamo Yethu. However, it seems doubtful whether this powerful moment of social cohesion can indefinitely withstand the wider partisan logic of state patronage and race politics.

 

Acknowledgements

We gratefully acknowledge the support of the UK Economic and Social Research Council and the Institute of Development Studies, University of Sussex, UK. We would like to thank Patricia Justino, Ana Arjona, Jaideep Gupte and the other members of the ESRC Governance and Non-State Actors research group for the discussions that helped inform our analysis. We would also like to thank PASSOP for partnering with us on the action-research component.

 

Notes

1 Action-researchers included Rory Liedeman, Bathulile Nthsingile, Laurence Piper and Joanna Wheeler.

2 Laia Balcells and Patricia Justino, Bridging micro and macro approaches on civil wars and political violence: issues, challenges, and the way forward, Journal of Conflict Resolution, 58:8, 2014, 1343-1355.         [ Links ]

3 Stathis N Kalyvas, Ian Shapiro and Tarek Masoud (eds), Order, conflict and violence, Cambridge University Press: Cambridge, 2008.         [ Links ]

4 See also Balcells and Justino, Bridging micro and macro approaches on civil wars and political violence; Kalyvas, Shapiro and Masoud (eds), Order, conflict and violence.

5 Ana Arjona, Wartime institutions: a research agenda, Journal of Conflict Resolution, 58:8, 2014, 1375.         [ Links ]

6 Diane Davis, Non-state armed actors, new imagined communities, and shifting patterns of sovereignty and insecurity in the modern world, Contemporary Security Policy, 30:2, 2009, 221-245.         [ Links ]

7 Laurence Piper, From party-state to party-society in South Africa: SANCO and the informal politics of community representation in Imizamo Yethu, Hout Bay, Cape Town, in C Bénit-Gbaffou (ed.), Popular politics in South African cities: unpacking community participation, Pretoria: HSRC Press, 2015, 21-41;         [ Links ] Laurence Piper and Claire Bénit-Gbaffou, Mediation and the contradictions of representing the urban poor in South Africa: the case of SANCO leaders in Imizamo Yethu in Cape Town, South Africa, in Bettina von Lieres and Laurence Piper (eds), Mediated citizenship, Basingstoke: Palgrave Macmillan, 2014, 25-42.         [ Links ]

8 Anthony Butler, Considerations on the erosion of party dominance, Representation, 45:2, 2009, 159-172.         [ Links ]

9 Daryl Glaser, Uncooperative masses as a problem for substantive and participatory theories of democracy: the cases of 'people's power' (1984-6) and the xenophobia (2008) in South Africa, in Bénit-Gbaffou (ed.), Popular politics in South African cities;         [ Links ] and Roger Southall, The black middle class and democracy in South Africa, The Journal of Modern African Studies, 52:4, 2014, 647-670.         [ Links ]

10 Piper, From party-state to party-society in South Africa.

11 Claire Bénit-Gbaffou, Party politics, civil society and local democracy: reflections from Johannesburg, Geoforum, 43:2, 2011, 178-189.         [ Links ]

12 Piper and Bénit-Gbaffou, Mediation and the contradictions of representing the urban poor in South Africa.

13 Ray Forrest and Ade Kearns, Social cohesion, social capital and the neighbourhood, Urban Studies, 38:12, 2001, 2125-2143.         [ Links ]

14 CR Browning, RD Dietz and SL Feinberg, The paradox of social organization: networks, collective efficacy, and violent crime in urban neighborhoods, Social Forces, 83:2, 2004, 503-534.         [ Links ]

15 Statistics South Africa, Census 2011, 2011, http://www.statssa.gov.za/?page_id=3839

16 Notably, Ovambo migrants, mostly men from Angola and Namibia who work in the fishing industry in Hout Bay, have lived in Imizamo Yethu from its formation in 1991. Most other migrants are from Zimbabwe, Malawi and the Democratic Republic of the Congo, and have settled in Imizamo Yethu over the years, and especially around the turn of the century. According to several respondents, Imizamo Yethu became a popular destination for foreign migrants because 'it was safe and close to jobs'.

17 Respondent 3, Zimbabwean national, 14 March 2012.

18 Respondent 4, taxi owner and leader of the South African National Civic Organisation (SANCO) faction elected 2015, Imizamo Yethu, 20 February 2015.

19 Statistics South Africa, Census 2011.

20 Respondent 5, former African National Congress Youth League (ANCYL) leader, Imizamo Yethu, 3 June 2014.

21 David Neves and Andries du Toit, The dynamics of household formation and composition in the rural Eastern Cape, Centre for Social Science Research (CSSR) Working Paper, 228, November 2008, http://www.cssr.uct.ac.za/sites/cssr.uct.ac.za/files/pubs/WP228.pdf        [ Links ]

22 Respondent 1, member of SANCO faction elected in 2007, Imizamo Yethu, 28 January 2012.

23 South African Police Service (SAPS), Crime statistics: April 2014 - March 2015, 2015, http://www.saps.gov.za/resource_centre/publications/statistics/crimestats/2015/crime_stats.php.         [ Links ]

24 Respondent 5, former ANCYL leader; Respondent 2, Community Development Worker (CDW), Imizamo Yethu, 27 January 2012.

25 Focus group 1, participatory action research workshop with South African residents on insecurity, 10 February 2012.

26 Focus group 3, participatory action research workshop with foreign residents on insecurity, 2 March 2012.

27 Respondent 3, Zimbabwean national.

28 Focus group 2, participatory action research workshop with women residents on insecurity, 17 February 2012.

29 Focus group 3, participatory action research workshop with foreign residents on insecurity.

30 Respondent 6, community artist, Imizamo Yethu, 24 November 2015.

31 Clive Ndou and Sapa, Taxis 'run by mafia, hit men', news24, 9 June 2005, http://www.news24.com/World/News/Taxis-run-by-mafia-hit-men-20050906.

32 Roslyn Bristow, City or their city? A case study of the Imizamo Yethu taxi industry and the MyCiti bus services in Hout Bay, MA thesis, University of the Western Cape, 2016.

33 Respondent 5, former ANCYL leader.

34 Development Action Group (DAG), Imizamo Yethu survey: report produced for the City of Cape Town, 2003, http://www.dag.org.za/images/pdf/research-reports/2003_Report_Imizamo%20Yethu%20Survey.pdf.         [ Links ]

35 Focus group 3, participatory action research workshop with foreign residents on insecurity.

36 Respondent 3, Zimbabwean national.

37 Usually localised protests about the failure of the (usually local) government to deliver houses, land or other so-called 'services'.

38 Respondent 4, taxi owner and leader of SANCO faction elected 2015.

39 Karl Von Holdt, The smoke that calls: insurgent citizenship, collective violence and the struggle for a place in the New South Africa, SWOP, 2011, http://www.swopinstitute.org.za/node/286.

40 Respondent 2, CDW.

41 Von Holdt, The smoke that calls.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License