The Use of Technology in African Elections: A Critical Overview

We now live in a digital age, where computers have in many areas replaced human beings in performing several tasks. It is hard to imagine a sector of modern human activity which has not been computerised to some degree. Many routine tasks such as paying for goods and services, packaging goods, and looking for directions are automated. The consequence of this has generally been efficiency and precision in the provision of services. As a result, the use of computers has become part of the global human culture and popular consciousness. In this popular consciousness and culture, computers have acquired an undisputed sense of infallibility about their accuracy and precision. Even in the context of politics, there was optimism that technology would contribute significantly to better governance and human rights through increased transparency and efficiency.

In the same manner, technology is permeating other sectors of human life, and it is increasingly being used in managing elections in Africa. Today across the African continent, computers are performing work that was previously done by humans in the electoral process. Over the last two decades, many African countries have computerised one or more areas of the electoral process. It is estimated that about half of the national level elections involve use of digital equipment in some parts of the electoral process. This could be through the establishment of a biometric voter register, automated voter identification or verification process, and the computerised tabulation, transmission of votes and dissemination of results. For example, Zambia introduced the biometric voter register in 2006; Ghana introduced a biometric register and biometric verification of voters for the first time in the 2012 elections; Nigeria used 

Smart Card reading devices to authenticate voters in the 2015 and 2019 elections; Uganda introduced the biometric verification of voters and electronic transmission of results in the 2016 election; Namibia introduced electronic voting machines in 2014; and Kenya, in 2013 and 2017, computerised most of their electoral process through the deployment of the Kenya Integrated Electoral Management System (KIEMS), which computerised the voter registration, voter identification and the transmission and tabulation of election results.

Apart from the desire to promote efficiency in the electoral process, the increased computerisation of the electoral process is usually encouraged as a panacea to election fraud or rigging. For example, following the 2007 violence occasioned by discredited elections in Kenya, a commission of inquiry (popularly known as the Kriegler Commission) was appointed to review the crisis. To enhance transparency and to eliminate the possibility of human interference and error in the electoral process, the Commission recommended in part that the Electoral Commission should start having developed an integrated and secure tallying and data transmission system, which will allow computerized data entry and tallying at constituencies, secure simultaneous transmission (of individual polling station level data too) to the national tallying centre, and the integration of this results-handling system in a progressive election result announcement system.

Similar faith in electoral technology is visible in case law. For example, the Kenyan Court of Appeal (2017), in relation to the transmission of election results, had this to say: The electronic transmission of results was intended to cure the mischief that all returning officers from each of the 290 constituencies and 47 county returning officers troop to Nairobi by whatever means of transport, carrying in hard copy the presidential results which they had announced at their respective constituency tallying centers. The other fear was that some returning officers would tamper with the announced results.

In this sense, technology is seen as a cure for many contemporary electoral problems. Cheeseman, Lynch and Willis have argued that there are at least three reasons for adopting electoral technology. These are to enhance election management efficiency, to reduce the scope for electoral manipulation, and to improve transparency and clarity regarding management of election results.

I contend, however, that in countries where elections are routinely disputed and genuine questions exist about the fairness of elections, increased digitisation of the electoral process is more likely to mask than resolve electoral problems. Where elections are liable to manipulation, where the electoral process is characterised by an unequal or lopsided playing field, the use of technology is more likely to be deployed to further those vices rather than enhancing the quality of the electoral process.[1] Odote and Kanyinga, in recently published research, have warned that electoral problems in Africa require political solutions rather than technological interventions. They assert that technology is a tool and like any political tool, it tends to reinforce the structural positions of those in power. It does not by itself provide a level playing field.

In the African context, election technology, judging from presidential election disputes seem to be vulnerable at three levels. The first is the stage of controlling the information the voters receive. For example, there have been revelations that Big Tech companies were involved in the 2015 Nigerian and 2013 Kenyan elections, whereby, through the manipulation of information and leaking of confidential data, public views are carefully influenced and tilted against one candidate in favour of another. Part of the problem here is that social media companies harvest information across the globe and use it to influence political processes in a manner that may not have been contemplated or agreed to by the sources of information. This gives such companies ‘a position of invisible power compared to those who do not.’ Although this does not directly relate to technology used in the electoral process, it may have a direct impact on electoral outcomes, which may not reflect the free will of the people. 

The second one has to do with the voter register and, concomitantly, how the voters are identified, verified, or authenticated on election day. Problems and allegations of manipulation of the technology around this issue have been the subject of litigation in many presidential election disputes. For example, the 2013 and 2018 Zimbabwean presidential election petitions raised questions about the integrity of the biometric voter register. The 2012 Ghanaian presidential election dispute revolved mainly around the failed biometric identification system, while the 2019 Nigerian presidential election revolved around the failed Smart Card reading machines for voter verification. Researchers who reviewed the use of the biometric voter verification equipment in Ghana during the 2012 election came to the conclusion that in polling stations that had election observers, biometric verification machines were about fifty percent less likely to break down as compared to polling stations that did not have observers. They concluded that the malfunction of the machines was as a result of human manipulation in order to facilitate election fraud and over-voting.

To continue reading the full paper, please visit the Cornell eCommons repository.