The multiple postponements of Somaliland’s presidential elections had begun to erode the legitimacy of Somaliland’s political institutions and created an environment of distrust that boiled over into violent demonstrations in September. Unlike in neighboring Kenya, however, major violence was avoided and within a year the people of Somaliland held peaceful elections. Nevertheless, as this paper on postelection violence in Kenya by Axel HarneitSievers reflects, the experience of successfully organizing elections is no guarantee that things will go well next time. The experience of Kenya, therefore, has lessons for the democratization process in Somaliland.
By Axel Harneit-Sievers,
Director, Regional Office for East & Horn of Africa, Heinrich Böll Foundation, Nairobi, Kenya
Introduction
The violence that followed Kenya’s presidential and parliamentary elections in 27th December 2007 was very much unexpected by most Kenyans and international observers alike. Despite the fact that Kenya had experienced instances of violence in elections since the re-introduction of multi-party politics in 1991, few people imagined that violence would erupt on the scale that occurred.
An estimated 1,133 people were killed[1]following the announcement of the disputed results on 30th December, several hundred thousand people were displaced, and the violence had major political and economic repercussions throughout the region.
With political protest degenerating into outright “ethnic cleansing” and inter-ethnic warfare, the crisis raged until the end of January 2008. By then, former UN General Secretary Kofi Annan, supported by Kenyan civil society, neighboring countries, and the international community, had begun a mediation process between the two main presidential candidates and their parties – President Mwai Kibaki (Party of National Unity – PNU) and his challenger, Raila Odinga (Orange Democratic Movement – ODM).
It took until 28th February 2008 before a “National Accord” was struck, which terminated the slide into outright civil war. This provided the foundation for a power-sharing arrangement between the opponents, known as the “Grand Coalition Government”, with Kibaki as President and Odinga as Prime Minister that continues in power today.
The Kenyan experience before, during, and after the 2007 elections contains some important warnings and lessons to consider when preparing for elections in countries that are characterized by long-standing ethnic-regional divisions, a political culture based on ethnic or “tribal” loyalties, and weak institutions.
Somaliland has some of these characteristics, but since 2002 the people of Somaliland and its political parties have demonstrated an admirable capacity to hold peaceful elections and avoid the levels of violence experienced in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya and elsewhere in the region. Nevertheless, this paper identifies lessons from the Kenyan experience that may be relevant for Somaliland in the future.
Relevance of the Kenyan experience
The Kenyan experience surely shows one thing: a long experience in successfully organizing elections by no means guarantees that things will go well next time.
Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya has been holding elections regularly since its independence in 1963. It even conducted parliamentary elections under one-party rule in the 1980s, when different candidates competed for seats under the Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya African National Union (KA NU) umbrella.
Kenyan elections have often been accompanied by fraud and rigging, but before the 2007 crisis, such malpractices had never fundamentally undermined the legitimacy of the overall electoral result. The same can be said about the instances of violence that accompanied the 1992 and 1997 elections after the return to multi-party democracy.
The successful and largely peaceful transition from Moi to Kibaki in 2002 was, and still is, widely regarded as the single most important success of electoral democracy in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya. Despite all these experiences, and despite the fact that institutional mechanisms (like an electoral commission and a judicial system to take care of dispute resolution) were in place for a long time, things went terribly wrong in 2007. This happened for specific reasons that I will turn to below.
In contrast, Somaliland has shown since the 1990s that having little to no experience in institutionalized democracy does not preclude the possibility of establishing democratic institutions.
Indeed in the Somaliland case, just the opposite may be true. Since its foundation Somaliland has been growing democratic institutions afresh and has constantly nurtured them with local knowledge and traditions on how to create peace and organize legitimate political representation.
This has produced some of the unique features of the political system in Somaliland, such as the guurti and the three-party system. These “home-grown” processes may actually provide a more solid foundation for democracy than simply a formally-institutionalized system. In the Somaliland case, this worked reasonably well for over a decade, although it was put at risk by the repeated delay of presidential elections from 2008.
The successful outcome of the elections in Somaliland underlines the viability of electoral democracies in Africa. Still, the negative experience of the 2007 elections in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya is worth taking note of. The Kenyan crisis did not arise as an earthquake “out of the blue”. It had identifiable reasons which, if understood correctly, provide some lessons about how to prepare for elections in other difficult situations.
I will now discuss the electoral commission in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya and the technical conduct of the electoral process, examine the reasons behind the outbreak of violence, comment on the role of election observation, by formal observers and by the media, and finally consider the need for strong mechanisms for dispute resolution and mediation.
Capability and credibility of the electoral commission
It is by now generally acknowledged that a major cause of the crisis in the 2007 Kenyan general election was the credibility of the electoral commission – or rather the lack of it.
For about a decade, the Electoral Commission of Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya (ECK) had appeared to be a well-established institution, successfully overseeing critical voting processes such as the 2002 general elections and the 2005 constitutional referendum. The foundation for this had been laid in 1997 by a cross-party agreement negotiated by the Inter-Party Parliamentary Group (IPPG) which established rules for the nomination of ECK councilors.
ECK chairman Samuel Kivuitu was widely regarded as a capable and trustworthy personality. Doubts about the credibility of the ECK began to arise when, in the autumn of 2007, President Kibaki appointed a number of new councilors to the ECK without the consultation process prescribed by the IPPG agreement. Still, many people hoped that the widely-respected Kivuitu would safeguard the situation. This was not to be.
Technically, the 2007 general elections in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya went well, up to the point of vote counting. On 27th December, more than 8 million Kenyans queued up and voted peacefully. The overall mood was optimistic, and the first results coming in the following day strengthened that optimism.
The fact that more than a dozen ministers lost their seats seemed to prove that the peoples’ vote was having an impact. In fact, in contrast to the presidential vote, the parliamentary results were generally regarded not to have been affected by large-scale rigging, and only a few of them were later disputed.
The crisis began during the counting of the presidential results, especially on the second day (29th December).[2] Communications between ECK headquarters and its officers up-country did not always work, making it difficult to address and clarify emerging problems (such as incomplete result forms). From a certain point in time, the ECK seems to have lost control over the process, as party agents and ECK officials debated the validity of returns from dozens of constituencies.
The resulting delay in the announcement of the results led to tensions in ODM majority areas, as fears grew that the election was about to be rigged in favor of the incumbent. The eviction of party agents and journalists from the ECK counting center in Nairobi only worsened the situation. When the ECK announced a victory for Kibaki on the evening of 30th December, with a margin of about 230,000 votes, and Kibaki was hastily sworn in as president, many Kenyans felt betrayed.
The violence started within hours. Pursued by the media, the visibly exhausted ECK chairman declared on the evening of 1st January 2008: “I do not know whether Kibaki won the election.” This put the last nail in the coffin of the ECK’s credibility and legitimacy. Following the recommendations of an independent review by the Kriegler Commission, the ECK was dissolved in late 2008 and replaced by a reformed body with entirely new personnel.
To have a credible and capable electoral commission is not merely a technical necessity, but is of critical importance for successful and peaceful elections. The electoral commission’s function is not only to prepare and hold elections but is also to ensure that the counting and collation process is conducted in an orderly and transparent manner.
The Kenyan experience shows how important (and how difficult) it may be for an electoral commission to withstand pressure from politicians and parties, and to keep control of the process, especially if difficulties emerge and tensions rise during and after the vote.
Somaliland’s first three elections were overseen by a national electoral commission (NEC) that is generally acknowledged to have been capable and credible. This was replaced by one that proved to be incapable and far from credible, which contributed to the postponement of the 2008 elections. A new electoral commission was formed after the crisis in September 2009, that proved to be more capable and restored the NEC’s legitimacy. The commission needs ongoing support and training.
Kenya’s experience in 2007-08 and the report by the Kriegler Commission and by civil society observer groups provide useful in-depth analysis of what can go wrong and how to avoid mistakes.
Reasons for the outbreak of violence
An important underlying reason for the outbreak of large-scale violence in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya in 2008 was the fact that political competition in the election reproduced and intensified the major ethnopolitical fault line in Kenya by pitching a Kikuyu and a Luo candidate (and their respective allies) against each other.
This “bifurcation” of the political landscape distinguished the 2007 elections from earlier ones, where important politicians of the same ethnic group had often found themselves on different sides. Political mobilization and voting patterns in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya have always had strong ethnic dimensions, but in the political configuration of 2007 (a result of the break-up of the 2002 coalition over the 2005 constitutional referendum) ethnic competition was particularly strong.
The immediate reason for the outbreak of violence was the widespread belief among Kenyans, and especially among ODM supporters, that the elections had been rigged in favor of President Kibaki. This perception was strengthened by the partial results published on the first day of the vote-counting which had indicated a strong lead for the challenger, Raila Odinga (more on this below). This seemed to reaffirm numerous opinion polls conducted since September 2007 which had placed Odinga in the lead.
Although this was much reduced by December, the margin of lead may have been narrow, but if there was any recognizable trend, it favored Odinga. In this context and, and whilst the ECK was unable to produce final results in time, opposition supporters took it for granted that rigging on a massive scale had taken place.
So far for the background, but how did the violence actually evolve? Certainly, there was a degree of “spontaneous” protest that quickly degenerated into violence, accompanied by what may be called “opportunistic crime” (looting, rape) encouraged by the general climate of tension, insecurity and lawlessness.
The government reacted to this with “violence from above”, employing its security agencies which at times exerted disproportionate levels of violence and firearms against protesters, looters and bystanders who were unarmed or armed with only primitive weapons. This was most marked in ODM strongholds, especially in Kisumu.
But beyond “spontaneous” violence and heavy-handed action by the security forces, it is important to note that various political actors were prepared to use violence – or the threat of it – as a political strategy, leading to an escalation of violence that went far beyond any “spontaneous” action.
- Immediately after the announcement of the results, ODM resorted to mass protests. While its senior party officials called for peaceful demonstrations, it was obvious that such protest would, in all likelihood, be accompanied by violence. The de facto threat of violence was part of ODM’s strategy. While unsuccessful in bringing down the government, it caused a degree of destabilization in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya that quickly brought about rapid international diplomatic engagement and thus forced the Kibaki government to the negotiation table.
- In parts of Rift Valley Province, evictions, attacks and “ethnic cleansing” directed against Kikuyu “settlers” (regarded as being Kibaki supporters) erupted right after the announcement of the election results. These events repeated patterns of violence from the 1990s, and they seem to have been organized by local (and perhaps even some national) political leaders.
- In the course of January 2008, violence escalated with attacks and counterattacks by ethnic militias which, again, are generally believed to have been steered by political actors. By the end of January, militia warfare had brought the country to the brink of a civil war that was only averted by the beginning of Kofi Annan’s mediation process. While details about the operation and dynamics of these militias are not yet publicly documented in detail, it is clear that political actors willing to use them could draw on the capacity for violence of existing underground structures that had existed for years, with the Kikuyu-based Mungiki being the most notable among them.
Thus, it was not so much a ‘spontaneous reaction’ to alleged vote rigging, but the readiness and ability of various political actors in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya to employ violence that played a fundamental role in escalating the post-election crisis.
Today, it is widely acknowledged that the use of violence had been encouraged by the fact that earlier instances of violence – especially during the 1992 and 1997 elections in the Rift Valley and in Coast Province – have never led to any prosecution of those responsible for it. Combating this “culture of impunity” (for acts of violence, as well as for corruption) should be at the heart of Kenya’s drive to reduce the risk of future electoral violence.
As attempts to establish local tribunals for this purpose have failed so far, the International Criminal Court has begun investigations that are expected to lead to the prosecution of a number of perpetrators.
Somaliland, by contrast, appears to be in a far better position than Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya as regards to risks of violence. Somaliland does not have long-standing deep ethno-regional cleavages that have fueled frustration and anger, such as between the Kikuyu and Luo ethnic groups in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya. Instead, Somaliland nationalism has been a strong integrating force since the 1994-1996 civil war and has been integral to building peace and democracy since the 1990s.
In Somaliland, there are no conflicts over land (between pastoralists, or between “indigenous” and “settler” groups) on the scale and historical depth as in Kenya’s Rift Valley. By contrast to Kenya’s political culture, which has put a premium on violence by providing immunity from prosecution for masterminds of violence, Somaliland’s political culture is integrative and has defined itself by consensus-building and the de-escalation of violence.
Still, the September 2009 crisis illustrated that peace cannot be taken for granted even in Somaliland politics. Opposition political actors demonstrated great restraint by not retaliating violently against the shooting of demonstrators by police, and by not resorting to the use of arms that are readily available in the country (although preparations to actually use them may well have been underway).
Instead, the negotiations that followed proved constructive. A repetition of this tragic course of events has to be avoided in the future at all cost, because the reactions of those who feel their rights have been violated and view themselves under attack may be different another time.
Election Monitoring and the Media
Independent media are a critical element of any democracy worth its name and have an important role to play during elections, by monitoring and reporting events from a variety of localities and perspectives that are independent of government and any political party. Formal election monitoring by national and international observer groups has also become a standard procedure in many countries.
Independent media and national as well as international observers were present in the 2007 Kenyan elections. Both of them played a largely constructive role in the monitoring and transparency of the electoral process. A number of vernacular radio stations were an exception to this, as they allegedly transmitted what amounted to “hate speech” against members of other ethnic groups before the elections and during the crisis. But this seems to have been limited to certain regions where ethnopolitical conflict has a history.
Overall, the national electronic and print media (and especially the private stations and papers) deserve much credit for having consistently tried to play a de-escalating and constructive role during the post-election crisis, by reporting fairly, calling for peace and dialogue, and conducting campaigns to support the victims of violence and displacement.
Still, there was a problem with the Kenyan media during the elections, arising from the delayed results. In the first two days after the voting, Kenya’s media published results from polling stations and collection centers, as well as the constituency-based results declared by the ECK. At times, it was unclear whether the results published were officially-declared, or just based on journalists’ observations and reports from up-country districts.
The results published (and added-up) by major news stations on the day after the elections seemed to suggest a clear ODM victory, but on day two Odinga’s lead reduced. While this “reversal of trends” seems to have been due (at least partly) to the fact that results from ODM’s strongholds came in earlier than those of the PNU, ODM supporters took it as an indication that the elections were in the process of being rigged.
The problem was, of course, first caused by the fact that the ECK was unable to produce the election results in time, resulting in rapidly rising tension during the counting exercise on day two. However, the media, by simply fulfilling their reporting duty, carried a degree of responsibility for the rising tension as well, by publishing incomplete results. They also missed the opportunity (and so perhaps neglected their duty) of carefully interpreting the preliminary data and warning their audience that the apparent “trends” may be misleading.
Similarly, an interview with the head of the EU observation mission was transmitted by a radio station on 29th December, as counting was still going on. In the interview, he hinted at irregularities in two particular constituencies (the results of which, later on, indeed turned out to be very much disputed).
While he may have been trying to bring transparency to the process, he contributed to the public uneasiness about the process, and the resulting tension. Thus, the Kenyan experience shows that media and observers’ reporting on an election process that is still going on may increase tension and lead to undesirable public perceptions.
The Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya media is strong and independent and forms a marked contrast to the situation in Somaliland, where independent media are restricted to newspapers, and where private radio stations are not licensed. Still, the Kenyan experience may provide some helpful lessons. Still, the Kenyan experience may provide some helpful lessons
Media and observers have a fundamental role to play in instilling transparency in an election. Instituting “rules of engagement” for observers and the media during an election would help to avoid the pitfalls experienced in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya in late December 2007.
These “rules of engagement” have to provide, of course, for the free access of observers and journalists to all stages of the election and counting process, but should also place limitations on reporting and publishing the observations while vote counting is still going on, especially as regards the publication of preliminary results.
Of course, there has to be a time limit on this embargo. It should only be applicable for a day or two, until such time as the electoral commission can be reasonably expected to deliver results. Similarly, formal election monitors should be restrained from making public statements on the conduct and the results of the election process until the election is completed, when they deliver their official report.
These “rules of engagement” should not be legislated upon, as this may also be misused or understood as intended to suppress press freedom. Instead, they should be based on a consensus among stakeholders, the details of which could be negotiated between media and election observers on the one hand, and the electoral commission on the other.
The results of these negotiations can be brought into a voluntary agreement (such as a “code of conduct”), based on the joint understanding that these rules were agreed in order to safeguard the transparency, sound, and peaceful conduct of the election, and in order to ensure that the transparency required for a successful election does not become a source of tension and possibly violence.
Institutions for conflict resolution and mediation
The postelection crisis in Somali region, and found that around 40% of samples failed to meet either national or WHO drinking quality standards [9]. In Kenya escalated because mechanisms of conflict resolution available within the country failed, despite well-established institutions and a vibrant civil society.
Right after the announcement of the election results, ODM refused to take legal action, professing it had no trust in Kenya’s judicial system to deal with what they saw as a large-scale political fraud. PNU argued that ODM should contest the ECK results in the courts. Given the balance of power, with the security forces under governmental control and killing opposition supporters in the streets, such an argument appeared hollow, if not cynical.
However, outside the legal system, there were no other institutions or personalities in the country able to play a mediating role. Splits along party lines were visible even within churches and civil society, which were rendered them largely ineffectual as potential mediators. It became nearly impossible for any local actor to play a role that would be recognized as truly non-partisan by either side.
External mediation became a necessity. Kenya, as the regional hub housing international organizations and media, could count itself lucky to attract immediate international attention.
Exploratory visits by potential mediators from other African countries started a few days after New Year. It is hard to see how the crisis could have been resolved without massive international engagement and pressure, and without consistent international support for a negotiated end to the crisis. Even with that support, the mediation process took several weeks to start in earnest and four more weeks to reach a conclusion.
In case of a political crisis, Somaliland cannot expect to generate comparable international interest. Still, during the September 2009 crisis in Somaliland, the Guurti, local civil society, neighboring Ethiopia, and actors of the international community did intervene and exerted pressure on Somaliland’s political parties to resolve the crisis. This resulted in the Memorandum of Understanding signed on 25th September that agreed a change in the leadership and members of the electoral commission.
However, as the International Crisis Group noted, “the parties only stepped back from the precipice when confronted by the likelihood that the alternative was indeed return to armed conflict.”[3] This experience should serve as a warning to have dispute resolution and (if necessary) mediation mechanisms in place that are as strong as possible and can be activated in the event of a major crisis during or after the election. In doing so, Somaliland can build upon its traditions of consensus-building and involve local actors such as the Guurti and respected civil society leaders.
Somaliland could also consider inviting the country’s trusted partners and friends in the international community to be on standby and to act, from the start of the election process, not only in an election monitoring role but perhaps also as part of a dispute resolution and mediation mechanisms, just in case advice and pressure from outside should become necessary. Both the Kenyan 2007-08 experience and that of Somaliland in autumn 2009 point to the fact that in times of crisis, the presence of outsiders (and pressure by them, if necessary) can play a very constructive role in conflict resolution; so why not build on it right from the beginning?
References:
Relevant documents on Kenya’s “National Accord”, including the reports of the Waki and Kriegler Commissions of Inquiry, can be found at http://www.dialoguekenya.org/.
An independent account of events during the critical phase of vote counting is the report by the civil society group “Kenyans for Peace with Truth and Justice” titled “Countdown to Deception: 30 Hours that Destroyed Kenya”, available, among others, at http://www.africog.org/reports/KPTJ-%20Countdown%20to%20 Deception.pdf.
The International Crisis Group’s report “Somaliland: A Way out of the Electoral Crisis” (Policy Briefing No. 7, Nairobi/Brussels, 7th December 2009) is available at https://icg-prod.s3.amazonaws.com/b67-somaliland-a-way-out-of-the-electoral-crisis.pdf
Notes
[1] According to the figures of the Commission of Inquiry into Post-Election Violence, popularly known as the “Waki Commission”. Relevant documents on Kenya’s “National Accord”, including the reports of the Waki and Kriegler Commissions of Inquiry, can be found at http://www.dialoguekenya.org/.
[2] An independent account of events during the critical phase of vote counting is the report by the civil society group “Kenyans for Peace with Truth and Justice” titled “Countdown to Deception: 30 Hours that Destroyed Kenya”, available, among others, at http://www.africog.org/reports/KPTJ-%20Countdown%20to%20Deception.pdf
[3] The International Crisis Group’s report “Somaliland: A Way out of the Electoral Crisis” (Policy Briefing No. 7, Nairobi/Brussels, 7th December 2009) is available at https://icg-prod.s3.amazonaws.com/b67-somaliland-a-way-out-of-the-electoral-crisis.pdf
About the author
Dr. Axel Harneit-Sievers is the director of the Heinrich Böll Foundation’s (HBF) Regional Office for East & Horn of Africa, based in Nairobi, Kenya. Born in 1957, he is a historian and political scientist specializing in African studies. He worked with various universities and research institutions in Germany, before moving to Lagos, in 2002, as director of HBF’s newly-founded Nigeria country office. In December 2006, he took over HBF’s regional office in Nairobi.
