Parties, issues and political fragmentation

Table 1: Levels of support for political parties

It will be seen that just before the 1994 election, the National Party (NP) was languishing at a low level to which it has subsequently returned. The effect of the I 994 election was to pull waverers towards the NP, voters which it retained for much of 1995 but which it has lost now. The question is whether, with the approach of the 1999 election, the NP can again peak when it comes to the crunch, or whether we are watching the party's slow decline.

The other features of Table 1 - the small but notable increase in support for the Democratic Party (DP) and Pan Africanist Party (PAC) and the African National Congress's (ANC's) stability - mirror what our own survey found.

However, the apparent rise in the Inkatha Freedom Party (IFP) vote is almost certainly illusory. It must be remembered that in April 1994 the IFP won over 10% of the vote, but that large numbers of its voters had been unwilling to reveal themselves to pollsters, hence their 5% showing in the table. They have merely become progressively more willing to identify themselves as IFP voters, with the proportion of 'don't knows-won't vote' shrinking correspondingly.

The NP had quit the Government of National Unity not long before our survey was held. We found that 55% to 60% of Africans were likely to believe that the impact of their departure was likely to be beneficial in virtually every area. But, of course, ANC party loyalty strongly affected these answers and it is perhaps more striking that over 40% of Africans thought their departure would be damaging.

Not surprisingly, 41% thought national reconciliation would be damaged. But 39% also thought economic development would be more difficult without the NP, 40% felt that reducing unemployment would be harder, 45% felt that it would be harder to achieve greater equality, 46% felt that it would be harder to improve governmental efficiency, and 51% felt that creating foreign confidence would be harder.

Thus although African pride is too great for the NP to be seen as indispensable (and for many this would be equivalent to 'white participation') the identification of NP or white politicians with efficiency, greater economic efficacy and ability to inspire foreign confidence, is quite striking. It should be realised, however, that in the popular mind these desirable attributes are probably linked more to whites as such than more narrowly to the NP. Thus it seems likely that President Nelson Mandela's attempt to woo the DP into the government was rooted in a similar concern to bolster the administration with these efficiency-efficacy-confidence factors.

Not surprisingly, large majorities of whites thought the NP's departure was damaging with regard to every issue area. However, it was notable that NP voters themselves - many of them coloureds and Asians - took a rosier view of how the government would do without the NP than did disaffected and alienated white voters who fell into the 'won't vote' and 'don't know' categories. Clearly, these groups include many voters who feel aghast at the loss of white power and are simply unsure how to come to grips with a situation in which this sine qua non has been lost or, as some of them would doubtless say, 'surrendered'.

There was a striking difference between Asian and coloured voters. Coloureds tended, by somewhat less dramatic majorities, to take the same view as whites - that De Klerk's departure would be generally damaging. But Asians were even more certain than Africans that it was a good thing and were less willing to give De Klerk or whites credit in matters of efficiency, economics and foreign confidence.

We then asked respondents: 'All racial groups are represented in the government. Are you happy with the balance between them?' In retrospect this loaded the question too positively. Many respondents would doubtless have been unaware of the actual racial composition of the Cabinet (one white, one coloured and five Asian ministers alongside 17 Africans), let alone that of the broader government {including deputy ministers), and clearly wished merely to express approval of multi-racialism.

Nonetheless, it was again striking that 67% of Asians were happy with the government's composition compared with only 44% of coloureds - despite the fact that these two groups contain roughly equivalent proportions of ANC voters. Clearly the differences in their representation at Cabinet level has not escaped either group. Only 32% of whites were happy with the balance: it seems clear that most whites do not see ANC whites as representing them at all.

On closer examination, one could see the difference was that coloureds' responses were more structured by partisan affiliation, with NP supporters far more likely to regret De Klerk's departure. Even they were clearly motivated to some degree by a racial pride, or at least a racial unwillingness to accept the 'necessity' of having whites in government. Among Asians, however, this was a far stronger factor, so that even the mass of NP-voting Asians were adamant that having the NP out of government would not damage government performance.

Those who were unhappy with the government's racial balance invariably wanted to see more of their own group in the government. Interestingly, although African voters had at an earlier stage expressed massive satisfaction with the constitution of a Government of National Unity with far greater white participation, they were just as satisfied now to have only one white minister. The most striking ethnic choice was made by African voters in KwaZulu-Natal, who not only wanted more Africans in government by 22% to 8%, but by a 17% to 4% majority wanted fewer Asians.

We then asked a series of questions aimed at discovering how far the transition to democracy in the country has been reflected in the growth of democratic attitudes at grass roots level. In any state in transition to democracy one would expect a lag in such attitudes, and this there indeed proved to be.

The peculiarity of the South African situation at present, however, lies in the enormous strength of party loyalty and mobilisation. So strong are party loyalties at the moment that they are a serious obstacle to a properly functioning democracy, precluding any real sense of alternative choice. Without doubt such loyalties and the high electoral mobilisation they produce will decline - but a great deal of South Africa's political history will be determined by the speed of that decline.

Some commentators, conscious of this gap between formal democracy and the weakness of grass roots democratic attitudes, have gloomily asserted that the mass of voters are extremely authoritarian. Thus the Mail and Guardian has suggested polls show that 60% of voters want no opposition at all.

Happily, this is not true: only 21 % of Africans, 19% of Asians and 10% of coloureds wanted to see no opposition. In fact, the African figure was depressed by the fact that only 2% of Western Cape and 15% of Eastern Cape Africans took this line - elsewhere the figures were much higher, peaking at 33% in the Free State.

But these authoritarian spirits were nonetheless outnumbered by those who wanted 'a strong opposition that subjects government to tough criticism' - 79% of whites, 51% of coloureds, 46% of Asians and 43% of Africans wanted this, so it was the choice of exactly half the electorate. The rest (31%) wanted 'a co-operative opposition that does not criticise too much'. If one lumps this latter category with the 'no-opposition' group we thus find the electorate splits right down the middle between those who want a strong and those who want a weak or non-existent opposition. This was not, however, a matter of ANC voters wanting to prevent or water down opposition to their government: the biggest single block of ANC voters wanted a strong and critical opposition.

Similarly, when we asked voters what sort of print media they wanted, only 27% of Africans (and 22% of all voters) thought that the press's job was to support government. Among voters as a whole, this ultra-supportive group was almost exactly balanced by a 19% ultra-critical group who wanted the press 'to watch for mistakes in government and be critical'.

But by far the largest section of every racial group took the balanced view that the press ought to 'praise or criticise the government as it sees fit1. Among all groups, those who actually read newspapers were more willing to see the press criticise the government than those who didn't read a paper.

Interestingly, IFP African voters were more likely than their ANC counterparts to say that the press must always support the government: traces of this more authoritarian and also more Africanist tendency were frequently to be found among IFP voters.

On the question of whether parties that lost an election would accept defeat and give up power peacefully, we found a great deal of pessimism: only 48% of voters thought all parties would do this, while 33% thought they wouldn't and 19% didn't know. The most pessimistic groups were whites - a 3:1 majority thought that not all parties would give up power peacefully - and coloureds, where 44% said they didn't know. Among the ANC's African voters 61% felt sure all parties would behave democratically, against only 27% who didn't.

Yet voters were extremely reluctant to say which parties would not behave democratically: 23% of whites pointed a finger at the ANC but 43% refused to name any party, while over 70% of all other groups refused to name any party. The real crunch, however, came with the question of whether voters would support efforts by their own party to stay in power even if it had been defeated in an election.

No less than 45% of whites said they would support such efforts, with that figure increasing as one moved towards the right of the political spectrum. Nonetheless, there was a remarkably high propensity for white supporters of all parties to give such an answer, betokening a sense of power loss that is clearly felt racially rather than just in party terms.

Overall, however, it was striking that while a 48% to 33% plurality had said they thought not all parties would accept defeat, a 60% to 28% majority said they wouldn't support attempts by their party to behave in such an undemocratic manner. That is, voters were much gloomier about how others might behave than they needed to be.

Among African voters those most likely to say they would support undemocratic attempts by their party to stay in power were PAC and IFP supporters: 69% of African ANC voters said they would not support such attempts against only 22% who said they would. It is possible, of course, that such voters feel so confident that their party will remain in power that they find it easy to reply in such a vein, but it is surely somewhat reassuring to find the majority party's core group taking a 'good loser' view by more than 3:1.

We then asked whether voters thought it better to have a large number of political parties so that everyone's views could get expressed or to have fewer parties 'so as to increase national unity and strengthen the country by not criticising the government too much'.

As it was phrased, this was an open invitation for one party or dominant party enthusiasts to show themselves, but once again the result was reassuring. African opinion was roughly equally divided: as many as 40% of African ANC voters wanted to have a large number of parties (against 49% who didn't), and it was actually only among whites and Asians that one found majorities wanting fewer parties. Given that all voters were presented with a menu listing the seven largest parties, even this was hardly an extreme choice.

Party loyalty was a different matter. Among Africans it was so strong as to qualify for the term 'secular religion'. Thus 48% of African voters, compared with a still high 21% of whites, said that they would 'support and stand by my political party and its leaders even if I disagree with many of its policies and actions' - and 26% said that they 'strongly agreed' with that statement.

This passionate avowal of party loyalty by Africans reached its highest level among the poorer and least educated groups - among women (52% agreement), the unemployed (57%), among squatters (62%) and in the Free State and Western Cape (67%). Such feelings ran notably higher among IFP supporters (66%) than their ANC counterparts (47%).

Some of the passion behind such answers became clear when we found that 65% of Africans, 45% of coloureds and 29% of Asians agreed that 'no matter how good their policies may be, I will never feel able to support parties that used to be supported by whites before 1994' - a formulation which, however unfairly, included the DP alongside the NP and the Freedom Front (FF). Interestingly, exactly the same proportion (67%) of ANC and IFP African voters gave this answer, but answers varied regionally. In the always politically correct North West, 83% of African voters gave this response compared with only 39% in the furiously discontented Eastern Cape and 29% in the Northern Cape.

Similarly, 45% of whites - 51 % of Afrikaans speakers, 38% of English speakers - said that 'no matter how good their policies may be, I will never be able to support political parties that were involved in the liberation struggle and violent protests before they were unbanned'. Clearly one must expect such figures to decline among all races as the events of pre-1994 subside into history. Already a majority of all non-Africans are willing to see their historic opponents as worthy of their political consideration.

Nonetheless, a majority of every racial group believed that South Africans would become one nation over time, with 19% of Africans insisting that they were already one nation now. Forty-six percent of whites averred that this would in fact never happen, with substantial minorities of coloureds (26%), Asians (19%) and Africans (14%) feeling the same.

Thus while all groups have signed on to the current theme of 'nation building', almost half of all whites and a quarter of coloureds are fundamentally disaffected from this African nationalist project.

On the other hand, almost no whites, coloureds or Asians agreed with the contrary statement that South Africa was mainly a country for blacks, to whom everyone else would have to take second place.

Encouragingly, only 24% of Africans agreed with this statement, with IFP supporters (31%) more likely to do so than PAC (29%), NP (28%) or ANC voters (24%). Africanist sentiment of this kind reached a peak among Africans in the Western Cape (38%) but was at its lowest among their brothers and cousins in the Eastern Cape (10%). The latter figure is support for the PAC in that region, and suggests a more general disillusionment with African rule.

The new nationalism is, however, intolerant of anything it feels smacks of tribalism: 54% of African voters felt that language and tribal groups concerned to preserve their own language or culture should be discouraged for the sake of national unity. (Large majorities of coloureds and Asians and virtually all whites disagreed with such a view.)

Given that Inkatha is frequently criticised by its opponents for 'manipulating ethnicity' and in general for being tribally based, it was interesting to note that intolerance of language-tribal groups was far higher (63%) among IFP than among ANC voters (51%). Those who most strongly disagreed with such views were found among the smaller black language groups, the better off, and among older Africans.

Controversy still rages among politicians as to whether South Africa should have a single capital city.

Among voters there is little debate: overwhelmingly all races and parties want things to stay as they are. If things are to change they would like to see everything moved to Pretoria. Almost nobody- even among Johannesburg residents - wanted the capital in Johannesburg.

The public is, however, considerably more exercised about law and order issues. Nonetheless, majorities of all groups were against vigilante organisations like Pagad that 'take law into their own hands'. Africans were most heavily against vigilantism (by 71% to 21%) and Asians the most sympathetic (56% to 35%), but there was no ground-swell of support for Pagad among coloured voters.

A real racial divide opened up, however, when voters were asked whether there was a low level of respect for law and order in the country: all non-Africans agreed emphatically but two thirds of Africans denied that this was so, perhaps because they felt that to say such a thing could be construed as a vote of no confidence in the first majority rule government.

Somewhat similar cleavages occurred in views of corruption. Whites and Asians thought there was more corruption now than under the previous regime, coloureds were less sure and African opinion was almost exactly balanced, with the consensus being that corruption levels were about the same now as then. The greatest confidence that there was now less corruption was evinced by African voters in the Free State, doubtless as a result of Patrick Lekota's anti-corruption campaign there. Lekota's sacking will have provoked grave disillusionment there.

Table 5: One national police force or provincial and city police forces?

Perhaps as a result of public anxiety over crime there has been a major reversal of opinion regarding federalising the police. Before the election, majorities took the centralist line in favour of a single national police force. Now there is a huge shift of opinion among all groups in favour of provincial and city police forces.

The high crime rate has probably also contributed to the unpopularity of illegal immigrants. Huge majorities of all groups wanted the toughest option of punishing illegal immigrants and sending them back home. Interestingly Africans, who suffer the most competition from such immigrants for housing and jobs, were somewhat more liberal than the other groups. It was also noticeable that Africans in the main cities and in squatter camps - that is, where the main concentrations of illegal immigrants are to be found - were more tolerant than others.

Opinion is more split on key economic issues. African and coloured opinion has far from caught up with the liberal assumptions of GEAR.

Table 6: What to do about illegal immigration?

Two thirds of Africans and coloureds were in favour of strict government regulation of business, while two thirds of whites thought that business should enjoy maximum freedom from government interference. Asians were more divided, but there was a pro-regulation majority there too. As Table 7 shows, a somewhat similar division of opinion exists over privatisation.

Table 8: Has every woman the right to legal abortion without condition?

One may perhaps glimpse the reasons why government has been so slow and hesitant in carrying out its privatisation pledges when one sees the huge African majority against such a policy, though it is worth noting that the most hostile groups - unemployed and rural voters - would be the least affected while blue collar workers, the most likely to be affected by such policies, were also the most positive about privatisation.

Finally, we found opinion among all groups crushingly opposed to abortion on demand. Coloureds were the most hostile group - partly because of the weight of Muslim feeling against abortion, with whites and Asians less hostile than others.

Among Africans, women were slightly more in favour of abortion than men, but among more liberal white and Asian groups men were likely to be at least as liberal as women on the issue. It was an issue around which religion played a significant role. Jews were by far the most pro-choice - indeed, they were the only group with a pro-choice majority - followed by Anglicans. Catholics were less opposed than one might have expected - perhaps a reflection of the church's decision not to campaign against the new abortion law as hard as it might - but African Zionists were particularly strongly opposed.

The dominating figure in our survey is, inevitably, the new African voter: Africans alone had no opportunity to vote before 1994 and they form three quarters of the electorate. Predictably, opinion among this huge group of voters has yet to crystallise on many issues. On the key question of the chieftaincy there is no consensus of any kind or among any group: there are ANC voters who want to strengthen the powers of the chiefs and, even more remarkably, 1FP voters who want to abolish the chieftaincy.

A similar lack of consensus emerged when we asked who such voters would like to see play a more prominent role in their local community: 42% of Africans wanted to see parties and politicians more involved while 58% opted for other groups - churches, civics, development forums and so on. But there were strong regional divisions: the Zionists of Northern Province wanted more power for the churches, Sotho speakers were hostile to chiefly involvement while Ngunis were relatively favourable to it, and so on.

It is worth pointing out that there has already emerged a 6% group of African voters who say they will not vote. This group is very different from the similar sized group of’ don’t knows' (who will also probably not vote). The deliberate abstainers are, in fact, a well educated and relatively affluent group, full of strong opinions and bitterly critical of the government. Many of their opinions coincide with those of the DP, but on some issues they take a decidedly Africanist line. It is a group to watch.

Table 9: What role for the chieftaincy?

It is important to note, however, that many African voters have strongly inconsistent views: they cannot be neatly sorted into conservatives, liberals and so on. In order to try to order this very disparate data we carried out a factor analysis - essentially, correlating every answer with every other answer.

The result does not entirely dispel the somewhat confused and confusing picture of myriads of African voters with discordant and sometimes incoherent views, but it does enable us to see clusters within that electorate, so that one can see particular bundles of attitudes cohering. These clusters are as follows:

• A group gripped by post-liberation euphoria: uncritical approval of government and its policies at every level.

• A group showing conditional satisfaction with the central government leadership: They are satisfied with Mandela, his government and some benefits they have brought - but want to see results.

• An authoritarian but non-racial group: scared of criticising government, they also want less opposition and government regulation of the economy. But they reject affirmative action and support racial equality.

• An appreciative, ethnically conscious group which tends to not blame government for non-delivery and to see it as less corrupt - but which is protective of language and cultural objects and which rejects blind party loyalty.

• A group showing disapproval and discontent, angry at non-delivery, dissatisfied with government, rejecting blind party loyalty

• and anti-abortion.

• Non-racial democrats who feel confident about the future of democracy, about holding opposing views themselves and who reject notions of Africanist dominance.

• A group characterised by democratic buck passing and a desire for patronage who support democracy in that they want a critical press but who are in fact very reluctant to imagine future electoral defeat for their party and who want it to implement affirmative action to their benefit.

• A group of strong democratic interventionists who accept the rules of democracy and thus the possibility of their party being defeated - but who reject privatisation and want government regulation of the economy.
  
  • Anti-democratic traditionalists who reject a critical press and Opposition and want more power for chiefs.

If one has to give a little more order to this mosaic one may perceive a general division between two kinds of political populism, one uncritical and symbolic, the other less symbolic and more conditional on getting concrete benefits from government. But what one does not find is groups nicely ordered along a spectrum: there is no cluster of African liberals and nor is there one of Africanists or African nationalists!

We carried out a similar factor analysis among whites, coloureds and Asians (taken together), producing the following clusters:

• Ethnic minority-based dissent: those rejecting affirmative action, critical of the government for non-delivery, rejecting blind party loyalty and committed to protection of culture and language.

• Comprehensive satisfaction: those satisfied with all levels of government and with government delivery.

• Authoritarians who want a compliant Opposition and a compliant press, who are themselves cautious of criticising government, who to promote unity want fewer parties and who reject privatisation and want government regulation of the economy.

• New South Africa enthusiasts who burst with admiration for Mandela, want a compliant press, do not blame government for non-delivery and think it less corrupt than its predecessors.

• Satisfied and optimistic democrats who feel their lives have improved and will improve further, who feel free to hold critical views and to express dissatisfaction with government and who reject traditional leadership

• Confident non-racial democrats who are confident of the future of democracy, are committed to non-racial equality, see South Africans as already one nation and who are comfortable about holding divergent or critical views.

• Authoritarian buck-passers - who want a critical press to safeguard democracy but are themselves cautious about criticising government, want fewer parties to promote unity and who don't want the party they support to accept defeat.

• Anti-abortion liberals, committed to non-racial equality and hostile to abortion on demand.
• A law and order group who have faith in the police, want local police forces and to promote unity want fewer parties.

The key point to note is, once again, the complicated mosaic of views and in particular the lack of an organising principle of general dissatisfaction. That is, the first group is strongly dissatisfied but this dissatisfaction is linked to ethnic minority concerns. There are a lot of clear-cut but small groups rather than a single large rallying of either pro-governmental or Opposition themes. Thus while these minorities predominantly support Opposition parties, the lack of a theme of general dissatisfaction affects the consistency of support for those parties.

Thus, for example, the DP is the most articulate and principled of the Opposition parties and its themes find a strong echo among many minority group voters - but at the same time the DP raises some issues or concerns which lack resonance entirely (for example, abortion on demand) with such voters and also fail to address the ethnic specificity of their perspective. The Freedom Front or Minority Front may, on the other hand, approach voters through that ethnic perspective but fail to project an image or policies capable of attracting support outside fairly narrowly defined groups.

Thus if one overall theme emerges from this survey, it is of actual or impending political fragmentation. At present a persistent cloud of political euphoria and the dominating figure of Mandela, as well as the sheer weight of the ANC bloc, give politics a monolithic feel. But as we have seen, Deputy President Thabo Mbeki seems unlikely to be able to replicate Mandela's unifying charisma, and the mood of euphoria - already significantly frayed - can hardly last forever.

The dominance of the ANC may continue, but increasingly the movement will have to reflect the fact that it is in reality a 'catch all' party, comprising a rag-bag of different and only partially coherent clusters of opinion. Similarly, the desire voiced on many sides for a united opposition cannot disguise the fact that Opposition politics too are a mosaic, with no single unifying theme yet developed.

It was the French Gaullists who first gave us the term 'catch all' party - when asked who the Gaullists were, De Gaulle answered 'the rush hour crowd' - and it seems likely that this sort of party is also best suited to South Africa's fragmented political reality.

At present only the ANC comes near fulfilling this role, which goes some way to explaining its dominance. But in fact this poses problems both for the ANC and its opponents. The ANC would still like to be a party of ideological discipline and is trying to force this mould on a reluctant reality. But already one can see splinters breaking off - large numbers disappointed over non-delivery, disillusioned voters of the Eastern Cape, the increased number of 'won't votes, don't knows', Holomisa, Lekota and Nkabinde supporters etc.

But the Opposition also has the problem that it is difficult for it to unite around, for example, a set of liberal principles when its target electorate is a rag-bag of disparate groups not driven by any single set of principles.

These are not, however, gloomy results. South Africa is a large and various country, bursting with pluralism of every kind. Both the juggernaut of apartheid and the answering juggernaut of the liberation movement have endowed South African politics with a huge symbolic dualism - liberation versus apartheid - which does not mirror or express the real social facts on the ground.

As that old cleavage recedes into history a different sort of political normalcy will take over - and there is no need to feel sorry about that.