Version classiqueVersion mobile

Kenya’s Past as Prologue

Christian Thibon
Marie-Aude Fouéré
Mildred Ndeda
et al.


Kenya’s 2013 General Election: A National Event Set Between ‘The Inescapable’ and ‘The Unforeseeable’

Christian Thibon, Marie-Aude Fouéré, Mildred Ndeda et Susan Mwangi

Texte intégral

  • 1 This undertaking follows a similar initiative launched during the 2007 Kenyan election by the Frenc (...)

1This volume is the fruit of joint research carried out by a group of academics drawn from the research network of the Institut français de recherche en Afrique or IFRA (French Institute for Research in Africa). Most of them work in major Kenyan and French universities: some are doctoral students while others are junior or experienced scholars in the humanities and social sciences, mainly political science, history and anthropology. All of them were aware of the tremendous civic issues that were likely to arise during the March 2013 general election. They had also participated in research on the 2007 general elections and post-election violence, or witnessed them as Kenyan citizens or foreign observers. Questions such as whether past mistakes could be avoided this time and what new faces might emerge after the presidential election results were foremost in their minds. They were therefore convinced that a national event of such significance in Kenya’s history and political life was a crucial research topic that needed to be both explored and explained.1 Their research was completed a few weeks after the vote and thus does not take into account the impact of the Westgate terrorist attack in September 2013 on the Kenyan political scene.

2This publication testifies to the desire to build a research program capable of offering insight into the recent past – by definition a complex past that calls for an approach at once narrative and explanatory. Exploring the recent past requires understanding the event – or its “momentum” – without, however, being able to view events from the distance normally demanded of the researcher or simply adopting a linear historical, cultural and structural perspective. Integrating all the necessary explanatory factors into an historical analysis can indeed be difficult at times. And in giving primacy to these explanatory factors, one tends to lose sight of the possibility of accidental trends, disruptions and discontinuities, commonly regarded as deceptive or delusive. All in all, the approach adopted in this volume situates the contingency of the event between the inescapable and the unforeseeable (Rémond, 1984). The general election, by nature a conflicting and passionate process, took place in a heated environment characterised by attempts to make political, partisan, ideological, media and even academic headway. At the same time, they sought to rectify previous political mistakes and justify the need to forgive and forget, a discussion that was – and still is – indispensable for the country as a whole.

3The approach and methodologies used in this work are undeniably academic: researchers collected data on the successive phases of the electoral process (the campaign, the political debate, the post-election period) and on the various actors involved. They have undertaken an analysis of the election and its outcome from multiple points of view, paying special attention to their different – though usually intermingled – national, regional and local dimensions. Most of the contributions in this volume draw extensively on findings obtained with ethnographic methods – in-depth interviews and on-site participant observation. They combine or compare these findings with data from nationwide field studies and data-collection techniques such as polls and surveys as well as scholarly literature on the historical and political scene since Kenya’s independence, particularly the 1992 and 2007 general elections. Reflections on colonial memory, especially as it relates to the land question and its attendant tensions, are also part of our understanding of the central issues at stake in Kenya.

Against electoral doom-mongering

4The breaking news that constantly punctuated the election period together with the unknown factors and worrying scenarios that were widely discussed – notably the polarisation of Kenyan society – prompted the academic community and even more civil society, the government and the international community to focus their attention on the places and populations that were prone to radicalisation. These “hotspots”, as they are often called, were targeted as areas in which potential political and ethnic violence had to be managed or controlled. These shared pessimistic scenarios were partly grounded in the trend towards electoral violence and its crescendo since 1992 as well as the workings of the political system and problems implementing the new national constitution. Due to the urgency and risks of the political situation, everyone, including the universities, felt compelled to take part in electoral monitoring. This broad-based engagement had its flaws, however: it mostly resulted in a restrained, consensual interpretation of some situations and the distortion of others, as if they were viewed through a magnifying glass.

5As a general rule, the sampling of observations gathered in this volume was made randomly. Despite insights into electoral sociology, the analysis lacks a representative sample of places and spaces of the various electoral constituencies, though this could still be done. Such a sample might have served as a useful database for estimating voting patterns based on early electoral results rather than on the results gathered at the polling stations after all the ballots had been cast. In other words, the assessment of voter choices and the electoral process, including electoral irregularities, could make use of other data besides the information obtained from nationwide exit polls (Ferree, Gibson and Long, 2014). We were aware of this gap and did not wish to fall into the trap of catastrophism that had characterised the predominant view of Kenya in the media and among academics during the previous election. More often than not, elections in Kenya and in Africa generally tend to bring out and intensify latent political crises rather than opportunities to overcome them. The research presented here set out to look at the elections differently, tracking the events in two distinct yet complementary directions.

Navigating the inescapable and the unforeseeable

  • 2 Longue durée history is defined by French historian Fernand Braudel as “a history to be measured in (...)

6Firstly, the researchers adopted a critical approach that some readers may find a bit cynical and pessimistic, in the sense that it tried to identify what we call “the inescapable” in the elections. This approach, which distinguishes between the objectives of academic research and political expertise, was focused on assessing the expected fault lines and disputes of the electoral campaign. It examines these issues at the local level in everyday practices, taking a close look at corruption, fraud, manipulation, etc., and at the national level through political actors (e.g. the ethnicising and politicising of identity, the use of violence to achieve strategic electoral goals, etc.). This second level could be called “the prisons of longue durée”, to borrow the words of Fernand Braudel (1958: 31).2 As explorers of electoral political life with its political elites and dynasties as well as its political entrepreneurs, researchers are familiar with the structural factors – defined as stable elements over the long term – that weigh on the course of the electoral event, limiting and shaping the ways it unfolds.

7Secondly, the approach took a more optimistic turn by paying attention to the overall trend and attempted to grasp what we call here “the unforeseeable”. It investigated the appeasing, or at least moderating, factors and hypotheses that might promise a happier election outcome. These factors include: cosmopolitan or mono-ethnic spaces, whether urban or rural; the crucial role of civil society; the significance of new, emerging actors – social actors such as the middle class and institutional actors; information and communication technologies, notably social media like Twitter, Facebook and blogs; and lastly, the institutional or practical capacity to manage an election requiring voters to cast six separate ballots on the same day. This approach was pursued against the backdrop of a broad reflection on whether Kenyan society could be considered an “electoral civilisation”, i.e. a civilisation featuring electoral civic practices and knowledge, in which voting is a peaceful act resulting from a historical process of self-restraint. Past experience and memory, particularly of 2007-2008 post-election violence, played a crucial role during the 2013 electoral year.

8These choices did not preclude an interest in exploring the specific challenges facing certain sensitive places ( “hotspots”) of the country. The sensitive areas are mostly peripheral, both in the geographical sense, i.e. the pastoral areas and semi-arid regions of the North and the Indian Ocean coast, and in terms of socio-cultural patterns (including in particular the marginalized populations, peoples and ethnic groups of Kenya). They also encompass historically identified ethno-political conflict areas. This was especially the case of the Rift Valley. In view of the new ethno-political Jubilee alliance between Uhuru Kenyatta and William Ruto, a Kikuyu and a Kalenjin respectively, most of these hotspots de facto lost their potential for conflict.

  • 3 The need to prevent violence was on everyone’s mind in 2012 and 2013, as the debates in the media a (...)

9The two-faceted approach being implemented by the IFRA research team became obvious at two interim programme meetings in January and February 2013. The first meeting was devoted to mapping the electorate and determining the relevance of opinion polls; the second centred on the ability of “soft power” to influence the conduct of the elections. The group discussed the increasing concerted efforts in December 2012 and January 2013 to mobilise citizens in the central areas of Kenya on the issue of land policy (which had already been the case for the 2010 referendum), whereas the peripheral areas remained poorly mobilised. This joint reflection enabled the research team to revise opinion poll projections that took only demographic data into account. The conclusions from the second meeting were more enigmatic. They addressed the likelihood of a calm election scenario under the management of civil society and several stakeholders. All these players seemed to tacitly agree that election-related ethnic and political violence loomed large over the election process and that it was crucial to avoid replaying the disputed electoral competition of 2007-2008 that had torn the country apart.3

  • 4 Some research and case studies not covered in this book are still in progress in individual doctora (...)

10But this half-hearted analysis suggested a certain interpretation of Election Day and the post-election period as the country awaited the validation of the results by the Supreme Court. Indeed, the peaceful conduct of the elections, the absence of political confrontation which might well have arisen in such a highly polarised electoral campaign, the voting itself and the settlement of post-election disputes all came as a surprise to many observers and analysts – particularly in light of numerous reports of electoral fraud and the malfunctioning of election-related technology. The questions guiding the group’s collective observation therefore changed in the course of events, confirming or rejecting the hypotheses formulated at the beginning of the electoral process. This book brings together these observations-cum-analyses,4 written from July to October 2013, when the terrorist attack on Westgate and its management by the Kenyan government brought the electoral period to a close in an apparent movement of national unity that cut across party allegiances.

Exploring the particular to reach the general

11Readers will encounter two different types of texts here. The first could be described as “impressionist”. These essays take stock of general or topical issues and highlight the structural factors underlying elections and voting in Kenya, as well as Kenya’s political system, culture and political transition process. They also examine new structural and short-term trends and the core issues at stake in the new political order. The second type of texts could be called “pointillist”. These chapters offer insights into specific case studies, situations and contexts and bring nuances and diversity into focus against the background of more systemic analyses in the “impressionist” texts.

12The introductory contribution to the volume presents a broad overview of the elections, seen from a middle- or long-term perspective. It provides keys to understand the combination of factors that shaped the specific patterns of the 2013 general elections in Kenyan electoral history since 1963. The other contributions focus on specific, localised topics. They seek to explore the role of electoral strategies – of individuals, political parties or alliances – at the domestic level but also with regard to international stakes; political loyalties and routine patronage practices; the sense of belonging instilled along clan, ethnic or regional lines; generational and gender dynamics; and patterns of inclusion in and exclusion or marginalisation from the political system. At this stage, such an event-focused, locally grounded approach to the elections may not show how these last elections fit into the broader perspective of Kenya’s long-term electoral and political history. Nevertheless, they do provide us with rich empirical findings that can serve as a basis for further – and later – theoretical reflection on the significance of the 2013 Kenyan elections – whether as a turning point, a mere continuation of the past, or a slight but real change – in Kenya’s history.

13The research collection nevertheless remains incomplete, both in terms of the topics covered and perhaps even more in its monographic section. A number of books, special issues and articles published to date (notably Cheeseman, Lynch and Willis, 2014) can serve as a useful complements to make up for the lack of studies on certain issues, areas and stakeholders. This does not mean, however, that the aim and general pattern of this volume come under the heading of applied research, like reports on how to manage and prevent electoral violence (Adenauer Foundation, 2013), let alone political expertise or consultancy. Nor does it fit entirely into broader, highly conceptualised theoretical discussions on elections in Africa. It explores specific case studies and topics mainly to give readers the empirical tools and alternative analyses they need to achieve a better grasp of Kenyan history and the country’s current political, economic and social situation. When the research presented here is compared to other national situations in neighbouring African countries, it can indirectly, and where applicable, satisfy the expectations of readers looking for usable, operational information as well as those seeking an entry to general discussions of elections in Africa.

The scope and limits of democratisation

14There are three main topics that cut across all the contributions, implicitly or explicitly. The first of these topics is violence. As indicated above, most scenarios indeed predicted, or at least feared, renewed violence in 2013. Yet, unlike 2007-2008, the 2013 general election in Kenya did not result in mass violence, though violent actions and turmoil were not totally absent. Clashes took place between and within communities in several regions of the country, especially in Kenya’s peripheral areas, resulting in more than 500 deaths and nearly 120,000 internally displaced persons, thus revealing ongoing latent tensions and contained violence. Deterring election-related violence – whether actual or symbolic – was a tremendous challenge and it turned out to be relatively successful, testifying to the growing ability of Kenya’s “soft power” to make a difference. A “virtuous environment” was introduced and politicians, civil society and the media were mobilised to prevent potential low or high levels of violence. In the run-up to the 2013 elections, the leading presidential candidates made a systematic, collective appeal for peace and stepped up their calls for tolerance; CSOs and NGOs – both secular and religious – launched programmes for civic education and peace strongly imbued with violence-prevention messages and engaged in election monitoring and conflict management; the media worked with government institutions to broadcast messages of peace and stifle “hate speech”, which had been a major trigger in previous violence and turmoil; and lastly, a dissuasive climate was fostered by the security forces and the judiciary. The 2013 election appears to have been exceptional in the sense that it contradicted the scenario anticipated by the international community and Kenya’s media. Hopefully it illustrated a step forward in the construction of public space, civil society and the rule of law, in other words, in political maturity and progress towards democratisation in Kenyan society.

  • 5 In the end, the electoral administration reached nearly 80% of its voter target. On the other hand, (...)
  • 6 While most observers expected a run-off between Raila Odinga (Cord) and Uhuru Kenyatta (Jubilee), K (...)

15A second theme that structured the research team’s questioning had to do with the reliability and credibility of the electoral process driven by the state. The organisation of free and fair elections that reflected the choice of the majority was a major concern throughout Kenyan society. It reflected underlying expectations that these elections could be a turning point in opening up a new democratic space. However, as in 2007-2008, manifest electoral irregularities point to a flawed electoral process, challenging once again assumptions about the maturity of Kenyan democracy and thus undermining the enormous hopes placed in Kenya’s 2010 constitution and national institutions. The legitimacy and reliability of the Independent Elections and Boundaries Commission (IEBC) to supervise the electoral process was called into question at a very early stage: delays in organising the election, hiring Election Day workers and determining the number of polling places eroded its credibility; technical malfunctions and multiple pressures in the form of intimidation and non-compliance with the rules were observed;5 and after the vote, the ballot-counting and compiling phase was marred by lack of transparency and inefficiency. The declaration of the results on 9 March 2013,6 and their validation on 30 March by the Supreme Court following a ballot recount for 22 polling places, did not convince the Kenyan citizenry of the integrity of the electoral process. Political strategies and manoeuvres adopted by political parties and politicians to garner votes, such as intimidation and vote-buying, were also crucial in assessing the solidity and reliability of national electoral institutions such as the IEBC and of the electoral process in general.

16Lastly, a topic that cut across the reflections of the research team was the ability of Kenyan society to prompt democratic political change. Central to this concern is the state’s ability to perform the tasks assigned to it – the last election showed that unpreparedness and urgency were the rule – and to counter manipulation or strategies of manipulation. It is crucial to avoid “diabolic causalities” (Rémond, 1984) or fears of plots and conspiracies, as they may skew the interpretation of electoral results and distort the picture of the political situation. It is also essential to question the capabilities of the elite before, during and after the elections. The elite may be able to mobilise through “democratic” persuasion – notably through the media – and thus offset the use of violence or the ethnic vote, which are no longer enough to win. Such new strategies work well in countries like Kenya where media coverage is high, while lack of control over electoral campaign funds enables their use. This means, however, that financial patronage remains at the centre of the post-electoral political configuration. The role of religion, as a belief but also as a political resource, is also one among such political strategies for gaining legitimacy and holding power.

17This concern intersects with the question of ethnic-driven voting behaviour and ethnic manipulation by the political class, which brings the researcher back to the debate on moral ethnicity (Berman and Lonsdale, 1992) in Kenya. It also relates to new social and economic dynamics, particularly the emergence of a Kenyan middle class that votes less and less along ethnic lines, but instead to protect its class interests, or to the position of women in Kenyan society set between domination and agency. The difficult position of marginalised communities and minorities whose voice remains muted or silenced at the national level reveals how voting is not simply a result of belonging, but is essentially the outcome of social, economic and political interests. And finally, further inquiry is needed regarding the less visible mechanisms used to reproduce and renew the political elite in Kenya; among them the revival of past loyalties based on family networks and transmission, which could be termed “dynastic”. The emerging aspect of coalition building among ethnic groups, especially for or against the ICC, is also highlighted.

18Questioning whether a given election should be viewed as a decisive turning point in a country’s electoral history, or on the contrary reveals the resilience of voting patterns and electoral functioning, is central to analysing its current political dynamics and foreshadowing its political trajectory. Undeniably, such questioning is highly relevant in the case of Kenya, where the possibilities for opening up a democratic space seem so great, but where, thus far, people’s hopes and expectations have seldom been fulfilled.



Konrad Adenauer Stiftung. ‘Summary Report on the Outlook of the Kenya Situation after the Election 2013’ (

Braudel, F. “History and the Social Science: The Longue Durée.” In On History (trans. Sarah Matthew): 25-54. Chicago: University of Chicago Press, 1982. Originally published in Annales ESC, no. 4 (Oct.-Dec. 1958): 725-753.

Cheesman, N., G. Lynch and J. Willis. “Democracy and Its Discontent: Understanding Kenya’s 2013 Elections.” Journal of Eastern African Studies 1, no. 8 (2014): 2-24.

Ferree, K. E., C.C. Gibson and J.D. Lang. “Voting behavior and electoral irregularities in Kenya’s 2013 Election.” Journal of Eastern African Studies 1, no. 8 (2014): 153-172.

Rémond, R. “Le siècle de la contingence?” Vingtième siècle 1, no. 1 (1984): 97-104.


1 This undertaking follows a similar initiative launched during the 2007 Kenyan election by the French political scientist Jérôme Lafargue in the framework of the IFRA programme “The Election Observatory in East Africa”, resulting in The General Election in Kenya, 2007 (2008).

2 Longue durée history is defined by French historian Fernand Braudel as “a history to be measured in centuries”, “the history of the long, even of the very long, time span” as opposed to the histoire événementielle approach (the history of events), concerned with a short time span, investigating the moment and the event, and characterised by the “headlong, dramatic, breathless rush of its narrative” (Braudel, 1958: 27).

3 The need to prevent violence was on everyone’s mind in 2012 and 2013, as the debates in the media and popular discussions reveal. The ethnically-targeted violence that broke out in 2007-2008 in various parts of Kenya after the announcement of President Kibaki’s electoral victory in December 2007, was directed first at Kikuyu supporters of Kibaki’s Party of National Unity (PNU) in the North Rift, then Luo and other allies of Raila Odinga’s Orange Democratic Movement (ODM) in the Central Rift. It had been a national trauma, leaving more than 1,100 dead, 300,000 injured and about 600,000 forcibly displaced.

4 Some research and case studies not covered in this book are still in progress in individual doctoral research, such as that of Joyce Kaguta, Christine Adongo, or Charles Khamala.

5 In the end, the electoral administration reached nearly 80% of its voter target. On the other hand, 44% of one voter age group remained excluded from the voter registration list (a significant unregistered population). This group corresponded to spaces and societies located in peripheral pastoral and rural marginalised areas with the highest poverty levels and the lowest human development indicators (including young potential voters and women).

6 While most observers expected a run-off between Raila Odinga (Cord) and Uhuru Kenyatta (Jubilee), Kenyatta was declared the winner of the presidential election with 6,173,422 votes (50.07%). Odinga received only 5,340,546 votes, about 830,000 fewer.

© Africae, 2014

Conditions d’utilisation :


Volume papier
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search