Chapter Six: Polling and Tabulation

Preventing Violent Disputes through the Free
Expression of the Electorate’s Will

This chapter discusses obstacles hampering the right to vote and the free expression of the electorate’s will, by examining critical operational and security challenges to the implementation of polling and tabulation procedures. Furthermore, it delves into key aspects that help to produce credible electoral outcomes, such as party agents’ access to polling stations; comprehensive, understandable voter-education campaigns; and observers’ freedom to scrutinise all stages of the process, including opening, polling, closing, and tabulation procedures. The second part of the chapter focuses on flaws in the transmission and collation of election results; a tense waiting period and voters’ mistrust of the tabulation process can easily trigger violent disputes, particularly when the contest is too close to call. In this regard, conducting parallel vote tabulations may help to deter fraud, build stakeholders’ confidence in the process, and prevent disputes over the credibility of official results.

Election Day

To a great extent, election-day results and the polls’ wide coverage by election observers and the media inform domestic and international perceptions of the credibility of the entire electoral process. The electoral authorities’ level of preparedness and political parties’ will to respect the legal framework are key to defusing tension. In contrast, after an election campaign and preparations marred by irregularities and incidents among stakeholders, instances of violence are more likely to flare up. Long waiting periods and queues account for many violent incidents at polling stations, particularly when organisational pitfalls exist along with poor crowd-management skills by polling staff or security officers on duty. In the populous neighbourhoods of African cities, an excessive ratio of voters and overcrowded polling stations often hamper polling. Violent incidents in overcrowded facilities prompt temporary or permanent closure of polling stations. The disenfranchisement of voters triggers political controversies, as opposition parties tend to enjoy more support in disadvantaged urban settlements where long queues and closure of polling stations usually occur (Vollan 2002; Carter Center 2002; OIF 2002c; EUEOM 2005b, 2006a).

Logistic glitches in the delivery and distribution of election material and staff to polling stations cause delays in the opening of polls. A wide range of incidents and disputes occur because polling staff negligently apply procedures, due to lack of training or partisan biases, such as non-eligible voters casting their vote, ballot stuffing, or double voting. Further delays in the transportation and transmission of results to the totalling centres spur anxiety and fears of manipulation among assembled crowds waiting for results, potentially leading to violent protests. New instructions by EMBs involving ad hoc changes in election procedures to address unexpected incidents generate confusion among election stakeholders. In addition, polling officials inconsistently apply last-minute instructions, due to communication shortcomings such as lack of mobile-phone coverage in rural areas.

Armed groups associated with political entrepreneurs and party organisers intending to disrupt the electoral process engage in violence or intimidation to prevent other parties’ supporters from voting (e.g., shootings to chase voters or party agents away from polling stations) or commit actions to rig the elections (e.g., ballot snatching). Security personnel near polling stations are needed in some countries to deter spoilers. Security officers are usually unarmed within the confines of the polling station, while rapid-response roaming brigades are on call to deal with potential emergencies. However, police personnel, unless deployed in sufficient numbers and with a clear mandate, cannot deter those who are willing to influence the election outcome by disrupting polling stations and intimidating officials, domestic observers, party agents, and voters.69

Party Agents

The presence of competent, independent party agents at polling stations is paramount to holding credible elections. In contrast, when party agents face obstacles to participating in the polls, the process is less likely to conclude peacefully, as stakeholders will hardly perceive the outcome as credible and legitimate. The legal or illegal barring of party agents from polling stations triggers tension and violent incidents. In some cases, security officers intimidate or arrest party agents before or on election day in order to prevent their presence at polling stations. More commonly, political parties face a wide range of challenges to get their party delegates accredited by EMBs, such as unclear instructions regarding where (at the central or local level) and when accreditation application forms must be submitted and will be processed. In addition to the challenges faced by electoral administrations to process accreditations on time, disorganised or resourceless political parties struggle to recruit their party agents prior to the end of the registration period. Most political parties also face constraints obtaining sufficient funds to provide transportation and permission for their delegates to be deployed.

Table 6.1. A Case Study on E-day Intimidation: Sudan 2010

A Challenge to Peaceful Elections:
Lack of Voter Education

Voter-education activities are crucial to guarantee that all citizens know their political rights, including their right to participate in the electoral process (European Commission 2008, 39–40). However, when resources are scarce, during the different stages of the electoral cycle EMBs prioritise other essential preparations over voter education. A common concern of EMBs is that voter-education activities concentrate mainly in urban areas, while remote, rural areas are often neglected (Carter Center 2002). Insufficient voter-education activities hamper the chances of vulnerable population groups to participate in the political process. As a result, the proportion of rejected ballot papers in rural areas is generally higher. The problem often stems from scarce resources, such as four-by-four vehicles and fuel needed to reach remote areas. Moreover, EMBs tend to produce voter-education materials only in the country’s main languages, depriving many speakers in rural areas of a clear understanding, in their local language, of the voting procedures. In other cases, materials are actually produced but kept in warehouses and never distributed on time (Carter Center 2010a). Generally, EMBs alongside state bodies responsible for civic education, NGOs, and civil-society organisations conduct voter-education activities. However, the bodies conducting voter education are not always well coordinated. Governments wary of losing political power may attempt to monopolise education activities by imposing restrictive legal measures banning any other bodies from conducting such activities (for further information on the role of political parties in the dissemination of voter education, see the section “Voter Education as a Campaign Strategy” in Chapter 5).

Table 6.2. Restriction on Voter-Education Activities

Obstacles to Observers­’ Accreditation
and Participation in the Process

Election observation serves as an effective conflict-prevention instrument by strengthening the credibility of the electoral process. The presence of competent election observers contributes greatly to building stakeholders’ confidence in the role of EMBs in electoral preparations; nonetheless, evidence gathered by international observer missions on polling days is not always statistically representative because the number of polling stations observed by their roaming poll watchers does not always exceed a representative sample of stations. In contrast, domestic observers can pool larger human resources, and if funded properly, local organisations can deploy fixed observers to most polling stations. However, financial, training, and logistical shortcomings constrain their capacity in terms of observation skills and reporting channels. Moreover, overzealous governments attempt to curtail the role of organisations in election observation tasks. Restrictions on observation activities stem from mistrust regarding alleged political agendas pursued by civil-society actors and foreign donors or simply because monitoring efforts expose incumbents’ manoeuvres to win elections by fraudulent means.

Table 6.3. Examples of Restrictions on Election Observation

Tabulation Flaws: A Conflict Trigger

African EMBs often lack resources to conduct the transmission and aggregation of results in an organised and efficient manner. Numerous factors provoke tensions during the tabulation exercise: lack of clear instructions, transparency shortcomings, fatigue of election officials, delays in the transportation of materials to collation centres, long queues to deliver materials to crowded or small collation facilities, interference or intimidating presence of security personnel, among others. Restricted access of observers and party agents to tabulation premises certainly indicates possible tampering with results and undermining the legitimacy of elected powers, thereby rendering the government and democratic institutions more vulnerable to instability and military interference in the long run. Delays in the announcement of results at the polling-station or collation-centre levels often instil public anxiety that quickly develops into protests. A common concern is that polling-station results can be manipulated in the tabulation process, particularly in complex multitiered collation processes.

Case Studies: Disputes Over Tabulation

Zanzibar

ZEC did not provide the results of the 2005 general elections in Zanzibar at the polling-station level, raising suspicions about the integrity of the process (NDI 2005). Voters perceived that the use of the army to transport and deliver election materials and results to the regional counting centres undermined the returning officer’s authority (EISA 2006c). Observers also expressed concern about the overwhelming presence of security forces (Commonwealth Observer Group 2005b). The cordoning off of Stone Town, where many supporters of the Civic United Front (CUF), the main opposition party, had concentrated in its central office, also heightened tensions, and the security forces fired teargas to disperse protesters (NDI 2005).

Democratic Republic of Congo

Uncertainty triggered tension between supporters of Joseph Kabila and Jean-Pierre Bemba during the tallying of the first round of the DRC elections in 2006. The announcement of provisional results led to violent clashes on 20 August 2006. Afterwards, to reduce tension, all stakeholders, including the media, agreed not to speculate about the results and to report only those provisional results announced by the Commission Electorale Indépendante (CEI). High-level meetings between representatives of the two candidates, the CEI president and the head of the United Nations Organization Mission in the Democratic Republic of Congo (MONUC), helped to calm people down. Runoff results were finally announced on 15 November 2006, giving victory to Kabila. Bemba did not accept the results but pledged to use only peaceful and legal channels to topple Kabila (EUEOM 2007c; EISA 2007a).

Kenya

It is widely accepted that the mishandling of election results did not primarily cause the wave of violence and chaos that broke out in Kenya; the root causes are multilayered, including long-standing rivalries between influential politicians, disputes over allocation of power and resources among dominant ethnic groups, high poverty rates, the proliferation of armed youth gangs, among others. Nonetheless, reports and studies on post-electoral violence agree that delays and severe problems of transparency and accountability in the transmission of presidential results helped to escalate conflict. While in some instances violence was orchestrated, the declaration of results also triggered spontaneous expressions of anger, as emphasised by the Commission of Inquiry into the Post Election Violence (CIPEV), which was established by the Kenya National Dialogue and Reconciliation Accord of 28 February 2008. On violence in the Rift region, its 529-page report concluded that “an aspect of the violence was spontaneous, and that the disaffection with the election results or the perceived illegitimacy of the process could have been the cause of the violence” (Waki et al. 2008, 159).

Serious transparency shortcomings marred the tabulation process. Party agents and observers were not granted permanent access to the national tallying centre. Constituency results received by telephone or fax and entered into the database were not duly cross-checked with the results forms submitted by returning officers. While observers deemed parliamentary results generally credible, various constituencies reflected serious inconsistencies between the presidential election results announced by the Electoral Commission of Kenya (ECK) and those initially announced at the constituency level (IRI 2008). Because the network of domestic observers, Kenya Election Domestic Observation Forum (KEDOF), was not allowed to monitor the tallying process at the ECK, observers could not mediate among the Party of National Unity (PNU), the Orange Democratic Movement (ODM), and the ECK regarding tallying disputes (Hoffman and Smith 2010). The ECK chairman declared Mwai Kibaki president, in a move of questionable legality, as the tallying was incomplete and many irregularities remained in the result sheets and in the documentary evidence received from returning officers. Election stakeholders, including observers and even four ECK commissioners, questioned the credibility of some results and called for an independent inquiry into whether the results had been tampered with prior to the announcement of the final results (EUEOM 2008).

Table 6.4. CSOs’ Role in Mediating Tabulation Challenges in Ghana 2010

Parallel Vote Tabulations

Parallel vote tabulations (PVTs), also called quick counts, play an increasingly important role in the prevention of election disputes in sub-Saharan Africa. Local volunteers usually conduct PVTs and collect data based on direct observation of voting and counting processes, including qualitative information on how the procedures were implemented and the actual count. All findings are collected through standardised forms and communicated to a central collection point. PVTs can be comprehensive, i.e., data from all polling stations or, usually, a statistically significant random selection of polling stations designed to produce a credible projection of election results. The main goals of quick counts are to deter or detect fraud, produce qualitative information on the process, and promote confidence in the official results by conducting an independent check of officially tabulated results or a forecast of the final results (Estok et al. 2002).

Donor agencies and international election observation groups such as the Carter Center, Democracy International, and the National Democratic Institute have developed expertise in training and technical support for domestic election observation groups to conduct quick counts countrywide or in potential hot spots. The NDI has thus far assisted civil-society groups in forty-six countries worldwide to carry out 180 PVTs (NDI n.d.b.). For example, NDI provided technical assistance to develop the Project Swift Count (PSC), a coalition of four civil-society groups (the Federation of Muslim Women’s Associations in Nigeria, Justice Development and Peace/Caritas Nigeria, the Nigerian Bar Association, and the Transition Monitoring Group) that conducted PVTs of the 2011 gubernatorial election results in Kogi, Adamawa, Bayelsa, Sokoto, and Cross River states (NDI 2012, n.d.a.). Networks of domestic observer groups and regional nongovernmental organisations such as the Electoral Institute for the Sustainability of Democracy in Africa (EISA) have recently developed their own expertise on PVTs, drawing on their own experience and on lessons learned from PVT exercises elsewhere in Africa.70

Quick counts can be particularly useful when election stakeholders’ and the electorate’s confidence in EMBs is low. In contrast, mistrust of the outcome of a questioned electoral process will likely grow if authorities do not permit civil society to conduct a parallel vote tabulation exercise. Nonetheless, quick counts tend to be politically sensitive since their results may condition the post-election environment and people’s reaction to the final results. Thus, organisations conducting quick counts should explain in advance the purpose and characteristics of the project to EMBs, political parties, and other key stakeholders and try to address reasonable concerns. On the other hand, organisations conducting quick counts should carefully assess whether to make PVT results public immediately or to share them first with the EMB. This issue becomes particularly crucial if substantial differences are already perceived between the provisional official results and the data emerging from the quick count. Alerting EMBs promptly to discrepancies and observed fraudulent practices before informing the public may give electoral authorities additional time to address shortcomings and modify the results accordingly. In contrast, prematurely announcing contested official results can trigger violent protests, as illustrated by the tragic events that followed the official announcement of controversial election results in Kenya at the end of 2007 (see Bjornlund 2004, 279–300).

Concluding Remarks and Lessons Learned

Tension runs high on election day, and if minor incidents are not promptly addressed, they can quickly spark violent disputes. Thus, as previous chapters have outlined, the orderly conducting of polls depends on many election preparations that must occur successfully during each phase of the electoral cycle. Nevertheless, deterring all individuals or organised groups with a political agenda to disrupt the polls is not always feasible given the size of most African countries and state resources available for securing the polls. There is no one-size-fits-all recipe to avoid trouble on election day. In certain circumstances, even when election preparations are on track and security arrangements are in place, authorities are powerless to deal with organised groups willing to disrupt the process for political purposes. On the other hand, although observers noted logistical glitches in most polls conducted in sub-Saharan Africa between 2000 and 2011, election days were relatively peaceful across the region, notwithstanding allegations of intimidation in countries such as Sudan, Angola, and Zimbabwe. Violent incidents resulting in fatalities occurred generally in conflict-torn countries such as Burundi, Côte d’Ivoire, and the Democratic Republic of Congo, or in those marred by a track record of violent elections, such as Nigeria, Kenya, and Togo. In this regard, as the following chapter explains, a prevailing culture of impunity for election crimes serves to perpetuate electoral malpractices in successive elections, such as ballot stuffing, ballot snatching, or violent disruption of polling stations and their surroundings.

Observer missions have provided recommendations to strengthen election-day procedures. In urban areas, reducing the ratio of voters per polling station may help to avoid overcrowding and chaotic queuing. Likewise, to implement tabulation procedures in an organised and efficient way, collation premises need to be adequate in terms of size and readiness (i.e., a suitable setup for collating results and retrieving sensitive materials, such as used ballot papers and completed result forms; a stable power supply; working equipment; sufficient and well-trained staff; among others).

Moreover, observers emphasise that election officials and party agents require comprehensive training on electoral procedures to address contentious issues in a peaceful and constructive way. Intensive and well-organised voter-education activities are also crucial, to instruct the electorate on polling procedures and avoid confusion and wrongdoing at the polls. These activities should target specific groups such as the elderly, the disabled, the illiterate, and first-time voters. In addition, competent authorities should facilitate the presence of duly authorised party agents, observers, and journalists at polling stations and tabulation centres, since their participation greatly benefits the integrity, transparency, and credibility of the process.

Last but not least, effective communication channels should be developed among EMBs, political parties, voters, observer groups, and security stakeholders to address potential security threats and other unexpected circumstances. An effective communication flow is paramount to ensure a smooth polling exercise and surmount operational challenges, such as delays in the distribution or retrieval of election materials, shortages of sensitive materials (e.g., ballot papers), inaccuracies in voter rolls, defective ballot papers, and so forth. In such circumstances, when EMBs deem it appropriate to make ad hoc changes in polling or tabulation procedures to address unforeseen glitches, election bodies must convey clear, timely instructions from headquarters to polling staff in the field, to preserve the integrity of the ongoing process.