Rejoinder: What does electoral violence signify?

In a recent article for the AfPak Channel, Megan Reif and Nadia Naviwala argue that the violence attending last month’s elections in Pakistan should be interpreted as "a sign that electoral administration is getting stronger and that democracy is maturing."  While they do not condone electoral violence, they argue that it is a normal part of the democratization process and signals the strengthening of institutions, reduced opportunities for fraud, and a crowding out of extremists from the electoral process.  To prove their points, they provide historical examples of electoral violence in France and the United States as a reminder that many of the world’s consolidated democracies also experienced violence as part of an incremental, centuries-long move toward democracy.

Part of Reif’s and Naviwala’s motivation may be to defend the democratic process under adverse conditions and to encourage nascent democracies to trudge forward in their pursuit of democratic consolidation.  However, the authors fall short when they argue that more reform can lead to more violence and, hence, violence can be interpreted as a sign of democratic and institutional progress.  Such a conclusion is troublesome and misleading.  Their thesis, while provocative, conveys a fundamental misunderstanding of the motivations behind the use of violence, which in many cases is a deliberate strategy employed by political actors.  

The Pakistani Taliban did claim responsibility for much, but not all, of the electoral violence that occurred before the national election in Pakistan. According to their statements, the violence was meant to disrupt and delegitimize the democratic process. However, there were also many instances where violence seemed to be employed as an electoral strategy by both the Taliban and political party agents.  

Commentators noted that Taliban-friendly political parties and candidates had been largely spared from violent attacks and that these parties, in fact, benefited from the violence. Violence was used to deter certain candidates — primarily those representing liberal parties — from participating in the election. Pamphlets were distributed that warned of the dangers of electing women and "infidels" into office.  There were also reports of voter intimidation and disenfranchisement by political party agents on Election Day.

The use of violence by political parties is not new to Pakistan. Indeed, the European Union’s Election Observation Mission report for the 2008 national and local elections notes that the Muttahida Qaumi Movement is known for using violence  Additionally, there is a precedent of Pakistani politicians using attacks by extremists as a cover for their own political gains; a new working paper by Paul Staniland of the University of Chicago details the collaboration between extremist groups and politicians. Thus, there is sufficient reason to believe that the electoral violence that occurred in Pakistan had many purposes besides preventing the election, one of which was to affect the outcome.  

The strategic use of violence may masquerade as an indicator that groups have no legitimate option by which to express their opposition to elections due to improvements in the electoral process.  But our research on Kenya shows that this use of violence suggests insufficient institutional reform, a critical failure of the state to protect the franchise of its citizens, and a high level of impunity for the past use of violence. The presence of electoral violence indicates weak institutional development, not democratic maturation. 

Much like the attacks in Pakistan, electoral violence in Kenya has historically taken many different forms and has been used for many different purposes.  It been used both to suppress and mobilize voters, deter aspirants from participating in elections, punish perceived political opponents, and to delegitimize the electoral process.  

Prior to elections in 1992, political elites allied with President Daniel arap Moi and the Kenya African National Union (KANU) party supported organized gangs, through which they orchestrated ‘ethnic clashes.’ While the clashes mainly pitted Kalenjin communities against non-Kalenjin and perceived KANU opponents, it also encompassed many different ethnic groups . Approximately, 1,500 Kenyans died as a result of the violence — first in a bid to demonstrate to Kenyans the dangers of transitioning from single party elections to multiparty elections, and later in an attempt to scare Kenyans into either voting for incumbent Moi or not voting at all. This pattern repeated itself in 1997 when approximately 100 people were killed in Coast Province (eastern Kenya) as part of Moi’s electoral strategy.  The Kikuyu ethnic group, which comprises approximately 22 percent of the population, was the main target of the violence, although other groups considered as non-indigenous to Coast were also affected.

After the 2007 elections, a serious political crisis broke out in Kenya.  The Electoral Commission of Kenya declared incumbent Mwai Kibaki the victor of a close and highly contentious presidential election, amid myriad allegations of fraud and vote rigging.  Violence — instigated by supporters of Raila Odinga, Kibaki’s opponent — began as a ‘punishment’ for supporters of Kibaki, a Kikuyu. Over the course of the next month, inter-ethnic violence claimed the lives of more than 1,300 Kenyans. 

This post-election violence was a turning point in Kenya. Its resolution required international mediation and led to the adoption of several major political reforms.  Most notably, a new constitution was approved in a 2010 referendum, significant electoral reforms were enacted to prevent fraud and increase confidence in the election, and steps were taken to reform the police — which were responsible for almost 40 percent of the fatalities. 

Elections in 2013 — Kenya’s first since 2007 and the major reforms that ensued –have been widely lauded for their relative peacefulness.  But here the world has focused only on the post-election period.  In fact, in the months leading up t
o the election, more than 300 people died as part of the campaign process.  Some of the violence, appearing as ethnic clashes, could be linked to aspirants vying for the new county-level positions, as documented by a Human Rights Watch report.  In other cases, the objective was to prevent the vote from taking place.  The Mombasa Republican Council (MRC), a separatist organization, attacked election officials and facilities to disrupt what they argued was an illegitimate election, much as the Taliban in Pakistan has denounced democracy as un-Islamic.  The MRC felt that by using violence and intimidation to lower voter turnout they would deprive the elections of their legitimacy and, ultimately, lend credence to the MRC’s claim that Coast Province should be independent of Kenya. 

The pre-election violence of 2013, which took place after considerable reform, is not a signal of institutional progress in Kenya.  Rather, it is a reflection of insufficient reform. Devolution spawned new competition for elected office, while an atmosphere of impunity for past attacks facilitated violence in other areas.  More to the point, ineffective and partisan electoral management bodies and a weak, corrupt judiciary have facilitated the use of violence as a part of the electoral process.  

To date, although the violent tactics employed by Moi and his associates are widely known, no one has been charged with a crime.  Furthermore, despite repeated calls from civil society and demands from the international community, none of the perpetrators of the 2007 electoral violence have been brought to justice.  In Kenya, electoral violence has never signified the strengthening and deepening of democracy, but rather it has served as an indicator of democratic fragility.

In addition to spotlighting institutional inadequacies, in rare but extreme cases, electoral violence may also be the first shot fired in what then becomes a deadly civil conflict. In Congo-Brazzaville’s 1994 legislative elections, the three leading candidates all had private militias, which clashed after the results indicated that Pascal Lissouba’s party had won.  The resulting violence left 2,000 dead.  After fighting broke out again between Lissouba and Denis Sasso-Nguesso in 1997 over electoral rules, 15,000 died before Lissouba took over.  Another 20,000 died from related violence over the next two years. 

Instead of accepting violence as a sign of democratic progress, we should learn from countries that have successfully navigated democratization without a call to arms.  There are many counter-examples of peaceful processes in Benin, Botswana, Cape Verde, and Mauritius.With greater understanding of the motivations behind electoral violence, we can do more to prevent it. Strengthening political institutions, improving electoral management, and ensuring those who commit electoral crimes are brought to account are all ways in which violence can be and has been successfully counteracted. 

Most importantly, violence is not an indictment of electoral democracy.  Instead, it should be seen as a means to help reformers identify where the breakdown of democracy is occurring.  Violence in Kenya has highlighted problems with electoral management, corruption in security forces, and judicial incapacity — all of which were targeted by significant reforms after 2007.  Kenya’s 2013 election was much improved from previous contests in terms of the integrity of electoral management and the monitoring of violence, but violence was still a part of the electoral process and it suggests that more reforms are necessary to protect the gains Kenya has made thus far. The lessons learned in Kenya can easily be applied to the Pakistani case. In both countries, we can congratulate courageous voters who cast their ballots under duress while still decrying violence and identifying areas for future reform. 

Stephanie M. Burchard and Dorina A. Bekoe, the editor of Voting in Fear: Electoral Violence in Sub-Saharan Africa, are research staff members at the Institute for Defense Analyses.  The opinions expressed above are those of the authors and should not be viewed as representing the official position of the Institute for Defense Analyses or its sponsors. Links to web sites are for informational purposes only and not an endorsement.