Abstract
Drawing on research from five peri-urban sites across South Africa on how local government is responding to mobility, this research explores how xenophobia is being produced by local governance processes and structures. Building a better understanding of the mechanisms of exclusion in local government is essential not only for planning interventions that may strengthen democracy, but to understand how the daily practices of local government can promote, or undermine democracy.
Introduction
Not only does xenophobia exist in a context of exclusionary local politics, it is a product of this violence. This article emerged from a year of fieldwork looking at local government’s responses to mobility in five peri-urban areas across South Africa, and begins by describing the bureaucratic mechanisms driving exclusion. It will then explore the causes and trajectory of these processes, looking at some of the outcomes and sociopolitical structures that have been produced, ranging from a fragmentation of citizenship expressions, to a repoliticisation of identity, to increased dysfunction in planning and consultation processes. Finally, it will explore some of the implications this has for local governance and social cohesion.
South Africa is just over 20 years into democracy, and over the past two decades, ongoing efforts have been made to strengthen local government as a means of strengthening citizen participation in decision-making. The logic was that by bringing political decisions closer to the community, citizens will be empowered to engage with political processes. However, the outcome has largely been the opposite. Local government is driving fragmentation, alienation, and conflict. The state, particularly at a local level, while enabling access of some citizens to certain socio-economic rights, is at the same time denying people political agency (Landau, Segatti and Misago, 2011).
Following the wave of xenophobic violence in 2008 in South Africa, a range of studies emerged that explored the causes and consequences of such violence in communities (Crush, 2008; Misago, 2009). Not only in the media, but also in related research, the blame is often placed on unemployment, inequality, and competition for jobs and resources (Human Sciences Research Council, 2008). However, this study contributes to existing research challenging this assumption, and instead finding that foreigners are less the subject of discrimination from local government, but instead find themselves on the fault lines of local political party politics (Misago, 2009; Monson, 2011). Localised bureaucratic structures have been designed in such a way that, while not often deliberately discriminatory, they are exclusionary of mobility in their nature, and they perpetuate the politicization of identity.
This study set out to explore how the daily practices of local government officials affected migrants through the delivery of social services, which would help explain the underlying process of xenophobic systems. However, what emerged is that xenophobic practices were less linked to discriminatory practices or attitudes at the level of individuals in local government, and instead part of the politics of patronage found within local government. This is entirely in keeping with Wilkinson’s argument that violent ethnic mobilisations are not an outcome of polarised politics of identity, but rather the means through which stakeholders use identity to achieve a political outcome (Wilkinson, 2006).
In each of five sites studied, there were strong connections between local political party structures in the local elite, which use mobility as a means of accessing resources (through provincial and national structures), and as a tool for excluding competing political role players. In all 5 sites, an inductive approach was used, which included an initial site visit that resulted in a mapping of the main actors in the community, followed by the development of research tools to use for each. Key stakeholders were defined, and drawn up into an interview plan. There was then a return to the site, and over 40 interviews were conducted over the course of two months each. Participants included local government employees, councillors, political party officials, trade union and business leaders, church, non-governmental organisation, residents associations and other community leaders, and regional or technical bodies as relevant. The results of this research have several limitations. Researchers faced opposition within certain parts of local government. Additionally, while the number of sites in this study is not large enough to draw wider conclusions about governance across South Africa or beyond, the common trends across different sites do point to certain broader considerations for social cohesion, local identity formation, and localised citizen–state relationships in peri-urban areas.
Background to local governance and mobility in South Africa
Local government in South Africa was configured in its current form as part of the transition to democracy in the early 1990s. However, since then, it has changed in a range of ways, and relationships between residents and local government structures are informed by a much longer history. The Local Government Transition Act of 1993 established negotiating forums with local authorities, to pave the way for the first local government elections in 1996. Since then, however, the dynamic policy terrain on which local government works has been far from stable. For example, in 2007, an Act was passed regulating the fiscal powers of municipalities. A municipal demarcation board was established in 1998 to redraw municipal boundaries in advance of the 2000 elections. Of the five cases studied, not a single municipality’s boundary has remained the same since 2000. This is just to highlight the fact that while democratisation involved a huge shift in South Africa’s public administration, things have not been standing still since then. This is one contributing factor to the difficulties municipalities have faced taking root locally, but it is far from the only one.
Local government was coming into a space where there was very low trust in local political institutions. Local authorities who worked with the apartheid regime often received technical skills and education, but lost legitimacy, while local authorities that opposed the apartheid regime were more grounded in the politically active people in the communities they serve, but this often came at the expense of experience in government bureaucracy. These divisions did not immediately disappear with the transition to democracy, and while the dynamics changed when the African National Congress (ANC) was overwhelmingly voted into power, it is perhaps at the level of local government that these historic dynamics become more apparent (Glaser, 1998). ‘Elements of consociational government, having been denied nationally, were ensured at local level, meaning that significant decisions could not be taken without the consensus of minority residents. Although this was later overturned, as Robinson has observed, local government became the site on which existing privilege was most robustly defended’ (Beall, 2005: 11).
Locally, the transition to democracy did not necessarily change relationships between neighbours. Competition for jobs, political support from local businessmen, and other less tangible forms of support or patronage have remained largely the same, even if the structures at the top have changed (Beall, 2001). As a result, the legitimacy of local government is always tenuous, which produces a culture of impunity, corruption, and violence. Because of tightly enmeshed allegiances between local political, business, traditional and other leaders, it has been very difficult to engender a culture of democratic opposition.
This is far from disconnected from the dynamics of migration locally. Migration was long used as a tool for economic and social control. From the way mining hostels deliberately promoted the politicisation of identity to forced removals, mobility has played a critical role shaping South African society. Geographic spaces in South Africa are strongly interconnected, but have different cultures and politics. Former homelands have high dependence ratios, high rates of out migration, but often more established comparably more stable and sometimes homogeneous communities with greater social cohesion. Townships are on the periphery of urban areas, and tend to be both more recently established, and also more cosmopolitan. Populations are more linguistically mixed, more educated, and while unemployment is still rife, employment opportunities for people living in townships far outstrip those in former homelands. These spaces were often combined with traditional urban areas or towns in newly drawn wards during the demarcation process in 1999. However, the different elements of a ward often have little in common historically, geographically, or socially. This process reduced the number of municipalities from 843 to 284, and was widely seen to threaten traditionally held areas of authority (Goodenough, 2002).
All peri-urban areas are politically significant, even though they may be economically peripheral, because support from local structures of the ANC is essential for provincial and national political interests. The ANC is the political party that has held a significant majority since democracy, and people in positions of power within the party and within national government usually require support from local party structures in peri-urban areas to legitimise their position. ‘If you have a senior job in the provincial government, but are a junior in the party, you are weak. If you are senior in the party, you may be a clerk, but the world is yours’ (anonymous, 8 November 2013).
This generates a culture of patronage, and ties in directly to localised experiences of xenophobia that will be explored later in the article.
You need to be on everyone’s good side. It’s not that you don’t want to criticise when you see something wrong. It’s that you don’t have that confidence, if you criticise, to believe others will see the good in what you are doing. If you criticise, someone important may get angry, and you will lose your job. Not only will you lose your job, you will lose the support and protection you had in the community. This might be your wife’s small business, or the health care your brother was getting. No one is confident enough in their legitimacy to take stands. People need local support, and you get that by being on everyone’s good side (anonymous, 8 November 2013).
Theoretical background
There is little question that the capacity of local government to plan and respond according to demographic shifts is going to be a key factor in the delivery of public services in coming years. Urbanisation has reshaped Africa’s demography over the past 20 years, and continues to shape secondary cities, though it remains an understudied phenomenon. Similarly, promoting social cohesion will be a central challenge. Building on previous research in the African Centre for Migration and Society (Landau et al., 2011), looking at urbanisation, migration, and the distribution of power within African cities, this study chose a focus on peri-urban areas, because they are often at the limits of state control. They are also hubs of local migration, often places of flux for individuals and families between urban economic centres, and rural roots of extended families; this places them at the forefront of the way migrants relate to the state, where identities, belonging, and norms around rights and entitlement are shifting.
Although South Africa is now just over 20 years into democracy, promoting social cohesion and equity remains a huge challenge to government. Protests feature heavily in the political landscape, and range in focus from labour disputes to service delivery, xenophobia to political disenfranchisement. The currency these protests have gained over the last decade demonstrates that participation and political voice are being expressed through contestation around access to services and migration. Both have become important points of mobilisation for South African citizens; a symbolic territory for expressing an alienation from democracy – and simultaneously, a space of contestation around rights. These forms of mobilisation have ranged from overt violence, to Bourdieu’s (1989, 193) ‘gentle, invisible forms of violence….the violence of credit, confidence, obligation, personal loyalty, hospitality, gifts, gratitude, piety.’
In addition to the literature on state formation, this study engages with the literature on ethnic identity and constructivism. Hale (2004) claims that, ‘Ethnicity, then, serves to structure such action by providing people with social radar that they use to efficiently identify or impose social possibilities and potential constraints in a world of immense uncertainty and complexity.’ Rejecting both strictly rationalist and strictly constructivist approaches, Hale (2004) looks instead at ‘why and how people tend to think and act in terms of macro-level identity categories in the first place.’ This provides a link to the anthropology of local governance (Bernstein and Merz, 2011). As the case studies demonstrated, both the reasons people call on certain identity categories, and the resultant actions have direct bearings on the process of state formation at a local level.
Case studies and key findings
Five case studies were chosen for this study; all were peri-urban, and were chosen according to the importance of mobility in their local political landscape. They ranged in location from Gauteng, to the borders of Zimbabwe and Botswana in Limpopo and Mpumalanga. The sites include: Lephalale; Bushbuckridge; Hammanskraal; Mamelodi; and Temba.
Lephalale, a district in Limpopo with about 20,000 inhabitants, is a relatively new settlement, with the main town of Ellisras having been established only in 1960. Ellisras is a small town, and is skirted by a township; both mainly serve the local coal mine. There are villages in the periphery, which historically served as labour reserves to the farming land that dominates the area. While its economy had always been based largely on mining, the discovery that it sits on one of the world’s biggest coal reserves led to the construction of Medupi, one of the world’s largest coal-fired power stations. Over the past decade, people have flooded to Lephalale in search of work, and the social landscape of the formerly sleepy mining town changed dramatically. In addition to the labour migration, there had been an influx of contractors, skilled foreign workers, and others hoping to take advantage of the booming economy. Rapid growth, with both its positive and negative elements, was visible everywhere. As the economy grew, so did the level of political contestation, with violence becoming a common feature in meetings of the democratically elected local council, or ANC. While there had been incidents of xenophobia, the locality was heavily governed by fear and intimidation.
Bushbuckridge is a municipality whose land is under tribal authority; this is important because unlike other areas studied, in Bushbuckridge, there is a historic and direct conflict between municipal authority and tribal authority that existed well before democracy. The area was formerly under Lebowa and Gazankulu administration, and these divisions, which were built around politicising identity, remain relevant to understanding the current local government landscape. While Bushbuckridge is currently part of Mpumalanga, it falls on the border with Limpopo, and was part of a redistricting that was only finalised in 2006. This demarcation only took place after a protracted period of often violent protests. It is a labour sending area, with a substantial portion of the population working on mines in the Northwest province. This leaves the municipality with few streams of revenue generation, as there is a very high dependency ratio in the community. One of the results of Bushbuckridge falling under traditional leadership is that residents living in areas classified as villages do not usually pay rates or taxes. The same level of services is also not provided, such as trash collection, and owners are only given a right to occupy land, as opposed to title deeds for full ownership; commercial farms in the nearby area which are liable to pay significant taxes, are zoned into different municipalities. Due to allegations of mismanagement, the municipality was placed under provincial administration at the time of fieldwork, which essentially places all functions of both the municipal manager and the mayor in the hands of provincial authorities.
Mamelodi, Hammanskraal, and Temba are all peri-urban areas within the Tshwane Municipality. All are heavily economically linked to Pretoria and Johannesburg. The populations of all three areas are diverse, with Hammanskraal and Temba falling along the former border between KwaNdebele, Bophuthatswana and Lebowa, and all near the current border between Gauteng, the Northwest province, Limpopo, and Mpumalanga. While they share many demographic characteristics, localised party politics as well as expressions of xenophobia and community-led protests are significantly different, making them illustrative cases of how citizen–state relations are shaped to include or exclude on the basis of identity.
The administration of the state and roots of alienation
To have a better sense of both the mechanisms and expressions of xenophobic exclusion and violence, it is important to contextualise local government’s connection to citizens. Since the onset of democracy, the government has made many concerted efforts to strengthen downward accountability and citizen participation at a local level. Coordinated administrative efforts were supposed to strengthen democratic structures by linking people more closely to their service providers (Wunsch, 1998). However, local government entered the political scene with low levels of trust and embeddedness in the community. Local political structures were already fragmented, with loyalties spread, sometimes competing, sometimes overlapping between tribal authorities, the apartheid regime, and the liberation struggle. While realignment occurred as part of the process of political transformation, a local government has never emerged that has had a widely agreed upon function and legitimacy (Pycroft, 2000). Not only do local government structures behave in very different ways across geography, department, and demography, but citizens have not reached a shared agreement on expectations and interactions (Lalloo, 1998). As a result, norms are both fragile and highly localised (Weber, 1978).
Understanding some of the drivers of legitimacy and capacity within local government institutions will also help uncover some of the ways it responds inclusively or divisively to residents. One example that helps explain how local planning is informed is the case of redistricting, which affected every single site of study over the course of the last two decades. In only 20 years, not a single municipality studied was serving the same geographic community delineated in 1994. Add to this a highly mobile population, and it is no wonder that systems of citizen accountability have not matured, and local government control is vulnerable to political contestation and capture. As one political leader in Bushbuckridge explained,
People from Bushbuckridge didn’t have political influence during the time we fell under Limpopo. There was no representative in the provincial government. There are changes since Bushbuckridge moved to Mpumalanga. But the politicians are the bigger beneficiaries. There are now more representatives in provincial structures from Bushbuckridge, including the MEC for Agriculture (anonymous, 21 July 2013).
1
While local government was established with a great deal of initial hope and optimism in the potential for a democratic transition, the routes local government has had available to connect with communities have been through the very same fragmented local channels of power, that rely heavily on patronage and silence. It is then understandable that these structures have been unable to function in robust and democratic ways, that there is strong intertwining between political and economic decision-making, and that there are low levels of trust with citizens and multiple forms of exclusion, and violent and oppressive responses to dissent.
Recent research points to this mistrust being justified (Landau et al., 2011; Pithouse, 2016). These findings hint that the state has been largely unable to transform colonial patterns of subjugation. The extent to which local government is able to relate to residents as rights bearing citizens remains limited. As Pithouse (2016, 254) describes, ‘most oppressed people in the city are denied the right to assert themselves as political actors, as rational actors, on a shared stage, we are in the presence of attempts to silence the present. Attempts to silence the present are, among other things, attempts to make a very strong claim about who is in charge…Protagonists in this drama were not just the state, capital, patriarchy or white supremacy, but most certainly included local political elites, operating in a matrix shaped by all these forces, along with others. It is clear that their arsenal includes forms of authority and power that, while in part authorised and legitimated via liberal democratic arrangements, also operate outside the law.’
As mentioned above, there is an increasing body of research pointing to the inherently political nature of the daily administration of the state (Chalfin, 2001; Ferguson, 1994; Trouillot, 2001). Furthermore, the way identity is articulated and used to build communities of power, access resources, or criticise actions by the state is playing an important role in shaping the relationship between residents and the state (Foucault, 1991). For the purposes of this study, the day to day operations of municipalities have implications for social cohesion, intergroup violence, the localised formation of identity, and the relationship between citizens and the state, which is central for understanding the nexus between identity and democratic processes (Gramsci, 1971).
While the nature of mobility and socio-economic municipal profiles of each of the study sites were varied, there were certain characteristics they all shared. Across all locations, these included:
- low levels of trust between citizens and the state.
- political parties playing both a divisive role in communities, and gatekeeping roles towards government resources.
- patronage politics and violence remaining an important tool of political and social control.
- a lack of shared expectations among local government leaders, community members, and traditional authorities about the roles and responsibilities of different parts of local government.
- some level of xenophobic violence directly linked to party politics.
Some residents in peri-urban areas were proactive in articulating gaps in their democratic relationship with local government. In most cases, people linked this to the multiple nodes of power and accountability at the local level (Andersson and Ostrom, 2008). However, identity issues were also raised. What emerged most clearly is that there is no commonly held vision of who the municipality is, who they serve, and what their responsibilities are. Local government was conflated, or put into conflict, in various ways in different places, with the ANC, the private sector, traditional leaders, and other local authorities. Several respondents reflect on the role and identity of the municipality in their community.
The municipality is closer to big business like Eskom and Exxaro than the community. What we are saying is that, as local employees and local businesses together with the multinationals and the big nationals, we are the players, and the municipality should be the referee….the municipality is not doing that (anonymous, 29 May 2013).
This demonstrates both the interlinkages between political and economic power, and also the confusion across the board among local role players about the role of the municipality. Another respondent also highlights this lack of clarity between powers of the municipality, and traditional leadership.
There is a very big challenge with regards to the relationship between the municipality and traditional leaders. The fact is that they own the land. When the municipality wants to formalise a village, there is always that scuffle, that competition and disagreements on what can be done by whom, and what is within the powers of the other party (anonymous, 23 June 2013).
A further respondent underscores the same issue, but highlights the blurred lines between political power in the municipality, and political party structures by saying, ‘do you mean the municipality? Or do you mean the ANC? Because I think you know the difference. But I’m sure they do not.’ (anonymous, 26 June 2013).
A final respondent speaks not only to a lack of clarity about the identity of the municipality, but points to the kinds of behaviour this produces on the part of elected representatives. It highlights exactly how this lack of clarity and lack of legitimacy affects the behaviour of local politicians. ‘They (municipal councilors) keep themselves busy when the community wants to discuss what is happening. Or if they do avail themselves, they will give answers such as “I don’t know what the municipality is waiting for”. Who is the municipality?’ (anonymous, 24 July 2013).
One clear dynamic that arose in every single site studied was the rapidly evolving and central role of the ANC in shaping local government politics. While membership in the ANC was clearly a kind of currency for access to political and economic resources (from jobs to tenders from the municipality to access to provincial and national leadership), it was not in itself sufficient to signal ‘in group’ connections, whether due to factionalism within the ANC itself, additional gate keeping mechanisms put in place by local party branches, or links to national or provincial party structures. As one respondent said, ‘You want a job? Forget studying. Forget applying. Join the youth league. The day you sign up, you will have job offers. If it’s in government, it’s because everyone thinks you’ll support them. If it’s in business, it’s because they think you can get them tenders’ (anonymous, 25 March 2012).
This research further found that, as local government failed to provide services and play other roles that would legitimise it in the community, people have been responding by legitimising more diverse bases of power, including (depending on the site) traditional leaders, private service providers, businesses, and other non-state centres of power. However, these actors can also get pulled into a relationship or co-option or competition. Several respondents pointed to the fact that local authority is a highly dynamic and contested space.
We know local government won’t do most things for us. You can go to their office, and they will send you from pillar to post. You can spend years chasing after anyone, but nobody will take responsibility. At least if you go straight to Eskom, you might not get what you want, but people will be straight with you. This is the person who decides. You talk to them, and you get a yes, or no (anonymous, 23 August 2012).
On the surface, poor service delivery within municipalities, or evasive responses by officials might seem to be a de-politicised, public administration challenge, possibly linked to capacity or skills development; however, it will become clear in the following sections that it is deeply linked to the generation of local identity, as well as both violent and non-violent expressions of xenophobia. This fits well with the school of rational choice theorists who claim that identity and expression are intertwined, not only in the decision-making processes of local government officials, but in how residents then use their own identity to engage in the local political sphere (Calvert, 2002).
South Africa has a long history of institutionalising identity in politics, in a way that is changing with the way democracy has evolved (von Holdt, 2013). This is in part because new local government structures have needed to take root through political parties, community groups, and other existing sources of local power that had been developed through the politicisation of identity; however, this has historically been, and remains, a highly contested process. South African citizens are challenging the authority of the state to exclude, sometimes by manufacturing exclusion that is at odds with the state’s progressive policies of asylum (Landau and Monson, 2008), but at other times contesting this expression of state power through participation, inclusion, and solidarity (Pithouse, 2009). This is an issue of renewed importance to scholars, some of whom have suggested that identity is being politicised in new ways (Neocosmos, 2010). Understanding these changing relationships is critical for understanding the new interactions between citizens and the state in South Africa. As Pithouse (2016, 254) summarises, ‘It is also clear that this is not an aberration, nor a hangover from the past, but is in fact a constitutive feature of the present. At the same time, there is also an exercise of power from below that, in certain moments and places, raises real questions – material and symbolic – about the extent to which those authorised to be in charge are in fact fully in control of the people and spaces they are attempting to govern.’ This contestation is clearly lively, and some of the discussion below will outline the way in which it is manifest.
Jobs are an important currency for municipalities, particularly in peri-urban areas where rates of unemployment are high. Several respondents noted that access to jobs is tied to politics of identity: ‘To get a job, it’s about who you know. Nobody will argue with that. But when will we start calling a spade a spade? It’s tribalism’(anonymous, 17 July 2013).
Another respondent pointed to the complex nature of defining in- and out-group members in peri-urban spaces. At one site, foreign migrants speaking Shangaan were seen as local, while migrants from Gauteng were outsiders. Similarly, counterintuitive dynamics existed in several sites, pointing to the complexity of any blanket conclusions around where access, power, and discrimination may lie. ‘Xenophobia? We don’t have that here. Foreign migrants aren’t a problem; they’re mostly Shangaan anyway. But people from the Gauteng, like me? Everyone asks me what I’m doing here’ (anonymous, 21 July 2013).
It is evident that the democratic transition in South Africa has meant a series of rapid changes in the role of the state, the ANC and the relationship between the state and citizens. While there may be an initial anticipation that the onset of democracy would engage a disenfranchised citizenry, the reality has been much more complex. One product of this has been the emergence of ethnic discourses in the political sphere, and the role of identity in legitimising access to power.
Local powers and the creation of identity
Municipalities have a role to play in social cohesion, but the precise nature of their responsibilities is unclear. Specific pieces of legislation link infrastructure development and service delivery to social cohesion, and the role of municipal and provincial government is deliberately linked to nation building (Chidester et al., 2003). 2 However, the responsibility for social cohesion sits somewhere between the Presidency and the social cluster, which is a group of government departments brought together to integrate their work around government wide objectives (many of which are most active at the local level).
As both a product and reflection of this complicated location within the public administration, it is not clear how municipalities should act to promote social cohesion. Given that local political institutions have historically played a strong role in manufacturing difference, deliberate efforts would be needed to shift practice. An excerpt from the Integrated Development Plan of one of the sites studied is a beautiful illustration of the contradiction in local government, between an explicit mandate to create unity contrasting with practices of manufacturing difference: ‘It is characterized by high unemployment rate, poverty, unregulated influx in particular that of the Somali, Ethiopian and other foreign nationals. The Municipality has in the past been able to address the issue of social cohesion in ensuring that the setup of placing people along tribal belonging not exist (sic)’ (Office of the Municipal Manager, 2010). While an increase in the politicisation of identity has been more widely acknowledged by South African scholars in recent years, the role of both mobility and the state in shaping these dynamics remain largely overlooked (Crush, 2008; Neocosmos, 2010).
There is no question that local government is currently in crisis; only 5 municipalities in the entire country received a clean audit in 2012. Protests about local governance and service delivery issues have erupted in every province in the country, and they are a direct reflection of the crisis of legitimacy held by local government authorities (Alexander, 2010). Far from being about services themselves, these protests demonstrate the lack of voice, or the ineffectiveness of current channels of participation in local government (Pithouse, 2009). This tenuous hold local government representatives have on power further undermines their incentives to act democratically, and instead pushes them to become even more entrenched in systems of patronage. While many government employees are genuinely dedicated public servants, they are contending with systemic issues that make it difficult to work effectively. These systemic issues link strongly to the way a localised identity is formulated, and the result of this has implications for how citizen and state relationships can be used to either build or break down social cohesion at the local level.
The importance of labour migration specifically and its impact on governance was raised in both sending and receiving communities. It is integrally linked to local government’s planning challenges, but also to certain assumptions about what an idealised citizenry needs in a space. These expectations are still very spatially grounded, while the reality of residents is not. Some respondents pointed to the conundrum mobility created for local government, to have even the definition of ‘resident’ confused. Some people worked in one area, where they are assumed to pay rates or otherwise contribute to the local economy, while their families were in another area. Others arrived as job seekers, but only intended to remain in the area if they found employment. In one family, the province of employment, house construction, children’s school enrolment, and social grant collection were all different. Many people were investing resources in communities in which they did not live, making them more likely to want to influence local politics in absentia. This differentiated use of space and government services makes it difficult for municipalities to plan.
While the nature of population dynamics at each site means the scenario in local government varies, there are common themes concerning the difficulty of planning and budgeting, challenges to social cohesion, data availability and quality, and accountability and expectations of the municipality.
For communities receiving migrants, no matter the scale, this was seen as a major challenge to municipal capacity. Overwhelmingly, receiving migrants was seen as something negative, ‘A big workforce has converged in a small town. The small town cannot cope…’ (anonymous, 18 June 2013). However, others also acknowledged that it was an opportunity for growth, ‘I can’t say the migration is bad, it’s just not what we are used to. We wouldn’t have the mall if the town hadn’t grown. Now there are more jobs. People complain because there is also the negative aspect, but I think on balance, it is a positive’ (anonymous, 19 June 2013).
Respondents were more balanced in their view of sending migrants, with a mix of positive and negative qualities discussed. The economic advantages of migration were clear in sending communities, ‘There are beautiful houses near the tar roads, they are built on migrant remittances’ (anonymous, 26 July 2013). However, so were the social disadvantages, ‘This is where people go to die. Anyone who can work, leaves. Imagine living in a place like this, full of people facing their graves’ (anonymous, 19 July 2013).
Municipal employees in communities that had migration patterns entrenched over generations knew not only the economic and social impacts of migration, but had a strong sense of how mobility was impacting on local governance. Some municipal workers spoke only of the socio-economic impact of mobility, ‘Migration affects our work, because it affects families. Some children apply for grants because their father left to find work in Johannesburg, but no one has heard from him ever since’ (anonymous, 18 July 2013).
Other respondents linked mobility directly to democracy and social cohesion.
What does democracy mean in a place like this? The people who are earning the money that builds this place, don’t live here. Do they vote where they stay? Do they vote here? They might not even know the candidates. Is it better they vote for their friends? Power here has been moved to the mines (anonymous, 21 July 2013).
Chandra (2012, 5) points to a cyclical link between democratic processes and identity; he claims that identity ‘can be a product of the very political and economic phenomena that they are used to explain…. Welfare spending and public goods provision can create or change the ethnic identities presumed to affect patterns of welfare spending and public goods provision. And violence in its many forms can create or change the ethnic differences presumed to cause violence.’
The specific situations across all 5 case studies varied tremendously, with a common thread across each place being the strong (but not similar) role the ANC played in shaping the local political landscape, and using patronage to control state resources. These systems and processes are strongly tied to identity historically, but are now returning in new forms. This is in keeping with existing scholarship that acknowledges a close relation between ethnic diversity and patronage. As Chandra (2012, 37) states:
Previous research on the relationship between ethnic diversity and patronage can be divided into two schools of thought. The first school, represented by a voluminous body of work, argues that ethnic diversity impedes the distribution of public goods, pushing a political system towards patronage goods instead (Alesina et al, others). The second school, represented by a handful of more recent work, argues that the causal relationship goes in the opposite direction: patronage-based politics can make ethnic differences more salient (Fearon 1999, Caselli and Coleman, Bates 1974, Chandra 2004).
In one site, shifting allegiances that came with democratisation had as a consequence the dominance of the ANC, and therefore municipal resources, by one ethno-linguistic group, that was closely linked to the Bantustan government. In another, geography was used as a proxy for ethno-linguistic allegiances, with municipality officials working with the (homogeneous, resident) village, at the expense of the (diverse, mobile) township. In another, there was a widely held impression that the ANC had been ‘captured’ by migrants, because branch meetings were held in a specific language. Each of these stories reflects how tenuously local government maintains its legitimacy, and demonstrates how many other possible counter-narratives of legitimate power there are at a local level.
This demonstrates that there is a tension between an expectation that citizenship is geographically rooted and expressed as such, and that migration has long been an integral part of shaping the political landscape in South Africa. In turn, this means that there are questions about who constitute legitimate residents locally. One result is an inconsistency in the way mobility and identity are expressed in the political space. Another is the importance of identity as a political currency, driven by an uncertainty about how citizenship is or can be expressed legitimately at a local level. This contributes to a vicious cycle of confusion about the legitimacy, role, and mandate of local government, and the patronage, weak institutions, and silence in the face of xenophobia, violence, and corruption are all products of this.
The bureaucracy of exclusion
While data contributing to this research did not primarily focus on attitudes and practices of public officials, it nevertheless emerged as a theme with important implications for identity, mobility, and political control. It is also an important contrast to xenophobic violence, which may emerge from the same crisis of citizenship, but which is expressed in very different local government spaces. Attitudes of officials in public service ranged from sympathy to apathy towards foreign migrants. However, it was not immediately apparent that bureaucratic blockages or indifference were very strong drivers of xenophobia. While a full analysis of the attitudes of local government service providers is beyond the scope of the paper, a sampling and some implications are set out below, with a need for more focused research on the subject.
While there were negative sentiments about foreign migrants expressed by service providers at the local level, they were not only an exception, but importantly delinked to actual acts of generating violence. This opens a much wider issue of ‘softer’ forms of discrimination, which, while certainly present, do not seem to be a driving force in xenophobic violence.
In one municipality, a local political leader suggested a linkage between social grant fraud and international migration, ‘Some reside in the neighboring countries, and only come to South Africa to collect social grants and return to their countries’ (anonymous, 18 June 2013). A Mozambican migrant familiar with this accusation pointed to the absurdity of this claim, ‘Do they know what the bus fare is to Maputo? Does the South African government think it is that generous?’ (anonymous, 21 July 2013).
There was certainly evidence of a general lack of awareness of how migration impacts on service provision, but the only overt exclusion seemed to be around having a South African identity document (ID book), required to access most public services. Given the difficulty of obtaining an ID book, both for certain nationals and certainly for international migrants, this is an example of the exclusion of bureaucracy. However, the more evidence that emerged of various forms of bureaucratic exclusion, the clearer it became that the mechanisms driving overt violence against foreign nationals were in fact very different from the mechanisms driving more subtle forms of exclusion. Further examination would no doubt uncover a relationship between the varying vulnerabilities and forms of exclusion.
At a service-office level, while occasional xenophobic attitudes were expressed, the overwhelming majority of officials professed goodwill towards foreign migrants, but seemed to think their hands were tied due to bureaucratic requirements of an ID book.
We want to help them (foreign migrants), but we can’t. We have to follow certain processes. They don’t have an ID book. If someone comes without an ID book, and needs a food parcel, the first time, we give it to them, because they are hungry. We will work with home affairs. But if you find the next month, they still don’t have an ID book, we can’t give to them again (anonymous, 18 July 2013).
Another respondent echoed a similar sentiment.
Migrants, it is hard for us to serve them. They have the worst time, because they have all the problems we do, with domestic violence and child neglect. But we don’t have capacity to respond to them positively most of the time, and so it creates a situation where most cases are not reported. It’s not good. But we don’t have the facilities. I think a special office should be set up to deal with them (anonymous, 23 June 2013).
It is a testament to a relatively professionalised bureaucracy, even at a local level, to see officials repeatedly voicing their personal opinion, whether it is of sympathy with insiders or sympathy towards migrants, but also immediately and deliberately recognising the official policy on serving people.
In my job, I don’t say anything, because everybody has rights, and all people are the same. But if you ask me, as a person, maybe I shouldn’t say it, but I think, South Africans are better. You should hire them. They’re our people, we should support our own. But I know that’s just in my heart (anonymous, 16 November 2013).
While looking at the relationship between local government officials attitudes and practices would no doubt be instructive, it is unfortunately outside the scope of this study. What did emerge with the methodologies employed, however, was information about how bureaucrats’ actions are changing citizenship expressions. Attitudes of people both within and about the municipality are important markers of legitimacy.
One migrant expressed frustration with a lack of available services, ‘The municipality can do nothing for us. We don’t bother going to them’ (anonymous, 6 July 2013). The sentiment was echoed by a local resident, who attributed the situation to politics of identity, ‘I’m sorry to say, it never used to be like this. But now, if you go in, and if you want something, you say you are Shangaan. You might get it’ (anonymous, 7 July 2013).
What these quotes demonstrate are not uniform processes or practices of discrimination, but more an anecdotal sample of the varying ways bureaucracy can discriminate against migrants. This supports the assertion that the state has a role to play in the manufacturing of difference.
Xenophobic mobilisation and local politics
Both migration literature and some media reports often implicitly put forward the view that xenophobic exclusion is a result of overwhelming anti-foreigner sentiments in the general public, and that violence and systematic exclusion of foreigners has been a manifestation of that (Crush, 2008). This study has found that while xenophobic sentiments are rampant, mobilisation against foreigners, in particular violent mobilisation, has generally been an offshoot of competition for resources, and reflects broader trends in local group mobilisation. While xenophobia attitudes are widespread, it did not follow that they were linked to violence. It has been criminality either perpetuated or condoned by local elites, rather than the general public. As one respondent said, ‘I don’t like the Somalis; they take our business. So I’ll pay one rand more for bread somewhere else. But violence? Look, I’m not crazy. Just because I don’t like something doesn’t mean I can do whatever I want’ (anonymous, 16 September 2012). This view points to a culture of impunity required for violence, which is linked to systems of patronage that are rapidly and newly re-politicizing ethnic identity.
In four of the five sites studied, foreign migrants were particularly prone to being scapegoats, since they were a ‘soft target’ for any political party branch looking for popular support. In the 5th site, most foreign migrants shared ethno-linguistic ties with the party in power, which meant identity was articulated in a different way. The state will need to play a proactive role in generating social cohesion to overcome the current violent and divisive tendencies. Through a stronger understanding of the relationship between citizens and the state, as well as how this relationship can contribute to the politicisation of ethnicity, could provide tools for shifting the violent nature of political contestation.
Again, this is suggested by Wilkinson (2006), who argues that ‘ethnic riots are best understood, not as the outcome of already high degrees of competition, polarization and hatred between solid ethnic groups, but rather as the means through which political parties and political entrepreneurs construct solid ethnic categories, however briefly, for a clear political purpose’ (cited in Chandra, 2012, 37). This is heavily suggested in much of the literature, so it is not a particularly surprising finding. What is more surprising, however, is the extent to which it was recognised by people involved in local political processes.
In one site, all foreign run spaza shops had recently been closed down, with their owners run out of town. It was common knowledge within the community as to who led the attacks on the shopkeepers. ‘We know who attacks foreigners, and it’s not the community….They are thugs. The mayor protects them, they’re his friends. The police are part of that’ (anonymous, 23 June 2013).
While the acts of discrimination against foreign nationals were linked to local political structures, it would be simplistic to write it off as the actions of one particular faction. Rather, it is a product of the same lack of legitimacy that causes divides across local power structures. ‘Sure, last week they shut down the shops of the Zimbabweans. That’s not xenophobia. That’s the ANC at it again’ (anonymous, 29 June 2013).
The ANC was often evoked to explain local mobilisations against foreigners. As the party dominating every municipal council, it was, in each site in different ways deeply embroiled in the often violent struggle for legitimacy.
I can tell you, there are two causes of violence in our community. One is alcohol. The other is politics. If someone was killed, first ask, was anyone drunk? If not, you can be sure it’s politics. And I say this as a member of the ANC leadership. I know what I am talking about. I’ve had bricks put through my window in the middle of the night. Xenophobia? It’s all politics (anonymous, 25 May 2013).
Some respondents went so far as to equate xenophobic violence to the perhaps more studied and better understood service delivery protests. Both forms of community mobilisation seem to be produced by the same patterns of governance.
The newspapers say we don’t like people from Zimbabwe, that we chase them. Really? They are our brothers. When we were fighting apartheid, they helped us. Sure, maybe some people have prejudices, you can say they are ignorant. But do they cause violence? No. Open your eyes. Each time someone has been chased, if it is not the mayor himself standing there, it is one of his men. It is lazy reporting. Like service delivery protests; it just looks like people are violent, but that’s not the real story. But they don’t want to look there. Knowing the truth is inconvenient (anonymous, 29 June 2013).
Many respondents across the board were comfortable labelling xenophobia as ‘political’. In some places, this was because the ANC or councillors in the municipality were at the nexus of conflict. In other sites, however, it seemed like there was a widespread awareness of how corrupt the local governance space was for opportunists to act criminally or violently, even if it was not necessarily linked to formal structures of government.
Xenophobia? No. We don’t have that here. Just look, you will find some of the most successful business owners are from Mozambique. On our street, people are from Lesotho, Swaziland. Our children go to school together. What you have, are criminals. Some may be foreigners, and they give the others a bad name. Like how nobody likes Nigerians because some are selling drugs. But the real criminals, they are South Africans. They steal money, we call them successful. They are politicians (anonymous, 19 June 2013).
In all 5 sites, foreign shopkeepers were seen as visible, soft targets for people trying to score quick political or economic gain. ‘The people in the community are not against us. In fact, they like us. It’s only the politicians who chase us out. Someone thinks they can get something. The police look the other way while they loot’ (anonymous, 25 June 2013).
This sentiment was echoed by many respondents, who were quick to say that xenophobia was not a widespread social problem. While some data dispute this, it does seem true that xenophobic public opinion is not itself sufficient to explain the levels of violence seen in South Africa.
For the people in the community, we don’t have any problems. We visit our neighbours, we celebrate festivals together. I can even say South Africa is a welcoming place. But the politics! You have to be sure to keep your head down. You know every time elections are coming, you are at risk. Next time, I’m going to lock up and stay with my cousin in Johannesburg until elections blow over (anonymous, 25 June 2013).
Others acknowledged that there were antiforeigner sentiments in the community, but still insisted that some political catalyst was needed for violence to spark.
The problem isn’t with the people. You read the paper, it says xenophobia, xenophobia, I want to say, there is no xenophobia in this town. Nobody cares where you are from. The problem is that they care who gets the money. We used to be renting out space from the spazas from ANC people. They agreed to give us the space, and we would pay to build. Now that we built, they keep increasing the costs, more and more. If we rent anywhere else, they drive us out of town (anonymous, 28 May 2013).
Migrants themselves also had a nuanced view of the roots of xenophobia.
Of course, there are South Africans who don’t like us. They don’t know how to run a business. About buying in bulk, accounting. They’d rather chase us than learn. But that’s not what gives us trouble. What gives us trouble is that the mayor wants to score cheap points. He shuts down our shops, and then tells all the youth, look, I’m doing this to create employment for you. Now you can have jobs. What can we do? There are many unemployed youngsters, and they vote. It doesn’t make sense, but it keeps them in power (anonymous, 19 June 2013).
What has come out of the interviews is that foreign migrants are easy targets in local political battles, whether it is for control of certain revenue streams that are specifically associated with migrant groups, or electioneering. This seems to be a far stronger explanatory factor for the acts of xenophobia than any attitudes or actions among service providers. This is not to imply that there have been no scenarios in which xenophobic acts were populist, or caused by a range of other factors not linked to party politics. Rather, it is to illustrate that causes of xenophobia are highly localised, and the role of governance in generating violence has not been researched sufficiently.
Conclusion
The literature on the bureaucratic forms of exclusion at the local level, and the political dynamics that underlie them are largely separate conversations. This research makes a case for bringing them together. In South Africa, a stronger understanding of both the bureaucratic processes of state exclusion at the local level, as well as the way power in the community is exercised is a key factor for understanding the root causes of xenophobic violence. All forms of exclusion should be understood in a context of how identity is being politicised at the local level. In each case study, foreign migrants and the politics around them were a subplot in a larger story about political violence.
This study comes out in support of the idea that identity is largely used to achieve political objectives, and that an absence of strong and cohesive local identities leaves foreign migrants vulnerable to violence. Furthermore, local government structures have not managed to overcome the use of identity as a fulcrum of exercising political power. In fact, to the contrary, changes associated with democracy have created incentives to politicise ethnicity in new ways, which in turn threatens the building of a cohesive vision of citizenship.
Overlapping layers of local authority, combined with a lack of a cohesive vision of citizenship, have created a bureaucracy open to exploitation by people jockeying for resources and control. This has been combined with many of the individuals, systems and structures that were designed under apartheid to politicise identity for the purposes of exclusion which are gaining currency as the state becomes an important vehicle for political and economic security; violent or exclusionary mobilisation around ‘outsiders’ is one manifestation of this, but a part of the same process is a tribalisation of identities that is happening through patronage politics within the ANC. As such, xenophobic violence, police brutality, service delivery protests, and the suppression of civil society dissent, should all be discussed as part of an interconnected democratic crisis.
Footnotes
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: The UK Department for International Development which supports policies, programmes and projects to promote poverty reduction globally provided funds for this study as part of that goal but the views and opinions expressed are those of the author alone.