USING CONFLICT MANAGEMENT PANELS TO RESOLVE TENSION IN THE SECOND POST-APARTHEID ELECTION: SOUTH AFRICA, 1999 - 2000 SYNOPSIS In 1994, South Africa's interim electoral commission accomplished a seemingly impossible task: navigating myriad technical and political challenges to hold the country's first post-apartheid election. Although the election ushered in a largely peaceful transition to majority rule, the months that preceded it had been plagued by political tension and violence. As the new and permanent Independent Electoral Commission prepared for the country's second national election, in 1999, it had to contend with the potential for renewed conflict and the weakening of many of the civil society organizations and peace structures the country had relied on in 1994. As part of a broader strategy, the electoral commission created conflict management mediation panels. Working with the Electoral Institute for Sustainable Democracy in Africa, the commission deployed respected community figures within each province's political hot spots as a way to resolve tensions on the ground. In 2000, for the local government elections, the commission extended the conflict mediation system into the country's 284 municipalities. Rachel Jackson drafted this case study based on interviews conducted in South Africa, in March 2013. Case published July 2013. For a detailed look at South Africa's first post-apartheid election, in 1994, see "Organizing the First Post-Apartheid Elections, South Africa, 1994." INTRODUCTION South Africa's 1994 national and provincial elections ended nearly 50 years of racially discriminatory government known as apartheid and ushered in a new era of majority rule. During the lead-up to election day, the country often teetered on the brink of civil war. South Africans breathed a collective sigh of relief when the voting ended peacefully and the contesting parties accepted the results. Although celebrated as a successful and important event, the 1994 election was not violence free. "The elections were marred with blood," said Rev. Gift Moerane of the South African Council of Churches. Moerane was a member of a vast network of clergy, civil society workers, and community leaders who had worked to mediate conflict during the tense negotiations that had paved the way for the country's first majority elections. In the months before election day, political parties jockeyed for political prominence in order to strengthen their negotiating positions. Clashes between the African National Congress (ANC) and the Inkatha Freedom Party (IFP) often turned violent. The confrontations occurred largely in the then tribal homeland of KwaZulu, which later formed part of the province of KwaZulu-Natal, and the East Rand townships, in what became the province of Gauteng, home of both the country's largest city, Johannesburg, and South Africa's seat of executive power, Pretoria. South Africa's second national and provincial democratic elections, scheduled for June 1999, threatened renewed violence. The ANC, which had decisively won control of the national government and of seven of the country's nine provinces, sought to solidify its parliamentary majority and its political brand as liberator of South Africa's majority black African population. To remain relevant, the IFP-which cast itself as a Zulu ethnic political party under former homeland governor Mangosuthu Buthelezi-needed to keep control of populous KwaZulu-Natal or make inroads into Gauteng to boost its share of national parliamentary seats. The New National Party, formed on the ashes of the apartheid-era, largely white National Party, sought to recast itself as multiethnic and a strong opponent of the ANC. As such, it would need to venture into territory that was then under ANC and IFP control. With the end of apartheid, many of the structures the country had relied on to dampen violence in the 1994 elections no longer existed. The country's new Independent Electoral Commission (IEC) would need a comprehensive strategy to keep the campaigns peaceful. In addition to improved policing and extensive information sharing with political parties to increase its credibility, the IEC used a network of respected community leaders across the country as mediators to help identify and resolve conflicts as they emerged. THE CHALLENGE An interim electoral commission ran the 1994 national and provincial elections and the 1995 local government elections as the country transitioned from apartheid to majority rule. That interim commission was replaced in 1996 by a permanent statutory body headed first by Judge Johann Kriegler, who also chaired the interim IEC, and then Brigalia Bam, former general secretary of the South African Council of Churches. The new IEC commissioners expected a resurgence of many of the interparty conflicts as well as the same types of challenges the commission had dealt with in 1994. In the national and provincial elections and the subsequent, 1995 local government races, the major sources of conflict and the biggest impediments to fair elections were no-go areas-or swaths of the country, in each of which a dominant political party sought to keep out other parties through intimidation or outright violence. In a report following the 1994 election, the IEC estimated that 66% of the no-go areas were concentrated in KwaZulu-Natal and the East Rand townships, split between the ANC and the IFP. "KwaZulu-Natal was high in political conflict and violence, and in most instances it would lead to killings and assassinations," said Mosery Mawethu, provincial electoral officer in KwaZulu-Natal. "Whatever political tensions were, we needed to minimize the loss of life." The tensions were so potent that the province postponed its local government elections until 1996 rather than risk large-scale conflict in the wake of the 1994 campaign. "That was the tendency emerging from pre-1994: the war zones, the no-go areas for other parties," Moerane said. "We had to break that culture. In a democracy, you cannot actually claim that space belongs to you. The people living in those areas have the right to choose who to vote for, and that was not the culture in South Africa." The South African Independent Electoral Commission that ran the 1994 election was a temporary institution focused on the immense technical challenge of running a credible election in a politically and administratively divided country. The temporary commission had set up a Monitoring Directorate, headed by Peter Harris, who had also worked on the 1991 National Peace Accord (NPA). To address the problem of no-go areas in 1994, the Monitoring Directorate launched Operation Access. Through that program, monitors and security forces entered communities -accompanied by candidates who then delivered their messages-and organized public meetings to discuss the importance of fair campaigning. To handle the violence in no-go areas, the interim IEC relied on countless civil society organizations. The 1991 National Peace Accord had created a countrywide network of 1 national, 11 regional, and more than 200 local peace committees aimed at resolving conflict and preventing the violence associated with transition toward democratic government. Those committees included representatives of (1) political parties that had signed the NPA, (2) religious organizations, (3) unions, (4) industry groups, (5) security forces, and (6) civil society organizations. Funding came from the national budget. "I had to travel in those no-go areas," Moerane said, describing his experience as a member of the peace committees. "It was difficult because you could be killed." He added, "[The 1994 election was] a serious kind of a conflict that was at a scale of civil war, in a way, because we lost a number of people before the elections." Less than six months after the 1994 election, however, the ANC-led National Assembly decided not to renew funding for the National Peace Secretariat, which had administered the peace committees. The parliament also repealed the Peace Institutions Act, which gave legal authority to the NPA. The post-apartheid government argued that it was in a better position to resolve community conflict than were the semi-independent NPA structures.1 Such developments meant that the IEC could not use NPA peace committees to deal with potential violence in 1999. Neither could the commission rely solely on the civil society organizations that had formed the backbone of many of the peace committees and voter education efforts. "[In 1994], civil society organizations were very vigilant," said Mosotho Moepya, who ran the commission's electoral logistics in 1999 and later became the new IEC's chief electoral officer. "They'd always been an alternative government under apartheid. They were very well funded by movements and governments overseas and were just as rich, almost, as government would be." But, he added, after the end of apartheid, that funding flowed instead to the government, and many experienced members of those organizations left to work for the state. Moepya said that civil society was cannibalized "for the skill it had and the financial resources it had, so when the next set of elections were coming, they were not in the position to be-in any way imaginative-close to what they had been in the past." On top of those issues, the new IEC, created in late 1996, had the added challenge of building credibility as a new independent electoral management body. It needed to avoid the technical flaws that had plagued election operations in 1994. And it needed to administer an entirely new set of electoral laws and regulations, which, if mishandled, could breed suspicion among political parties. "We knew from the experience of the '94 elections-with all the difficulties, challenges, and, to some extent, chaos that had been experienced with election management-that the new approach in our election management had to be known, had to be clearly understood by voters, and had to be known and understood by the chief stakeholders: in this case, the political parties," Mosery said. "That posed a number of challenges. Any process that was not understood was seen to be interfering with the fairness of the election, and it would lead to a protest. In the context of the protests, violence would flare up and the whole environment would become unmanageable." In late 1998 and early 1999, the potential for election-related violence was made all the more clear. KwaZulu-Natal witnessed sporadic outbreaks of politically linked violence, primarily between ANC and IFP party supporters; the Western Cape and Eastern Cape had similar incidents between ANC and United Democratic Movement (UDM) members. The January 1999 killing of Sifiso Nkabinde, a former ANC activist who had become the KwaZulu-Natal provincial leader of the UDM, and the subsequent related deaths of 11 others sparked calls for a peace process between the three parties. As efforts to reach an accord stagnated, the IEC needed to be ready for further violence. FRAMING A RESPONSE The interim commission's experience in 1994 heavily informed the new IEC's strategy. In late 1998 and early 1999, Bam and her fellow commissioners devised a structure to deal with election conflicts and head off violence. First, they partnered with the newly restructured South African Police Service to ensure adequate security in volatile areas. Second, IEC staffers knew that many of the 1994 protests had arisen from incomplete understanding or from suspicion of the interim commission's procedures, due in large part to the last-minute nature of many of those policy decisions. To avoid similar problems in 1999, the IEC formalized and expanded the role of political party liaison committees, based on a temporary structure used in 1994. The liaison committees would give political party leaders at the national, provincial, and municipal levels a chance to weigh in on regulations and policy changes and to raise objections. Finally, to adjudicate legal cases quickly, the commission relied on the electoral courts, established through the same legislation that had created the permanent commission. The courts would work around the clock during the campaign, the voting, and the immediate postelection periods to resolve quickly any legal challenges or complaints and to penalize criminal offenders who disrupted elections. The courts also oversaw the IEC and drew their authority from the 1996 Electoral Commission Act, replacing the special court that had overseen the 1994 interim commission. The IEC leadership knew, however, that improved policing, the party liaison committees, and the electoral courts would not be enough to dispel the types of conflict the commission anticipated. The commissioners recognized the need for a new structure to take the place of the expired National Peace Accord committees. Like the former Monitoring Directorate, the new structure would reside within the commission itself and would handle disputes and complaints before they reached the threshold for police or Electoral Court intervention. The commission's leaders also knew that the new body would have to engage civil society organizations to succeed. "Our chairperson in 1999, Dr. Brigalia Bam, came from civil society, and she was sitting on the other side looking out and realized that one of the things we needed to think about was how we were going to help civil society," Moepya said. The then Electoral Institute of Southern Africa (EISA, which later became the Electoral Institute for Sustainable Democracy in Africa)-a South Africa-based nonprofit organization aimed at working toward more-peaceful and more-credible elections-approached the commission in late 1998 with a proposal to put mediators in each province. The organization's founders all came from labor dispute mediation backgrounds and wanted to use that experience to resolve election conflict. "We thought to ourselves, What we need is some kind of mechanism-like we had in the labor field-where we have people who work through the election management body but are part of civil society, who the community trusts, who could help resolve election conflict and keep the temperature down," Ilona Tip, EISA director of operations, said. "We approached the electoral commission and said, 'What about putting mediators in every province and across the country, who could help the commission and be resources to assist in resolving election-related conflict?'" In February 1999, four months before voting day, the commission's Legal Services Department assembled representatives from civil society organizations that worked on conflict resolution and election monitoring, including EISA, to formulate a single interorganizational strategy to handle electoral conflict and avoid duplication of effort. The meeting's participants recognized that remaining civil society structures-such as the Network of Independent Monitors in KwaZulu-Natal and the Urban Monitoring Awareness Committee in the Western Cape-already served as effective early warning mechanisms to identify conflicts. They were less well equipped, however, to resolve those conflicts on a countrywide basis. Based on those meetings, the IEC's Legal Services Department and EISA developed the concept of conflict management mediation panels. Each of South Africa's nine provinces would have its own panel comprising respected, nonpartisan community members experienced in mediation. The IEC planned to draw on lawyers, scholars, clergy, teachers, and other such professionals. During the design stage, the IEC and EISA initially considered using a national structure to manage-and delegate to-the conflict management mediation panels in each province. There was not time, however, to amend electoral legislation to create a national body, so the commission leadership instead decided that each of the provinces would administer its own panel. EISA also proposed establishing a national call center to handle incoming complaints and divert them to the appropriate panel. The Electoral Court, the semi-independent mediation panels, and the IEC's political party liaison committees would work together to filter and respond to any complaints, political tension, and conflicts that had the potential to derail the 1999 election. The mediators' work in curtailing political party misbehavior would be aided by the legally binding nature of South Africa's Electoral Code of Conduct contained within the 1998 Electoral Act. Each political party had to sign the code to be eligible to compete in an election. The legally binding nature of the code of conduct meant that, though the mediators themselves could not impose penalties on erring parties or individuals, they worked in the shadow of the Electoral Court and the potential sanctions it could impose. GETTING DOWN TO WORK Following the February meetings, the IEC had less than six months to identify and train mediators, secure political parties' buy-in, put together a media strategy to inform citizens about the mediators, and administer the panel system in each province. Realizing that the IEC staff did not have the administrative capacity to implement the system, the commission turned to EISA to carry out those tasks. Setting up the panels To identify mediators to fill the panels in each province, EISA partnered with the provincial electoral officer to establish a conflict management committee with representatives from election- and conflict-management-related nongovernmental organizations (NGOs), representatives from security services, and one mediation expert. EISA drew from its pool of labor mediators and worked with those conflict management committees to identify in each province those individuals who were respected in their communities. IEC and EISA staff members chose mediators from diverse cultural and professional backgrounds, particularly people who spoke a variety of South Africa's 11 official languages. Though the IEC and EISA had intended to staff each panel with experienced mediators, in practice many of the people recruited were community leaders with no formal training, though they all had prior informal dispute resolution experience in their local areas. The commission and EISA also used a large number of religious leaders and lawyers as mediators. Many of the mediation panelists had participated in the peace committees and other dispute resolution efforts in the 1994 elections. Using high-profile community and religious leaders immediately gave the mediation system visibility and created "moral obligations for party leaders to cooperate with the mediation panels," KwaZulu-Natal provincial electoral officer Mosery said. The number of mediators chosen per province depended on the level of conflict that each province had experienced in 1994 and the issues anticipated in the 1999 campaigns. KwaZulu-Natal, the Eastern Cape, and the Western Cape-provinces with the highest levels of violence in 1994-had approximately 20 mediators each. The IEC assigned 14 mediators to Gauteng, and the Northern Province (renamed Limpopo in 2003), the Northern Cape, and the Northwest Province had fewer than 10 mediators each. Numbers were not available for Mpumalanga and Free State. Because the IEC leadership had not begun planning for the mediation system until early 1999, the program was unbudgeted. Instead, EISA took the initiative to work with international donors to secure a modest budget for the program, obtaining grants from the Swedish International Development Cooperation Agency and the United Kingdom's Department for International Development. Funds covered one coordinator in each provincial office; the program otherwise relied on existing staff in those offices and in the Legal Services Department at IEC headquarters. The mediators themselves worked on a largely volunteer basis, with small reimbursements for each intervention: 100 rand to 150 rand per hour (US$16.50 to US$25 in February 1999). Because of that reimbursement policy, for the mediator to be compensated the IEC offices had to approve a mediator's intervention before the intervention took place. Securing political buy-in After determining the structure of the conflict management mediation panels, the IEC had to convince the political parties to support and use the system. (The commission had already secured the support of security services and relevant NGOs through their participation in the conflict management committees.) The IEC turned to the provincial political party liaison committees, which comprised high-level representatives from each political party. The provincial election offices called meetings of their respective political party liaison committees to introduce the concept of mediation panels and the proposed mediators. The political party representatives had a chance to weigh in on the system and to object to specific mediators if they suspected bias. "It really offered an opportunity for people to assess the political histories of the people we were proposing, but it also provided us buy-in, which we would not have had had we not given people an opportunity to object," said Sy Mamabolo, deputy director of party liaison in Gauteng province during the 1999 elections. Training the mediators The IEC and EISA were unable to complete the lengthy process of identifying mediators and securing political party approval of their choices until late April. Due to the delays, EISA held two-day training sessions in each province in early May. The training course aimed to teach the mediators how to identify and resolve conflicts and disputes, and it included an intensive unit on South Africa's various electoral laws and regulations, including the electoral code of conduct. The EISA trainers helped participants identify which conflicts could be mediated and which ones would need to be referred immediately to the police or the IEC. EISA staff prepared guides and handbooks for the mediators to refer to at any point during the election. The IEC Legal Services Department also used those training sessions to have mediators review and sign a code of conduct, pledging they would act fairly and with integrity. Getting the word out and referring disputes In the two months before the election, the commission began a media campaign to publicize the availability of mediators in each province. In many provinces, the IEC and EISA organized large public events to introduce the mediators to the public, and they invited the political parties to make public pledges to adhere to the code of conduct. The provincial offices also included reports from the mediation panels in their regular broadcasts and described the panels in voter education campaigns. Because of the campaign's late start, however, the IEC was unable to get the information to the entire population, a postelection review found. The IEC's publicity also focused on alerting voters to the commission's National Call Center. Housed in the IEC's headquarters and administered by EISA personnel, the call center operated 24 hours a day. Staff handled incoming complaints and conflicts, resolved what they could over the phone, and referred the remaining conflicts or disputes to the mediators in each province. The call center was not the only intake mechanism for conflicts; political parties and voters instead could report issues directly to provincial electoral offices or the Legal Services Department at IEC headquarters. The IEC and EISA had originally intended to use the conflict management committees that had originally identified mediators as referral mechanisms that would both anticipate and identify existing conflicts and refer them to the mediation panelists. In practice, however, that system was too cumbersome and the IEC abandoned the plan in favor of direct referrals to the panels. As election day drew near, the IEC also accepted complaints through its operations nerve center. Mediating conflicts and resolving disputes The mediators began working in May 1999, only one month before voting day. The majority of the issues they handled involved no-go areas and intimidation of political party supporters, though they also received complaints about destruction of campaign posters and other materials, violations of election procedures, and confusion over regulations and laws. A person could make a complaint either to the National Call Center or to the conflict management panel administrator in the provincial electoral office. A receiver who decided the case could be mediated-as opposed to handing it over to the police or the electoral commission staff-would refer the case to a mediator in that province. The mediator would then investigate the issue, gathering information from the involved parties and the community before assembling the parties for mediation. The training program included lessons on conflict resolution, on comprehensive knowledge of electoral laws and regulations, and on the penalties for transgression. However, mediators relied largely on their previous experience to navigate racial and political tensions while calling out improper behavior. "Some people just by appearing were able to cool down situations of conflict because people respected them," said Moerane, who was a mediator in Gauteng. Community presence was also important. "Initially, we thought being a member of the panel, you sit at home, you wait for a call, but no," Moerane said. "With us, you are a field-worker, you are on the ground, you visit polling stations, you see how people vote, you observe party agents, whether they understand, [whether] they follow the electoral process." As in 1994, many of the no-go areas centered on large, migrant-worker hostels throughout the country, which were typically controlled by Inkatha Freedom Party supporters, though some belonged to the ANC. "I had to travel in those no-go areas, which was difficult because you could be killed," Moerane said of the 1999 election. "It was a war zone. I had to interact with those warlords to allow people to vote the way they liked." The political party supporters would often block any other party from campaigning in and around hostels, neighborhoods, or public gathering spaces they controlled. Attempts to break the blockades could turn violent. According to Mosery, in KwaZulu-Natal, the provincial office counted approximately 55 no-go areas in the period leading up to voting day. When it came to hostels once the mediators received a complaint, they often approached the nduna-or local traditional leader, who controlled the hostel-to negotiate entry for the political party in question. The formal request from the mediator sometimes immediately secured entrance. "Often the nduna would say, 'Those people came here without first talking to me. If they make proper arrangements, I will allow them in,'" Mamabolo said. Other times, the mediator had to facilitate protracted negotiations. "Some of them took quite a while before they came to an agreement, but by and large, once the head of the hostel-the nduna-agreed to allow supporters of one or another political party to campaign, then other people would listen," Tip said. "You had to get commitment from people who had some authority." Not all traditional leaders were so cooperative. In some instances, traditional leaders would continue to attempt to block from entrance into their territories any parties they didn't support. In those cases, the panelists would remind traditional leaders of their legal obligations. "We would tell that chief: 'Get your senses together. That's not acceptable, and it's unlawful,'" Moepya said. Threats of legal action were often enough to open the door to further mediation. In many instances, misbehavior or electoral code violations by political party supporters sparked incidents of violence or intimidation. However, negotiation with the supporters themselves was rarely productive. Instead, mediators often resolved the situation by reminding the party leadership of its obligations to control supporters and of the associated penalties if it failed to do so. "I would talk to the leaders of all the participating parties to say, 'Please make sure that you remember what you signed at the party liaison committee. You have signed this code of conduct that you will never do the following things, because I can submit a report about your behavior and you will find you can be arrested,'" Moerane said. Moerane described one incident in which a candidate from the Democratic Party faced physical intimidation and threats of violence when she attempted to campaign in an ANC stronghold. "They were labeling people such as [that candidate] as people who were betraying their struggle, bringing the white parties into the townships," he said. Moerane reached out to the local ANC party leadership, a move that ultimately proved effective. "Leadership was running around making sure nothing serious happened to her, because they knew now that the IEC was aware of the culprits in this and they knew the penalty would come," he said. "When you read to the people the Electoral Act and explain to them the implications of their actions-realizing that you mean what you say-they will always listen and change their positions." Working with the police Mediators often needed to work closely with the police to identify which issues required mediation and which required police intervention. "It was important not to create a sense of impunity," Mamabolo said. "We did not want to use the conflict management panel as an alternative remedy to criminal sanctions if, in fact, a crime had been committed." Norman du Plessis, deputy chief electoral officer, added that establishing the right relationship between mediation and policing was sometimes difficult. "It's a balance between pure policing, as it should be-you cannot do a, b, or c because it's against the law-and getting people to agree; and it has to be a balance, because if you purely want to talk your way out of things, you just create platforms for people," he said. Du Plessis added that the police were the only ones who could create a basic level of security, after which mediation could work. "You cannot let situations evolve where people walk with spears and pangas [machetes] by the hundred marching down the road and expect that they're not going to come across somebody somewhere that's going to lead to a conflict." Tip said the police were often "quite open to sitting around the table" with mediators and political parties. She attributed those good relationships to early police involvement in discussions about the mediation system. "At the top levels [of law enforcement], people were in on the discussion and agreed that this is what the role of the mediators would be. We made it clear that we don't take the job of the police away, nor do mediators replace the legal system," she said. The police, the army, and the National Intelligence Agency had formed national and provincial bodies to identify potential election flashpoints and share information. Those structures worked with the IEC to distribute resources, including the mediators, where they were most needed. The mediators also sometimes relied on the police for security when entering dangerous areas but occasionally found police presence an impediment to their ability to resolve conflict. Mediators often had to determine whether a conflict was in an "area where you should been seen with the security forces or areas where you should just be alone and take a brave face to confront the warlords," Moerane said. The mediators also had to monitor police enforcement of election laws to ensure that the police adhered to the code of conduct. "The police can also contravene some electoral codes of conduct because of the way they will push themselves over the perpetrators, so you need to make sure even the intervention by the police is not disruptive," Moerane said. Voting day and the postelection period By the time South African citizens went to the polls on June 2, 1999, the conflict management panels across the country and the National Call Center had received and logged hundreds of complaints. Due to citizens' confusion over the intended role of the call center, many of these complaints had to do with procedural concerns or objections rather than disputes or conflicts between political parties. As election day dawned and the voting booths began to open, the mediators were mobile, moving from polling station to polling station to monitor procedures and make themselves available to resolve conflicts that might arise. "We were always on the phone talking to the IEC, giving them feedback on how things were developing," Moerane said. He added that on election day, it was important for the mediators "to be vigilant as conflict management teams to make sure that people don't relax at the expense of systems' collapsing." According to international election observer reports, there was not a single incident of election-related violence or intimidation that day.2 When the IEC announced the results shortly after the election, the ANC had won decisive control of the National Assembly, with 66.5% of the vote, securing 266 of its 400 seats. The remaining third of the vote was split mostly between the Democratic Party, the IFP, the New National Party (NNP), and the UDM, each with less than 10% of the votes. The ANC also won the vast majority of seats in seven of the nine provinces. The party managed to secure control of the Western Cape legislature, with 42.1% of the vote and 18 seats to the NNP's 38.4% and 17 seats. In KwaZulu-Natal, the IFP retained control of the legislature, with 41.9% of the vote and 34 seats to the ANC's 39.4% and 32 seats. No political party contested the official results. OVERCOMING OBSTACLES The mediation panels were able to resolve conflicts and disputes as they arose and prevent further escalation. In some instances, the mediation panels were also able to act as early warning and information-gathering mechanisms, allowing the commission and security forces to divert resources to unstable areas. The mediation panels were not, however, an effective conflict prevention mechanism, because they were permitted to enter a situation only once a dispute had already begun. To fill that gap, IEC headquarters and some of the provincial electoral officers turned to the political party liaison committees. The committees-with structures at the national, provincial, and municipal levels-comprised two representatives from each contesting party. The committees met regularly-at least once a month and sometimes several times a week as circumstances warranted. The interim IEC had also used the liaison committees in the 1994 election. The new commission reestablished them with the goals of sharing information and increasing the transparency and credibility of the commission's procedures, especially because the IEC had changed electoral regulations close to the beginning of the 1999 campaign period. Once the commission had established the mediation panels, the political party liaison committees became avenues to flag conflicts and channel them to the panels. But as problems around no-go areas and misbehavior by political party supporters increased close to election day, some liaison committees' members began to work among themselves to handle conflicts. "[The liaison committee] reduces the propensity for conflict, but at the same time it has gained a lot of practical experience to deal with conflict," Moepya said, adding that the committees' informal mandate to deal with conflict helped them respond flexibly. Du Plessis, who chaired the national-level committee in the 1999 election, noted the unusual willingness of political party representatives to work together. "They were almost like a brotherhood in trying to accommodate each other," he said. "You would find accommodations there amongst opponents that I have never seen anywhere else. That is where the heart of goodwill was." In Gauteng, the political party liaison committee (PLC) was particularly helpful in resolving such disputes. "The PLC was an information mechanism, but because people build relationships, you find people who are archrivals, when they are together in the PLC, they have good rapport. With time, the representatives of the two parties will just talk the matter over and say, 'It's not even necessary for us to refer it to the panels,'" Mamabolo said. He said the Gauteng provincial liaison committee would refer issues to the panels only when the party representatives were unable to resolve them. The Gauteng committee and mediation panel had "a dynamic relationship," whereby the panelists would flag recurring issues for the committee members to address. Rev. Courtney Sampson, provincial electoral officer for the Western Cape, also used the provincial party liaison committee meeting to head off potential problems between political parties and between political parties and the IEC. "It really minimized for us the levels of conflicts we have internally with political parties," he said. He added, though, that it took time to get the committees to function in that way. "If you nurture the relationship with the PLCs and they begin to trust you sufficiently, then you can work together." Not every province used its political party liaison committee in this manner. In KwaZulu-Natal, which by far experienced the highest level of conflict, the committee stuck to its original purpose as an information-sharing platform. "There were a number of changes in the electoral framework so the time [in the liaison committee] was also spent getting [parties'] buy-in on the electoral processes," Mosery said. He added that the focus of the PLC "was also very much about the acceptance of the election outcome at the end of the election. We rarely used the PLC for managing conflict in the province, save where we could-through the PLC-organize a joint intervention into a particular area with all the parties to go and appease and calm the people." Separate from the political party liaison committee, the KwaZulu-Natal legislature had established in 1997 a bilateral ANC-IFP peace process to deal with the ongoing political violence in the province, which significantly contributed to reducing conflict. In March 1999, that process appeared to falter as the IFP became concerned it might lose control of the province, but in May, less than a month before the election, the two parties agreed to sign a bilateral provincial code of conduct and publicly pledged to end political violence. Both Mosery in KwaZulu-Natal and Sampson in the Western Cape-who had to contend with the most-competitive provincial elections-also relied on outside civil society groups to supplement the conflict management mediation panels and the party liaison committees. In KwaZulu-Natal, the KwaZulu-Natal Democracy and Voter Education Forum played a key role as a reporting mechanism for bringing issues to the IEC. In both KwaZulu-Natal and the Western Cape, the Electoral Code of Conduct Observer Commission (ECCOC)-headed by the bishop of Natal of the Anglican Church in the former case and the archbishop of Cape Town, South Africa, in the latter-brought political party leaders together to pledge publicly that those leaders would abide by the code of conduct and work toward peaceful elections. ECCOC began in KwaZulu-Natal, and the Western Cape quickly adopted the structure. In both provinces, ECCOC played a key role in organizing political parties to work toward peaceful elections. According to Sampson, in the Western Cape, ECCOC, the provincial party liaison committee, and the mediation panels formed a three-pronged system of informal conflict management. ECCOC's independence from the IEC and the status of its members enabled it to organize high-level meetings to deal with political violence, ongoing provincewide problems, and conflicts that went beyond interparty dynamics. ECCOC also served as a forum where political parties could lodge complaints about the IEC itself. The party liaison committee handled disputes related to the rules and procedures of the election and to basic interparty coordination problems. The IEC then sent the mediators out into communities to deal with specific local-level problems and respond quickly to developing situations. "While ECCOC was very much, 'You come to us, we're here, we're available, you can bring your complaint,' the conflict management panelists were for deployment," Sampson said. Such a multifaceted strategy enabled the Western Cape provincial electoral office to handle most issues without police involvement. "The state security structures have to be a last resort," Sampson said. "It is important that they be on high alert at all times, but if you throw them in too early, you can create more problems. The best way forward is to talk, to find common ground, to deal with moral authority." Learning from 1999 and adapting for local government elections Two months after election day, the commission's Legal Services Department organized a debriefing to gather individual experiences with the mediation panels in the 1999 election and propose improvements for the future. The meeting participants concluded that the commission needed legal backing for the panels to give mediators greater credibility but also to make it easier to regulate their activities. The participants also decided it would be more cost-effective to retain old panelists and retrain them for each election rather than seek out new mediators. The commission also needed to define clearly the scope of the mediators' intervention, which would make it easier to channel disputes quickly to the mediators, to the commission, or to the electoral court. The debriefing also allotted significant time to how to prepare and adapt the mediation system for the upcoming local government elections in December 2000. National- and provincial-level mediation panels would not be sufficient to address the types of localized conflicts the country had experienced in the 1995 local government elections. In addition to using the 1999 mediators, the commission called on the municipal electoral officer to identify a local community leader to serve as a local mediator. "We started looking at the mediators to be as local as possible rather than what we were doing for national elections, wherein we were looking for persons at the provincial level who had some level of integrity and status across the province," Mosery said. That effort was largely successful, although in some rural areas of the Northern Province, the municipal electoral officer was unable to identify a qualified individual, according to the commission's postelection evaluation. The commission employed 166 municipal mediation panelists and brought back 67 of the provincial panelists, deployed across the country's 284 municipalities. Again, the mediators handled conflicts primarily around no-go areas. "It's about local, small constituencies trying to close each other down and closing the political space for others," Mamabolo said. The IEC relied less on a central conflict-reporting mechanism in the local government elections. Instead, it encouraged voters to report to their local municipal offices. In KwaZulu-Natal, the provincial electoral office also encouraged reporting to the province's NGO Election Monitoring Network and the Electoral Code of Conduct Observer Commission, which would then refer matters for mediation if necessary. Although the commission's Legal Services Department had originally planned to ask parliament for the legislative authority to implement mediation panels ahead of the 2000 local government elections so as to strengthen the authority of the panelists, the commission was not able to do so until 2003, when parliament passed several amendments to electoral laws. The amended Electoral Act stated that the commission had the authority to attempt "to resolve through conciliation any electoral dispute or complaint" brought to it related to the code of conduct. ASSESSING RESULTS To evaluate the conflict management mediation panels' achievements, it is first necessary to consider the underlying shifts in electoral politics in South Africa in that time period. The nature of election-related violence and political tension in South Africa shifted dramatically from 1994 to 1999. International election observers noted "a striking difference in the tenor of the political environment during the 1999 election campaign period in South Africa . . . On the whole, the parties adhered to the Code of Conduct and all the party activists whom we met commented on the new air of political tolerance and the impact this had had on their ability to campaign freely in all parts of the country."3 It is difficult to determine the exact cause of that decreasing tension in South African electoral politics. Although behavior improved, opinion surveys conducted at the time showed that the political tolerance of supporters of opposing parties had not changed dramatically from 1994 to 1999, though more citizens identified as independent voters.4 The drop in incidents of political violence from 1994 to 1999 was due at least in part to decreased competition. The ANC had established decisive control of the country's national legislature and most of the provinces in 1994. "If we had a 51-49 split between the two parties and a real contention for who was going to be the next government, you might have a different situation," du Plessis said, adding, however, that there was still uncertainty as the country prepared for 1999. "The governing party was tense. Would the center hold? Would it not hold? It held, it held solidly." Even though decreased competition for control of the national legislature might explain lower levels of violence in many of the provinces, there was still a real fight for control of the provincial legislatures of the Western Cape and of KwaZulu-Natal. In those two provinces, the provincial electoral officers attributed the decrease in violence and tension to a number of factors: One was better policing; a new national South African Police Service had replaced the disorganized and often biased transitional security forces of 1994. A second factor was what Sampson called informal structures of conflict management: the political party liaison committees, outside organizations such as the Election Monitoring Network and the Electoral Code of Conduct Observer Commission, and the conflict management mediation panels. Following both the 1999 election and the 2000 election, the IEC commissioned an independent evaluation of the conflict management mediation panels. The 1999 study concluded that the program was generally useful but that its effectiveness was compromised by the short timeline for the panels' operation. Additionally, the IEC's media strategy to publicize the panels was limited, meaning that many political party representatives, many supporters, and many voters were confused about the role of mediators. That confusion led to the high volume of procedural complaints, which often could not be mediated. The IEC's 2000 local government elections publicity strategy was more effective, though some political party representatives had only limited knowledge of the mediation system. The conflict management mediation panelists played important roles in dealing with particularly fraught issues such as no-go areas, which several interviewees cited as the largest impediment to fair elections. In the 1994 election, prior to implementation of the conflict management panel system, the IEC received 3,558 complaints of no-go areas, political violence, and intimidation, spread across the country. The Human Sciences Research Council logged 165 no-go areas across the country-70 of them in KwaZulu-Natal alone.5 In 1999, the IEC received reports of 1,113 conflicts; and no-go areas were limited to KwaZulu-Natal (55) and the East Rand area of Gauteng. In the 2000 local government election, the IEC received reports of 314 complaints, most of them related to alleged biases on the parts of IEC staff, destruction of campaign posters, harassment, and procedural objections. That downward trend continued for the 2004 national and provincial election period, during which the IEC received 253 complaints through the conflict management program and did not identify a single no-go area. By that point, the majority of complaints related to procedural issues rather than violence or intimidation. 1999 Provincial distribution of complaints Complaints received by the National Call Center, IEC Nerve Center, and IEC Legal Services Department KwaZulu-Natal 444 Northern Cape 137 Western Cape 61 Eastern Cape 144 Northern Province* 15 Gauteng 259 Mpumalanga 20 North-west 15 Free State 19 Complaints referred to panels (resolved through mediation/facilitation) KwaZulu-Natal 37 (35/2) Northern Cape 6 (0/6) Western Cape 11 (9/2) Eastern Cape 17 (17/0) Northern Province* 1 (1/0) Gauteng 51 (1/50) Mpumalanga 3 (2/1) North-west 8 (6/2) Free State 5 (5/0) Complaints referred to police, resolved by telephone or resolved ,without formal reporting KwaZulu-Natal 407 Northern Cape 131 Western Cape 50 Eastern Cape 127 Northern Province* 14 Gauteng 208 Mpumalanga 17 North-west 7 Free State 14 *The Northern Province was renamed Limpopo in 2003. Source: Independent Electoral Commission and Electoral Institute of South Africa. "The Independent Electoral Commission of South Africa Report on Conflict Management Committees & Mediation and Arbitration Panels for the National Assembly and Provincial Elections, June 1999," October 1999 Due to incomplete reports, problems with the reporting format, and informal resolution of some conflicts, the numbers were likely conservative estimates of the number of conflicts reported and resolved in each election. Many complaints were recorded at intake but were not tracked through the referral or mediation processes. The inconsistent reporting of complaints and record keeping across the nine provinces prevented the IEC from conducting a full evaluation of the effectiveness of the conflict management mediation panel system, according to a series of IEC reports to parliament.6 In KwaZulu-Natal, which experienced the greatest volume of complaints, 80% involved intimidation, harassment, and political intolerance. "Where [the panelists] were successful was really in the hard-core political issues around campaigning-in particular, for the 1999 elections in no-go areas and intimidation," Mosery said. "But other, related campaign matters of poster defacing-it was not possible to mediate any of those. They always came up as areas [in which] to try to build a different dynamic for future campaigns." He added that many political parties and voters appeared to see the mediators as the place to report election management issues, which clogged up the system and reduced their ability to focus on more-serious issues of violence and intimidation. Tom Lodge, who consulted for the IEC on conflict management and on establishment of the IEC panels, wrote in a review of the 1999 election that the delayed establishment of the mediation panels meant they would not have been effective had South Africa experienced levels of political violence similar to those of 1994. "Had the stakes been higher [in 1999] and had activists behaved less well, the election authorities would have been unable to exercise much influence, for their conflict resolution arrangements were established much too late," he wrote.7 In the local government elections in 2000, provincial panel coordinators resolved 167 complaints over the telephone-without the intervention of a mediator. Most of the complaints (109) were in KwaZulu-Natal. Provincial and municipal mediation panelists recorded 193 complaints, though some of them were multiple reports of an ongoing issue. Of those recorded complaints, the mediators resolved 157 either through formal mediation, via a simple conversation with the party complained against, or by raising the matter with the local party liaison committee. Seven complaints were referred to the police for investigation, and an additional 23 lacked reported outcomes. Unlike in 1999, the majority of these complaints involved alleged IEC staff bias-charges that an independent evaluation could not substantiate.8 Of the 193 complaints addressed by panelists, only 29 involved violence, murder, death threats, or intimidation-a dramatic shift from previous elections. REFLECTIONS Based on its experiences in 1999 and 2000, the IEC continued to use the mediation panels in subsequent national, provincial, and local government elections as part of its broader conflict management strategy. A number of other countries in sub-Saharan Africa-including the Democratic Republic of Congo, Kenya, Lesotho, Malawi, Mozambique, and Zambia-took inspiration from the system and, under the guidance of EISA, implemented conflict management mediation programs of their own. Even though the South African experience was not without obstacles, those involved pointed to important lessons. In KwaZulu-Natal, Mosery said, the mediators were not themselves problem solvers but, rather, helped the political parties solve their own problems. "They are there to facilitate a dialogue between those people who must change the environment," he said. "The mediator can come in order to facilitate that political tension is minimized, and if the tensions are minimized, then there will be less intimidation, which affects the broader community." The mediation panels' successful interventions relied on the foundation the IEC had built as a credible, independent institution before the 1994 election. "If an electoral commission has no credibility, if it's fraught with problems, and you put a conflict management panel in, do you think people will trust it?" Moepya said. The mediators' work was made easier by a genuine, countrywide desire to move past previous violence and build a more peaceful future for South Africa. "We have learned the hard way that we don't want to repeat the mistakes of the past," Moerane said. "We don't want to repeat the violence that ravaged and destroyed our communities; we try by all means to make sure that can be prevented." Moerane attributed his own success as a mediator to his personal drive to work toward a more peaceful South Africa. "You should have the heart, as a mediator, to work 24 hours," he said. "That spirit of determination has made us respected . . . When people know me, they say, 'Oh, the peacemaker has arrived."' In the 1994 and 1999 elections, the country focused on reconciliation. "Part of the discourse of reconciliation was, How do we manage conflict? How do we keep peace together?" Sampson said. "We had a deep sense of how critical peace and the achieving of peace were [and] how much we had to lose if we had an uncontrollable flare-up of violence and conflict during the [1999] election." As time went on, however, the nature of electoral conflict in South Africa shifted from interparty conflict to intraparty conflict, attributed to party-run primaries and internal nominations. The IEC generally did not intervene in those issues. Sampson noted that in 2000 and subsequent years, as political dialogue in South Africa shifted away from post-apartheid reconciliation, the mediation panel system became a less effective tool to resolve election conflict References 1 Phiroshaw Camay and Anne J. Gordon, "The National Peace Accord and its Structures," South African Civil Society and Governance Case Study No. 1, Cooperative for Research and Education, 2000, http://www.nelsonmandela.org/omalley/index.php/site/q/03lv02424/04lv03275/05lv03294/ 06lv03321.htm. 2 "The National and Provincial Elections in South Africa, 2 June 1999," Report of the Commonwealth Observer Group, Commonwealth Secretariat, United Kingdom, 1999, http://www.thecommonwealth.org/shared_asp_files/GFSR.asp?NodeID=141347. 3 Ibid. 4 Robert Mattes, Helen Taylor, and Cherrel Africa. 1999. "Public Opinion and Voter Preferences 1994-1999." In Election '99 South Africa: From Mandela to Mbeki, edited by Andrew Reynolds. Cape Town, South Africa: David Philip, 5 Report of the Independent Electoral Commission, South African Elections of April 1994, Johannesburg, 1994, 53. 6 Independent Electoral Commission and Electoral Institute of South Africa. "The Independent Electoral Commission of South Africa Report on Conflict Management Committees & Mediation and Arbitration Panels for the National Assembly and Provincial Elections, June 1999," October 1999. Independent Electoral Commission, and Electoral Institute of South Africa. "Conflict Management Programme Report: South African Local Government Elections, August-December 2000," 2001. 7 Tom Lodge. 1999. Consolidating Democracy: South Africa's Second Popular Election. Johannesburg: Electoral Institute of Southern Africa and Witwatersrand University Press, 202. 8 Independent Electoral Commission and Electoral Institute of South Africa. "Conflict Management Programme Report, op cit. Innovations for Successful Societies makes its case studies and other publications available to all at no cost, under the guidelines of the Terms of Use listed below. The ISS Web repository is intended to serve as an idea bank, enabling practitioners and scholars to evaluate the pros and cons of different reform strategies and weigh the effects of context. ISS welcomes readers' feedback, including suggestions of additional topics and questions to be considered, corrections, and how case studies are being used: iss@princeton.edu. Terms of Use In downloading or otherwise employing this information, users indicate that: a. They understand that the materials downloaded from the website are protected under United States Copyright Law (Title 17, United States Code). This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/. b. They will use the material only for educational, scholarly, and other noncommercial purposes. c. They will not sell, transfer, assign, license, lease, or otherwise convey any portion of this information to any third party. Republication or display on a third party's website requires the express written permission of the Princeton University Innovations for Successful Societies program or the Princeton University Library. d. They understand that the quotes used in the case study reflect the interviewees' personal points of view. Although all efforts have been made to ensure the accuracy of the information collected, Princeton University does not warrant the accuracy, completeness, timeliness, or other characteristics of any material available online. e. They acknowledge that the content and/or format of the archive and the site may be revised, updated or otherwise modified from time to time. f. They accept that access to and use of the archive are at their own risk. They shall not hold Princeton University liable for any loss or damages resulting from the use of information in the archive. Princeton University assumes no liability for any errors or omissions with respect to the functioning of the archive. g. In all publications, presentations or other communications that incorporate or otherwise rely on information from this archive, they will acknowledge that such information was obtained through the Innovations for Successful Societies website. Our status (and that of any identified contributors) as the authors of material must always be acknowledged and a full credit given as follows: Author(s) or Editor(s) if listed, Full title, Year of publication, Innovations for Successful Societies, Princeton University, http://successfulsocieties.princeton.edu/ (c) 2019, Trustees of Princeton University Rachel Jackson Innovations for Successful Societies (c) 2013, Trustees of Princeton University Terms of use and citation format appear at the end of this document and at www.princeton.edu/successfulsocieties ISS invites readers to share feedback and information on how these case studies are being used: iss@princeton.edu