The 7 July 2012 elections in Libya were hailed locally and internationally as a significant turning point toward peace. The polls, barely a year after the violent ousting of Colonel Muammar Gaddafi, saw 200 candidates elected to the General National Congress, which would serve both as an interim parliament and as a constitutional assembly to guide the country into a new era of democracy, peace and stability.

Remarkably, the elections were comparatively peaceful. A small but significant boycott centred on the oil-rich city of Benghazi. The elections were carried out mostly by Libyans, reflecting a high degree of local ownership, and were hailed as a “light footprint” success for the United Nations Support Mission in Libya. People who had served in the Gaddafi regime were banned from participation, but the election was otherwise largely inclusive. Peace, however, has not come easily to Libya. Armed militias continue to hold considerable sway and the central government has yet to extend its authority across the country. Without a central system of accountability, crimes and insecurity continue, including widespread torture in informal prisons run by militias. The elections do not appear to have made much difference to most Libyans.

Do post-war electoral processes advance peace and democracy? Or do they lend artificial legitimacy to dubious governments? Do they even induce further violence – instigating conflict, exacerbating social differences, and abetting ethnic, sectarian or religiously inspired “entrepreneurs” who mobilise around society’s fault lines in pursuit of personal power? 

Elections can be a turning point either away from or toward peace: much depends on the why, when, and how of specific electoral processes. From the celebrated 1989 elections that brought independence to Namibia, to Libya in 2012, to the looming presidential elections in Afghanistan in 2014, debates continue about whether and how electoral processes help advance legitimate governance in countries emerging from conflict. 

The controversial elections in Côte d’Ivoire in December 2010, verified by the UN but disputed by local parties, led to violence that cost an estimated 3,000 lives and displaced nearly half a million people. The genesis of the poll can be traced to a commitment in the 2003 Paris Peace Accords. Delays in holding the elections exacerbated divisive factionalism among the principal contenders for the presidency.

In the past, international actors have prematurely pushed election processes. The 1996 elections in Bosnia and Herzegovina, for example, were held a year after the Dayton Peace Agreement according to a timetable that had more to do with domestic electoral politics in the United States than any objective assessment of the conditions on the ground. The Bosnian elections arguably exacerbated conflict as the electoral process effectively legitimised nationalist political elites and factions with a highly dubious commitment to peacebuilding.

But it is not the case that post-war elections are always externally driven. The impetus for quick elections after the guns fall silent often comes from within, including from those parties and factions who believe they will win and hope that their rule can be legitimised by an external pronouncement that the vote credibly reflected popular preferences. In fact, donors often seek to slow down the election timetable, to improve the technical aspects of elections (particularly voter registration processes) and to encourage the parties to adopt an electoral system that will yield inclusive parliaments and realise other international standards, such as the inclusion of women.

Political elites in societies divided along ethnic, sectarian or religious lines may have incentives to “play the ethnic card” in electoral processes as a way to induce fear among the population and to manipulate support for more extreme positions, as Jack Snyder has previously described. This poses a security dilemma for other groups, who counter with their own claims, thereby generating a centrifugal or outward spin to the political system. Under such conditions of deep social division, elections can become something of an ethnic census.

The problem of elections inducing conflict is directly related to three additional factors. First, losing an election may jeopardise personal or group security. This problem is particularly acute in presidential elections like in Côte d’Ivoire, which are perceived as winner-takes-all. In Iraq, insurgents who expected to be systematically excluded from power mobilised to disrupt governorate or provincial elections in 2013.

Second, parties may resort to political violence as a strategy to influence the electoral process or outcome. In parliamentary elections in Afghanistan in 2005 and 2010, insurgents targeted election workers (both international and Afghan) and sought to disrupt balloting as a way to undermine the legitimacy of the process and of the regime of President Hamid Karzai.

Third, where capturing state power brings opportunities for enrichment, such as access to revenues from natural resource exports, this creates an incentive to use violence, intimidation and electoral fraud. Sudan’s elections in 2010 are a case in point: the regime used a wide array of tactics to ensure that the ruling National Congress Party would stay in power and retain access to revenue derived from exports of crude oil.

Post-war elections can also provide a cloak of legitimacy to governments that have won militarily on the battlefield. For example, the Economist reported that 2013 parliamentary elections in Rwanda were used to legitimise the rule of the Rwandan Patriotic Front and President Paul Kagame in a poll in which opposition forces had been imprisoned or otherwise suppressed for fostering ethnic “divisionism”.

The Rwandan elections are indicative of a broader concern that many governments with dubious democratic credentials have used elections to mask autocratic rule. Such cases of “facade democracy” reflect how ruling elites have learnt the electoral game and devised ways of controlling, managing and ultimately determining the outcome of elections well before the first vote has been cast.

In Angola, Cambodia and Zimbabwe, elites who prevailed in conflict have used the levers of state power – such as control of the judiciary, manipulation of the electoral system, restrictions on or vetting of candidates, or falsifying the voters’ roll – as a means to prevail in elections. International efforts to monitor elections and prevent electoral fraud are also often highly constrained or “stage-managed”. In Zimbabwe, monitors with the Southern African Development Community have been prevented from revealing electoral fraud by the sensitivities of regional and neighbourhood politics.

When electoral processes fail to be sufficiently inclusive, or when they are managed and controlled by wily governments seeking to predetermine their outcome, elections can lend a facade of legitimacy to otherwise undemocratic ruling elites.

International law in effect provides for a right to democracy. Instruments such as the Universal Declaration of Human Rights and the International Covenant on Civil and Political Rights (particularly Article 25) include specific rights to freedom of association, to choose representatives, and to participate in elections. It is well-nigh impossible for external peacebuilders working for the UN, for example, to say that any country or any person is not ready for democracy when it is codified in international norms – including in the charters of European, Latin American, and African regional organisations.

But electoral processes can provide a turning point toward more “normal” politics and peace. The 1994 elections that signalled an end to apartheid in South Africa are often seen as archetypal. Despite a very bloody transition in which some 14,000 people lost their lives in widespread political violence, the elections led to the creation of a constitution-making body and interim parliament that eventually produced a new constitution in 1996, and which many saw as the basis of a new social contract for a long-divided society. There had been widespread fears that the elections would generate new conflict; in fact, violence diminished dramatically and President Nelson Mandela’s new government enjoyed broad internal and external legitimacy.

Electoral processes can include public decision-making – usually by majority rule – through a referendum process designed to lend direct public legitimacy to a peace agreement. Following 27 years of conflict in Northern Ireland, a 1998 referendum revealed widespread public support (71 per cent) for the Good Friday Agreement (Belfast Agreement) reached earlier that year. Likewise, the 2011 referendum on independence in what was to become South Sudan was the culmination of a multi-year peace process that also produced the Comprehensive Peace Agreement of 2005. Nearly 99 per cent of the southern electorate cast votes in favour of independence from the north, lending strong (but perhaps not enduring) legitimacy to the ruling Sudan People’s Liberation Movement regime.

Properly designed and implemented elections can lead to broader inclusion in governance for marginalised and historically disadvantaged groups, particularly women and minorities. Rwanda’s managed elections have helped women to join the political elites – Rwanda has the highest proportion of women in government according to the International Parliamentary Union.

Similarly, previously excluded minority voices can be brought into politics through mechanisms such as low-threshold proportional representation. In Nepal, following the 2006 Comprehensive Peace Accord, the 2008 Constituent Assembly elections led to the inclusion of 25 political parties in the constitution-making process, among which were parties representing Madhesi groups from the Terai region and indigenous and minority groups that had suffered discrimination during the royalist ancien régime. That process failed in the end, but the 2013 elections, which were mostly peaceful, may give Nepal a new chance at creating a Constituent Assembly to draft a new charter for this deeply divided society.

Electoral processes are too complex and diverse, with too many moving parts and too many variations of experience and examples, to warrant generalisation. Much depends on the nature, trajectory and manifestations of the conflict, the ways in which elections are defined and agreed in peace accords, and the nature of international–local partnerships for election administration. Each and every context deserves very specific assessment across a wide range of factors, and strategies for intervention to tip the balance of any electoral process toward peace must be carefully designed based on the outcomes of such assessments.

The electoral system is, as Giovanni Sartori has said, the most manipulable element of politics. Whichever electoral system is ultimately chosen for any given electoral cycle, it must be based on a broad consensus to ensure it is acceptable – across political parties, civil society, academia, and traditional or informal leaders. Such acceptance is only likely to be found when the electoral system is highly proportional.

The choice should belong to local stakeholders. External partners can encourage a system that provides for the broadest possible inclusion in post-election governing bodies as the most important ingredient to sustainable peace. This means promoting a mindset for inclusion and proportionality and paying attention to how the choice of election system leads to stable and capable governing coalitions. As in Liberia, majority-rule systems that are centrist and inclusive can lead to a more “developmental state”. 

In Nepal, failure to build a stable, inclusive nationally-oriented governing coalition has impeded development. Political polarisation along identity lines meant that the Constituent Assembly ran aground over whether Nepal should adopt an “ethnic federation”. The most important challenge for Nepal is not about whether to have elections, but how to balance the need for inclusion and representation across a wide range of social cleavages with the need for stable government that is focused on development and can provide inclusive economic growth and a shared national vision. Given the likelihood that there will be no majority in the new constitutional assembly, the focus in Nepal will likely progress from elections to coalition building.

It is vital that elections do not themselves encourage violence. Preventing election-related conflict requires a continuous programme of work focusing on a range of issues including the legal and constitutional framework, election-related security, electoral management bodies, civil society training and engagement for monitoring and observation, dispute resolution, informal “infrastructures for peace”, and public awareness campaigns demanding a peaceful and credible election. New social media approaches have rightfully gained attention in recent years as innovative ways to more effectively link citizens directly to monitoring and accountability for the electoral process.

The credibility of the state runs much deeper than the fleeting credibility of elected elites. Yet Libya shows that electoral processes are essential components of modern democracy for many people. The internationally recognised National Transitional Council led the election process in Libya, and the UN was able to provide critical technical and administrative assistance. But Libya also challenges common assumptions around the need for statebuilding before democracy or peace. Perhaps it would have been wise for the Libyans to have engaged in security sector reform (integrating the militias into the state), rather than rushing into elections so quickly after the end of the civil war. In the end, it was their own decision.

Today, the confluence of global norms and local expectations means that elections are essential to peace agreements and their successful implementation. But the Libya case acts as a reminder that elections are but one of many turning points in transitions from war to peace and a democratic and capable state.