COMPROMISE AND TRUST-BUILDING AFTER CIVIL WAR: ELECTIONS ADMINISTRATION IN MOZAMBIQUE, 1994 SYNOPSIS Mozambique's first multiparty presidential and parliamentary elections in October 1994 followed 16 years of civil war. Because neither side had won the conflict, the stakes of the contest were high. Mutual distrust characterized the run-up to the vote. A new electoral law in early 1994 created a multiparty election commission that forced the parties to work together on overcoming the many operational challenges of running elections in a sprawling country severely damaged by war. The commission succeeded in damping the risks of violence that are often associated with competitive elections in such situations, building consensus among members of different political parties. When the election results were announced, all parties accepted them. However, Mozambique struggled after 1994 to overcome the legacy of the institutional arrangements forged during the peace process. The country's problems demonstrate the challenges that post-conflict countries face in designing processes and procedures to meet the immediate goal of maintaining peace while serving the longer-term aim of developing mature democratic institutions. This memo examines the 1994 elections and the impact that the initial design of the election commission had on subsequent elections. Amy Mawson drafted this policy note on the basis of interviews conducted in Maputo, Mozambique, in January 2010. INTRODUCTION When Mozambique's new Comissão Nacional de Eleições (CNE, or National Election Commission) chose him as its head in early 1994, Brazão Mazula understood that the commission's role encompassed much more than elections administration. Less than two years earlier, in October 1992, Mozambique's governing party, the Frente de Libertação de Moçambique (FRELIMO, or the Liberation Front of Mozambique) had signed a general peace agreement with its wartime foe, the Resistência Nacional Moçambicana (RENAMO, or the Mozambique Resistance Movement), ending a 16-year civil war. In 2010, Mazula recalled that he cautioned his fellow commissioners, "We have a large part to play in ensuring that the peace is maintained, that we don't go back into war again." Under the peace agreement, RENAMO and FRELIMO agreed to call a ceasefire, demobilize their armies, form a new unified army and hold national multiparty elections within a year. The elections would cap the peace process and be the final act in the country's transition from a one-party state to a multi-party democracy. More importantly for FRELIMO and RENAMO, however, the elections would determine which of them would govern the country-an issue they had already fought a long civil war over. Mazula faced the tough job of leading a politically charged election commission that comprised 10 members from Mozambique's existing government (dominated by FRELIMO), seven RENAMO members and three representatives of smaller political parties that had formed after the war ended and were known in Mozambique as the unarmed opposition. The electoral law that created the commission required that all of the group's decisions be made by consensus. Yet even if consensus could be reached within the commission, the parties' mutual suspicion and the high stakes of the election ensured that every decision would attract intense scrutiny. Technical decisions that had political implications, such as the voter registration process, were especially vulnerable to attack. If the commission failed to perform successfully and the elections ran into problems, the entire peace process might collapse. "All eyes were on the CNE," Mazula said. "I saw it. I felt it." Because of steps Mazula and others took, the 1994 election occurred without significant violence, and all parties accepted the final results. As a result, the episode offered lessons to others about how to dampen the potentially divisive effects of competitive processes in post-conflict situations. Over time, however, some of the initial choices had less desirable longer-term consequences, deepening conflict in subsequent elections. In early 2010, Mozambique still struggled to find an alternative electoral administration set-up that was politically palatable to all sides. THE CHALLENGE Mazula and the election commission in 1994 had good reason to worry that an electoral contest would rekindle violence. Although the civil war had ended, neither party was fully reconciled to the terms of the United Nations-sponsored peace agreement. The process outlined in the agreement was already off-track because of the two sides' inability to work together effectively. Finally, the country's geography created significant logistical challenges, which, if handled poorly, could have created the appearance of unfairness. The long and complex history of the conflict underscored the commission's difficult task. In 1962, three nationalist groups combined to form FRELIMO and fight together for Mozambican independence from Portugal. FRELIMO achieved this goal in 1975 and established a one-party state based loosely on Marxist ideals. The same year, the minority white Rhodesian government helped set up RENAMO as an anti-communist political organization. Rhodesia, later known as Zimbabwe, was keen to prevent collaboration between the FRELIMO government and pro-independence fighters in Rhodesia. Two years later, tensions between FRELIMO and RENAMO boiled over and full-scale civil war began. Both sides were financially dependent on external backers. Initially, the Soviet Union supported FRELIMO and Rhodesia continued to support RENAMO. In 1980, however, Zimbabwe declared independence and RENAMO found a new regional partner in the form of the minority white government in South Africa. RENAMO had employed violence against civilians as a tactic in the civil war. A 1986 report by the U.S. Committee for Refugees stated that "heinous and widespread atrocities against noncombatants-skulls crushed, infants boiled alive, throats slit, and women and girls brutally raped-have been reported by numerous eyewitnesses, ascribed most often to insurgent forces [RENAMO]." In 1988, a report by the U.S. State Department's Bureau of Refugees Programs said, "That the accounts are so strikingly similar by refugees who have fled from northern, central and southern Mozambique suggests that the violence is systematic and coordinated and not a series of spontaneous, isolated incidents by undisciplined combatants." By the late 1980s, shifting regional and international political currents brought an end to external support for both sides. Conditions in Mozambique worsened in 1990-91, when a severe drought swept through the country, exacerbating the chronic famine that large parts of the country were already experiencing due to the war. The FRELIMO government faced a humanitarian crisis while RENAMO was unable to feed its soldiers. Under the circumstances, both sides realized that a decisive victory was impossible and the only option was to forge ahead with peace talks. When FRELIMO and RENAMO signed the October 1992 peace accords in Rome and brought an end to the 16-year civil war, they did so with some regret. Although both realized that the peace talks were necessary, neither side had achieved the decisive victory it had sought. RENAMO's regret was eased somewhat by the promise of substantial international funds to transform itself into a political party; the RENAMO Trust Fund would eventually disburse US$18 million to the party. FRELIMO's regret was tempered by its confidence that RENAMO would fail at the ballot box due to its brutality during the war. At the time, RENAMO controlled about a quarter of the country's territory. When Mazula was appointed, the elections were already a year behind schedule, and the delayed transition was costing the U.N. peacekeeping mission a million dollars a day. The two parties had spent the latter half of 1993 arguing over the electoral law. To complicate matters, the only way the two parties had managed to come to any agreement on the law's terms was to fill the commission and its sister institution, the Secretariado Técnico de Administração Eleitoral (STAE, or Technical Secretariat for Elections Administration) with political appointees. The two organizations were therefore set up as delicately balanced political institutions. This staffing strategy was especially at odds with the original plan for the technical secretariat. Initially, the FRELIMO government had wanted an institution that would remain outside the political fray to work solely on technical issues while the election commission took on political issues. The commission faced geographical challenges as well. Mozambique is a large country, roughly the same size as Turkey, with a coastline of 2,500 kilometers. It also has a geographically dispersed population. Reaching population centers in 1994 was particularly difficult due to the cumulative damage that the long civil war had inflicted on the country's infrastructure. Roads were badly damaged, and the countryside was littered with an estimated two million landmines. The war had displaced 4.5 million people and forced 1.5 million refugees to flee the country. Many had not returned home or were in the process of returning when the commission's work began. The new election commission had to reach all parts of the country to run civic and voter education campaigns, register voters and set up polling stations. In mid-February 1994, Mazula and other members were sworn in and met for the first time. Mazula immediately began the task of persuading his new colleagues to set aside their differences and work together to administer a peaceful founding election. FRAMING A RESPONSE Mazula spent his first days building a rapport with commission members and taking stock of the decisions that had to be made. Tensions surfaced at the commission's first meeting. Given that the commission had to decide all matters by consensus, discord could paralyze election planning, delay the vote and rekindle conflict. Mazula felt that time spent building a rapport was time well spent. He encouraged each member to trust the others. "I had to trust all of those present," he recalled, "firstly because I was voted by all of them, and secondly, you can't work with someone you don't trust. You have to be able to trust in someone to be able to work with them properly." Mazula encouraged everyone to discuss, cool off, reconsider and discuss again-a time-consuming process that frustrated some commission members. Leonardo Simbine, the government-nominated vice president of the commission, who went on to become the commission's president for the 1998 local elections, said, "Because of the consensus [rule], we had discussions that would take too long-so long that sometimes we would take days discussing a certain issue." Looking back on the process, Mazula observed that the election commission gradually came to act as a cohesive body. Mazula worked to cultivate a role as a humble and honest broker of thoughts and ideas. He understood the tightrope he was walking as the head of an institution subject to extreme scrutiny. When asked what advice he had for other electoral administrators, he opined, "If [as election commissioner] you are invited for lunch, if you are invited for dinner, ... you go and you discuss, but you always maintain an independence from political affiliation. So you eliminate all possibilities of showing any political inclination, so that others can get the idea that you don't bend one way or the other politically." The stress of those long debates and the politically fraught context were particularly hard on Mazula. "I think he aged many years during those two years," said Miguel de Brito, country director of the Electoral Institute of Southern Africa in Mozambique, who was a researcher at the International Relations Institute of Mozambique in 1994. "He would get phone calls at three o'clock in the morning from leaders of party A or party B, you know, putting pressure on him to steer the commission in one direction or the other, and it was very difficult for him to manage this pressure," said de Brito. Mazula recalled getting several threatening visits from armed RENAMO members, trying to influence the direction of the election commission. Aware of the sensitivities of the electoral process, the commission members decided early on that they had to maintain a united public face. To this end, they chose a single spokesperson to transmit information from the election commission to the media. Initially they chose Ismael Valigy, one of the 10 FRELIMO government appointees to the commission, for this role, but three months later they gave the job to one of the unarmed-opposition representatives on the commission, Dionisio Simbe. The commission decided that none of the other members would give interviews concerning the commission's business and that even the spokesperson would report only the commission's final decisions. Internal debates would remain private. The election commission's early decisions-which included setting the final election dates for 27-28 October, deciding when to conduct voter registration, buying cars to transport electoral officials, preparing voter education materials and establishing provincial and district election commissions-were not particularly controversial. Further, many of the purely technical aspects of the elections were already worked out by the time the members started their work. In 1993, the FRELIMO government had set up a 15-person Technical Commission for the Preparation of the Elections to put together the bulk of the technical program while the electoral law was still being negotiated. Abdul Carimo Issa, an adviser to the minister of justice and a former judge, headed the technical commission. Issa laid the groundwork and then brought in people from different spheres of government. The technical commission sent members on global fact-finding missions to examine how other countries administered elections. It also worked closely with experts from the U.N. RENAMO's appointees on the electoral bodies had little choice but to accept the work prepared by their political opponents, because the electoral preparations were already behind schedule. The fact that the U.N. was heavily involved in the work of the technical commission helped alleviate suspicions. After the parties established the election commission, the original technical commission came to an end and transferred the preparatory work it had produced to the new technical secretariat, the STAE. The preparatory work became known as the STAE Action Plan. When the election commission started its work, the STAE Action Plan provided a framework for tackling the different aspects of the electoral process. The action plan described the process through five stages: preparation and inception, voter registration, preparation for the actual voting, the voting and the vote counting, and the publication of results and termination of the process. Members of the technical secretariat and the election commission, who had worked on the government's technical commission, presented the plan to the other election commission members in a series of training sessions during February 1994. Despite the preparatory work that had already been undertaken, however, the United Nations Development Programme reported that "in practice, virtually everything still had to be done."i Valigy recalled that the commission set deadlines for the different stages by working backward from the election dates. The commission also divided the work up so that RENAMO, FRELIMO and the unarmed opposition each had a commissioner working on every issue. Yet despite their collective efforts to organize the elections in a technocratic manner, problems emerged that neither the election commission, the STAE, nor the government's technical commission had foreseen. Among them were running a successful voter education program, registering voters, settling on a reliable vote-counting system that was acceptable to RENAMO, and keeping RENAMO committed to the elections. GETTING DOWN TO WORK Because the 1994 elections were the country's first multi-party elections, it was clear that a substantial civic and voter education campaign would be crucial to ensure people registered and voted. It took the commission more than a month, however, to agree on the voter education materials and an additional two months for the U.N. to pay to print the materials. Civic education finally started on 20 May, just 10 days before voter registration was due to start. The commission's civic and voter education program mostly employed students and young people to run the education program, as they were willing to travel to rural areas where security was patchy. Shortly after the program started, however, the commission realized that it was not working as planned. Mazula explained that the people the election commission had employed simply did not have enough credibility in the communities they visited. "A youth, a young person that will go into a community and stand before his elders, he doesn't have a voice; his voice doesn't carry any strength," Mazula said. He realized the only solution would be to work with the régulos, or traditional leaders, who commanded the respect of the country's largely rural population. Once the election commission started reaching out to the traditional leaders, and the youth educators had the leaders' approval to work in their communities, it was much easier to implement the civic education program. Deciding to work through the traditional leaders was a controversial decision within the election commission. During the fight for independence and throughout the civil war, FRELIMO had marginalized traditional leaders, viewing them as anti-revolutionary figures who had collaborated too closely with the Portuguese. Conversely, RENAMO had used the traditional leaders during the war to spread its message. When FRELIMO and RENAMO signed the peace agreement in 1992, Mazula explained, "FRELIMO actually opened its eyes and realized that RENAMO had a strength with the régulos to get the message out to the people." FRELIMO therefore started to examine how it might use the régulos in the future. But within the election commission, less than two years after the parties had signed the peace agreement, it was not easy to get FRELIMO members to agree to use the régulos for the election commission's work. After much debate, the FRELIMO members agreed. "Once we started using the régulos, everything moved very quickly, very successfully; the information got out," Mazula said. Voter registration was the commission's next stumbling block. Although registration was the second stage of the STAE Action Plan, the plan was scant on details. Recognizing that registration officials would need to travel around the country to reach a dispersed population, the commission worked closely with the parties to arrange an amendment to the electoral law that would allow mobile voter registration brigades to run the registration process. Mazula explained the importance of this type of interaction with the parties: "It wasn't just a decision taken by the CNE and imposed on others; it was a decision taken by all." António Carrasco, who worked at the STAE, also remembered working closely with the parties. "We organized many meetings-large, wide meetings-to establish mechanisms to hear what they had to say and mechanisms to find solutions to what they had to say," Carrasco said. The amendment to the electoral law allowed 1,600 voter registration brigades, each with five members, to work throughout the country for two and a half months. In order to reduce transportation costs, the STAE recruited brigade members as close as possible to the areas where they would be working. Each brigade was expected to register an average of 80 voters a day. If the targets were met, all of the country's 8.5 million estimated voters would be registered by the end of the registration period in mid-August.ii On 1 June, shortly after the civic and voter education campaign began, the commission kicked off the voter registration process. Due to political and logistical constraints, however, the process proved even more difficult than the commission had anticipated. Logistics were hampered by the continuing landmine threat, the poor state of the country's infrastructure, unseasonable rains in parts of the country, and delays in procuring vehicles. For the first month of the voter registration campaign, electoral officials had just 20 cars to get around the entire country. An additional 120 cars had been ordered but were held up at the port.iii The UNDP also reported problems with local suppliers who failed to deliver vehicles on time.iv Faced with these challenges, the commission relied on international donors-using money from an electoral trust fund that UNDP had set up to coordinate donor support for the elections-to lease two light planes and six helicopters. Air transportation proved essential for accessing remote areas that were cut off from urban centers by bad roads and landmines. The decision to rely on air transportation for the voter registration process-and subsequent stages of the elections when the trust fund paid for additional planes and helicopters-greatly increased costs. By the end of the elections, the hire and fueling of planes and helicopters accounted for US$10.8 million, or 17% of the total elections budget, all financed by donors. In the absence of air transportation, however, sticking to the electoral timetable would have been close to impossible; polling would have slipped into the country's rainy season, and the success of the election would have been further in jeopardy. Yet even with access to air transportation, the commission struggled to distribute registration materials and food supplies to the registration brigades, especially in rural areas.v Politically, the registration brigades were constrained at the local level by RENAMO's continued distrust of FRELIMO. In 1994, RENAMO still controlled about a quarter of the country and local operatives within these areas were wary of the election commission and its registration officials. After almost 40 years of FRELIMO's one-party state, low-level RENAMO members found it difficult to accept that the commission was not just another FRELIMO organization masquerading as a multiparty body. Voter registration brigades and government health workers found it difficult to work in these zones, and there were reports that RENAMO officials tried to prevent people from talking to election officials.vi However, when early registration figures showed that registration in RENAMO areas was proceeding more slowly than in traditional FRELIMO strongholds, the commission came under intense pressure from the national-level RENAMO party structure to step up registration efforts in RENAMO areas. The election commission used various strategies to overcome voter registration hurdles. To better reach voters in rural areas, the commission instructed several brigades to travel further into rural areas by foot. To improve food supplies to the brigades, the commission worked with the U.N.'s World Food Programme, which arranged to distribute maize, beans, oil and sugar.vii Finally, to ensure that logistical constraints at the beginning of the registration did not adversely impact the numbers of voters registered, the commission continued to work closely with the parties to extend the registration deadline. With the parties' cooperation, the commission arranged an amendment to the electoral law that prolonged the registration period by two weeks, to 2 September. By the final deadline, the commission had managed to register over 80% of the estimated total number of voters-which had been revised downward from 8.5 to 7.5 million. Another significant problem was RENAMO's opposition to using computers to tally votes. Valigy, one of the FRELIMO members of the election commission, recalled that RENAMO became especially concerned when it found that the company that won the tender to provide the computers and software turned out to be the same company that had been involved in Angola's failed elections two years earlier. Although the selection had been made by the European Union, which was one of the major international donors supporting Mozambique's transition to democracy and which was paying for the vote-tallying equipment, RENAMO did not trust that the computer system would be impartial or reliable. On this issue, Valigy recalled that the election commission relied on local diplomats, especially the Italians, and the European Commission to persuade RENAMO that there was little cause for concern. Setting up and staffing the polling stations was the commission's next task. On the basis of the voter registration data, the commission decided to establish 7,417 polling stations at 2,600 locations. Each polling station would process a maximum of 1,000 voters, and each voter was registered to a specific polling station. The location of each polling station was determined by STAE officials at the district and provincial level, who worked with the U.N. to find suitable sites. Every polling station required five officers and two assistants, meaning the election commission, through the STAE, had to recruit and train more than 50,000 polling station workers. The UNDP helped STAE, at both the central and provincial levels to implement a training initiative whereby teams of trainers would train other teams, which would go on to train others. To monitor the polling staff, international NGOs hired and trained 35,000 Mozambican election monitors. Votes were initially counted at polling stations in front of party agents and election monitors. After the count, all party agents and polling station officials signed the public notice of results, which officials then transmitted to the district level; a copy was also posted outside the polling station. At the district level, the results from the various polling stations were compiled, and officials were to repeat the validation and transmittal procedures. At the provincial level, officials were to log data from the district level into a computer program to calculate the provincial results. Provincial election officials were then supposed to transmit that information to Maputo, where the election commission would validate the final results. OVERCOMING OBSTACLES The election commission often had to work with the international community to keep the electoral process on track because it lacked the mandate or the resources to manage some of the challenges that arose in keeping political parties committed to the process. "I personally established permanent contact with the ambassadors from African countries, from the United States, from all different countries," Mazula said. He met with the ambassadors and key people at the U.N. every two weeks at the World Bank's offices in Maputo "to ensure that everything was running smoothly," he said. According to de Brito at the Electoral Institute of Southern Africa, RENAMO was keen to involve the international community at all stages and would often turn to the U.N. or local diplomats when it was unhappy with an aspect of the peace process or the electoral preparations. The crucial role of the international community came into sharp focus on the eve of the election in October 1994, when RENAMO's leader, Afonso Dhlakama, announced that his party would boycott the elections. Dhlakama alleged that FRELIMO was planning massive electoral fraud. The diplomatic community went into overdrive to bring RENAMO back on board. Early on the second day of the elections, RENAMO agreed to rejoin the electoral process. "Nobody doubts that had it not been for the special representative of the secretary general of the U.N. and the Western ambassadors plus the South African ambassador putting pressure on Dhlakama all night long, the situation would not have been resolved," said de Brito. In a 1995 report, the UNDP credited the commission for playing a pivotal role: "The firmness, cohesion and coherence of the [election commission] in the face of this difficult situation contributed substantially to the confidence of the people and the international community in the electoral process." The commission in charge of supervising the peace process, which was chaired by the U.N., guaranteed RENAMO that it would monitor the elections, and donors added a further US$1 million to the RENAMO Trust Fund to sweeten the deal. After RENAMO rejoined the elections, the election commission added a third day of voting to make up for the day the party had lost because of its boycott. The international community's role, however, was sometimes criticized after the fact. The readiness to bribe RENAMO to stay on course in 1994 encouraged the party in subsequent elections to drop out of the electoral process at key moments and make various other threats to attract international attention and support. In early 2010, commentators said RENAMO's political influence was so limited that the international community no longer paid much attention to its leader's attempts at brinkmanship. However, many did attribute RENAMO's apparent inability to develop into a viable political party, a failing that worked against efforts to build Mozambique's democracy, at least in part to the significant financial and political support the party received in the early 1990s from the international community. MID-COURSE CORRECTIONS Occasionally the election commission found that the electoral law, although 277 articles long, did not have the answer to a specific situation the commission faced. Mazula learned to take a flexible approach. "Law has to be dynamic," he said. "It has to move with the times, we have to see whether in practice there are phenomena or facts which identify weaknesses in that law. ... Good sense and intelligence have to be used in those areas where weaknesses have been identified." When Mazula and his colleagues in the election commission found a weakness in the law, they worked with the parties to seek consensus to change it. When consensus was achieved, changes to the law could be approved quickly in the legislature, as happened during the voter registration process, when amendments to the law approved mobile voter registration brigades and extended the registration deadline. ASSESSING RESULTS The international community and Mozambicans alike hailed the 1994 elections as a success. War did not break out again, as some had feared, and the last-minute attempt to boycott the elections by RENAMO was resolved. In the end, all parties accepted the results, even RENAMO. By the time the polls closed, 87% of registered voters had cast ballots. The EU's election observation mission reported a few days after the election that, "Counting procedures at polling stations were characterized by a high degree of transparency, conscientiousness and diligence on the side of the Mozambican polling staff."viii Still, the success was not total. Because of problems during the registration and counting of votes, the announcement of the official results was slow in coming. One of the problems involved the registration of special case and contested votes. Special case votes were those of the poll workers, police and election monitors who were permitted, through another last-minute amendment to the electoral law, to vote at a polling station where they were not registered. There was not enough time to train all 52,000 poll workers on the procedures for dealing with special case voters and as a result, protocols were applied erratically. Poll workers' uncertainties about how to register contested votes revealed a weakness in training. Problems with the contested and special case votes blocked the computerization of the results at the provincial level. Recognizing that the problems threatened to undermine the credibility of the election-particularly for RENAMO, which had been wary from the outset about using computers to assist in tallying the votes-the STAE proposed a change in procedures, which the election commission approved in the face of UNDP opposition.ix Instead of computerizing the results at the provincial level and faxing the information to Maputo, the STAE arranged to transport, by land and air, all materials to the central STAE office in Maputo. It was this change in procedures that substantially slowed the calculation and declaration of the final results. Before the official results could be announced, the election commission had to validate the results produced by the STAE. It was the last decision the commission had to reach by consensus. Valigy remembered the meeting that preceded Mazula's announcement of the results: It started at 10 a.m. on 18 November-20 days after the polls closed-and finished early the next morning. The government cleared the results first, and the unarmed-opposition representatives quickly followed. But RENAMO members were reluctant to give their assent. According to Valigy, Dhlakama had called and urged them to reject the results. Yet during the long commission meeting, the RENAMO members slowly came round, one by one. The RENAMO vice president of the commission was the last to agree, at around 5 a.m. on 19 November. "It was a very, very difficult process," Valigy reflected. "What we wanted to preserve at that time was the peaceful situation-the reason which we pushed for discussion until we got consensus." Joaquim Chissano, the FRELIMO incumbent, won the presidency with 53% of the vote, while RENAMO's Dhlakama won 33%. Other small-party candidates received the rest. In voting for the assembly, RENAMO won 38% (112 seats) against FRELIMO's 44% (129 seats). A third-party coalition, the Democratic Union, won 5% (9 seats), and the rest went to smaller parties. REFLECTIONS The election commission's efforts in Mozambique's successful 1994 elections were bolstered by donors and the United Nations, which together provided financial, technical and moral support throughout the peace process and the elections. In total the elections cost US$64.5 million, of which $59.1 million was provided by 17 donors. Still, the election commission played a significant role, managing its affairs and public efforts in a way that inspired faith in the underlying fairness of the voting process. The commission's strategy for communicating with the public was effective. The requirement for consensus in decision making helped to build trust in an atmosphere of distrust, in the view of observers interviewed. "I think that RENAMO accepted so readily the elections because they knew that, at least in terms of the management of the CNE, things had not been imposed on them," said de Brito, Mozambique country director of the Electoral Institute of Southern Africa. But de Brito also pointed out, "When we say there was consensus, it doesn't mean there was consensus from day one, that everyone was a big happy family. It was hard, it was difficult." Mazula agreed: "Consensus takes time." Mozambican observers sounded a note of caution about inferring lessons from the 1994 election, however. The decision to assemble a commission dominated by party appointees was part of a political deal and was not designed with long-term goals in mind. De Brito said this approach made it "more difficult for us to go towards a modern conception of electoral management." Indeed, the ACE Electoral Network, an online resource for electoral administrators created by NGOs and donor organizations, says, "Where electoral bodies represent a balance of political forces, there may be time lags in decision-making." x Decision-making delays certainly affected the Mozambican commission in 1994, where reaching consensus was mandated in the electoral law. The consensus rule did not last. In preparation for the 1998 local elections, the FRELIMO-dominated legislature passed a new electoral law that dropped the requirement for consensus. The commission subsequently took decisions by simple majority. This change could have functioned well if a genuine political balance had been retained in the election commission. But the new electoral law also changed the rules by which the parties could appoint commissioners. The new law linked political representation on the commission to a party's share of seats in the National Assembly. As the dominant party, FRELIMO was therefore able to pass a law that gave it more representatives on the election commission while establishing a decision-making procedure that ensured the party's dominance in electoral management. At least in part because of these changes, which RENAMO parliamentarians voted against, RENAMO boycotted the 1998 elections. Reflecting on the changes, Mazula said, "I think it's actually bad that subsequent election laws removed the mandatory consensus." Referring to his own experience in 1994, he said, "The members of the CNE are men. So if they don't have to reach a consensus, it will create conflict." It took 15 years for Mozambique to move away from an election commission dominated by political appointees. Although in 1999 a new electoral law permitted civil society groups to nominate the election commission's president for the first time, it was not until the 2009 elections that civil society representatives outnumbered political appointees. In preparation for the 2009 vote, the FRELIMO-dominated legislature agreed to an election commission composed of 13 members. Five would be chosen by parties in the legislature, and eight more would be chosen by the original five from a list of civil society candidates drafted by a coalition of civil society organizations in Maputo. This decision represented little risk to FRELIMO, because most civic organizations had originated with the party and were only slowly gaining an independent foothold. i UNDP, Assistance to the Electoral Process in Mozambique: Final Report. April 1995, p.11 ii Ibid. iii Hanlon, Joseph. Mozambique Peace Bulletin. July 1994 iv UNDP, Assistance to the Electoral Process in Mozambique: Final Report. April 1995 v Ibid. vi Ibid. vii Ibid. viii European Union Election Observation Mission in Mozambique, Press Release. November 2 1994, p.1 ix UNDP, Assistance to the Electoral Process in Mozambique: Final Report. April 1995 x ACE Electoral Network, accessed 08/27/2010 http://aceproject.org/ace-en/topics/vo/vo30/onePage 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/ Amy Mawson Innovations for Successful Societies (c) 2010, Trustees of Princeton University (c) 2019, Trustees of Princeton University