In South Africa, the first formal meeting between the ANC and the government began on 2 May 1990 at Groote Schuur in Cape Town, and lasted for three days.

The discussions were characterized by openness on both sides. A commitment to making the maximum effort at finding common ground to eliminate tension, and a desire to make a success of the meeting were equally evident in both parties. By the second day, delegates were asking why these talks had not taken place years ago. According to Thabo Mbeki, within minutes of sitting down ‘everyone understood that there was nobody there with horns.’

The meeting produced an agreement, hailed as a major breakthrough.

The Groote Schuur Minute showed that problems, which were previously perceived as being intractable, really could be amicably resolved. Within three days, a remarkable rapport had been established between parties in an experience described as ‘cathartic.’ The government delegation learnt of the frustrations of being banned and why the ANC was forced to take up arms, while the ANC learnt about the economic implications of sanctions. Both sides realized that they had to rethink some of their assumptions.1

And:

Humour was also a great binding force, with jokes made about the most unlikely subjects. Former prisoners chuckled with their captors over the circumstances of their arrest; terrorists showed off their prominent ribs, and joked that they had grown thin evading the security police; captor and captive reminisced about mutual acquaintances.2

The Sowetan — the only black paper in South Africa with a mass circulation – commented in an editorial that: “Last week’s breakthrough talks between the ANC and the Government have shown it is possible for even the fiercest antagonists to sit down and discuss their problems.”3

In August 1990, the government and the ANC met in Pretoria. The result was the Pretoria Minute, an agreement that all the obstacles identified by the ANC as obstructing negotiations would be removed or addressed.4 Before the meeting took place, Joe Slovo had proposed to the ANC NEC that the ANC should unilaterally suspend the armed (just the armed or the armed struggle?), giving the ANC the moral and political high ground. Mandela himself was at first dubious, but on reflection fully concurred, and subsequently persuaded the NEC to carry the motion unanimously. When the Pretoria meeting began Mandela announced that effective immediately, the ANC would suspend the armed struggle “in the interest of moving as speedily as possible towards a negotiated settlement and in the context of the agreement reached.”5 Mandela, however, was to say over and over again to the ANC membership – and perhaps it was necessary to do so to keep the hawks in the ANC who had opposed the NEC decision to suspend the armed struggle – that “we suspended action; we did not terminate the armed struggle.”6

This had a number of salubrious effects. It took the issue off the negotiating table and thus precluded the spectacle of the ANC having to accede to the NP’s demands in this regard since there was little prospect of the government negotiating under circumstances that would countenance the continuation of the ANC’s armed struggle while negotiations were under way. Second, it gave De Klerk more political breathing space to maintain the cohesive support he needed in the NP to ensure that the process would continue without it appearing that the?? would not come from the “barrel of a gun,” it would have been foolhardy for the ANC to take an uncompromising position on an issue that would not advance their strategic objectives, despite the fact that the armed struggle was essentially a myth, a powerful symbol of mass mobilization, particularly for black youth for whom militancy was an essential part of the culture of resistance. Advancing the process had become more important than the promulgation of putative symbols, which had served their purpose.

The D.F.Malan Accord, confirmed at the D.F.Malan Airport in Cape Town on 12 February 1991,effectively removed most obstacles, especially with regard to the meaning “of the suspension of the armed struggle,” that stood in the way of a multi-party conference. The ANC undertook not to carry out armed attacks or infiltrate South Africa with men or weapons. Recruitment of cadre within the country for training would also cease. There would be no statements inciting violence, threats of armed action or the creation of underground structures. There was no agreement, however, on the surrender of ANC weapons, the identification of arms caches, the demobilization of cadres, or the establishment of self-defense units. Furthermore, it was agreed that membership of Umkonto We Sizwe would not be illegal. That participation in the democratic process implied and obliged all political parties to participate in the process peacefully and without resort to the use of force, and, hence that both parties “accepted the principle that in a democratic society no political party or movement should have a private army.7 Where do you want to close the quotes?

However, the goodwill that had been so pervasive between the ANC and the government in their initial contacts following Mandela’s release began to erode when an eruption of violence, almost all of it black-on-black, began to envelop the country and spread at frightening speed. Random shootings, mid-night massacres, civilians, commuting to and from the townships, shot dead at train stations or on the trains themselves, in sheebens, or common places of congregation. The violence was aimless, yet obviously carefully orchestrated to instill much fear. The appearance of aimless violence leaves everyone feeling vulnerable, has a crippling impact on how society runs, and creates a degree of uncertainty among ordinary people that ultimately paralyses social interaction. If the objective was to make people feel that no one was safe and that there was no one on whom they could rely to protect them, the perpetrators were spectacularly successful. If the objective was to drive a wedge between the ANC and the government, they were even more successful.

Mandela blamed the government for the violence and for the failure to bring it under control. Although he did not implicate De Klerk personally, he pointed his finger at the Nation Intelligence Service (NIS), the Civil Cooperation Bureau, and Military Intelligence and other clandestine security groups behind the carnage.8 He even provided De Klerk with affidavits supporting this claim.9 On one occasion, he angrily declared, “While the Government is talking about peace and negotiations, it continues to wage war against us.”10

He could not understand why De Klerk, with the far-reaching powers of the state president at his disposal, could not get to the bottom of matters, why with the immense resources of the state at his command, he could not bring recalcitrant elements in the security forces under control. He wondered whether the government had actually lost control of the defense forces or whether the violence was a deliberate ploy on the part of the government to destabilize the ANC, undermine its support base, and prevent it from organizing ANC branches on a countrywide scale. In short, Mandela believed – and repeatedly charged that the government was doing nothing to counteract the atrocities. Perhaps what bothered him the most was what he perceived as De Klerk’s apparent lack of concern about the loss of black lives in contrast to what he believed De Klerk’s re-actions would be if similar attacks were being directed against whites. Black lives were cheap and counted for nothing.

The ANC NEC met to consider the situation and issued a statement:

If the government does not carry out its duties [it said] we will have to find ways and means to defend our people against these criminal attacks…. Then we will have no alternative but to concede to the demands of our people for arms. We are reluctant to do this because we are committed to the idea of peace. But we will not stand down and see our people mown down like dogs. We will have to defend them.11

And an ANC document, For the sake of our lives, bluntly stated that “in the wake of the ugly violence unleashed against our people by security forces, vigilante groups, and hit squads, it is imperative that our liberation movement takes responsibility for guiding and building people’s self defence movements.”12

As a result, Self Defence Units (SDUs) were formed in volatile townships and rural areas.13 They were comprised, for the most part, of youths and by the “comrades,” many of whom had been active in the UDF’s campaign to make the country ungovernable in the latter part of the 1980s. Few had had any training in the use of weapons, and many were as intent on using their arms to enrich themselves at the expense of the communities they were supposed to protect as they were in protecting the communities from violent onslaughts. Many formed gangs and fought among each other for territorial control of sections of their townships; many townships were virtual hostage to rampaging SDUs and the surrogates they spun off.14 In the following years, the ANC found itself saddled with a problem it found difficult, and at times, impossible, to handle.15 The genie had gotten out of the bottle and had no intention of passively crawling back to its former habitat. Moreover, the arming of the SDUs flooded the townships with weapons creating a stash of the wherewithal necessary for engaging in criminal activity. If a weapon couldn’t be used for one purpose, there was always another in waiting.

In South Africa, the first multi-party constitutional talks took place on 20 and 21 December 1991 in the World Trade Centre at Kemptom Park, Johannesburg. This forum was called the “Convention for a Democratic South Africa” (CODESA). Its deliberations continued until May 1992 when the failure of the ANC and the NP to agree on the weighted majority that would be required in an elected constituent assembly to ratify a new constitution brought negotiations to an abrupt halt. Nevertheless, CODESA 1 achieved much, and most of the measures agreed to in its working groups were incorporated in the final settlement16.

The Boipatong Massacre in June 1992 collapsed the process.17 In order to maintain some conduit of connection between the government and the ANC, the two decided to establish a “channel bilateral” 18 — a line of communication between Cyril Ramaphosa, Secretary General of the ANC and the ANC’s chief negotiator, and Roelf Meyer, Minister for Constitutional Affairs in the NP government and the government’s chief negotiator. As a result of their personal and professional rapport19 and the fact that Derek Keys, the avowedly apolitical Minister for Finance, had informed their principals, Nelson Mandela and FW De Klerk, that the economy was tottering on the edge of bankruptcy and that unless they got their acts together, and swiftly, there would be precious little for either of them to govern, Ramaphosa and Meyer were able to convince their principals to participate in an indaba that led to the signing of the Record of Understanding in September 199220.

This agreement led to a further series of indabas to iron out remaining details in early 1993, and the resumption of negotiations at Kempton Park in April 1993 (CODESA 11). These negotiations resulted in an agreed settlement in November 1993, the country’s subsequent first non-racial elections in April 1994,21and South Africa’s first Government of National Unity (GNU), although the final text of the agreement called for voluntary not constitutionally mandated power sharing.22

Intra-party rivalries for political control of the community on one or other side of the divide often prove to be more troublesome to whatever parties are predisposed to negotiations than inter-communal rivalries over the division of power. Thus, moderate parties in negotiations are frequently obsessed not with positions taken by their counterparts on the other side of the divide, which they may share, but by how their being seen to share similar reconciliatory positions may be used against them by parties opposing them within their own community to undermine political support among their constituents. Hence, they find themselves, having to take measure of proposed initiatives not out of their being opposed to them but to preclude the perception of their being weak, easily manipulated, and not representing the best interests of their community. This makes the process more intricate, progress slow and incremental. It underscores one of the most important principles that is the hallmark of successful negotiations: always put yourself in the shoes of your opponent, for without an understanding of the difficulties he faces in his community, you cannot help him overcome them, and hence, you cannot advance your own position.

In this sense — given the complicated dynamics that drive negotiating processes, dynamics that for all their manipulative tactics, no protagonist is able to manage and control to his own advantage — the refusal of the Conservative Party to participate in CODESA 1 and CODESA 11 – was a catastrophic mistake on its part, which worked to the advantage of both the government and the ANC, and later to the advantage of the NP.

In March 1982, a rupture occurred in the National Party (NP) when 16 MPs, under the leadership of Dr. Andries Treurnicht, leader of the NP in the Transvaal, abandoned the NP – they saw the structures under consideration in the NP for some limited form of power-sharing with Coloureds and Indians as a fundamental betrayal of the principles of Verwoerdian apartheid23 – and formed the Conservative Party (CP). In the 1997 elections for whites only — Coloureds and Indians had their own elections; the CP increased its representation to 22 MPs. In the 1989 election, with De Klerk at the helm of the National Party, the CP further increased its representation to 39 seats and secured over 50 per cent of the Afrikaner vote.24 If an additional 600 votes in each of nine marginal NP constituencies had swung to the CP, the NP would have wound up with 82 seats, one short of an absolute majority, leaving the balance of power with the liberal anti-apartheid Democratic Party (DP). While De Klerk hailed his victory (93 seats out of 165) as an endorsement of the NP’s platform, the fact that both the CP and the DP had made inroads into its constituency meant that it was being squeezed from both the right and the left, and that the results did not constitute a radical shift in white opinion, but rather that support for the right was growing, albeit at a slow rate, slower than might be expected, given the broad outline of reforms the NP had proposed.

When CODESA 1 opened, the absence of the CP gave the government more maneuvering space than it would otherwise have had. Hence the degree of consensus that was emerging in the early months of 1992 from the five working groups augured well for eventual agreement on the way forward. The negotiations were not strung out by the presence of a right-wing party that would have been but an obstacle in the way of every proposal advanced by either the NP or the ANC, and stretched the concept of “sufficient consensus” for decision making to its limits. Indeed, such was the level of collegiality among the working groups that in February 1992, “the ANC argued that sufficient progress had been made, constituting a major breakthrough. It seemed possible to complete CODESA’s work within the next six weeks.25 The government, however, did not share this optimism.”26 Since the crucial by-election in Potchefsroom, a hitherto bastion of National Party support, was due to be held the following day, to which the media had cast as a mini-mandate on the government’s performance at Kemptom Park, it might be expected that the NP would go out of its way to squash any suggestion that the country might be ready for elections within six weeks or that it was anywhere close to agreement with the ANC on the major constitutional issues.

But even though the CP were not part of the process itself, it’s continued barrages of criticism that the NP was selling out the Afrikaner nation began to exact a toll on the NP. De Klerk:

Every effort that we made to launch negotiations and every agreement that we reached [with the ANC] were dismissed in Conservative Party propaganda as concessions and deviations under pressure that the election promises that we made during the 1989 election campaign. The ongoing political violence and the ANC’s inability to control its followers were energetically and often deftly exploited by the right wing to sow fear, suspicion and doubt in the minds of white voters.

At the end of 1991 the National Party suffered a painful defeat in a by-election in Virginia, a gold-mining community in the Orange Free State. In 1989 we had held the seat with a narrow majority. In the 1991 by-election the Conservative Party had won with a majority of more than 2,000. It was a large swing and, according to the experts, indicated that the Conservative Party would be able to win a general election among white voters.

The mandate that I had received from the white electorate was visibly slipping away from me and the National Party. Our credibility was being seriously eroded.

A second by-election in Potchefsroom, which had for decades been a safe NP stronghold was scheduled to take place on 19 February 1992. Both the NP and the CP used the election as a referendum on the course the NP’s reforms were taking. The NP was trashed by the CP; the result was widely interpreted as an indicator of the level of white antipathy toward the NP’s constitutional reform proposals, and as a compelling sign that a country wide election among whites would lead to a Conservative Party victory. The Conservative Party demanded an immediate election.27

De Klerk, cleverly, opted for a referendum, and put a simple question to the white electorate: ” Do you support continuation of the reform process that the state president started on 2 February and which is aimed at a new constitution through negotiations?” The wording was brilliant in its vagueness, yet so implicitly direct in the implications of a no vote that it didn’t give whites a choice, rather it presented them with a subliminal ultimatum: Armageddon or else. Indeed, De Klerk was confident enough of the outcome to promise that if he lost, he would dissolve the NOP government and force an election.28 And he had the support, albeit grudgingly given, of the ANC and the media – indeed, of all institutional organs of opinion and influence. 29

De Klerk won the referendum convincingly30 and no longer had to look over his shoulder. The CP returned to the trenches and complained about the unfairness of it all. But the CODESA participants did not want the CP to marginalize itself, or perhaps fearing right wing violence lurking in the hinterlands of the Transvaal and Orange Free State, attempted to woo it into the process. Indeed, the working group dealing with constitutional principles agreed to discuss the principle of self-determination and its application in the South African context in the hope that it would alleviate the concerns of the CP and others further to the right and provide the space they needed to convince their constituency that there might be merit in joining the negotiations, if only to act as a watch guard for their interests.31

But the referendum altered the chemistry of CODESA 11: Rather than seeing both the government and the NP as being unchained from the shackles of the right, De Klerk according to a number of political observers,32 interpreted the results to mean that he could slow down the pace of negotiations and dictate the course of the transition.

“Coetzee’s [Kobie] role changed noticeably after the referendum,” says one WG1 participant. ‘It was as if he had decided the government could not get either the decisions it wanted in the group, or scupper it. It was probably the size of the yes vote that did it. Had the result been narrower, we would probably have made more progress. Do you want to close this quote here?‘ Asmal [Kadar] agrees: “The referendum sent inappropriate signals to government. A week before they were trying to resolve issues at CODESA. Thereafter there was a perceptible change in attitude.”33

De Klerk had run on the platform, “Vote yes, if you’re scared of majority rule,” On the campaign stump, his unconditional guarantee was that no NP government would be party to a negotiated settlement that would mandate majority rule. Perhaps, the ANC’s acquiescence to the referendum and the fact that it did not publicly repudiate De Klerk on the question of majority rule led De Klerk to think that the ANC would be less than insistent on simple majority rule, or that he could cobble together an anti ANC coalition in a power-sharing government. Hoping that the NP would at last rid itself of the albatross the CP had become with a convincing victory, the ANC could not afford to be seen to be at odds with the NP since that would only buttress the position of the CP that the ANC were only interested in a winner-takes-all settlement.

Less than a year earlier, the ANC’s soon-to-be former Secretary General Alfred Nzo had provided delegates to the ANC’s first annual congress held in South Africa in 33 years with a confidential assessment of the shape the ANC was in: “We lack creativity, energy, and initiative,” Nzo wrote. “We appear very happy to be pigeon-holed within the confines of populist rhetoric and cliché.” Attendance at ANC rallies had plummeted. Preparations for rallies were incompetent; they were poorly advertised. Organization was shoddy, and rallies frequently started hours after the time they were supposed to. Recruitment of new members was an ongoing problem. In the first year since its unbanning, the ANC had managed to sign-up a mere 200,000 members. Among many veterans of the struggle there was a joke: it had been easier to join the movement when it was banned than when it was legal. And Nzo concluded with the damning observation that “Clearly we have not utilized our full potential to mobilize millions of our people into effective action,” and a warning that the ANC itself was in danger of being “removed from the leadership pedestal it now occupies.”34 One can easily understand why De Klerk thought that he might, perhaps not have to surrender power at all or that if he did he could do it under circumstances highly favorable to the NP. Indeed, a number of polls, taken around the time CODESA 11 was winding up, indicated that ANC support had, at that point leveled off at 45 per cent, reinforced that kind of thinking.35

Or perhaps, the fact that his actions heretofore – the gamble to release Mandela, unban the ANC and the SACP, his popularity in the townships, the approving receptions he received in various capitols around the world when he first ventured overseas, his even more daring gamble on the referendum – had made him feel invulnerable, had him believing that his political instincts always paid off in handsome dividends, and that he could still coerce the ANC into substantial compromises on basic constitutional issues.

The question that brought matters to a head leading to an irresolvable stalemate in CODESA 11 was the question of the percentages that would be required in a constituent assembly to pass the new constitution. After a great deal of haggling and compromises on both sides, WG2, which dealt with constitutional issues, agreed that the CODESA parties would write an interim constitution, and that they would also agree on constitutional principles that would bind the elected constituent assembly, that is, principles that no party or combination of parties in the constituent assembly, no matter what proportion of delegates they represented could amend. These principles envisioned as being carved on tablets of stone, immutable, and above the constituent assembly’s mandate to rewrite the interim constitution as a final constitution.

Hence, the bottom-line decision for WG2: what percentage of delegates in the constituent assembly would be required to ratify the final constitution, and in the event of deadlock what deadlock-breaking mechanism would the assembly employ to break the deadlock?

Obviously, the ANC would be looking for the lowest threshold it could bargain for, the NP for the highest. Obviously, the ANC would want to see as few principles as possible, drawn up by the unelected and unrepresentative CODESA in which De Klerk could count on as many allies, entrenched in the final constitution drawn up by the elected constituent assembly in which the ANC could reasonably expect to have a representation that would be much higher than its CODESA representation, and the smaller parties’ overrepresentation at CODESA much whittled down. Obviously, the government/NP would see matters in a diametrically opposite way. The ANC wanted to agree to a set of procedures that would ensure that the final constitution unequivocally mandated the rule of the majority;36 the NP government a set of procedures that would result in a final constitution that would mandate entrenched power sharing.

Both government and the ANC faced each other across a political chess board, each player keenly aware of what the other’s thinking was, each looking at every proposal put forward by the other, no matter how tempting or conciliatory from an analytical perspective that sought to determine not what was in it for themselves, but what hidden advantages might accrue to their opponent. The key to success, like in chess, was in being able to think through the implications of a permutation of moves, both on your part and your opponent’s, and to discount those implications before you made your next move. And like chess there could be no trusting of your counterpart’s intentions. Indeed, if you gave in to the impulse that a gesture to accommodate did not have a hidden agenda, you had already lost. In this sense, neither side could win since whatever tenuous bonds of trust that had been established between the two in the run-up to the plenary session of CODESA 11 had to be discounted. The fact that the continued proliferation of violence, which each side blamed on the other, added to the climate of ‘prudent’ suspicion and wariness, was not conducive to compromise emerging, especially when the deadline for reporting back to the plenary session that passed and could no longer be ignored.

Three percentage markers were put on the table during the final frenetic days – and – hours of negotiation. First, the NP (and the IFP) insisted on a 75 per cent majority for the adoption of the constitution, 80 per cent for the bill of rights, and the entrenchment of power sharing among the immutable constitutional principles. None of these proposals were acceptable to the ANC. It countered with an offer of a two-thirds majority for the ratification of the constitution and a 75 per cent requirement for the bill of rights, and no provision for mandated power sharing. The latter was simply out of the question. The government came back with a new offer: a two-thirds vote to pass most clauses in the new constitution and a 75 per cent majority for matters referring to a bill of rights, devolution of power and multi-party democracy. But the offer was contingent on the ANC agreeing to there being a Senate representing minorities, whose membership would be appointed by CODESA, that would also have to pass the constitution by a two thirds majority. Effectively, this would give minorities a veto over critical clauses of the constitution, meaning the adoption of a final constitution could drag on ad infinitum and making amendments to the interim constitution almost impossible to achieve unless the ANC had more than 75 per cent of the votes in the assembly – also almost impossible to achieve. For all intents and purposes the interim constitution would become the final constitution. Once again, the ANC rejected out of hand the government’s proposals.37

At this point the ANC had come to the conclusion that CODESA 11 was going to fail, that their differences with the government over the question of the percentages that would be required to pass the final constitution had become irreconcilable, but they did not want to be seen as the spoilers, as the party that brought the process to a halt, thus damaging the image of the ANC as the party that, on the one hand, upbraided the government for not moving forward fast enough, and on the other, seemed quite prepared to pull the plug on the process. Ramaphosa and his colleagues brought the matter to Mandela late in the evening on 14 May. Having heard their analysis, Mandela made his fateful decision: postpone CODESA 11.

Hence Ramaphosa’s machinations: make an offer to the government that it would have to refuse, thereby letting the ANC off the hook. On the morning of the last day of the plenary session, 15 May 1992, Ramaphosa agreed to accept the NP proposal that a 70 per cent threshold would be required to ratify all clauses of the constitution except for the bill of rights which would require a 75 per cent majority. In addition, the government would have to drop its demand for an appointed Senate in favor of an elected one. But he added a rider which he himself concocted without consultation with the leadership of the ANC: If, after six months, the constituent assembly failed to agree on a constitution, a referendum would be held at which point the votes of a simple majority of the electorate would suffice to pass the new constitution. The offer, of course, was rejected by the government, which foresaw a situation where the ANC would sit idly by for six months, have their referendum, and get a constitution of their own choosing. Stalemate. And to mix metaphors, advantage to the ANC.

De Klerk staunchly denies that the government was responsible for CODESA 11’s collapse. On the contrary, he lays the blame squarely on the shoulders of Ramaphosa. Indeed, in one sense, the final meeting of WG2 to try and resolve its differences was a charade. Ramaphosa had his instructions, and Mandela had little doubt that Ramaphosa’s not inconsiderable ingenuity would rise to the occasion. Ramaphosa himself admits to engineering the collapse of CODESA 11, saying that he wanted to show “the people of South Africa that they were dealing with an enemy that would not give in easily.”38

Mandela preferred a more contextual explanation: CODESA 11 ended in stalemate because of the “National Party’s continued reluctance to submit their fate to the will of the majority. They simply could not cross that hurdle.” 39 “The essence of the problem is not one of percentages or arithmetic,” he told an audience in Sweden less than a week later. “It is that the National Party is trying to hold on to power at all costs.”40

But CODESA did not end in a bout of recriminations and invective. Agreements made would be honoured; the Management Committee was instructed to make arrangements for CODESA 11.

A number of different dynamics were at work, often at cross purposes. Once De Klerk had rid himself of the threatened white backlash, he should have been able to reach accommodation with the ANC on the outstanding issues. But the context of the government’s thinking had changed. It was no longer thinking in terms of securing the best deal it could from the ANC, and was no longer willing to concede an outright victory to the ANC in a non-racial election. Rather it had begun to think in terms of building an anti-ANC coalition that would defeat the ANC. Hence the need to appease the ANC was no longer essential: it now began to lay the basis for a strategy that would take more heed of the parties it hoped to “woo” into the grand coalition it envisioned – parties which were already resentful of having to dance to the tune of the ANC/NP duet.

On the other hand, the ANC, having seen the NP shrug off the right wing – with its help, albeit reluctantly given, no longer saw the need to make the kind of compromises they had been making in order to help the government conciliate the right. Now it could be more demanding. Moreover, to an increasing extent, it had to look to its own, to pacifying elements within the ANC who bitterly opposed the suspension of military operations — a breach of the Harare Declaration, taken without consultation with the grass-roots, and especially ill-advised in view of the escalation of violence; elements who were not proxy to the goings-on behind closed doors and the agreements being brokered that had to accommodate the realities of real politics; elements who only learned of such agreements from second-hand sources which were prone to the distortions that such communication encourages; elements who believed that the opinions of people on the ground who had borne the brunt of apartheid and were at the forefront of the struggle to bring down the government were being ignored by an elite that had spent the apartheid years in exile in comparative comfort and were now negotiating their future while practically ignoring their existence.

Hence, simmering discontent had poisoned the well of the initial goodwill that had greeted the beginnings of the CODESA negotiations. The rumblings of grass-roots rebellion in many ANC strongholds began to penetrate the confines of Kemptom Park, much to the dismay of the ANC negotiators. The need for transparency in the process was exigent. This “tug of war” between factions in favor of a more aggressive, militarist approach to speed up the process and factions in favor of negotiations for the soul of the ANC had reached crisis proportions. The ANC needed to stand back and reassess where it was going, how it was going to bring its constituency along with it, what bottom-lines were simply non-negotiatiable, what it was willing to concede to bring negotiations to an expeditious but satisfactory conclusion and getting a grip on the spiraling violence before it developed a self-sustaining momentum that would make the holding of elections impossible and the transfer of power problematic without the further spilling of blood.

In South Africa, blacks and whites did not trust each other. Their emotions were far more raw. Whites despised or feared blacks, and blacks hated the white-imposed system that oppressed them and feared the white-man’s power over their lives. But raw and obscene prejudices, being so obvious, have a certain integrity, and can be dealt with, once they are acknowledged. Outright hatred lends itself to an antidote; lingering dislike does not.