Violent xenophobic episodes in South Africa , 2008 and 2015

A comparison between the 2015 and 2008 violent xenophobic episodes in South Africa is made, identifying both similarities and differences. Since government and other commentators, in both 2015 and 2008, differ considerably in deciding whether these episodes ought to be considered as essentially xenophobic in character, a number of prominent interpretations employed by these commentators will be identified and discussed. The perpetrators mobilised during these events are drawn from the same pool as the protesters mobilised during current country-wide violent service delivery protests. Accordingly, this article concludes by suggesting that an insightful interpretation covering both categories of actors belonging to South Africa’s urban working and under-class may be made. The violent collective behaviour that is becoming widespread country-wide stems, to a significant extent, not from their deep-seated xenophobic attitudes, but rather from the unfulfilled expectations of what they believe should be their just entitlements.


Introduction
Before the main theme of this article is identified, an anecdote that plays the role of preamble is recounted.It foregrounds a number of issues directly relevant to the main theme.
In June 2015, during a visit to Cameroon in West Africa, I was travelling south of Yaoundé, the capital city, with a doctoral student and his brother to visit the two men's parents.Since Cameroon is one of the countries confronted with the Boko Haram threat, there were several roadblocks in the rural forested region through which we were travelling.At each, I was requested, as a foreigner, to step out of the car and show my passport to the uniformed gendarme responsible for passing cars and their passengers."You are a South African, monsieur?"asked one such gendarme and when I replied in the affirmative, he continued by telling me that he and his fellow security officers treated foreign visitors to Cameroon with courtesy."Do you in South Africa do the same thing, for foreigners from Africa?" he then asked with perhaps a touch of scepticism.I had no ready and credible reply to this.
On the 18 th of April 2015, Emmanuel Sithole, a cigarette and sweet hawker and Mozambique national living and working in Alexandra Township in Johannesburg, was attacked and stabbed in the heart by four young male residents.He later died after journalists rushed him to hospital.It is probable that the Cameroonian gendarme knew of South Africa's less than courteous treatment of foreigners through Emmanuel Sithole's sad story.A South African photojournalist's sequence of remarkable photographs of the build-up to the murder, some of which appeared on the front pages of South African newspapers, were also broadcast globally through social media.This deadly incident that took place during the series of closely-knit violent attacks on the persons and property of foreigners at that time, in the words of a South African journalist, gave "a face, a name, a life and a personality" to one of the targets of this violence, thus providing "humanising elements that have been absent in much of the (media's) coverage" (Harber 2015).They also led to criticisms by senior government members that wide-reaching broadcasting of such material damages South Africa's international reputation.
That South Africa's image as a safe refuge for African asylum-seekers and as a hospitable territory for economic migrants from the developing world has been further diminished in 2015 is common cause.This article will compare the 2015 series of closely-knit violent events to that of 2008 when a similar series took place.The comparison between what hereafter will be referred to as the 2015 and 2008 xenophobic episodes, will identify both similarities as well as differences.The focus of the comparison will be on the geography and chronology of the events, on both civilian perpetrators as well as civilian victims, and on government reaction to the continuing outbursts.In the second place, since government and other commentators, in both 2015 and 2008, differ considerably in deciding whether these episodes ought to be considered as essentially xenophobic in character, a number of prominent interpretations employed by these commentators will be identified and discussed.
The perpetrators mobilised during these events are drawn from the same pool as the protesters mobilised during current country-wide violent service delivery protests.Accordingly, this article will conclude by suggesting that an insightful interpretation covering both categories of actors belonging to South Africa's urban working and under-class may be made.It will be argued that the violent collective behaviour that is becoming widespread country-wide stems, to a significant extent, not from their deep-seated xenophobic attitudes, but rather from the unfulfilled expectations of what they believe should be their just entitlements.

Conceptual Framework and Methodology
The rise of xenophobic attitudes in South Africa is not a new topic of research (Nyamjoh 2006).The 2008 series of violent events in urban South Africa shifted the research focus toward analyses of violent collective action and the various causes of such action.This in turn required a wider conceptual framework, in particular to include analyses of cross-border migration streams, state migration policies, and the attitudes and actions of the members of the South African Police Service (Hassim et al. 2008, HSRC 2008, Misago et al. 2009).It also required an approach to analyse violent collective action itself, leading to debates regarding the utility of relative deprivation and resource mobilization approaches (Bekker 2010).In the analysis below, particular attention is given to the approach employed by Donald Horowitz (2001) in which social psychological as well as sociological concepts are employed and the roles of both precipitating factors and of rumours are foregrounded.
The methodology employed during these analyses begins with a comparison between the 2015 and 2008 xenophobic episodes where each episode comprises a series of closely-knit violent events identified in the print media.The validity of such event data sets may be taken to be acceptable whilst acknowledging that missing data may be the most serious form of description bias (Earl et al. 2004).Subsequently, a comparison is drawn between primary stakeholders' interpretations of the causes and nature of these 2015 and 2008 episodes.These interpretations are identified both in publications by members of the research community as well as in the print media.In the third place, a preliminary analysis of a recently constructed event data set of service delivery protests in South Africa (Powell n.d.) is made with the aim of revealing resemblances between this analysis and those undertaken earlier.

Outbursts in KwaZulu-Natal and Gauteng, March and April 2015
On Monday the 30th of March, the 2015 violent xenophobic episode began in Isipingo, an industrial and residential area in the eThekwini metro south of Durban's CBD.The probable precipitating factor was the rumour that local businesses were hiring foreigners to replace local workers involved in industrial action and wage disputes.After a series of attacks on individuals in the area, some 240 displaced foreigners reported to the local police station and were subsequently moved to a tented transit camp.This first outburst was followed by a series of others in the metro, particularly in the northern township of KwaMashu and in the inner city residential area of Verulam -more specifically, in its informal residential neighbourhoodswhere foreigners' shops were looted and burnt down and foreigners themselves attacked.This onslaught continued for two weeks and included large crowds of locals marching on a number of occasions, with the ostensible purpose of driving migrants out of Durban.In response, local foreign migrants organised a demonstration in the metro's CBD during which clashes took place between the demonstrators and the South African police.
Temporary shelters to accommodate the displaced foreign nationals were established in Isipingo, Chatsworth and Greenwood Park (located in Durban North)."The municipality has supplied tents, electricity, showers, ablution facilities and primary health care in the form of mobile clinics where the displaced foreign nationals have been accommodated," an eThekwini spokesperson said (Sosibo 2015).With seven people killed at the height of the violence, it was reported that many foreign nationals still feared for their lives.The temporary shelters were replaced by four camps in the province -in Chatsworth, Phoenix, Isipingo Beach and Pietermaritzburg -which together accommodated about 5 000 people.The camps were situated on grounds that were not intended for habitation and, as a result, sanitation was a problem (Sosibo 2015).Some two weeks after the commencement of these outbursts in KwaZulu-Natal, in anticipation of a similar wave in Gauteng, foreign shop owners in the CBD of Johannesburg closed their doors in a bid to protect their stock amid rumours circulating on social networks that attacks were imminent.This was three days before the murderous attack on Emmanuel Sithole in Alexandra Township.
The anticipated assaults soon materialised with foreign-owned businesses in the inner city areas of Jeppestown and Cleveland, and later in the East Rand, being attacked and looted.Simultaneously, as was the case earlier in Durban, locals blocked roads with rocks and burning tyres, purportedly to drive foreigners out of South Africa.The violence was reported to have claimed the lives of seven people, three of them South African.Simultaneously, camps for foreign nationals fleeing the violence were established close to the CBD and in Germiston on the East Rand.
The South African police were called upon, in both Durban and Johannesburg, to contain the violence.Much scepticism about police effectiveness and neutrality was expressed both in the media and by many on the ground (Sosibo et al. 2015, News24 2015).Eventually, during the third week of April, the South African National Defence Force was deployed to "volatile areas" to prevent attacks on foreigners, as stated by Defence Minister Mapisa-Nqakula.She was reported to have said that the army was intervening because an "emergency" had developed (BBC News, 2015).The first deployments were to Alexandra, the poor township north of Johannesburg where Emmanuel had been hawking his wares.

Build-Up and Immediate Outcomes of the 2015 Episode
Though numerous violent outbursts have taken place during the past seven years, these have not been as intense and as closely-knit as the two episodes during April 2015 and May 2008 that are compared here.However, seemingly isolated outbursts ought not to be seen as once-off and unrelated to earlier and later events as a local history of violence against strangers, mixed with social and mass media coverage of such recent violence elsewhere may become a potent combination.Such an event took place in Soweto, Johannesburg's largest township, in January 2015.What was described as a looting frenzy of foreign-owned shops in Soweto broke out and later spread to other townships in Gauteng.It left six dead and large numbers of Bengalis, Pakistanis, Ethiopians and Somalis displaced (Sosibo et al. 2015).
Gauteng provincial politicians put these attacks down to economic reasons: "The recent attacks are because township entrepreneurs feel demoralised, frustrated, and they feel they cannot thrive as business owners in their own communities," Gauteng's economic development MEC said a week after the outbursts (Sosibo et al. 2015).National Small Business Development Minister Lindiwe Zulu added, "You cannot run away from the fact that there are underlying issues and that our people are being squeezed out by these foreign shop owners" (Sosibo et al. 2015).On Monday 16 March, some two weeks before the start of the outbursts in KwaZulu-Natal, a public announcement by the Zulu King called for foreigners to "take their bags and go" (Ndou 2015).While addressing the rural Pongolo community in isiZulu during a moral regeneration event, King Zwelithini was reported to have accused government of failing to protect locals from the "influx of foreign nationals"."Most government leaders do not want to speak out on this matter because they are scared of losing votes.As the king of the Zulu nation, I cannot tolerate a situation where we are being led by leaders with no views whatsoever" (Ndou 2015).
The dissemination of credible rumours, whether these reflect the precise statements of influential public figures or not, can prove to be a potent form of contagion (Horowitz 2001: 74f): once a violent event has taken place, violence may spread to another place where precipitants at the subsequent location are less significant than they were at the first.Violence does not occur in isolation; it derives intellectual impetus from events regarded as comparable elsewhere (Bekker 2010: 144).The interpretations at local community level made of Minister Zulu's statement in Gauteng and the Zulu King's speech among the predominantly isiZulu population of eThekwini are probably cases in point.
Two developments in the aftermath of the 2015 episode will be identified.They have been chosen since comparable outcomes did not take place after the 2008 episode.They have also been chosen since they may be seen as different sides of the same coin: from foreign African governments' points of view, the South African government's response to the 2015 episode has been visibly unapologetic.Conversely, the South African government condemned the Nigerian government for the withdrawal of its envoys and established a special body to search out criminals (both foreign and South African) for extradition and prosecution.
First, there was a wide-ranging and deeply critical reaction to these events by a number of sub-Saharan African governments, some of which went as far as threatening reprisals through the deportation of South African nationals and their businesses from their territories.Political leaders from Malawi, Mozambique and Zimbabwe publicly condemned the attacks, with both Malawi and Zimbabwe sending buses to repatriate citizens following the violence.Mozambiquan workers at South African gas companies downed tools in protest.Nigeria's acting high commissioner in South Africa and its consulgeneral were called home to brief the Nigerian Parliament about the welfare of Nigerian citizens in South Africa following what was called "the antiimmigrant violence" in Durban and Johannesburg (Visser 2015).
Second, one week after the termination of the 2015 episode, there was the establishment of a national government programme aimed at ridding "our country of illegal weapons, drug dens, prostitution rings and other illegal activities" (Hunter 2015).Named Operation Fiela-Reclaim and initially carried out by the South African National Defence Force, it was launched by a national inter-ministerial committee on migration.During its first week of operation, 265 suspects had been arrested and charged in relation to 150 cases of public violence around the country, according to Minister Radebe.Simultaneously, he announced that government was working hard to ensure that 1 507 documented foreign nationals awaiting repatriation would be sent home as soon as possible and stated: "We have thus far repatriated a total of 1 997 undocumented migrants from both KwaZulu-Natal and Gauteng temporary shelters" (Hunter 2015. The South African National Defence Force (SANDF), as mentioned above, was first mobilized about a month after the 2015 episode began.It was subsequently requested to launch Operation Fiela.In 2008, during the violent xenophobic episode of that year, the SANDF was also called upon to intervene toward the end of the episode.However, no subsequent operation was launched.We now turn our attention to 2008.

Outbursts in Gauteng and the Western Cape, May and June 2008
During the second half of May 2008 (and continuing through the first week of June), a series of short violent outbursts took place in neighbourhoods of numerous South African cities and towns.The violence during these outbursts was committed by civilians and was inflicted on the property and the person of civilians.The perpetrators were largely young and middle-aged poor black South African men; the targets were largely the property and businesses of foreign African nationals as well as these civilians themselves; and the locations were predominantly urban informal settlements, townships and hostels.The series of outbursts began in Gauteng and, about a week after the first serious event, spread to other urban areas of the country, Cape Town and the Western Cape in particular.Initial state reaction was evasive, essentially denying the scope and seriousness of these events.Subsequently, as the series of events spread across the country, the state sought explanations in criminal and mob behaviour.This geographic spread of outbursts was accompanied by widespread coverage in the mass media -television, radio and newspapersof these events and their possible causes.Since the reaction of many of the victims was flight from their residential areas, a number of temporary refugee camps were established (in Gauteng and Cape Town, in particular).During the aftermath of these outbursts, more than 20 000 refugees were accommodated in this way, numerous African foreign nationals were reported to have left the country, and government urged refugees in camps to return to the residential areas from which they had fled since these were said to have calmed down (Bekker et al. 2008;Bekker 2010).
In order to distinguish the violent events that took place during the first week from those that occurred subsequently in May and early June, news articles on the 2008 episode captured from the print media were filed under an event name (typically the locality where the outburst took place) and a standardized set of event data was assembled for each.These data included date, duration, type of settlement, nature of violence, reported precipitants and rumours and the nature of police intervention.In Tables 1-4 below, events have been classified in terms of the province in which they occurred, the date on which they took place, whether they were major or minor in intensity and whether they included violence against persons or solely against property.The first three tables depict events that took place in the periods 10 to 20 May, 21 to 31 May, and during June 2008, respectively.The fourth table aggregates these results for the entire period.
As is clear from Table 1 (first phase), the majority of outbursts during the first phase took place in Gauteng.In addition, all major outbursts were situated in Gauteng and comprise more than a third of all events in that province.In fact, the most reported deaths during the three-phase period under scrutiny took place in this province during the first phase.Four events were reported to have had identifiable precipitants, Alexandra -a black township in Johannesburgbeing the most detailed.Significantly more reports identified rumours that were claimed to have fuelled the violent outbursts.A minority of reports on these events indicated police presence and police intervention, most often detailed as attempts to disperse crowds.During the middle phase, comprising ten days (see Table 2), the print media reported a total of sixty-three events.Thirty-two of these events took place in the Western Cape, thirteen in KwaZulu-Natal, seven in Gauteng, and eleven in other provinces.From the media reports, it appears that xenophobic violence in this phase was not as violent as during the first phase.The majority of events were classified as minor events on property alone.Police intervention was reported in a number of cases, most often involving the arrest of perpetrators.These events appear to involve opportunistic behaviour more often than in the earlier phase since media coverage and rumours involving earlier events led many foreigners to depart from their residences and abandon their property.This anticipation of possible attacks created a context in which locals could vandalize and loot homes and shops belonging to those who had fled.
In the final phase (see Table 3), ten of the eleven reported events were minor with half involving assault on persons.This final phase -effectively comprising of the first half of the month of June -reflects a diminishment in the frequency of events in all four classes as well as a (late) diffusion to new provinces.This phase may be seen, at least in terms of print media coverage, as the petering out of the series of outbursts country-wide.In the aggregated data of outbursts in Table 4, it is worth noting that most major events took place in Gauteng during the first phase, and that most outbursts during the whole period under scrutiny were minor events involving vandalizing of property and looting rather than assault on persons.The

and 2015 Episodes: Similarities and Differences
There are a number of striking similarities between the 2015 and 2008 series of violent events.To begin with the geography and chronology of the two episodes, the outbursts were principally confined to urban townships, inner city residential areas and informal settlements of South Africa's metros.It is worth noting that local perpetrators also employed the CBDs of Durban and Johannesburg in the 2015 episode.The location of these events are often at a substantial distance from one another, implying that communication regarding justifications for violent action by word-of mouth would have been well-nigh impossible.In both episodes, chronologically, outbursts began and, in large measure, spread rapidly within one metro for some ten days before shifting into a second metro.In both cases, after two to three weeks, the series of closely-knit outbursts petered out at approximately the same time, as the state decided to mobilize the South African Defence Force to deal with what they belatedly called an emergency.
The second striking set of similarities is found in the rapid diffusion of information across neighbourhoods of metros as well as between metros themselves.Rumour, justifications for violent actions, and shared belief in immunity from punishment were communicated within the mass medianewsprint, television and radio (in various languages) -as well as via social media.The latter probably played a more important role in 2015 than in 2008 since more residents were capable of using social media tools during this later episode (Worldwideworx n.d.).In both Cape Town in 2008 and Johannesburg in 2015 (the second metros to be affected in each respective episode) foreigners who had been informed via such media of outbursts elsewhere anticipated attacks a day or two before they began.However, the impact of these outbursts in Cape Town, where most events were opportunistic since foreign residents and shop-owners had fled their homes and shops before the attacks began, differed from the impact in Johannesburg where events appear to have been as violent as in Durban.
Similarities regarding the content of information being diffused before as well as during the outbursts are also striking.Rumours form an essential part of the process leading up to an outburst, Horowitz (2001: 74-75) argues They mobilize ordinary people to do what they would not ordinarily do.They shift the balance in a crowd toward those proposing the most extreme action [… But] a rumour will not take hold unless there is a market for it […] What is remarkable is not that an interested agitator starts a rumour but that the rumour is spread, believed, and acted upon.
In 2008, research identified "(s)hared rumours in different sites […] regarding the 'stealing' of jobs and 'unfair' business competition," issues clearly identified during the 2015 episode in Durban as well as in Johannesburg (Bekker 2010: 139).It would appear that the frustrations of many urban dwellers regarding access to housing, municipal services, jobs and the like, access to what they as South Africans regard as their entitlements, have persisted and enabled rumours spread in 2008 to be, once again, believed in and acted upon in 2015.
Likewise, the widespread criticism of South Africa's state police during 2008 was repeated in 2015.Criticisms ranged from incompetence and lack of resources to tacit complicity in the violence.However, criticisms of this nature appear to be insufficient in explaining the widely reported perceptions of immunity from punishment and incrimination that perpetrators across urban South Africa revealed during both episodes.The claim of insufficient penetration of state police presence into the informal settlements, inner city neighbourhoods and townships of urban South Africa is at issue here.If policing is defined as "any organized activity, whether by the state or civil groups, that seeks to ensure the maintenance of communal order, security and peace," and if security is considered from the point of view of the resident "rather than from the governance perspective of the political authorities,", then the resident may be thought of as having a choice of police services, beyond that provided by the state police (Baker 2008: 22, 27).The difficulty arises when the state no longer has the capacity to bring non-state policingsuch as vigilante groups, civics, traditional authorities, informal security groups and so on -under effective accountability while community or local political organizations endorse intolerance of outsiders and associated violence.This appears to have taken place during a number of outbursts in both 2008 and 2015.In short, the issue is not solely that of perceptions regarding state police but, probably more importantly, local residents' perceptions regarding the role and legitimacy of bodies such as vigilante groups, civics, traditional authorities and informal security groups, bodies that promise some form of communal order to local residents.
There are, however, clear differences that emerge from this analysis of the two episodes.The first difference regards the manipulation of xenophobic attitudes by influential public figures.The Zulu King's speech two weeks before the outbreak of the 2015 episode was publicly supported soon afterwards by the son of the South African President who was reported to have "come out in full support of King Goodwill Zwelithini's controversial call to deport foreigners from South Africa" (Khoza 2015).Earlier, after the January 2015 looting of foreign-owned shops in Soweto, the national Small Business Development Minister stated that foreign-business owners in South Africa's townships could not expect to co-exist peacefully with local business owners unless they shared their trade secrets.She was quoted as saying: "Foreigners need to understand that they are here as a courtesy and our priority is to the people of this country first and foremost" (Mail & Guardian 2015).Three weeks into the 2015 episode, in mid-April, both the Minister and the King called for an end to the outbursts, the King claiming that he had been misinterpreted.
In sharp contrast, before and during the 2008 episode, such manipulation 'from above' of xenophobic attitudes was absent.In the words of a South African political scientist, The country's leaders may bear indirect responsibility through policy failure and acts of commission, but I see no evidence that the marauding crowds are taking their cue from government immigration policy or from corrupt cops […] We do not have the active antixenophobic leadership we need, but at least […] we do not have leading politicians manipulating anti-foreigner sentiment as they have recently done, with calamitous consequences, in the Democratic Republic of the Congo and Ivory Coast (Glazer 2008: 54).
A second major difference concerns what was earlier termed the two faces of the same coin: foreign sub-Saharan African governments' angry response to authorities' handling of the 2015 episode, on the one side, and the South African government's rejoinder by establishing a national 'operation' aimed at ridding the country of criminals and illegal immigrants, on the other.There were no such accusations and ripostes after the 2008 episode.Some reasons for this difference include the greater visibility, in 2015, of South Africa on the African continent.This visibility was not only in terms of trade, but also within the African Union where a South African is the African Union Commission Chair, as well as militarily through South African participation in peace missions.Simultaneously, however, the image of South Africa as a regional hegemon has shrunk: Nigeria has now been crowned as Africa's largest economy and it is generally known that President Zuma is struggling to maintain a positive reputation during his second term, in contradistinction to the robust and sometimes controversial image that former President Mbeki enjoyed on the African continent in 2008.
The third difference to be noted here is the change in the initial and subsequent metropolitan epicentres of the two episodes.In 2008, greater Cape Town proved to be particularly vulnerable after the series of outbursts in Gauteng.In 2015, in Cape Town, there were no visible outbursts classified as xenophobic.On the other hand, in the case of the predominantly Zulu-speaking metro of eThekwini and its later migration to Johannesburg, the series of events making up the 2015 episode may be put down, at least in part, to shared frustrations converted into group aggression, to the manipulation of antiforeigner sentiment, as well as a less than effective state police reaction to looting in Soweto and neighbouring urban places.However, what did take place before and during 2015 in the Western Cape (as well as in Gauteng and Durban) were continued violent outbursts related to service delivery issues.These will be discussed in more detail below.The point to be made here is that, while disregarding the rumours and justifications for violent actions against foreigners that flooded the mass and social media in 2015, residents in the Cape Town metro mobilised often in a violent fashion, by venting their frustrations on municipal targets instead.

Prominent Interpretations of the 2015 and 2008 Episodes
Upon his return from an official international visit in June 2008, President Mbeki was clear in his interpretation of the 2008 events as well as in reprimanding how this episode was being reported: The dark days of May which have brought us here today were visited on our country by people who acted with criminal intent.What happened during these days was not inspired by a perverse nationalism […] resulting in our communities violently expressing the hitherto unknown sentiment of mass and mindless hatred of foreigners -xenophobia -and this I must also say -none in our society has any right to encourage or incite xenophobia by trying to explain naked criminal activity by cloaking it in the garb of xenophobia (Hassim et al 2008: 4).
Government officials at the time also argued that much of the violence was well-coordinated and, accordingly, probably organised.The Minister for Intelligence Services was quoted as stating that "we cannot ignore […] that there were reportedly meetings held in hostels, that this prairie fire of hate seemed to move fast as if planned, and that there were printed pamphlets" (Bekker 2010: 132).This suspected underground organisation -an imagined third force -was never identified.
This official response to the 2008 episode -emphasising criminal rather than anti-foreigner intentions behind the violence -remained state orthodoxy throughout the next seven years.In addition, this orthodoxy stood firm despite fractures caused by various public statements from government officials and politicians at public events that proved difficult officially to place within a criminality framework.The statement by the Minister of Small Business Development and the speech by the Zulu King, as examples, are difficult to locate within such a framework.
In April 2015, Police Minister Nhleko said he found it hard to view the attacks as just xenophobia.He was reported to have said that "In a sense, what we are witnessing are afrophobic kinds of activities and attacks, resembling elements of self-hate among Africans.The evidence shows the attacks are mainly against the Congolese, Zimbabweans, Malawians, Somalis and some South African nationals as well" (Gqirana 2015).
Strict orthodoxy appears to be returning: according to the joint ad-hoc committee set up to investigate the 2015 xenophobic attacks, police in Alexandra told the committee that they have no evidence suggesting that the violence between foreigners and locals in April was xenophobic.In a reprimand reminiscent of Mbeki's 2008 speech, committee chair Ruth Mbengu was reported to have said "it's not only wrong but extremely irresponsible of the media to incorrectly label the attacks on foreign nationals in April as xenophobic" (Eye Witness News 2015).As of September 2015, the committee had yet to table its final report.
Few commentators beyond the South African state endorse interpretations within such a criminality framework.Equally, few endorse a framework within which xenophobia emerges as the primary and dominant determinant for violent outbursts.A recent SAMP publication that distinguishes between xenophobia denialism, xenophobia minimalism and xenophobia realism and opts for the third interpretation is an exception (Crush & Ramachandran 2014).Rather, most commentators recognise the importance of xenophobic attitudes during the outbursts without placing this at the heart of the interpretation.There appears to be agreement among these commentators on the well-researched nature and breadth of xenophobic attitudes among South Africans (Adam & Moodley 2013) as well as on the deprivationunemployment, poverty, inadequate shelter and basic services -facing South Africa's urban poor (Simkins 2011).Where disagreement surfaces is in the interpretation of the intentions underpinning the behaviour of members of this urban poor when that behaviour turns violent and targets persons and property in their neighbourhoods.
Two influential interpretations -both focusing on rational elements of behaviour -will be summarised.The first is the relative deprivation interpretation: poor black urban residents are experiencing competition regarding jobs, inadequate provision of housing and service delivery in their informal settlements, little effective government communication regarding these issues in their residential areas and corruption from government officials and the police.The relative deprivation they experience then comprises the frustrations they develop as a result of their expectations in these regards not being realized.These sentiments that combine to generate shared generalised anger, are converted into violent aggression against those they perceive as competitors and as the immediate cause of their frustrations, partially since they appear to be better off than themselves.Their targets accordingly are the 'foreigner' and the 'stranger' in their neighbourhoods (HSRC 2008).
The second interpretation is that of resource mobilisation, where the focus shifts from attitudes and frustrations to the political economy in the immediate neighbourhoods of the urban poor.For instance, collective action before and during an outburst targeting foreigners and their property is considered to be a strategy employed to extract benefits.Local leaders who may lack institutional access to either the local economy or to the political process aim to acquire material and/ or political resources by mobilizing local residents.For example, such resources could comprise local trading and job opportunities, or access to municipal and local council positions.Such an interpretation focuses attention on local leadership institutions; on how violence is organised and why locals participate (Tarrow 1994, Misago et al. 2009).
In both interpretations, it is clear that while the violent targeting of foreigners becomes a destructive instrument used against innocent civilians, it is not necessarily the primary goal -aggression may well be redirected (toward the local municipality, as an example) and local leaders' attempts to extract benefits from those who control or own resources may be pursued by means other than attacking foreigners and their property.
The last alternative interpretation identified here shifts the analysis from rational to emotionally-suffused behaviour.Perpetrators, Horowitz (2001) argues, mix calculation with passion during their uninhibited violent behaviour.The insight that the discharge of aggression is a satisfying experience leads to the question of what violent xenophobic behaviour offers young urban South Africans who suffer deprivation and are keenly aware of unfulfilled promises of entitlement made in post-apartheid South Africa.At least in part, such behaviour, which often appears to offer pleasure to perpetrators, enables the reversal of humiliation.Horowitz (2001: 536) notes: […] the violence that aims to thwart domination […] is suffused with affect born of humiliation.Much of the pleasure that violence brings springs from the mastery that reverses dishonour […] In summary, interpretations of the 2015 and 2008 violent episodes range from an investigative spotlight illuminating the xenophobia of many South Africans to the state's insistence that the spotlight ought rather to be focused on criminal behaviour.In both cases, it is probably true that this terminology that refers in singular fashion to 'xenophobia' or to 'criminality' conceals as much as it reveals.Accordingly, it is fitting to remark here that the perpetrators mobilised during these events are drawn from the same pool as the protesters mobilised during the rising number of violent service delivery protests across the country.Both are members of South Africa's urban working and underclass.

Service Delivery Protests that Turn Violent
In order to broaden our events-based analysis of violent collective behaviour in urban South Africa, a profile of the growing number of violent service delivery protests in the country will be developed here.A research project named the Civic Protests Monitor (CPM) has such events-based data covering the eight year period from 2007 to 2014.Civic protest is defined as "referring to organised protest action within a local area which directly targets municipal government or targets municipal government as a proxy to express grievances against the state more widely" (Powell n.d.: 5).These protests, more generally known in the popular discourse as service delivery protests, generally involve urban residents from poor neighbourhoods of both metropolitan and nonmetropolitan municipal areas.Three of the general trends over this period are highlighted in the CPM (Powell n.d.: 3):  The number of protests in 2014 reached an all-time-high of 218.The previous maximum was 204 in 2009. In the three year period of 2012 -2014, Gauteng experienced more protests than any other province.Since 2007, Gauteng's share of protests has been rising more rapidly than in other provinces.Cape Town was the most protest-prone municipality with 84 protests, followed by Johannesburg, eThekwini, Tshwane and Ekurhuleni.
Between them these five metro municipalities accounted for half of all the protests recorded. The prevalence of violence associated with protests has continued to increase.The number of violent protests reached a record high in 2014.
In 2007, just less than half the protests were associated with some violence.In 2014, almost 80% of protests involved violence on the part of the participants or the authorities.
Violent protests have been defined in this study "as those protests where some or all of the participants have engaged in actions that create a clear and imminent threat of, or actually result in, harm to persons or damage to property" and the research reveals that the number of violent protests has risen at a faster rate than the total number of protests.In 2007, the total number of violent protests was 44.Seven years later this number had increased more than fourfold to 181.
Five forms of violence are distinguished in the event data set: intimidation, personal attacks, arson, damage to property and looting.Intimidation was the most frequently cited form of violence (376 protests) associated with protest in the 2012 -2014 period.Physical attacks on individuals were less prominent (315 protests).The destruction of property (including arson) was recorded more often than attacks on individuals (a combined total of 372 protests).Thus, two thirds of the types of violence recorded at protests went beyond "mere" intimidation and involved the destruction of property, assault, looting and even death (Powell n.d.: 9).
What emerges from these trends is a continuing series of violent events across urban South Africa during which residents mobilise and target municipal officials, local councillors and municipal property, in a phrase, elements of the local state.The demands made by protesters captured in the database comprise improved municipal services (such as housing, electricity and water), improved non-municipal services (such as education and policing) as well as demands for employment opportunities.
A relative deprivation interpretation of this rising number of violent events would root residents' frustration in the same structural conditions as in the case of violent xenophobic events.However, the aggression, as it shifts from generalised anger toward the identification of targets blamed for the causes of this frustration, is directed at elements of the local state rather than at foreigners and strangers and their property.In equal measure, a resource mobilisation interpretation would have it that local leaders, with the aim to acquire material and political resources, exploit the frustrations of local residents by promoting violent action against the local state rather than against foreigners and strangers and their property.

Conclusion
A profile of South Africa's urban poor has been sketched that depicts them as both deprived and frustrated.One way to conceptualise their common state of mind is to attribute to them a shared generalised anger.This generalised anger may be seen through the lens of unfulfilled expectations: entitlement promises made by the African National Congress during the continuing struggle for transformation and liberation, promises made to the newly urbanised, to the urban poor and to those against whom apartheid discriminated.The frustrations that surface through the inability of many urban dwellers to realise or receive their fair share of entitlements -housing, municipal services, jobs and the like -are easily converted into aggression that is directed at those held responsible for these failures.
In the case of the recent xenophobic outbursts, this aggression appears to have been directed against those who, according to widespread rumours and widely broadcast elite discourse, appeared to be largely responsible for much of this failure.These small foreign groupings also appeared to be both isolated as well as unprotected.Aggression against superiors is converted into aggression against unranked groups in the immediate neighbourhood, with little fear of retribution.
In the case of violent service delivery protests, this aggression was directed against local municipal officials and councillors and, often, against their municipal offices and town halls.These local elites are often perceived to be significantly more privileged than most residents in their local community as well as incompetent, corrupt or both.Here too, local protesters have little fear of retribution, particularly after the Marikana wildcat strike that turned into a massacre with more than 40 miners killed by the South African police.This has given the police pause in the presence of protest incidents.In addition, the simultaneous emergence of a new political party -the Economic Freedom Fighters -that supports many service delivery protests adds a measure of legitimacy to such actions.
The generalised anger may also be seen through the lens of resource mobilisation where a focus on attitudes moves to a focus on the political economy of violence.In both cases, xenophobic as well as service delivery events, it is local leaders (both informal and formal) who are considered to be at the heart of the mobilization of perpetrators and protesters during the outbursts.They do this by playing to the frustrations of local residents with the aim of extracting benefits from those who control or own resources -be they local shopkeepers and artisans or officials and elected councillors.This shift identifies an overlap much debated recently by senior officials in the South African government between outburst, protest and criminal activity.
The looting during outbursts and protests indicates that one dimension of this collective behaviour is undoubtedly criminal, probably significantly more of a petty rather than of a hardened and professional nature.
These two lenses focus our attention on a South African urban poor who reveal less of a deep hostility to the foreigner and stranger than a generalised shared sentiment of exclusion, sometimes approaching hopelessness regarding their current urban lives.The lenses also reveal an urban working and under-class that is frustrated by the failure over the past two decades of what they have been led to believe are their just entitlements.Simultaneously, as they become more aware of the growing inequalities between themselves and the new local as well as national elites, this frustration is converted either into aggression against those deemed to be responsible for their current state or into mobilisation in pursuit of some of these privileged entitlements.
Lastly, this analysis reveals a new form of agency that the South African urban poor are developing.This agency emerges from rational as well as emotional motives, from aggression against those seen to be responsible for limitations on entitlements as well as ways to convert long-standing humiliation into mastery through violent behaviour, particularly gratuitous violence.Moreover, over the past few years, access to new forms of social media has facilitated the broadening of this agency.The 2015 violent xenophobic episode has revealed that it is not only South African groupings that possess agency in the country: foreign African immigrants in both Durban and Johannesburg organised and expressed their anxiety and displeasure in public and the voices of foreign African governments after the episode were public and forceful in South Africa, on the African continent and internationally.South Africa may well remain chiefly a limited access order (North et al. 2013) in which elites take most of the national decisions that affect South Africans' lives, but both the urban poor as well as foreign actors are demonstrating, through their agency, that these elites do not have it all their own way.