Linking governance and xenophobic violence in contemporary South Africa

Dr Jean Pierre Misago is a researcher with the African Centre for Migration and Society (ACMS) at the University of the Witwatersrand, Johannesburg. His research interests include migration, identity and belonging; xenophobia and violent outsider exclusion; humanitarianism; and governance of migration and human mobility.

Abstract

By demonstrating that local governance facilitates the occurrence of xenophobic violence by providing what I term favourable micro-political opportunity structures, the article argues that governance is a key determinant of xenophobic violence in South Africa and of collective violence generally. Research evidence (from extensive comparative empirical data and the global literature) informing this argument sits incongruently with the common and widely accepted understanding of governance and its relationship with collective violence. It shows that some aspects of this relationship are misunderstood and others are yet to be examined. Indeed, theoretical predictions in this regard indicate that collective violence and other forms of contentious collective action tend to occur in societies where mechanisms of social control have lost their restraining power. This article challenges these predictions by illustrating that, in most cases, xenophobic violence occurs in areas where social controls are strong and actually a facilitating factor. Further, the article indicates that the biggest misunderstanding of the relationship between governance and collective violence lies in interconnections yet to be examined. Such an examination would reveal the predominant role of governance, not only as a determinant, but particularly because of the significant role it plays in the making of violence co-determinants.

1. Introduction

Xenophobia in its various manifestations continues to threaten the lives and livelihoods of foreign nationals in South Africa. In particular, the August 2018 attacks on foreign nationals in parts of Johannesburg and North West province were yet another reminder that xenophobic violence, which generally refers to any acts of collective violence targeted at foreign nationals or ‘outsiders’ because of their being foreign or strangers (Dodson 2010:7), has become a perennial feature in post-apartheid South Africa (Landau 2011:3). Indeed, since 1994, tens of thousands of people have been harassed, attacked, killed or displaced because of their status as outsiders or foreign nationals (Misago 2016a:444). The main characteristics of this violence in South Africa include murder, assaults, looting, robbery, arson attacks (burning of people and property), displacement, intimidation and threats (Misago 2017:40).

The violence has prompted academics, political leaders, migrant rights organisations and other analysts to explain its causal factors and propose measures to stop/prevent its occurrence or at least mitigate its effects. This article aims to contribute to these ongoing efforts by providing a comprehensive empirically based and theoretically informed causal explanation for this violence. It does so by exploring the linkages between governance and the occurrence of such violence. Drawing on extensive comparative empirical data and the global empirical and theoretical literature, this article argues that governance (and more specifically local governance) plays a defining role in the occurrence of xenophobic violence in South Africa by providing a favourable political opportunity structure and by using social and political controls to facilitate violence rather than prevent it. While the focus is on governance factors that facilitate xenophobic violence, note is also taken of governance factors that are non-receptive of violence and can help to prevent it despite other violence determinants that may be present.

It is however important to emphasise that ‘governance favourable to violence’ does not necessarily mean or imply absence or weakness of governance. This insight could be the article’s main contribution to the literature on xenophobic and collective violence. Indeed, the article uses empirical data to interrogate widely accepted and time-honoured theoretical predictions that collective violence and other forms of contentious collective action tend to occur in societies where mechanisms of social control (particularly institutional leadership and authority) have lost their restraining power.

Building on the increasingly recognised understanding of governance as ‘the hybridisation of modes of control that allow the production of fragmented and multidimensional order within the state by the state, without the state, and beyond the state’ (Levi-Faur 2012:3), the article, for present purposes, uses the term ‘local governance’ broadly to refer to all formal and informal systems of order in a given locality or polity, i.e. the integration of – or interaction between – all localised systems of controls (social, economic, normative, legal, and political) and leadership, authority and power regimes.

2. Methods and approach

The analysis presented here is based on extensive comparative qualitative empirical data collected by the African Centre for Migration and Society (ACMS) at the University of the Witwatersrand from 2008 to date (cf. African Centre for Migration and Society (ACMS) n.d.). It draws more specifically on data collected in sixteen locations across the four South African provinces most affected by xenophobic violence (Gauteng, Western Cape, Eastern Cape and KwaZulu-Natal).

The specific aim of this on-going research is to explain the occurrence of xenophobic violence in affected areas. The focus is not so much on xenophobic attitudes as such, i.e. why many South Africans dislike or distrust foreigners and other ‘outsiders’ (Crush 2000:103). The research does aim, however, to explain why long-standing negative attitudes suddenly turn into organised and mass violence, to identify immediate triggers and conditions, and to explore why certain groups are targeted. What is also investigated, is why violence breaks out in some areas and not in others. To achieve this goal, the study adopts the ‘most similar systems’ approach by selecting research sites affected by the violence and sites that did not experience the violence despite having similar socio-economic indicators and dynamics. This approach is informed by the conviction that ‘no enquiry into riots [in this case xenophobic violence incidents] should fail to account for their absence’ (Horowitz 2001:xiv).

At each site, the above-mentioned research teams have been conducting in-depth, qualitative interviews with South African residents, foreign nationals, relevant government officials, community leaders, and representatives of different civil society organisations. In addition to individual in-depth interviews, the teams conducted an average of two focus group discussions at each research site. Thus far, the study counts more than 760 participants.

3. Local governance as a political opportunity structure for xenophobic violence in South Africa

In this section, I illustrate that local governance (both official and informal) regimes can provide a political opportunity structure for xenophobic violence in two ways. First, in many cases, official community leadership uses its authority and legitimacy to mobilise community members for violence. Second, in areas where official community leadership is weak or absent, violent alternative governance by non-state actors defines new forms of social control and authority that conceive violence as, not only a tool to consolidate their power and legitimacy, but also as legitimate means for protecting or restoring threatened local socio-economic and political orders. I start with a brief discussion of the general causal relationship between the political opportunity structure and collective violence to set the scene for the discussion of the critical role local governance plays in the occurrence of xenophobic violence in South Africa.

3.1 On Political Opportunity Structure and collective violence

Used more prominently in the social movement literature, the Political Opportunity Structure (POS) concept generally refers to a complex compound of formal and informal social and political conditions that facilitate the formation and/or operations of a social movement (McAdam 1999; Tarrow 1994; Tilly 1978). Vermeersch (2011:9) argues that the concept (and the related theoretical model) ‘responds to an intuitive feeling that social movements will act in accord with the institutional opportunities and constraints with which they are confronted in a given political system’.

With regard to contentious politics, which usually involves some form of collective action, POSs refer to ‘those aspects of the political system that affect the possibilities that challenging groups have to mobilize effectively. In this sense, opportunities are options for collective action, with chances and risks attached to them, which depend on factors outside the mobilizing group’ (Giugni 2009:361). As such, a POS is a balance between facilitation and repression. Indeed, Tarrow (1998, cited in Meyer 2003:19) notes that repression or facilitation of dissent by the state is a key aspect of the POS. According to Tilly (1978:100) ‘repression is an action by another group which raises the contender’s costs of collective action. An action which lowers the group’s costs of collective action is a form of facilitation’. Therefore, ‘Social movements must examine opportunity and threat and they must decide whether to act or not, based on that opportunity and/or threat’ (De Búrca 2009:6; see also Tarrow 1998).

Scholars have also identified POS as a key variable or determinant of other forms of collective action, particularly collective violence in its different forms which include political violence (Tilly 2003; De Búrca 2009), race riots (Lieberson and Silverman 1965), religious violence (De Búrca 2009), ethnic riots or pogroms (Bergmann 2011) and genocide (Bond 2007). De Búrca (2009:1), for example, explains that political violence by Hamas and the Irish Republican Army was in each case made possible by a favourable POS in terms of social sanction and political support by their constituencies. Similarly, Bond (2007:39) argues that mobilisation (by political leadership) for genocide in Rwanda succeeded due to a receptive socio-political setting. Bergmann (2011) further identifies the POS as a key variable in the outbreak and escalation of ethnic riots or pogroms. In his study of anti-Jewish violence in Europe, he provides evidence that pogroms ‘require a favorable political opportunity structure, in which the behavior of the government, the police, public opinion, and bystanders fulfills a key function in the outbreak and escalation of the violence in intergroup conflict’ (Bergmann 2011:489).

3.2 Local Governance playing a critical role in the occurrence of xenophobic violence in South Africa

Micro-political opportunity structures

Analysts usually use the POS concept in reference to national level socio-political factors or conditions that facilitate collective violence, with a particular focus on the state capacity (or lack thereof) to regulate and contain violent conflicts (see, for example, Meyer 2003 and Tong 1991). However, to understand the role of local governance in the occurrence of xenophobic violence in South Africa, I propose to extend the meaning and application of the POS model to subnational, local, community level socio-political arenas and their governance regimes. This is in recognition of the fact that national political systems often nest subnational power and authority regimes with significant relegated or appropriated autonomy that often translates into variations in authority patterns, institutional structures and political incentives.

As I argue elsewhere (Misago 2011:106), these variations mean that we must start thinking of the state in the ‘plural’ rather than the ‘singular’ in terms of (in the South African case, for example) its role, responsibilities, capacities and incentives at national, provincial and local levels (see also Boone 2003). The recognition of these subnational variations similarly challenges those who continue to speak of politics as fundamentally a set of national processes, and rather confirms the adage that ‘all politics is local’ (O’Neil and Hymel 1994:xv). Building on increasing recognition that POSs vary among subnational spaces within the same national political systems (Kitschelt 1986:63) and that there are often recognisable differences in the conduct of local state agencies and organs of control (Bergmann 2011), I argue that the POS model maintains its relevance and that its explanatory power (with regard to collective violence) prevails even when applied to local, community level socio-political factors and systems of order (i.e. local governance regimes). Indeed, the analysis of xenophobic violence in South Africa shows that local governance regimes present (independently or in conjunction with the wider national level socio-political systems) opportunities for – and constraints to – xenophobic violence. It is these locally generated opportunities that I term ‘micro-political opportunity structures’.

As the discussion below indicates, local governance provides a (micro-)POS for xenophobic violence in two ways: (1) in many cases, official local authority facilitates and is directly involved in the violence, and (2) in areas where official community leadership is weak or absent, violent informal leadership groups are provided with an opportunity to act.

Official local authority’s direct involvement, complicity and inaction: A perfect opportunity for xenophobic violence

The research this article draws from provides detailed evidence that, in many violence affected areas such as Alexandra, Diepsloot, DuNoon, Madelakufa II, Ramaphosa, Durban, the institutional local authority and official community leadership structures (e.g. local police, local Community Police Forum (CPF) branches, ward and street committees) permit or provide the means and incentives for xenophobic violence. It specifically shows that, in those areas, these structures permit the occurrence of the violence in a number of ways. They either (1) directly organise the violence and/or are actively involved in the attacks, (2) are complicit with instigators/perpetrators and sanction their actions, (3) passively encourage or tolerate the violence, or (4) do not make any effort to prevent the attacks despite visible warning signs (Misago 2016b:211).

In these cases, by commission or omission, the local leadership and authority provide a perfect socio-political opportunity for xenophobic violence. Indeed, using their clout, institutional authority and moral legitimacy, local governance regimes easily secure social sanction and normative acceptance; legitimise the violence by framing it as a necessary act of solidarity in legitimate self-defence; and lower the violence costs while raising its (real or perceived) normative and/or material benefits.

All these are acts of facilitation, characteristic of a favourable POS.

Examples from elsewhere also show that official community leadership structures can indeed provide a facilitating social and political opportunity and therefore play a critical role in the occurrence of collective violence. For example, Worsnop (2013:2) notes that strong community leadership structures are crucial in both starting and sustaining rebellions. Such structures are able to mobilise community members for participation by employing status rewards based on solidarity, enforcing social controls and norms, controlling the flow of information leading up to and during the collective action, and ensuring monitoring and concomitant sanctioning of undesired behaviour (Worsnop 2013:2).

Absent or weak institutional authority: An equally de facto POS for xenophobic violence

In addition to areas where the official leadership and authority is directly involved or tacitly supportive, the research finds that xenophobic violence also occurs in areas where local institutional authority and community leadership is weak or absent. The absence or weakness of the official community leadership is demonstrated by its inability to exercise its bestowed authority, and its subsequent inability to enforce the rule of law, and by its lost legitimacy, and lack of public trust. In these areas, the community leadership and authority have been usurped by informal leadership groups which residents considered more legitimate, more competent and more worthy of their trust.

The absent or weak official local authority and community leadership provides a favourable opportunity for xenophobic violence in three related ways: (1) its absence means there are no effective conflict resolution mechanisms able to diffuse tensions before they escalate into violence; (2) its absence also means impunity and lack of accountability for the perpetrators and instigators of xenophobic violence, and (3) its absence leads to the emergence of informal leadership groups that use violence to further their economic and political interests.

Many of the places where xenophobic violence occurs lack conflict resolution mechanisms capable of channelling or solving concerns in ways that could diffuse the socio-political tensions inherent in any diverse and dynamic community. Without denying that South Africa’s townships have a documented history of using violence as a means of solving problems, communities largely resort to violence, vigilantism and mob justice when relevant institutions and existing conflict resolution mechanisms have failed to adequately address issues of concern. The words of a respondent in Itireleng are telling in this regard: ‘If there are no other ways of resolving these problems even after several meetings, violence seems to be the only voice we have left’.

Respondents across all affected areas reported that the members of the community took the law into their own hands because they did not trust the local authorities and leaders or the police and the criminal justice system. With regard to xenophobic violence, the lack of effective conflict resolution mechanisms is particularly evident in local authorities’ failure to engage communities during the events that precede the attacks. In some affected areas, violence is fuelled by people’s frustrations over the inability or perceived unwillingness of local authorities (such as police, ward councils and CPFs) to address communities’ concerns/complaints (substantiated or not) with regard to the presence of foreign nationals in their communities.

In the face of the local authority’s inability or unwillingness to address communities’ concerns, instigators start organising mass meetings during which attacks on foreign nationals are publically planned. That the police and local authorities are aware that the attacks are being organised and do nothing to prevent them is further evidence of lack of effective mechanisms to resolve conflicts in communities. By allowing the public collective discontent and resentment towards foreign nationals in affected areas to fester and mobilisation for violence to take place and succeed, the lack of effective conflict resolution mechanisms presents a favourable opportunity for the occurrence of xenophobic violence.

Similarly, this research identifies impunity as a facilitation for xenophobic violence. It finds an endemic culture of impunity with regard to perpetrators and instigators of xenophobic violence. As indicated earlier, foreign nationals have been repeatedly attacked in South Africa since 1994 but few of the attackers have been charged and fewer convicted. Perpetrators are rarely arrested and where a few arrests are made, suspects get released without charges and in some cases with the assistance of local and provincial authorities. Further, repeated government promises to set up ‘special courts’ to deal swiftly with xenophobia-related crimes have never materialised. As other authors have observed (see, for example, Monson and Misago 2009), the National Prosecuting Authority (NPA) seemed to share – with political leaders of different levels – the lack of interest or incentives to hold the offenders of the xenophobic violence accountable. Monson and Misago (2009:30) note ‘…there is an evident lack of strong determination to hold the perpetrators of the violence accountable. … The actual and perceived impunity with which instigators and perpetrators of xenophobic violence are seen to act can only continue to encourage the ill-intentioned to attack foreigners and outsiders’. The inability or unwillingness of relevant organs of control to hold perpetrators and instigators accountable perpetuates a perceived sense of impunity that in turn encourages the continuation and the spread of the violence.

Studies elsewhere also confirm that impunity is common for acts of collective violence when the state organs of control and the population majority are not the primary target or when control organs and representatives of the majority are to a certain degree involved (Bergmann 2011). Black (1998:40) notes that ‘These crimes which are often perceived as “collective self-help” are usually treated comparatively mildly’. With ‘brutality greeted by impunity, and impunity greeted by indifference’ (Monson 2011:46), the lack of accountability in terms of prosecution and restorative justice (i.e. impunity) provides a favourable opportunity structure for violent attacks on foreign nationals. Indeed, impunity proves to be an excellent act of facilitation that lowers the hostile group’s costs of xenophobic violence.

In many affected areas, absent or weak local authority and community leadership leads to the emergence of violent alternative governance in the form of informal leadership groups that use xenophobic violence as means to consolidate their leadership legitimacy and consequently further their political and economic interests. That the absence of official local authority provides a POS for xenophobic violence by allowing these violent groups to emerge and use violence to articulate their interests is probably obvious. What is perhaps less obvious is that in these instances, local governance provides a double opportunity for the violence to occur.

First, by its absence, institutional governance indirectly permits the violence by allowing violent groups and other interested parties to organise and carry out attacks against foreign nationals. Second and perhaps most importantly, the absence of institutional governance does not mean the absence of governance altogether (see details in the sections to follow). It rather ushers in an era of alternative governance by non-state actors that defines new forms of social control and authority that see violence as not only a tool to consolidate their power and legitimacy, but also as ‘legitimate’ means of protecting or restoring threatened local socio-economic and political orders. This new and violent alternative governance provides an opportunity for xenophobic violence by its direct involvement and by its mobilisation of communities for participation. Therefore, the absent ‘old’ institutional governance and the present ‘new’ alternative governance provide (indirectly and directly, respectively) a double opportunity for xenophobic violence in affected areas.

4. Local Governance as an effective constraint to xenophobic violence

In previous sections, I discussed how local governance provides a favourable POS for xenophobic violence in affected areas. Here I wish to briefly highlight this research finding that without such an opportunity, xenophobic violence does not occur even when other determinants are present. The research provides detailed evidence (see Misago 2016b:141) that local authority and community leadership are instrumental in preventing xenophobic violence in potentially volatile areas by not only discouraging potential perpetrators from within but also and most importantly by successfully mobilising communities to stand against actions and influence from outside violent elements.

In these areas, local governance constrains rather than facilitates xenophobic violence and these cases present clear evidence that mobilisation for xenophobic violence does not succeed without a favourable POS provided by local governance. Local governance is the most significant distinguishing factor that explains the absence of xenophobic violence in those areas. It does not provide the micro-POS needed for violence to occur. As Monson (2011:189) confirms:

…the spread of [xenophobic] violence appeared to depend on the strength of leadership institutions in the surrounding areas. Arguably, more strongly democratic forms of leadership created firebreaks against the conflagration, while adjacent areas of weakly institutionalised leadership or leadership autonomous from the state presented softer boundaries, more easily penetrated both by political instigators and by the depoliticised spread of recidivism.

In sum, research evidence indicates that local governance (formal and/or informal) provides a favourable POS for xenophobic violence. This is an indication that xenophobic violence in South Africa occurs both in areas where official local authority is present and strong and in areas where it is weak or absent, and this finding is in line with research findings elsewhere. For example, Bergmann’s study on ethnic violence in 20^th century Europe concluded that waves of pogroms occur more often during periods when the state authority has either suffered a loss of power, making it less effective in exerting its control, or when it has become a party to the conflict (Bergmann 2011).

5. On social controls and collective violence: Interrogating the existing theory

The concept of ‘social control’ generally refers ‘to structures, mechanisms and strategies whose purpose is to cause society’s members to adhere to its valid norms and standards’ (Kirschner and Malthaner 2011:13). In its original and wider sense, the concept denoted a society’s capacity to regulate itself through social influence, and as such was contrasted with forms of state control that applied coercion (Kirschner and Malthaner 2011:13). Currently and increasingly however, the concept is used in a narrower sense to denote formal, state-sanctioned mechanisms and instruments as the police, the justice and prison system and institutional governance (Kirschner and Malthaner 2011:13; see also Horwitz 1990).

Regarding collective violence, there are widely accepted and time-honoured theoretical predictions that collective violence and other forms of contentious collective action tend to occur in societies where mechanisms of social control (particularly formal state mechanisms) have lost their restraining power. Indeed, a number of prominent theorists of collective violence (see for example Useem 1998; Smelser 1963; Horowitz 2001 and Tilly 2003) have long predicted that collective violence occurs when and where social controls are weak or no longer have deterrence power. Noting that ‘social control involves the institutionalising of respect for the law and for orderly means of resolving grievance’, Smelser (1963:261) argues that the occurrence of collective hostile outbursts signals the failure of agencies of control to prevent, interrupt, deflect, or inhibit the accumulation of the determinants that eventually produce such outbursts (Smelser 1963:261). The theoretical predictions of Tilly (2003:232) equally imply the correlation between the weakness or incapacity of state organs of control and the occurrence of collective violence. He notes that collective violence is more likely to happen in ‘low-capacity’ and ‘undemocratic’ states that allow violence entrepreneurs to operate.

This study on xenophobic violence (a form of collective violence) in South Africa challenges the above-outlined predictions on two main accounts. First, it proves these predictions not entirely correct by demonstrating that in most cases, it was the formal community leadership and authority that used its clout, institutional authority and moral legitimacy to successfully mobilise communities for violence (Misago 2016b). Here official, institutional leadership and authority was not absent or weak but rather it was its presence and power that facilitated the occurrence of the violence. In these cases, local official or state-sanctioned social controls had clearly not lost their restraining power. Instead of attempting to ‘restrain’ it, official social and political controls rather sanction and ‘facilitate’ the violence. This proves that collective violence also occurs in societies where official leadership and authority are strong and social controls are in fact a mobilising factor. Formal social controls provide a perfect POS for xenophobic violence to occur in areas where it does. Examples from elsewhere are plentiful. Analysts (see, for example, Hinjens 1999 and Mamdani 2001) note that the 1994 Rwandan genocide was not a result of the collapse of social controls or the inability of the national authority to enforce them. It is rather now common knowledge that state-sanctioned social controls (including government, the army and state-sponsored militia) supported and facilitated the violence rather than trying to constrain it. Similarly, Bergmann’s (2011) study of ethnic violence in Europe confirms this study’s finding that collective violence indeed happens both in areas where formal social controls are present and strong and in areas where they are absent or weak.

Second and perhaps more importantly, I argue that even in communities where institutional leadership and authority are absent or weak, social controls are not necessarily absent or weak – i.e., have not necessarily lost their meaning and power. The lack of state-sanctioned social controls does not necessarily mean that the entire control system (i.e. governance regime) has collapsed. Instead, the ‘unoccupied’ space allows the emergence of alternative, informal governance regimes that create new forms of social controls and socio-political order that supplant the older ones. Official leadership vacuums created by absent or weak institutional governance lead to the emergence of powerful informal community leadership structures that take over the authority of the state in their respective locations. These leadership groups forged their own laws and/or new law enforcement mechanisms (see Monson 2011).

This means that where institutional leadership is not trusted, its legitimacy and related socio-legal controls are questioned, confronted and where possible replaced by newly defined modes of social order that are perceived more legitimate and more relevant. Social controls are not always in line with state authority or institutional leadership regulations. This means that in those locations where official leadership is absent, new forms of social control emerge (or existing ones are redefined); new modes of enforcement are adopted and new custodians entrusted. As Monson (2011:172) correctly notes, during xenophobic violence these new custodians of local governance create new social controls and new enforcement mechanisms by either ‘making the law, breaking the law or taking the law into their own hands’. Demonstrating that xenophobic violence involves various levels of departure from the state-sanctioned social order, Monson argues that making the law, breaking the law or taking the law into own hands are three sub-national forms of sovereignty and political authority that help understand ‘…xenophobic violence as local-level appropriations of – or incursions into – one or both of the dual components of state sovereignty: legitimacy (or recognition as the “lawful source of socialpredictability”) and capacity to regulate (for instance, through its theoretical monopoly on mobility and coercive force)’ (Monson 2011:173).

These findings not only challenge the theoretical predictions outlined above but also the assumptions informing them. Indeed, predictions that collective violence tends to occur in societies where social controls have lost their regulatory capacity seem to be informed by two main assumptions that are equally poorly supported by empirical evidence: (1) the assumption that collective violence is an aberrant behaviour that social controls are there to prevent; and (2) the assumption that legitimate governance is a state monopoly (i.e. that the state is the sole producer and arbiter of legitimate governance and authority).

While increasingly rejected, the assumption that collective violence is an aberrant or anti-social behaviour which strong social controls (particularly those state-sanctioned) should be able, or should at least try, to prevent still persists and ‘still shakes the field’ (Roche 1996:98), as many ‘continue to characterize collective violence of some kinds by some people not only as deviant behaviour but also as undesirable and blameworthy – irrational, pathological, or criminal’ (Roche 1996:98). Horowitz (2001:35) similarly notes that there is an enduring ‘assumption that collective violence or crowds act in contradiction to values accepted in the wider society’. This article adds voice to calls that reject the assumption as mostly unfounded. It indeed illustrates that, (1) collective violence is not always regarded as anti-social or deviant behaviour but rather often enjoys social approval and is facilitated by the very same social controls; and (2) when collective violence enjoys social approval, those state-sanctioned controls that would attempt to prevent it would no longer be relevant or legitimate.

This research finds that, for many of those who are involved in the xenophobic violence – and for many who are not – attacking foreigners is a legitimate means of protecting South African lives, livelihoods and systems of order; a means of extending official law by other means (Monson 2011). Although some express sympathy with the victims, most respondents report that the communities in general support the attacks and feel satisfied that foreigners are finally being removed from their space and society.

A respondent in Alexandra, for example, states, ‘…others were crying with excitement; they were saying “at last action is taken against foreigners”’. In this case (and undoubtedly in many others), collective violence was considered as a legitimate means of protecting or restoring threatened local systems of order. Here collective violence acts as a form of social control or as a collective behaviour for which social controls are the dependent variable (Black 1990).

In line with this finding, this article affirms the emergence of an increasingly recognised but still unpopular conception of some types of collective violence as a form of social control (albeit not always in line with state authority and regulations), a form of self-help by a group, a form of protest, a quest for justice (no matter how justice is defined). For Roche (1996:101), for example, ‘collective violence is often an extreme form of self-help, a species of social control that entails the handling of a grievance by unilateral aggression’. Similarly, Gurr (1989, cited in Roche 1996:98) argues that ‘collective violence is now commonly regarded as a form of protest, a quest for justice, and the purposive expression of real grievances over underlying social, economic, and political issues’. Drawing from the ‘functionalist’ theory that explains collective violence in terms of its purpose and motives, Aya (1979:49) argues that, as an indicator of severe underlying social discontents and maladjustment in the community, collective violence is often a legitimate attempt to protect or restore threatened social, structural and material orders. It is motivated and triggered by the pressing need for redress of grievances. Looking at and analysing specific types of collective violence, different analysts have reached similar conclusions. For example, Bergmann (2011:488) defines pogroms as ‘a one-sided and non-governmental form of social control, as a form of self-help by a group that occurs when no remedy from the state against the threat which another ethnic group poses can be expected’; while Tilly and others (1975:85), drawing upon the analysis of European crowd violence, concludes that ‘justice lies at the heart of violent conflict’. Xenophobic violence in South Africa certainly is a form of social control at least in the eyes of perpetrators, sympathisers and custodians of the local authority. It is one of those newly and locally designed forms of social control, and given the extent of mass participation, popular support and social approval it receives, there is no doubt that it is often considered duly legitimate and effective by the communities concerned.

The second assumption informing the predicted effect of social controls on collective violence is that legitimate governance is a state monopoly or that the state is the sole producer and arbiter of legitimate governance and authority (particularly the legitimate use of violence as social control). This assumption flows from the Weberian understanding of the state as the only institution in the society, which has the monopoly on legitimate use of violence on society’s members (Wulf 2007). This understanding implies that state-sanctioned social controls are relevant or have currency in communities across the state’s territorial jurisdiction.

By demonstrating that xenophobic violence in most communities represents a local appropriation of state authority and the redefinition, reinvention and reclaiming of social controls (see also Monson 2011), this article argues that legitimate governance and use of violence are not the monopoly of the state. In those communities, governance and authority are exercised by non-state, informal actors that gain their legitimacy from the assent of both the governed and other local power holders. These alternative systems of authority are at liberty to use violence (e.g. xenophobic violence, vigilantism) whenever it is deemed to serve their interests and/or those of their constituencies. The use of violence to achieve societal goals has proven to be an effective type of service provision that confers authority and legitimacy to these non-state actors. Helping communities expel unwanted foreign nationals appears to be a highly appreciated service which the state has failed to deliver.

In many of the areas effected by xenophobic violence, the institutional authority has no normative power as a result of lack of public trust due to poor service delivery; has no coercive power due to weak and incompetent law enforcement agencies; and has no economic power due to lack of control over material resources. In other words, in the ‘Weberian’ sense, the local state has no power at all. The person who – or group that – governs those spaces is one that can claim and/or dispense at least one of those forms of power. Xenophobic violence provides that power because (1) it is in itself a form of coercive power in addition to other forms of vigilantism characteristic of those areas; (2) it mobilises the normative values of the communities (normative power); and (3) it distributes material resources through direct distribution of material incentives for participation in violence (e.g. housing or cash for the hired youth) or through the elimination of business competition (economic power).

Things are not necessarily different in contexts beyond South African borders. Indeed, many scholars (see, for example, Clunan 2010 and William 2010) note the emergence of alternative forms of authority and governance in a context of softening state sovereignty and show that increasingly ‘social structures exert authority over the control of violence’ (Clunan 2010:8). Others (see, for example, Baylouny 2010 and Arias 2010) demonstrate that, state incapacity and/or the exclusion of certain communities from mainstream official economic, social and political space creates room for alternative forms of governance to emerge, particularly over the provision of goods of policing and conflict resolution. Arias (2010, cited in Clunan 2010:8), for example, notes that in Latin America, ‘non-state actors wielding violence have become authoritative governors along with the police and unveils hybrid systems of urban governance’ which he labels ‘violent pluralism’. Similarly, Baylouny (2010) discusses how violent non-state actors in the Middle East gained authority and legitimacy through the provision of basic services of security to residents of marginalised spaces.

In sum, the case of xenophobic violence in South Africa and examples from elsewhere clearly demonstrate that the state is not the sole producer and arbiter of legitimate governance and has no monopoly over the legitimate use of violence. I therefore agree with those who argue that instead of calling locations where the state is absent or has limited influence ‘ungoverned spaces’ (see the discussion in Clunan and Trinkunas 2010 or Keister 2014); we should refer to them as places governed by alternative authority and governance structures led by non-state actors (Clunan 2010; Keister 2014).

These are unconventional ways of governance but governance nevertheless. As Keister (2014:2) correctly puts it, ‘Ungoverned spaces are actually not ungoverned, but exist under authorities other than formal states’. Similarly, Landau and others (2010:168) note that ‘the absence of state-centered, stable regulatory regimes does not reflect an ungoverned space, but a space that is alternatively governed’. Those areas demonstrate multiple layers of authority and ‘shared monopoly’ of legitimate use of violence (Wulf 2007) and render outdated ‘conventional accounts of the monopoly of force concept in which the nation-state is conceived as the sole appropriate agent’ (Wulf 2007:16).

6. Conclusion

By demonstrating that local governance (formal or informal) facilitates the occurrence of xenophobic violence by providing a favourable micro-POS, the article argues that governance is a key determinant of xenophobic violence in South Africa and of collective violence generally. Research findings supporting this argument sit incongruently with the common and widely accepted understanding of governance and its relationship with collective violence. It shows that some aspects of this relationship are misunderstood and others are yet to be examined. Indeed, the current understanding of this relationship revolves around the role social controls play in preventing collective violence. Many analysts define this relationship in terms of the capacity (or the lack thereof) of the state organs of control to prevent collective violence often perceived as anti-social behaviour. Theoretical predictions in this regard indicate that collective violence and other forms of contentious collective action tend to occur in societies where mechanisms of social control (particularly institutional leadership and authority) have lost their restraining power.

This article interrogates and challenges these predictions on two accounts. First, it proves these predictions incorrect by demonstrating that in most cases, xenophobic violence occurs in areas where local official or state-sanctioned social controls are not absent or weak. In these cases, instead of attempting to ‘restrain’ it, present and strong official social and political controls rather sanction and facilitate the violence. Second and perhaps more importantly, it demonstrates that even in areas where institutional local authority is absent, social controls are not necessarily weak. The lack of state-sanctioned social controls does not necessarily mean that the entire control system (i.e. governance regime) has collapsed. Instead, the ‘unoccupied’ space allows the emergence of alternative, informal governance regimes that create new forms of social control and adopt new modes of enforcement. The new local custodians of social controls use their authority to mobilise communities for violence. In both instances therefore, social controls are a facilitating rather than a restraining factor in the occurrence of xenophobic violence. This finding is further evidence that collective violence is not always regarded as a deviant behaviour which social controls are meant to prevent or contain.

Perhaps the biggest misunderstanding of the relationship between governance and collective violence lies in interconnections yet to be examined. Indeed, in addition to the often misunderstood effect of social controls (as discussed above), the analysis of causal factors of collective violence critically fails to detect the role governance plays in the making of other determinants. I argue that the inability to detect that aspect of the relationship makes current analyses incomplete. For example, while there is value in understanding the effect of governance on collective violence in terms of social controls ‘quashing the rebellion’ or preventing aggrieved and discontented group members from carrying out a collective violent act, there is also need to investigate the role governance plays in the making of that rebellion and collective discontent as well as in the framing of the target group as the source of group members’ frustrations in the first place. Such an investigation would probably reveal that governance plays a predominant role in the occurrence of xenophobic/collective violence, not only because of its role and efficacy as a determinant but particularly because of significant roles it plays in the making of violence co-determinants and their interconnections.

A more accurate understanding of governance and its multiplicity of modes, nodes, levels and actors would provide a more solid foundation for a better analysis and understanding of the causal relationships between governance factors and the occurrence of different types of collective violence including xenophobic violence in South Africa. Such an understanding would be the first and indispensable step towards more effective measures to address xenophobic/collective violence wherever it occurs.

Sources

1. African Centre for Migration and Society (ACMS) n.d. South African Chair in Mobility and the Politics of Difference. Available from: <http://thearmada.co.za/launch/sarchi/> [Accessed 31 July 2018].
2. Arias, Enrique D. 2010. Understanding criminal networks, political order, and politics in Latin America and the Caribbean. In: Clunan and Trinkunas eds. 2010, pp. 115–135.
3. Aya, Rod 1979. Theories of revolution reconsidered: Contrasting models of collective violence. Theory & Society, 8 (1), pp. 39–99.
4. Baylouny, Anne M. 2010. Authority outside the State: Non-state actors and new institutions in the Middle East. In: Clunan and Trinkunas eds. 2010, pp. 136–152.
5. Bergmann, Werner 2011. Ethnic riots in situations of loss of control: Revolution, civil war, and regime change as opportunity structures for anti-Jewish violence in nineteenth- and twentieth-century Europe. In: Heitmeyer, Haupt, Kirschner and Malthaner eds. 2011. pp. 487–516.
6. Black, Donald 1998. The social structure of right and wrong. San Diego, CA, Academic Press.
7. Bond, Michael H. 2007. Culture and collective violence: How good people, usually men, do bad things. In: Drozdek, Boris and John P. Wilson ed. Voices of trauma: Treating psychological trauma across culture. New York, Springer. pp. 27–58.
8. Boone, Catherine 2003. Political topographies of the African state: Territorial authority and institutional choice. Cambridge, Cambridge University Press.
9. Clunan, Anne 2010. Ungoverned spaces? The need for reevaluation. In: Clunan and Trinkunas eds. 2010, pp. 3–13.
10. Clunan, Anne and Harold Trinkunas 2010. Conceptualizing ungoverned spaces: Territorial statehood, contested authority, and softened sovereignty. In: Clunan and Trinkunas eds. 2010, pp. 17–33.
11. Clunan, Anne and Harold Trinkunas 2010. Ungoverned spaces: Alternatives to state authority in an era of softened sovereignty. Stanford, CA, Stanford University Press.
12. Crush, Jonathan 2000. The dark side of democracy: Migration, xenophobia and human rights in South Africa. International Migration, 38 (6), pp. 103–133.
13. De Búrca, Aoibhín 2009. Suicide missions: The ‘political opportunity structure’ and constraints of constituency. ConWEB – Webpapers on Constitutionalism and Governance beyond the State. Vol. 1. University College Dublin. Available from: <www.bath.ac.uk/esml/conWEB> [Accessed 20 November 2017].
14. Dodson, Belinda 2010. Locating xenophobia: Debate, discourse, and everyday experience in Cape Town, South Africa. Africa Today, 56 (3), pp. 2–22.
15. Giugni, Marco 2009. Political opportunities: From Tilly to Tilly. Swiss Political Science Review, 15 (2), pp. 361–368.
16. Heitmeyer, Wilhelm, Heinz-Gerhard Haupt, Andrea Kirschner and Stefan Malthaner 2011. Control of violence – Historical and international perspectives on violence in modern societies. New York, Springer.
17. Horowitz, Donald L. 2001. The deadly ethnic riot. Berkeley, CA, University of California Press.
18. Horwitz, Alan V. 1990. The Logic of Social Control. New York, Plenum Press.
19. Keister, Jennifer 2014. The illusion of chaos: Why ungoverned spaces aren’t ungoverned, and why that matters. Policy Analysis, (766), pp. 1–24. Available from: <http://object.cato.org/sites/cato.org/files/pubs/pdf/pa766_1.pdf> [Accessed 31 July 2018].
20. Kirschner, Andrea and Stefan Malthaner 2011. Control of violence. An analytical framework. In: Heitmeyer, Haupt, Kirschner and Malthaner eds. 2011, pp. 3–44.
21. Kitschelt, Herbert P. 1986. Political opportunity structures and political protest: Anti-nuclear movements in four democracies. British Journal of Political Science, 16 (1), pp. 57–85.
22. Landau, Loren B. 2010. Loving the alien? Citizenship, law, and the future in South Africa’s demonic society. African Affairs, 109 (435), pp. 213–230.
23. Landau, Loren B. 2011. Exorcising the demons within: Xenophobia, violence and statecraft in contemporary South Africa. Johannesburg, Wits University Press.
24. Landau, Loren B. and Tamlyn Monson 2010. Immigration and subterranean sovereignty in South African cities. In: Clunan and Trinkunas eds. 2010, pp. 153–172.
25. Levi-Faur, David 2012. From ‘Big Government’ to ‘Big Governance’? In: Levi-Faur, David ed. Oxford Handbook of Governance. Oxford, Oxford University Press. pp. 3–18.
26. Lieberson, Stanley and Arnold R. Silverman 1965. The precipitants and underlying conditions of race riots. American Sociological Review, 30 (6), pp. 887–898.
27. Mamdani, Mahmood 2001. When victims become killers: Colonialism, nativism and the genocide in Rwanda. Kampala, Fountain Publishers/Oxford, James Currey.
28. McAdam, Douglas 1999. Conceptual origins, current problems, future directions. In: McAdam, Douglas, John D. McCarthy and Mayer Y. Zald eds. Comparative perspectives on social movements: Political opportunities, mobilizing structures, and cultural framings. Cambridge, Cambridge University Press. pp. 23–40.
29. Meyer, David S. 2003. Political opportunity and nested institutions. Social Movement Studies, 2 (1), pp. 17–35.
30. Misago, Jean Pierre 2011. Disorder in a changing society: Authority and the micro-politics of violence. In: Landau ed. 2011, pp. 89–107.
31. Misago, Jean Pierre 2016a. Responding to xenophobic violence in South Africa: Barking up the wrong tree? African Human Mobility Review, 2 (2), pp. 443–467.
32. Misago, Jean Pierre 2016b. Migration, governance and violent exclusion: Exploring the determinants of xenophobic violence in post-apartheid South Africa. Unpublished Ph.D. thesis, University of the Witwatersrand, Johannesburg.
33. Misago, Jean Pierre 2017. Politics by other means? The political economy of xenophobic violence in post-apartheid South Africa. The Black Scholar, 47 (2), pp. 40–53.
34. Monson, Tamlyn 2011. Making the law; breaking the law; taking the law into our own hands: Sovereignty and territorial control in three South African settlements. In: Landau ed. 2011, pp. 89–108.
35. Monson, Tamlyn and Jean Pierre Misago 2009. Why history has repeated itself: The security risks of structural xenophobia. SA Crime Quarterly, 2009 (29), pp. 25–34.
36. O’Neil, Tip and Gary Hymel 1994. All politics is local, and other rules of the game. Holbrook, MA, Adams Media Corporation.
37. Roche, Roberta S. 1996. Collective violence as social control. Sociological Forum, 11 (1), pp. 97–128.
38. Tarrow, Sidney G. 1998. Power in movement: Social movements and contentious politics. Second edition. Cambridge, Cambridge University Press
39. Tilly, Charles 1978. From mobilization to revolution. Reading, MA, Addison-Wesley.
40. Tilly, Charles 2003. The politics of collective violence. Cambridge, Cambridge University Press.
41. Tilly, Charles, Louise Tilly and Richard Tilly 1975. The rebellious century, 1830–1930. Cambridge, MA, Harvard University Press.
42. Tong, James 1991. Disorder under heaven: Collective violence in the Ming Dynasty. Stanford, Stanford University Press.
43. Useem, Bert 1998. Breakdown theories of collective action. Annual Review of Sociology, 24, pp. 215–238.
44. Vermeersch, Peter 2011. Theories of ethnic mobilization: overview and recent trends. Centre for Research on Peace and Development (CRPD) Working Paper No. 3. University of Leuven. Available from: <http://www.kuleuven.be/crpd> [Accessed 11 July 2018].
45. Worsnop, Alec 2011. Insurgent organizational structure and the control of collective violence. Précis, Fall 2013. Available from: <http://web.mit.edu/cis/precis/2013fall/insurgent_org.html#.VFdbNTSUfZM> [Accessed 10 August 2018].
46. Wulf, Herbert 2007. Challenging the Weberian concept of the state: The future of the monopoly of violence. The Australian Centre for Peace and Conflict Studies Occasional Papers Series. Number 9, December 2007. Available from: <http://www.wulf-herbert.de/ACPACS-occ-paper9.pdf> [Accessed 21 October 2018].