What went wrong with the "rainbow nation"? The shocking attacks against foreign nationals in South Africa - mostly immigrant workers and asylum-seekers from other parts of the continent, including Mozambique, Zambia, Malawi and Zimbabwe - raise the question in sharp new form. These attacks escalated in the days after 11 May 2008, since when around sixty-two people have been killed and 670 injured, and many thousands more forced to flee their homes in a number of South African townships (initially from around Johannesburg, but later spreading to Cape Town and other areas). Both the violence and the expressions of xenophobia that accompanied it, have focused attention on failures of political leadership and social policy. More broadly it has forced South Africans to look inwards and ask whether something more fundamental is happening that needs to be addressed.

Faten Aggad is a researcher at the South African Institute of International Affairs (SAIIA), based at Wits University in Johannesburg

Elizabeth Sidiropoulos is the national director of the South African Institute of International Affairs (SAIIA), based at Wits University in Johannesburg It is not that attacks against foreign nationals in South Africa are unknown. The first such killings after the end of apartheid and the installation of one-person-one-vote government in 1994 occurred in the mid-1990s when members of the Somali community were targeted in the Western Cape province. These, however, were sporadic and did not reach the levels seen in 2008.

This precedent, however, is itself a reminder of a fact that can easily be forgotten in the rush to analyse and judge the attacks: namely, that immigrants - especially those of African descent, and from far countries as well as near - have always been a component of South Africa's social fabric. In particular, many migrants found work in South Africa's mines - and still do. The loose assertion that South Africans are inherently xenophobic or are disposed to be hostile to foreigners (a feature of commentary on the May attacks) ignores this longer, larger history. But if "xenophobia" begs as many questions as it answers, what did trigger these days of carnage?

The policy hole

The first place to look for an explanation is in the South African government's own policy record: especially its inability to address the problems of poverty and unemployment, and its lack of leadership on the issue of immigration and refugee policy.

The record has its positive elements: the African National Congress (ANC) government has since 1994 made substantial progress in delivery of social and economic services (such as water, sanitation, electrification). But many South Africans remain economically marginalised (the expanded unemployment rate, a key indicator, is about 36%). Though the government's post-1996 neo-liberal macroeconomic policy is largely credited with stabilising the economy, it is also blamed by many (especially the ANC's trade-union allies) for favouring the new black business elite at the expense of millions of everyday citizens.

The groundswell of dissatisfaction with the government - particularly over service-delivery - was reflected in the protests that surrounded local elections in many parts of the country in 2006. Jacob Zuma's victory in being elected ANC president at the December 2007 party conference at Polokwane - in a campaign supported by the Congress of South African Trade Unions (Cosatu) - was a further indication of a growing constituency that felt the government had become too remote from people's daily lives.

These discontents may have reached an invisible tipping-point with the effects on South Africa's poor of the global economic crunch, most notably in rising food prices and inflation. Many citizens were already excluded from the benefits of economic progress; when their already straitened circumstances became even worse as worldwide conditions as well as local realities pressed more tightly on them, the outcome was violence against an available though blameless target.

The migrant dimension

There is also, however, an immigration dimension to South Africa's troubles. The country may have its economic problems, but it is wealthy in comparison to many of its neighbours. This means that - its strict entry and settlement conditions notwithstanding - it has acted as a magnet to many people from the rest of the continent, who enter the country both legally and clandestinely. The number of illegal immigrants has accelerated over recent years, and now constitutes anything between 3 million and 5 million (in a country of around 47 million); Zimbabwe's deteriorating economic and political situation in particular has accelerated the flow. These new arrivals have had a huge impact on South Africa's social landscape, including demand for jobs, housing, and other services; the results of the inflow include periodic outbreaks of tension that threaten to break out into open violence, though (before the explosion of May 2008) very rarely do.

Most immigrants succeed in finding employment with farmers and businesses (albeit without documentation, and on very low wages). But this influx does nothing either to address South Africa's shortage of skilled labour or to alleviate the predicament of its own large pool of unskilled labour - many of whom cannot be absorbed into the formal economy.

Some of these African migrants have been in the country long enough to acquire citizenship, or have married South Africans and thus acquire a route to legal residence. At the same time, high levels of corruption in South Africa - especially in the department of home affairs - mean that some illegal immigrants are able to secure identity cards, which then allow them to apply both for government grants and for housing.

The confusion surrounding these different categories of migrants is exacerbated by Pretoria's failure to develop an effective policy on refugees and asylum-seekers. There have long been calls by international organisations for South Africa to introduce a stronger refugees bill that would help it to accommodate the vast influx from Zimbabwe especially, but it was only in March 2008 that proposed amendments were put before parliament.

Also in openDemocracy on South Africa's politics:

Nahla Valji, "South Africa: no justice without reparation"
(2 July 2003)

Achille Mbembe, "South Africa's second coming: the Nongqawuse syndrome"
(15 June 2006)

Achille Mbembe, "Whiteness without apartheid: the limits of racial freedom"
(4 July 2007)

Roger Southall, "South African lessons for Kenya"
(8 January 2008)

Roger Southall, "South Africa and Zimbabwe: the end of ‘quiet diplomacy'?" (29 April 2008) This revised bill was long overdue. But the measures were further delayed by the political sensitivities that have surrounded many aspects of South Africa's policy towards Zimbabwe: for example, giving refugee status to Zimbabwean nationals would imply admitting that there is a crisis in Zimbabwe, which (to this policy's architects) would have unwelcome knock-on effects.

This lack of a coherent approach to immigration in response to accumulating migration pressures reinforces the government's failures in other areas, and forms part of the context for the outbursts of violence.

The warning-sign

In these difficult circumstances, there is an urgent need for comprehensive, creative and sustainable government action, on three fronts:

* addressing the fundamental problems of high unemployment, poor living conditions and endemic corruption in many local governments

* adopting a refugee policy that follows international best practice, and an immigration policy that seeks long-term solutions not short-cuts

* helping to bring peace and stability to conflict-ridden countries such as Zimbabwe.

None of these is a panacea; for example, a peace-building policy towards Zimbabwe would not in itself guarantee reduced migration flows, and to solve the unemployment problem would not make prejudice or violence disappear.

Moreover, the effects of the attacks in South Africa and the region have created yet further problems. The capacity of neighbouring governments to cope with a large number of returning immigrants who have fled South Africa in fear is one; Mozambique, for example, has seen an estimated 36,000 of its 2 million nationals go back, creating a social problem of reintegration and an economic one of the loss of remittances from those individuals who were in work.

The regional impact of the attacks is bound to be damaging on all sides. This includes the influence on other African countries' perceptions of South Africa's own much-touted Africa agenda and continental leadership. And closer to home, the violence has shown that a domestic environment where poverty is rife and resources to help the poor scarce, and where key foreign-policy concerns such as Zimbabwe are badly managed, is in important respects no longer "domestic" at all.

The boundaries between "internal" and "external" policy are no longer as clear as they once were (see Saskia Sassen "Globalisation, the state, and the democratic deficit", 18 July 2007). The flow of economic and political refugees to South Africa from Zimbabwe and other African countries; the global financial crisis and an overall rise in the price of staple foods (which falls disproportionately on the poor) - these factors have direct social ramifications, which make it even more urgent for national and local governments at home to correct their flaws. The context in which South Africa is involved is changing fast. Its problems can, if they are not handled well, destroy much of what South Africa has tried to build - both at home and on the continent - since the end of apartheid. The anti-immigrant violence is both a morbid symptom and a warning-sign.