1. Introduction

1The United Nations (UN) Global Compact for Safe, Orderly and Regular Migration (GCM) has recently been labelled a ‘depoliticised document’ marked by major internal contradictions (Pécoud, 2021). There is at least one respect, however, in which it is neither contradictory nor depoliticised; the agreement of signature states to eliminate all forms of discrimination against migrants and their families. Objective 17.33 of the GCM makes a commitment ‘to eliminate all forms of discrimination, condemn and counter expressions, acts and manifestations of racism, racial discrimination, violence, xenophobia and related intolerance against all migrants in conformity with international human rights law’ (UNGA, 2019, 24). To further this objective, states commit to establishing mechanisms ‘to prevent, detect and respond to racial, ethnic and religious profiling of migrants by public authorities, as well as systematic instances of intolerance, xenophobia, racism and all other multiple and intersecting forms of discrimination’, to promote awareness-raising campaigns to inform public perceptions about the evidence-based positive contributions of migration, and ‘to end racism, xenophobia and stigmatization against all migrants’. This may be idealistic and unachievable, but it is certainly not contradictory or depoliticised; instead, it puts the onus on states to deal with evidence of a growing challenge accompanying increased global mobility and migration: the growth of xenophobia in countries of migrant destination (Crush and Ramachandran, 2010).

2The character, drivers, and impacts of xenophobia and policy responses to this phenomenon have been a topic of recurrent scholarly interest in recent decades, generating a sizable literature, most of which has focused on negative attitudes towards and the discriminatory treatment of migrants in Europe and North America (see d’Appollonia, 2017; Gorinas and Pytliková, 2018; Peterie and Neil, 2020; Rensmann and Miller, 2017). Much less attention has been paid to xenophobia in the migrant-receiving countries and regions of the global South (Crush and Ramachandran, 2010). This mirrors a broader research and international policy disinterest in South-South migration and its role in the development of countries of origin and destination in the global South (Fiddian-Qasmiyeh, 2020). At the same time, evidence of intensifying xenophobic sentiment in the global South is beginning to accumulate in disparate settings including in India (Adibe 2017; Ramachandran, 2019), Singapore (Gomes, 2014; Yang, 2018), the Gulf (Ullah et al., 2020), Latin America and the Caribbean (Gill and Danns, 2018; Meseguer and Kemmerling, 2018; Jones, 2020), and a number of West and Southern African countries (Akinola, 2018; Miran-Guyon, 2016; Campbell and Crush, 2015; Crush and Pendleton, 2007; Whitaker, 2015). One extremely common xenophobic trope associates migrants with threats to the health of citizens by bringing disease and using up scarce health resources. There is evidence, for example, of a global upsurge in COVID-19 related xenophobic reaction (Ahuja et al., 2020, Castillo and Amoah, 2020; Chan and Strabucchi, 2021; Reny and Barreto, 2020). The significance of the new literature prompted by the pandemic is that it draws close attention to the politics of xenophobia and, in particular, the ways in which the national and local state is imbricated in the development, reinforcement and reproduction of xenophobia.

3State responses in the global South to evidence of xenophobia in the general population range across a broad terrain. Populist political parties invariably embed anti-immigrant rhetoric and policy proposals in their election platforms and, if and when they come to power, enact policies that are demonstrably xenophobic. The ruling right-wing Bharatiya Janata Party (BJP) in India, for example, has long advocated a punitive approach to ‘infiltrators’ from neighbouring states and has enacted a range of new policies designed to exclude millions of Muslim migrants (both internal and international) (Ramachandran, 2019). In post-colonial Africa, as Fourchard and Segatti (2015, 6) point out, ‘the attention paid to exclusionary discourses and practices that haunt the politics of belonging throughout the continent is not always balanced by an interest in countervailing discourses and practices (reconciliation, diffusing ethnic oppositions, everyday conviviality etc.)’. The main reason why there has been little interest in countervailing discourses is that there are few, if any, good examples to study. The nation-building projects of most post-colonial states tend to focus more on the perceived threats to national sovereignty posed by incoming migrants and immigrants. Opportunities to take up permanent residence and citizenship in other countries are generally extremely constrained. As well as the governance of xenophobia by mitigation (rare), disinterest (more common) and intensification (extremely common), the South African case suggests that there is a fourth alternative; governance by displacement. This can take a variety of forms including denial that xenophobia exists and is a problem (xenophobia denialism), and displacing blame for xenophobic actions onto their victims or onto so-called criminal behaviour of anti-social and marginalised citizens.

4The apartheid state was deeply racist and xenophobic in its immigration policies, only allowing white immigrants from Europe to settle in the country while simultaneously barring all Black immigrants other than temporary contract workers in the mining industry (Crush, Jeeves and Yudelman, 1991; Crush, 2000; Klotz, 2016; Peberdy, 2009). Apartheid South Africa also did not accede to the UN and African Union (AU) refugee conventions, although the Mozambican civil war in the 1980s displaced several hundred thousand people to South Africa. The state did everything in its power to keep these refugees confined to border communities far away from the large urban centres (de Jongh, 1994). One of the effects of racist immigration and refugee policies was that very few South Africans came into direct contact with migrants from other African countries. The collapse of apartheid in the late 1980s and early 1990s brought a decisive shift in perceptions of South Africa as a desirable destination within and outside Africa: the official census number of migrants from other countries in the South increased from 780,000 in 1990 to 2.5 million in 2019 (UNDESA, 2020). The range of countries from which migrants came also expanded from neighbouring Southern African countries to more distant countries in West Africa and the Horn, as well as Asian countries such as Bangladesh, China and Pakistan. During the first decade of apartheid rule, and despite the limited personal contact, attitudes hardened towards the new migrants who were widely perceived as bringing crime and disease and using resources (housing, services, jobs) that should have been reserved for citizens (Crush and Peberdy, 2003).

5Between 1994 and 2005, the post-apartheid state deported over one million migrants (mainly to neighbouring countries such as Zimbabwe and Mozambique) (Sutton and Vigneswaran, 2011). In order to justify these increasingly harsh and unconstitutional measures of arrest, deportation and exclusion, the stereotype that the country was being ‘flooded’ by migrants was popularised by the media and the pronouncements of government officials, including the Minister of Home Affairs, with constant exaggerated references to the millions of ‘illegal aliens’ supposedly storming the country (Crush and Williams, 2001; Danso and McDonald, 2000). State discourse and policy reinforced the increasingly widespread perception that all migrants (simply by being from elsewhere) were a fundamental threat to citizens. Very early on, critics of state, media and popular representations attached the label xenophobia to the phenomenon, calling it a ‘new pathology’ and the ‘dark side of democracy’, adopting the standard dictionary definition of the term as extreme fear, hatred and prejudice towards strangers or people from other countries (Crush, 2001; Harris, 2002). Early attitudinal surveys in the late 1990s confirmed that xenophobic hostility and prejudice, directed particularly towards migrants from Africa and Asia, had become widespread (Mattes et al., 1999).

6There were isolated incidents of violence against migrants and refugees in the first decade or so of South African democracy after 1994, but systematic, orchestrated, nationwide violence targeting the lives and property of migrants did not occur until May 2008 when over 70 people were murdered and over 100,000 people driven out of their homes across all of South Africa’s major cities (Crush, 2008). Since then there have been regular upsurges of this extreme manifestation of xenophobia, in 2011, 2015 and again in 2018-19 (Crush, Ramachandran and Pendleton, 2013; Landau, 2012; Misago, 2017). Hardly a day goes by, however, without reports of isolated attacks or local mob violence directed at migrants in different parts of the country (Crush, Chikanda and Skinner, 2015). The drivers of recurrent xenophobic violence in South Africa have been debated at length without a great deal of consensus about its fundamental causes (Gordon, 2020; Hassim, Kupe and Worby, 2008; Landau, 2012; Matsinhe, 2016; Neocosmos, 2008; Nyamnjoh, 2006; Steinberg, 2012 and 2018; Tevera, 2013; Misago, 2016). Less attention has been paid to the responses of national, provincial and municipal governments and whether these responses mitigate or exacerbate xenophobia (Misago, 2017; Musuva, 2015). In terms of the proposed governance typology, are state responses to xenophobia best characterised by disinterest, intensification, mitigation or displacement? This chapter focuses primarily on national government responses to xenophobia in South Africa and argues that displacement (and its defining characteristics of scapegoating and denialism) best characterise the reactions of government to post-1994 xenophobic attitudes and behaviours. Deliberate policies of intensification have been more evident at sub-national provincial and local levels. However, in the South African case, displacement and the failure to mitigate has arguably intensified xenophobia in the populace at large. The first section of the chapter provides an overview of the evidence for the existence of xenophobic attitudes and violence in the country. The next section provides an overview of the violence that appears to the victims and most independent commentators as motivated by xenophobia. The chapter then turns to the response of the South African government and its policy of displacement as a response to such xenophobia. The chapter then concludes with a discussion of the implications for implementing the GCM commitments to eliminating xenophobia in the country.

2. Profiling Xenophobia

7In the absence of globally comparative surveys of public opinion, it is hard to say whether South Africa is exceptional or typical in its antagonistic response to immigration and immigrants. However, Wave 6 of the World Values Survey (2010-2014) provides a preliminary answer around the single common question in all its representative country surveys of whether government should allow migrants into the country under one of four conditions: prohibit entry, strict limits, as long as jobs are available, and letting anyone come. Table 6.1 clearly shows that South Africans have the most negative attitudes towards immigration of all global South countries surveyed. As many as 30% want completely closed borders (the same percentage as in India), while 78% support closed borders or strict limits on entry (compared to only 55% in India). Only Malaysia, Thailand and Indonesia have comparable or greater scores on both metrics combined. The South African Migration Programme (SAMP) has conducted three national surveys of South African attitudes towards migrants, refugees and migration policies (in 1998, 2006 and 2010), in which the same question elicited very similar results for South Africa (Crush, 2001 and 2008; Crush, Ramachandran and Pendleton, 2013; Debrosse et al., 2016). The survey also collected representative data on a broad range of attitudes and their demographic, social, economic and cultural determinants.

Table 6.1 Comparative citizen attitudes toward immigration

Country	Prohibit Entry (%)	Strict Limits (%)	As Long as Jobs Available (%)	Let Anyone Come (%)
South Africa	30	48	16	6
India	30	25	22	23
Egypt	26	43	25	5
Jordan	25	46	28	2
Malaysia	18	72	8	2
Mexico	17	25	45	12
Zambia	15	44	30	11
Thailand	14	65	16	5
Morocco	11	20	41	28
Brazil	11	33	47	9
Trinidad & Tobago	10	55	32	4
Chile	9	35	50	6
China	8	21	51	20
Indonesia	8	72	15	6
Guatemala	7	21	55	17
Ghana	6	36	39	18
Peru	6	21	50	23
Argentina	6	34	45	15
Ethiopia	5	27	28	40
Mali	4	16	46	34

Source: Inglehart et al. (2014).

8The 2010 survey used Principal Components Analysis to develop a composite measure of xenophobia based on a range of attitudinal questions (the SAMP Xenophobia Index or SXI) (Auger et al., 2011). SXI scores range from 0 (no xenophobia) to 10 (intense xenophobia). Figures 6.1 to 6.3 cross-tabulate the SXI with three independent variables showing high overall levels of xenophobia but variations in the intensity of xenophobic sentiment by racial group (with Black South Africans being least xenophobic and Indian/Asian South Africans most xenophobic), by household income (with levels of xenophobia increasing with household income), and by amount of contact with migrants (with levels of xenophobia decreasing with increased amounts of contact). The SXI did not vary significantly with age, sex, education or employment status of citizens.

9Xenophobia also expressed itself in negative attitudes toward refugee protection, willingness to take action against migrants in the neighbourhood, and perceptions of reasons for xenophobic violence against migrants. Table 6.2 shows strong opposition to taking in more refugees (57% opposed, 11% supportive), using taxpayer money to support refugees (46% opposed, 16% supportive), and granting permanent residence to long-term refugees (44% opposed, 18% supportive). Contrariwise, there was strong support for repatriating refugees when they are no longer at risk (56% supportive, 14% opposed) and for mandatory HIV testing of all refugees (41% supportive, 29% opposed). Next, as Table 6.3 shows, around a third of South Africans said they would take various actions against migrants in their neighbourhood, with 15% saying they were prepared to force migrants to leave the area and 11% being prepared to use violence to do so. Finally, the survey asked respondents why they thought the 2008 nationwide violence had occurred. Around two-thirds of residents of hotspots experiencing violence agreed that migrants were largely to blame by engaging in crime, taking jobs from locals, being culturally different, using health services for free, and ‘stealing women’ (a misogynistic reference to inter-marriage between migrants and South Africans) (Table 6.4). As many as 62% agreed that ‘migrants do not belong in South Africa’.

Table 6.2 Attitudes toward refugee protection, 2010

	Support (%)	Oppose (%)
Grant asylum to those escaping war and persecution	38	23
Increase refugee intake in South Africa	11	57
Grant permanent residence to refugees in South Africa for > 5 years	18	44
Send refugees back when they are no longer at risk	56	13
Refugees must live in special camps near the border	31	32
Use government budget to look after refugees	14	46
Allow refugees to work in South Africa	25	35
Test refugees for HIV	41	29

Source: SAMP (2010).

Table 6.3 Likelihood of taking action against migrants, 2010

	% Likely	% Unlikely
Report them to police	36	39
Report them to employer	27	45
Report them to community association	27	45
Combine to force them to leave	15	73
Use violence against them	11	72

Source: SAMP (2010).

10Other attitudinal surveys conducted since 2010 confirm the persistence of high levels of xenophobia in the country (Claassen, 2017; Dube, 2019; Facchini, Mayda and Mendola, 2013; Gordon, 2015; 2016; 2017a; 2017b; 2018a and 2018b; Gordon and Maharaj, 2015; Ruedin, 2019). Xenophobic attitudes translate readily into routinised xenophobic behaviour involving negative stereotyping, exclusionary language, verbal denigration, denial of access to services such as health and education, and insistent demands from citizens that government rid their communities and the country of ‘foreigners’. Xenophobic attitudes have also been closely linked to xenophobic actions including collective violence. South Africa has experienced intense nationwide rounds of violence targeting the businesses, homes and lives of migrants and refugees in 2008, in 2015 and again in 2019 (Bekker, 2015; Burke, 2019; Desai, 2015; Hayem, 2013). These moments constitute ‘a heightened form of xenophobia in which hostility and opposition to those perceived as outsiders and foreigners is strongly embedded and expressed through aggressive acts directed at migrants and refugees (and) recurrent episodes of violence’ (Crush and Ramachandran, 2014, 3).

11Collective violence is also manifested on an almost daily basis in particular localities with the looting and destruction of migrant-owned businesses in the informal sector and bodily injury and murder of business owners and their employees (Crush, Chikanda and Skinner, 2015; Ramachandran, Crush and Tawodzera, 2017). Episodes of collective violence targeting small businesses include combinations of written or verbal threats and insults, public intimidation through protests or marches, forced shop closures, physical assaults and the murder of migrant store owners or their employees, looting of store contents, arson or other damage to the physical structure of shops, damage or destruction of business properties including homes and vehicles, temporary or permanent forced displacement, and extortion for protection by local leaders, police and residents. Crush and Ramachandran (2015) document 220 episodes of collective violence against migrant and refugee businesses in various locations around the country between 2005 and 2014 (excluding the violence of 2008). The frequency of collective violence has increased over time (Table 6.4). In December 2018, Xenowatch reported 529 xenophobic violence incidents in post-apartheid South Africa resulting in 309 deaths, 901 physical assaults, 2,193 shops looted and over 100,000 people displaced (Mlilo and Misago, 2019).

Table 6.4 Frequency of collective xenophobic attacks

Year	No. of Incidents	% of Total
Pre-2005	9	4
2005	4	2
2006	9	4
2007	9	4
2008	19	8
2009	17	7
2010	46	20
2011	22	10
2012	25	11
2013	36	16
2014 (to end-August)	32	14
Total	228	100

Source: Crush and Ramachandran (2015).

3. Governance by Denial and Displacement

12This section of the chapter discusses the response of the South African national government to the evidence outlined in the previous section for the existence and persistence of xenophobia in South Africa. The chapter argues that the dominant response is governance by displacement which has two complementary elements: denialism and scapegoating (or blaming the victim) (Crush and Ramachandran, 2014). First, with regard to xenophobia denialism, the Mandela and Mbeki governments did little if anything to acknowledge or address growing xenophobia in the late 1990s and early 2000s (Crush, 2001). However, in 2007, the AU’s African Peer Review Mechanism’s (APRM) report on South Africa said that xenophobia was a serious issue for South Africa and urged the government ‘nip it in the bud’ (AU, 2007, 26–27). The review came at the same historical moment that Mbeki was espousing pan-Africanism and the idea of an African renaissance on the one hand, and engaging in the HIV and AIDS denialism that cost hundreds of thousands of South African lives on the other (Fassin and Schneider, 2003; Bongmba, 2004; Nattrass, 2003). Xenophobia directly undermined his case for a South African leadership of the African renaissance and undercut the ideology of pan-Africanism. Xenophobia denialism was also perfectly consistent with his tendency to ignore scientific evidence and avoid taking responsibility for catastrophic social outcomes.

13Against the backdrop of mounting AU and international scrutiny and condemnation, Mbeki denied any connection between attacks on migrants and xenophobia. Under his leadership, xenophobia denialism became official government policy and was initially deployed to explain away nationwide violence against migrants and refugees in 2008. In a public address meant to commemorate the more than 60 people who had died, Mbeki announced that he had never met a xenophobic South African. Furthermore, he stated that anyone who called South Africans xenophobic was themself guilty of xenophobia: ‘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’ (Mbeki, 2008). The idea of ‘naked criminality’, rather than xenophobia, took root and became a central plank in the state orthodoxy that continues to the present (Gerber, 2019). Mbeki’s successors abandoned HIV and AIDS denialism, but reinforced xenophobia denialism. From 2008 onwards, politicians from the ruling party carefully avoided the term xenophobia when referring to violence against migrants, which they instead blamed on criminality. As some have pointed out, the rate of arrest and prosecution of these ‘criminals’ over the years has been abysmal at best.

14In 2010, the Minister of Police characterised attacks against migrants as ‘crimes of opportunity’ where criminal or anti-social elements ‘take advantage of the situation to engage in such misdeeds’ (Sapa, 2010). Commenting after a Zimbabwean migrant was stoned to death in 2011, police spokesperson Zweli Mnisi echoed this view: ‘Once you start talking about xenophobia and Afrophobia, you are talking about semantics. It is crime disguised under xenophobia’ (Isaacson, 2011). On another occasion, Mnisi is quoted as saying that ‘holistically speaking, South Africans are not xenophobic and many cases are merely crime’ (Bauer, 2013). In mid-2013, following an upsurge of violent assaults on Somali refugees, Minister of International Relations and Cooperation, Maite Nkoana-Mashabane, announced that ‘the looting, displacement and killing of foreign nationals in South Africa should not be viewed as xenophobic attacks, but opportunistic criminal acts that have the potential to undermine the unity and cohesiveness of our communities’ (Patel, 2013). The South African Cabinet also issued a public statement on the violence, noting that ‘Cabinet is cautious not to label this violence as xenophobia because preliminary evidence indicates that these acts may be driven primarily by criminality’ (RSA, 2013).

15At an African Union meeting in Johannesburg in June 2015, President Jacob Zuma reiterated the government’s position by arguing that ‘South Africans are not xenophobic. We do not believe that the actions of a few out of more than 50 million citizens justify the label of xenophobia’ (News24, 2015a). Zuma also publicly declared that ‘millions of peace loving South Africans are in pain also because they are being accused of xenophobia, which is not true. South Africans are definitely not xenophobic. The actions of a small minority should not be used to wrongfully label and stereotype more than 50 million people’ (Zuma, 2015). In a parliamentary debate following the upsurge of xenophobic violence in late 2019, Defense Minister Mapisa-Nqakula restated the official line that the attacks were ‘mostly acts of criminality irrespective of the nationality of those involved. Crime is crime. It is not South African to hate thy neighbour’. In the same debate, Police Minister Bheki Cele stated that ‘for us it’s nothing to do with xenophobia, it is criminality’ (Gerber, 2019).

16The second component of governance by displacement is scapegoating or assigning blame for the violence to migrants themselves. In 2015, for example, Defence Minister Mapisa-Nqakula expanded the definition of criminality to include migrant ‘criminals’: ‘While government is going to be taking resolute actions against South Africans who attack foreign nationals, we are equally determined to take action against all foreign nationals who commit crime in our country’ (Merten, 2015). The ruling party’s Secretary-General, Gwede Mantashe, openly blamed the rising numbers of migrants for the violence and said the solution was the ‘tightening [of] immigration laws’ and ‘if need be, establish[ing] refugee camps’ to geographically segregate migrants from citizens (Finnan, 2015). However, scapegoating has been most evident in government responses to the country’s recurrent episodes of violence, looting and destruction of the premises of migrants and refugees operating small informal businesses (called spazas) in the informal sector. First, an Inter-Ministerial Committee on Migration (IMC) was constituted in 2015. Fifteen government ministers sat on the IMC, an indication of how seriously government viewed the crisis. The official brief of the IMC was ‘to promote orderly and efficient migration and peaceful co-existence between citizens and non-South Africans, as well as to consider social, economic and security aspects of migration’ (PJC, 2015, 19). Second, an Ad Hoc Parliamentary Committee was constituted by both houses of parliament with the mandate to investigate the causes of the violence (the Ad Hoc Parliamentary Joint Committee on Probing Violence Against Foreign Nationals, PJC) (PJC, 2015).

17The Chair of the IMC, Minister Jeff Radebe, briefed the Ad Hoc Parliamentary Committee on the findings, stating that the primary cause of the violence against foreign nationals was ‘increased competition arising from the socio-economic circumstances in South Africa’ and the ‘business models used by migrants to discourage competition such as forming monopolies, evading taxes, avoiding customs and selling illegal and expired goods’ (PMG, 2015). Competition had been heightened by ‘a decade of poor implementation of immigration and border controls’. Furthermore, foreign nationals were placing a strain on government services such as health, housing, education and social grants and ‘dominating trade in certain sectors such as consumable goods in informal settlements which has had a negative impact on unemployed and low skilled South Africans’. He also blamed the victims for the attacks: ‘They roam, they go to townships to occupy the economic space. We never invaded economic space in exile’ (News24, 2015b). At a press conference he further observed that ‘as the Inter-Ministerial Committee, we've concluded that South Africans are not xenophobic’ (Davis, 2015).

18The Ad Hoc Parliamentary Joint Committee’s investigation went even further, repeatedly asserting that xenophobia as a phenomenon does not exist in South Africa (News24, 2015b; Nicolson, 2015; PJC, 2015). The Committee claimed that South Africans do not hate or loathe migrants and refugees. In the parliamentary deliberations leading up to the adoption of the report, it was recommended that the term ‘xenophobia’ be omitted completely from the report because no convincing evidence had been found that the phenomenon existed (PMG, 2015). The final report notes that ‘Parliament had not yet come to the conclusion that the incidents of violence against foreign nationals were due to xenophobia as per the dictionary definition of extreme, irrational hatred of foreign nationals’ (PJC, 2015, 19). The Ad Hoc Parliamentary Committee concluded that ‘the main causes of the violent attacks were criminal actions that started with stealing of goods from foreign owned spaza shops by South African criminals who are often drug addicts. The spaza shop owners would react by shooting at those who steal from their spaza shops using unregistered firearms rather than reporting to the police. When this happens and someone is killed, local communities retaliate by looting spaza shops owned by foreign nationals rather than reporting to the police’ (PJC, 2015, 35).

19The idea that migrant spaza owners were responsible, by their very presence, for the attacks on their persons and premises resonated strongly with the views of many citizens and even beyond.

20Independent commissions of enquiry, such as that headed by the former UN High Commissioner for Refugees (UNHCR), argued that the immediate cause of the outbreak of violence was ‘deliberate efforts of select individuals, some of whom had interests in the informal trading sector, to drive away competition by foreign national-owned businesses. […] These deliberate efforts sparked the outbreak of widespread incidents of criminality, violence and looting of properties owned by foreign nationals’. Furthermore, ‘many of the perceptions of foreign national traders, although largely unfounded, contributed to heightened tensions’ (SRG, 2015, x). However, the Special Reference Group on Migration and Community Integration in KwaZulu-Natal (SRG) studiously avoided labelling the violence ‘xenophobic’ or seeing xenophobia as a contributing or even motivating factor. At most, it conceded that ‘the violent attacks against foreign nationals were, in some measure, fuelled by dominant and negative perceptions that exist amongst locals and foreign nationals about one another’ (SRG, 2015). This implies that the attitudes of foreign nationals were responsible for their own victimisation. In addition, none of the mob violence that took place was perpetrated by migrants against South Africans. An anti-xenophobia protest march organised by NGOs and migrant groups to Durban’s City Hall on 7 April 2015 was declared illegal and the police used water cannons, teargas and rubber bullets to disperse the crowd.

21In April 2015, in the wake of the violence, the IMC began to implement its strategy in the form of the controversial and militaristic ‘Operation Fiela’. Operation Fiela was described on the government website as ‘a multidisciplinary interdepartmental operation aimed at eliminating criminality and general lawlessness from our communities. As the word ‘fiela’ means to sweep clean, we are ridding communities of crime and criminals so that the people of South Africa can be and feel safe. The ultimate objective of the operation is to create a safe and secure environment for all in South Africa’ (South African Government, 2015). The central objective of Operation Fiela was not, in fact, to protect migrants or arrest the perpetrators but instead to launch a nationwide campaign by the police and army to harass migrant-owned businesses, locate undocumented migrants, and deport them. By the end of 2015, the government boasted that Operation Fiela had searched 460,000 people, 151,000 vehicles and 38,000 premises. A total of 41,000 arrests had been made. Between April and June 2015, 10,242 migrants were deported, of which Zimbabweans constituted over a quarter (SRG, 2015). The absence of due process in Operation Fiela prompted Lawyers for Human Rights (LHR) to (unsuccessfully) challenge its constitutionality in the North Gauteng High Court in June 2015. An application for leave to appeal to the Constitutional Court was dismissed in December 2016 (Constitutional Court of South Africa, 2016). LHR characterised Operation Fiela as institutional xenophobia which ‘have the unfortunate trend of creating a link between foreign nationals and crime, which is both misleading and inaccurate. It does nothing to address the core problems around xenophobia’ (LHR, 2015).

4. Conclusion

22The global migration and development agenda puts great emphasis on the positive development impacts of migration. As the former UN Special Representative for International Migration, Louise Arbour, noted: there exists a global ‘virtuous circle’ in which ‘migration is overwhelmingly positive for migrants and their communities, both origin and destination […] [and] a potent motor of development’ (Arbour, 2018). In order to achieve these outcomes, governments and citizenries need to openly recognise and acknowledge the existence of this virtuous circle, work assiduously to ensure its realisation, and remove institutional and attitudinal obstacles that stand in the way. In post-apartheid South Africa, there has been little recognition of the positive relationship between migration and development, and even fewer concrete efforts to mainstream development in migration policy and migration in development policy. The country’s omnibus National Development Plan 2030 does make isolated references to the need for greater openness for skilled migrants and a ‘more progressive’ immigration policy (NPC, 2012). On just one occasion it observes that well-managed migration can ‘contribute positively to South Africa’s development’, but this is followed almost immediately by a statement about the burden of migration (NPC, 2012, 105). Nor is there a systematic plan for how to make migration work for development, other than by recruiting high-level skills. While the still unimplemented 2017 White Paper on International Migration claims to be that plan and pays lip-service to the migration-development relationship (DHA, 2017), its managerial framework focuses more on how to better control, monitor and manage migration and refugee flows. Neither document mentions xenophobia as an obstacle to the development of a well-managed migration system, nor to making migration work for development.

23A major under-explored barrier to realising the full development benefits of migration is negative attitudes, shading into passive and active xenophobia, amongst the citizenry of migrant destination states. In this context, the South African case has particular relevance. While the country occupies an unenviable position at one extreme of the spectrum of attitudes to migration, xenophobia is also a growing phenomenon in many countries in the global North and South. In South Africa, as this chapter suggests, there is incontrovertible evidence that xenophobia is rife throughout the country and in the corridors of state power. State responses to xenophobia vary considerably but can basically be distilled into three models of governance; mitigation, intensification and displacement. South Africa has eschewed mitigation and, in contrast to countries where populism is on the rise, none of the country’s political parties have sought to use anti-migrant hostility as a central policy platform, promoting and intensifying xenophobia. Instead, the South African response has been characterised by displacement; first, in the face of ferocious collective violence against migrants, government denies that xenophobia exists or is responsible in any way for the mayhem. Blame is invariably displaced onto criminals and criminality. The acts are certainly criminal, but they are not generally perpetrated by organised crime groups or habitual criminals. Since 2015, however, blame has increasingly been displaced onto migrants themselves, not least by all of the government ministers on the influential IMC which launched a punitive campaign against migrants in the aftermath of collective xenophobic violence.

24One obvious objection to this characterisation of the governance response is the acceptance by Cabinet of a long-awaited National Action Plan to Combat, Racial Discrimination, Xenophobia and Related Intolerance (RSA, 2019). This plan, which fulfilled a longstanding commitment made by South Africa to develop and implement the Declaration and Programme of Action adopted by the 2001 UN World Conference against Racism, Racial Discrimination, Xenophobia and Related Intolerance in Durban, took almost 20 years to develop and was riven with internal debate about whether xenophobia should even be included in the Plan. Earlier drafts from the Department of Justice removed it entirely. The National Action Plan might, however, be read as signalling that Cabinet may finally have accepted that xenophobia is a real phenomenon and needs to be combatted. Nonetheless, government ministers have simultaneously continued to deny the existence of xenophobia and to displace blame onto criminals and migrants themselves. The Plan treats xenophobia in a perfunctory manner, providing no information about the nature and extent of the phenomenon and proposing no proactive steps to deal with it. The proposed remedies are largely reactive: condemning violence when it occurs, enacting hate crime laws, strengthening law enforcement and prosecuting offenders. One sentence is devoted to the need to monitor and report on attacks and another to ‘promot[ing] a spirit of integration through engaging communities where xenophobia is rampant’. The primary concrete measure proposed is to implement the recommendations of the Parliamentary Joint Committee and the SRG Report (RSA, 2019, 61). It remains to be seen whether government will act on the SRG’s more progressive and proactive remedies (SRG, 2015,172–178) or the calls for greater controls on migration and the disadvantaging of migrant businesses proposed by the PJC (2015, 36–39).

25One consequence of two decades of xenophobia governance by displacement in South Africa is, in fact, intensified xenophobia on the ground as there is no countervailing discourse about the benefits of migration. The police and justice system seem generally unable or unwilling to bring perpetrators of xenophobic violence to book, and xenophobic sentiment is licensed by policies toward refugee protection and migrants in the informal sector that are generally extremely hostile to both (Crush, Skinner and Stulgaitis, 2017). At best, the authorities (aided and abetted by international organisations) have brokered ‘deals’ to limit the number of migrants working in the informal sector, a dispute resolution move that Gastrow (2018) finds completely unconstitutional. With official policies of xenophobia denialism and blaming in place, there seems little hope that South Africa will address one of the core commitments of the Global Compact on Migration; that is, ‘to, condemn and counter expressions, acts and manifestations of […] xenophobia, and related intolerance.’ Indeed, if xenophobia does not exist, then, by definition, there is nothing to condemn and counter. In this environment, the consequences for migrants in South Africa will continue to be extremely deleterious and deadly.