Series Editor: Jonathan Crush

Southern African Migration Project

Migration Policy Series No. 22

PLEASE NOTE: Readers are welcome to reproduce and reference
this article as long as appropriate acknowledgments are given.


In May 2001, President Thabo Mbeki observed that all South Africans must be vigilant against “any evidence of xenophobia” against African immigrants. He noted that it is “fundamentally wrong and unacceptable” that South Africans should treat people who come to South Africa as friends as though they are enemies. This is a long-awaited and critically important statement from the highest level of the South African government. In the aftermath of the World Conference on Racism, Racial Discrimination, Xenophobia and Related Intolerance, the President’s words will hopefully be acted upon by all South Africans.

What is the evidence of xenophobia against African immigrants to which the President refers? Evidence of xenophobia can be seen in high-profile violent assaults on immigrants by hands of citizens (in which a number of refugees and others have lost their lives). But how typical are these xenophohic acts? Perhaps, as in some other countries, these are just the actions of a small group of extremists and are untypical of mainstream attitudes? What do South Africans really think of non-citizens, of African immigrants, of refugees and asylum-seekers? Are these attitudes reflected in, or contrary to, official thinking? And how and why does the media influence attitudes, for better or worse?

In 1997, the South African Human Rights Commission (SAHRC) identified xenophobia as a major source of concern to human rights and democracy in the country. Then, in October 1998, the SAHRC (in partnership with other agencies) launched a public and media education programme known as the Roll Back Xenophobia (RBX) Campaign. The campaign is designed to send a message, in the words of its founding document, that “South Africa needs to send out a strong message that an irrational prejudice and hostility towards non-nationals is not acceptable under any circumstances” (Appendix A). The powerful Congress of South African Trade Unions (COSATU), the country’s largest labour federation, has also condemned the growth of xenophobia in South Africa in no uncertain terms (Appendix B).

This paper first examines various reports and studies which made the claims about growing xenophobia in the mid-1990s. The paper shows that these claims were not based on systematic, national research, hut rather on anecdotal evidence or generalization from small and unrepresentative samples. In 1997, SAMP set out to rectify this problem with a series of nationally-representative surveys of citizen and non-citizen public attitudes towards immigration and immigrants. To date, the project has conducted three public opinion surveys of South African attitudes towards immigration; national public opinion surveys of immigration in 5 other SADC states; and two large surveys of non-citizens living in South Africa. Together, these surveys provide a unique data base for accurately assessing the attitudes of citizens and immigrants towards a wide range of immigration-related issues in the Southern African region.

The results of these particular surveys, and SAMP research into xenophobia more generally, have been published elsewhere (see end-notes and Appendix C). This paper presents an overview of the findings of public opinion surveys conducted from 1997-2000. The basic aims of this paper are as follows:

• to critically review the evidence for official and other claims that South Africans are intolerant of outsiders and African immigrants in particular;

• to summarize the results of various SAMP research into public opinion (citizen and non-citizen) on immigration issues;

• to analyse the extent and character of “xenophobia” amongst the populace at large;

• to provide concrete suggestions to government, the ANC, unions, NGQ’s and CSO’s and others for public education and other initiatives to counteract xenophobia and intolerance;

• to inform the public education strategies of initiatives such as SAHRC’s Roll Back Xenophobia Campaign.

In order to devise workable and effective strategies for countering xenophobia, it is extremely important to have a good sense of the extent of the problem and how it manifests itself. This paper is pub­lished in the belief that a greater understanding of this troubling phenomenon will lead to better and more workable counteracting strategies and policies.

The paper addresses four basic themes:

• the basic level and character of human rights awareness amongst the South African citizenry;

• citizen views of immigration and the presence of non-citizens in the country;

• the kinds of rights citizens are willing to extend to non-citizens including refugees; and

• migrant perceptions of their own treatment in South Africa.

The research shows that government and other agencies have a major task ahead of them if they are to convince South Africans of the value of a more open and inclusive immigration policy that is actually in the interests of the country. Attitudes are currently very negative and political leadership and public education need to confront this reality at the outset. The evidence for this assertion is as follows:

• South Africans as a whole are not tolerant of outsiders living in the country. The surveys revealed strong support for policies that would place strict limits on or prohibit immigration altogether. Fully a quarter favour a total ban, considerably more than in any other country for which there is comparable data. Nearly 80% favour a total ban or very strict limits. One in five actually feel that everyone from neighbouring countries living in South Africa (legally or not) should be sent home. Attitudes have also hardened. Between 1995 and 1999, for example, support for a highly restrictionist policy increased from 65% to 78%. Support for a policy that tied immigration to job availability declined dramatically from 29% in 1995 to only 12% in 1999.

• Between 1996 and 2000, government offered generous legal amnesties to longstanding contract workers, undocumented SADC country citizens resident in South Africa, and ex­Mozambican refugees. In total, over 350,000 people benefited from this effort to compensate black non-citizens for the actions of apartheid. However, South Africans, in general, do not support the idea of immigration amnesties. Antagonism is particularly intense amongst white respondents (at 76%). Black South Africans, perhaps more mindful of the history of the anti-apartheid struggle, are more generous with 40% unopposed to granting amnesties to undocumented migrants.

• The majority of South Africans currently believe that immigration and migration impact unfavourably on the country (with nearly 60% believing that they “weaken” society and the economy, and over 60% that they put a strain on South African resources). Fear of crime, threats to jobs and the economy, and disease are the leading reasons given for opposition to immigration. These are the same arguments advanced by those who oppose immigration everywhere.

• South Africans favour forceful approaches to controlling immigration. Respondents were asked their opinion of control-oriented policy measures such as turning on the electric fence on South Africa’s borders; putting more money into border protection; using the army to patrol borders; increasing taxes to pay for border patrols; requiring foreigners to carry identification; giving police the right to detain suspected illegal immigrants and penalizing those who employ illegal immigrants. With the notable exception of raising taxes, each of the measures enjoyed wide support with whites again more supportive than blacks.

• Respondents were asked what action they would take against people from neighbouring SADC countries. A third would be prepared to personally try and prevent migrants from moving into their neighbourhood, operating a business, becoming a fellow worker or having their children in the same classroom. They were also asked what they would do if they found out someone was “illegally” in the country. Forty-seven percent said they would report them (with 3% saying they would band together to force the person to leave the area).

• The South African Constitution guarantees basic rights and freedoms to everyone living within the boundaries of the nation-state. Many South Africans are clearly in disagreement. Around 40% are opposed to Africans from elsewhere enjoying the same access to health and educational services as South Africans. Rather more (54%) oppose giving the same right of access to housing. On the positive side, the survey found that 47% of respondents feel that Africans from other countries should still be allowed to vote in elections. Whites are significantly more negative than black South Africans on all of these issues.

• South Africans were asked about their attitudes to giving certain basic rights (freedom of speech, freedom of movement, legal protection, police protection and access to services) to legal and unauthorized migrants. There is a consistent pattern of conditional support for rights for temporary migrant workers. While only a quarter of the population thinks that these rights should always be accorded to legal migrants, around half are prepared to see these rights extended in certain circumstances.

• When it comes to unauthorized migrants, the picture changes dramatically. Around 85% of respondents feel that these migrants should have no right to freedom of speech or movement. Some 60-65% also feel that they should not enjoy police or legal protection or access to services. There is clearly a feeling, certainly not confined to South Africa, that by being in a country without official permission one sacrifices any entitlement to basic rights and protections, even if (as in South Africa) those are guaranteed by the constitution.

• The SAMP surveys show that South Africans accept that many newcomers are indeed genuine refugees. They also agree with the general proposition that refugees warrant protection (with 70% in favour). However, they distinguish between the general principle of protection and their own government’s responsibility in offering that protection. Only 47% feel that the South African government should give asylum and protection to refugees. When asked whether they would personally support the South African government paying for the cost of sheltering refugees, the response was decidedly lukewarm with only 17% in favour.

• An important question is what people understand by “protec­tion” and what rights they are prepared to extend to refugees. “Protection”, as understood by South Africans, does not extend to granting basic rights to refugees. Nearly 70% feel that refugees in the country should never have the rights of freedom of speech and movement, with only 3% feeling that these are automatic entitlements. Support for other refugee rights is only marginally more solid with less than 20% of respondents of the opinion that refugees should always enjoy legal and police protection in South Africa, or access to basic services. None of this indicates a citizenry well-educated in the circumstances and plight of refugees.

• One hypothesis in the literature is that proximity to and direct social interaction with non-citizens will impact citizen attitudes (negatively or positively). What emerges from the SAMP surveys is that many South Africans have no direct interaction and experience of foreigners, even from neighbouring states. In the 1998 survey, only 4% of respondents said they had “a great deal of contact” with people from countries in Southern Africa; with 80% having little or none. Those who have no contact are statistically most likely to have negative opinions of foreigners. The more contact they have, the more likely they are to have tolerant opinions. Type of contact is also critical. South Africans who are friends with foreign citizens are more likely to have positive views than those who live next to, work with or buy things from them.

• Citizens of neighbouring states are evenly divided on whether they are viewed positively or negatively by South Africans. A significant minority of people interviewed (30-50%) feel that South Africans have positive or very positive views of people from their home country. This suggests that not all migrants have personally experienced hostility and intolerance. Less than 30% expect bad or very bad treatment from South Africans. Asked, for example, about their general experiences in South Africa, 64% said it had been positive or very positive, with only 20% saying it was negative or very negative. The majority of migrants and immigrants are very much aware of the negativity that surrounds their presence in the country. However, only those who have had direct personal experience of hostility, abuse or prejudice are prepared to translate that general awareness into a firm belief that South Africans are intolerant and hostile.

This paper shows that South Africans are not tolerant of outsiders. These feelings are widespread and cut across indicators of age, education, gender, economic status and race (although whites are generally more hostile than blacks towards African immigrants). Many migrants and immigrants are aware that South Africans are not favourably disposed towards them. Yet they are also surprisingly generous in their expectations of South Africans. They expect to be treated well and, with the exception of those who have had direct negative experiences, they believe that they will be, not only by ordinary South Africans but by the police as well. But the situation is finely balanced.

The majority of South Africans are attitudinally hostile to outsiders but that they are not yet prepared to translate those attitudes into action; at worst they are “latent xenophobes.” The single biggest mitigator of negative stereotyping is personal familiarity. In other words, as South Africans become more socially familiar with non-South Africans their attitudes begin to change positively. This, in turn, suggests that public education programmes alone (the preaching of tolerance and good neighbourliness in the abstract) are unlikely to be successful. Of particular concern are attitudes to “refugee protection.” There should be great cause for concern that the reluctance to grant rights to refugees is uncomfortably close to the set of responses given for “illegal immigrants.” South Africans clearly continue to have difficulty distinguishing in their own minds between refugees and migrants. Government, NGO’s and refugee organizations have a major task to turn some latent sympathy for refugees into widespread popular support for genuine refugee protection that is consistent with South Africa’s convention obligations.

South Africa has made enormous strides since 1994 in building a non-racial, human rights culture appropriate to the new democratic order. But there are clearly considerable obstacles to be overcome before the citizenry is prepared to embrace the notion of equal treatment for foreigners and to ensure that African migrants (whether legal or undocumented) are constitutionally entitled to basic human and labour rights, simply by virtue of being on South African soil.

It is hard to see how even the best of public education campaigns can, in isolation, effect the necessary shift in public attitudes. In that context, the World Conference provides an opportunity for South Africans to reflect and seek advice on how to turn back the insidious tide of racism and xenophobia. What is required from those in government, civil society and the media is a new approach. Instead of isolating and stigmatizing all migrants as “aliens” and “foreigners” or preaching against xenophobia in the abstract, there needs to be acceptance and promotion of the presence and contribution that non-citizens are, and can, make to the country’s growth and development.