The battle for Central: ethnicity, urbanization and citizenship in Kenya’s 2022 general elections

ABSTRACT

In Kenya’s 2022 general elections, presidential candidate William Ruto did something unusual – he dominated the vote in the heartlands of one of his main political rivals, President Uhuru Kenyatta. In turn, Ruto’s ability to win power by mobilizing support outside of his own Kalenjin community has led to speculation that the political power of ethnicity is waning. Drawing on surveys, economic and demographic analysis and interviews during the campaign in Central Kenya – a predominantly Kikuyu area and Kenyatta’s supposed political homeland – we cast doubt on this argument. Evidence from the campaign trail suggests that while structural changes within Kenyan society, including the emergence of a more urban, educated and critical electorate, played an important role in Ruto’s electoral success, it was also underpinned by historic ethnic allegiances, inter-group prejudice, and the use of ethnically-rooted clientelist strategies to build political networks. In particular, past campaigns saw Odinga depicted as an unsuitable leader based on his history, personality, and Luo ethnicity, generating formidable barriers to winning support. The implications of this argument extend beyond our understanding of Kenya to recent debates about the impact of urbanization and political socialization on political subjectivities and behaviour in Africa.

KEYWORDS:

• Kenya
• elections
• ethnicity
• citizenship
• parties
• urbanization

In Kenya’s 2022 general elections presidential candidate William Ruto did something that seemed impossible just a decade earlier – he dominated the vote in the heartlands of his rival, President Uhuru Kenyatta. Ruto’s ability to mobilize support across ethnic lines, persuading a majority of the predominantly Kikuyu electorate of Central Kenya – the densely-populated area immediately north of Nairobi and west of Mount Kenya that was Kenya’s Central province until the 2010 constitution – to throw their weight behind a Kalenjin presidential candidate, was particularly impressive for four main reasons. First, Kenya has a reputation for ethnic politics,¹ and Kenyatta’s bailiwick among the Kikuyu community had voted in lockstep in the 2013 and 2017 general elections. Second, Ruto was accused of committing crimes against humanity by the International Criminal Court (ICC) for orchestrating attacks on the very same Kikuyu community during post-election violence in 2007/8. Third, political competition was particularly fierce in Central Kenya due to the high number of registered voters and its status as a key election battleground. Finally, although he had previously promised to support Ruto’s presidential ambitions, President Kenyatta turned against his former ally and deputy, going out of his way to claim that Ruto could not be trusted.

Ruto’s ability to overcome these barriers and outmaneuver Kenyatta’s favoured candidate, long-time opposition leader Raila Odinga, has led some analysts to speculate that 2022 was the country’s first “post-ethnic” election.² Others have stopped short of this claim, but have nonetheless concluded that the outcome of the polls reflects the waning power of ethnicity in the face of urbanization and an increasingly educated and sophisticated electorate. Mary Kinoti and James Mwita, for example, forecast ‘a class-based contest along the lines of narratives about the “haves and have nots”’.³ Hervé Maupeu has argued that ‘In 2022, we did not see the usual ethnic polarization’, as ‘Kenya’s multi-ethnic democracy took on a new “face”’.⁴ Meanwhile, Peter Lockwood’s analysis highlights a ‘backlash’ in the Kiambu region as voters sought ‘to punish Uhuru for his brazen attempts to court votes in the region, against the backdrop of economic malaise’, and Ken Opalo suggests that in the future key ‘enabling economic and institutional factors may continue to undermine the role of ethnicity in Kenyan politics’.⁵ This article evaluates these claims on the basis of a combination of interviews, survey data, media reports and economic and demographic data – as set out in greater detail below.

We argue that taken together, this evidence suggests that structural changes within Kenyan society played an important role in Ruto’s electoral success, but also casts doubt on the idea that Kenya’s 2022 elections represented a dramatic break from past practice. We agree with Maupeu, for example, that the combination of Ruto’s focus on hustlers and economic difficulties led to a rise in prominence of “cadets sociaux” (social underdogs) both as candidates and as a more assertive bloc within the electorate. But we also argue that Ruto’s success depended on four main factors, many of which were rooted in ethnic politics. More specifically, the first factor we discuss – namely the emergence of a more urban, educated and critical electorate increasingly aware of the damaging impact of corruption – supports the idea that the last decade has witnessed significant changes in Kenyan political subjectivities. Three other factors were also important, however, and tell a different story. First, as Lockwood has argued, following John Lonsdale and Nic Cheeseman et al.,⁶ one reason that Kikuyu voters in areas such as Kiambu were so disappointed in Kenyatta was that they expected to benefit to a far greater extent from his time in office. Understood appropriately, therefore, demands for better government in Central appealed to a “patrimonial” register of virtue just as much as they did a “civic” one.⁷ Second, Ruto made extensive use of ethnically constituted clientelistic structures, which provided a “Kikuyu” face to his campaign in Central that was critical to the growth of his support in the region.

Third, a history of Kikuyu leaders “decampaigning” Odinga through personal attacks had encouraged voters to see him as an unsuitable leader based on his history, his personality, and his Luo ethnicity, generating formidable barriers to winning support in Central Kenya. Thus, while Kikuyu voters may have voted across ethnic lines, this does not necessarily mean that some people did not vote – at least in part – “ethnically”. Thus, while we agree with Lockwood that some voters were more “against Kenyatta” than they were “for Ruto”, we suggest that this analysis needs to be nuanced to reflect the reality that in at least some cases people were less against Kenyatta than they were against Odinga. Our findings therefore suggest that while the forces that shape Kenyan politics are gradually changing, ethnic politics played a critical role in the battle for Central. Thus, while we agree with Opalo that ‘Structural economic changes (e.g. from subsistence agriculture to trade) and urbanization may loosen voters’ ties to ethnic groups’, we are more skeptical about the prospects for the emergence of ‘strong inter-generational non-ethnic parties’,⁸ at least in the next decade.

We develop this argument in four stages. The first part of the paper introduces the sources we draw on. The second then surveys the political economy of Central Kenya, before providing a brief overview of the campaign. After this, the next three sections review, in turn, the factors identified above. The conclusion then reflects on the critique of Kenyatta’s presidency that emerged in Central, and what it means for Kenya’s political future. On the one hand, we identify the emergence of a nascent form of the “insurgent citizenship” that James Holston describes in the urban peripheries of São Paulo, Brazil.⁹ As Holston argues, the combination of urbanization and democratization with a political settlement that is relatively inclusive in terms of national membership but highly inegalitarian in the distribution of rights and wealth in practice, can generate “productive encounters” in which citizens join forces to challenge the status quo. Kenyan society may thus be changing in a way that will make it more likely that inclusive and accountable forms of “moral ethnicity”¹⁰ will overcome the divisive impact of “political tribalism”, as previously suggested by both Jacqueline Klopp and Stephen Orvis.¹¹

On the other hand, we echo the recent literature on urban politics, which has stressed that neither urbanization nor the expansion of the middle class will necessarily undermine existing political logics.¹² Instead, clientelism, corruption, and the politics of belonging can survive and even flourish in more urban and educated environments.¹³ As Karuti Kanyinga has argued, aspects of ethnicity, patronage and class have long intertwined in Kenya’s complex political arena.¹⁴ Further urbanization, education, and political socialization into democratic practices – coupled with the transfer of power in 2022 – have the potential to strengthen demands for accountability. But ethnic politics continues to be a critical factor in Kenyan politics, and may return as a dominant consideration.

Studying the moral economy of elections

The data collected for this article include a wide range of sources including nationally representative survey data and analysis of demographic and economic trends. Unless otherwise stated, the surveys we report were nationally representative and conducted face-to-face in the language of the respondent’s choosing by respected organizations including the Afrobarometer and IPSOS Kenya. We also draw on interviews and observations from the start of the campaign up to and including election day itself undertaken by both authors. While Cheeseman was present in Kenya on five occasions during the elections, including for the two weeks around polling day, Kamencu lives in Kenya and so was present throughout the campaign. Both authors attended rallies conducted by William Ruto in Central Kenya during this period, including both larger set-piece rallies and smaller “road shows”. Where an interview is cited, the conversation was conducted under formal conditions, often in a café or office. As far as possible we attempted to engage with a broad cross-section of activists and party leaders across the political divide, from Secretary Generals down to grass roots activists. In almost all cases, however, interviewees requested full anonymity due to the sensitivity of the topic, and so their names are not included here.

We also report more informal conversations with some of the individuals we met at rallies. While all of these individuals understood that we were present to conduct research on the election campaign, they were usually more informal exchanges. In some cases, the conversation took place in English so that both authors could understand. In other cases, people addressed Kamencu in Kikuyu, so that Cheeseman was effectively excluded. In some of these occasions, interlocutors suggested motivations – including ethnically based attitudes – that they had not mentioned previously. We discuss these different contexts, what these conversations revealed, and why people chose to engage with us in different ways in greater detail below.

We report both types of conversations because in all cases it was made clear to interviewees that we were conducting academic research on the motivations individuals had to vote for Ruto (or Odinga), and interlocutors were therefore fully aware of the use to which the information that was collected would be put. Our intention in sharing these conversations is not to judge those we interviewed, but rather to illuminate the ways in which some of the motivations for the pro-Ruto vote in Central may be hidden from external researchers, especially those largely conducting formal interviews. At the same time, we recognize that individuals are motivated by a wide range of factors, and that the conversations we experienced cannot be taken to be representative of the wider universe of Kenyan voters. Our interviewees tended to be younger males, and so we must be particularly careful of generalizing these findings to women. For this reason, we blend interviews and insights gained from attendance at rallies with survey data and other sources to ensure that our evidence base is robust.

The politics of Central Kenya

Central Kenya consists of Kiambu, Murang’a, Kirinyaga, Nyandarua and Nyeri counties, whose voters make up more than one-fifth of the national electoral register. The region is widely seen to be Kenya’s seat of power and wealth, having produced three of the country’s four presidents prior to the 2022 polls. The communities that live in Central are generally perceived to be effective at promoting their own economic interests, through organizations such as the Gikuyu, Embu and Meru Association (GEMA), which sought to spur development in the 1970s.¹⁵ This narrative is backed up by demographic data, which show that Nyeri and Kiambu have the highest life expectancies in the country – more than fourteen years longer than the worst performing county.¹⁶ It overlooks, however, deep inequalities and divisions within Central Kenya that stretch back to the colonial period.¹⁷ Although Mau Mau’s resistance is often remembered in terms of nationalism, on the ground it was also fought out within an increasingly stratified Kikuyu community, leading some scholars to posit that it had similar features to a civil war.¹⁸

Those divisions did not heal post-independence. Instead, President Jomo Kenyatta’s hostility to land redistribution led to continued tensions between landed and landless individuals, and between wealthier areas such as Kiambu and poorer areas such as Murang’a.¹⁹ This has led to a complex set of dynamics in which both economic status and region plays a role. Following the reintroduction of multi-party elections of 1992, for example, Mwai Kibaki was also seen to represent the interests of wealthy Kikuyu, while Kenneth Matiba was extremely popular with a poorer demographic, but there were also pronounced regional voting patterns that did not always map onto economic variations.²⁰ These divisions – as well as Oginga Odinga, Raila’s father – enabled President Daniel arap Moi to win against a fragmented opposition in 1992 and 1997.²¹ The “bloc voting” of 2013 and 2017, when well over 90% of Central Kenyan voters cast ballots for Uhuru Kenyatta,²² should therefore not be understood as the norm.

Kenyatta’s ability to temporarily unify the vote in Central owed much to the specific context of the 2013 elections. The 2007 elections had revolved around accusations of ethnic favouritism; the violence that followed the polls encouraged the sense that the Kikuyu community was under attack and allowed Kenyatta to recreate himself as the unlikely saviour of Central Kenya. Having previously been humiliated at the ballot box in 2002 due to the perception he had mainly been chosen to be Moi’s successor to protect his interests,²³ Kenyatta’s reputation as an authentic community leader was rehabilitated by the belief – ironically fuelled by the ICC proceedings – that he had deployed the notorious Mungiki gang to protect Kikuyus and carry out revenge attacks during the clashes that followed the 2007 elections.²⁴

Again, however, this period of unity was short-lived. Once the ICC proceedings collapsed, internal dissent and competition began to re-assert itself. As Opalo has argued, two factors that have contributed to this process are economic and institutional change: ‘economic shifts away from subsistence agriculture and urbanization are weakening attachments to ethnic groups. Devolution and statutory allocations of funds for legislators’ constituency service are eroding the powers of ethnic kingpins while also reducing the stakes of presidential elections’.²⁵ The 2010 constitution also impacted on the everyday practice of politics in other ways, most notably by creating 47 directly elected governors with greater power and resources than members of parliament. In turn, this facilitated new opportunities for competition, both between governors and with the county assemblies that were elected alongside them. The second governor of Kiambu Ferdinand Waititu, for example, rose to prominence on the back of his populist appeal only to be impeached by the assembly and succeeded by his deputy.²⁶ The impact of devolution, and Waititu’s rapid rise and fall, provide further evidence of just how quickly political fortunes can change within Central Kenya, and the challenges that leaders have had in maintaining the loyalty of its voters.²⁷

The 2022 campaign in Central Kenya

Even before the presidential campaign officially began, Ruto and his allies had begun to tour the country emphasizing their “bottom-up” message. This included distributing the wheelbarrows that would become the emblem of Ruto’s United Democratic Alliance (UDA), symbolizing the idea that a Ruto government would reward hard working “hustlers”.²⁸ During rallies in Central Kenya, Ruto highlighted the unusually high level of economic difficulty and unemployment in one of Kenya’s wealthiest areas, stating that he was shocked it had happened under a Kikuyu-led government. The implicit message was clear: Kenyatta had failed his own community, and so his preferred candidate could not be trusted to look after Kikuyu interests. Ruto, by contrast, pledged that he would support those who had been overlooked, boosting jobs and making loans available to “hustlers” on favourable – though often unspecified – terms.

This was not the first time a Kenyan leader has mobilized a message along these lines. Indeed, Opalo has identified²⁹ a number of forerunners to Ruto's 2022 campaign. These include J.M. Kariuki who railed against the growing inequality of post-colonial Kenya; Mike Sonko, the charismatic former Governor of Nairobi, who gained a nationwide profile through his carefully orchestrated support to those who had been let down by the state; and Waititu, briefly introduced above. Ruto was therefore not building on virgin ground. What was different about his campaign, however, was the way in which such sentiments were connected to a nationwide mobilization effort.

During rallies, Ruto’s message was typically delivered in a sober manner, with multiple references to facts and figures, but only after a populist narrative designed to warm-up the crowd. The first phase of his performance often revealed how much Ruto has learned about Kenyan audiences over the years. It also demonstrated his recognition that it was important to entertain as much as campaign. At a rally in Nyeri in February 2022, for example, well-known musicians such as the Kikuyu singer Muigai wa Njoroge performed and were given a prominent place in the seating arrangements. This served a dual purpose, engaging the audience and generating the sense that Kenya Kwanza would create economic opportunities for artists and other aspiring entrepreneurs.

For its part, the Odinga campaign did not always appear to have a clear message. Early on, Odinga emphasized economic development and unity, harking back to the nationalist era. This may have been a ploy to make him appear statesman-like, but it played into concerns that Odinga was too old to understand the needs of the youth. It was also a message that echoed the themes of Kenyatta’s campaigns in 2013 and 2017. This continuity – and Odinga’s inability to be highly critical of Kenyatta now that the two men were allies – made it harder for the Luo leader to present himself as the change candidate that many voters desired.³⁰

As we discuss below, both Ruto and Odinga relied heavily on Kikuyu allies and intermediaries to make their case in Central. The challenge that this raised for Odinga was that the most prominent Kikuyu figures in his camp was Kenyatta, which only served to remind voters that an Odinga presidency would likely adopt similar policies to the sitting – unpopular – government. Although Odinga appeared to be boosted by the announcement that his running mate would be Martha Karua, the fiercely independent Kikuyu leader and former Minister of Justice,³¹ some of the highest profile moments of the Azimio campaign were set piece – and much hyped – speeches by Kenyatta. It is unclear exactly why Azimio pursued this strategy. It may be that those leading the campaign were unaware of the extent to which deploying the president in Central was a double-edged sword. They may also have recognized the need to explain why Kenyatta had turned against Ruto, who had served him loyally in 2013 and 2017, which was certainly an issue on the ground.

As Ngala Chome and Justin Willis argue in this cluster, many voters believed that the president owed Ruto a political debt. Yet Kenyatta’s appearances failed to provide the “smoking gun” that would explain his change of heart. In an address at Sagana State Lodge in Nyeri County on 23 February 2022, for example, Kenyatta appeared to suggest that the main reason for his rift with Ruto was that he was corrupt, while making a number of attacks on the character of his former Deputy. The problem with this strategy was that corruption accusations against Ruto go back to his time as Agriculture Minister (2008-2010),³² which meant that Kenyatta already knew of the issue before selecting him as a running mate in 2013 and 2017. Conversations with young male matatu drivers and young female passengers on a route from Nyeri to Nairobi four days after the address, for example, suggested that at least some Kikuyu voters did not believe Kenyatta was telling them the whole truth.

In response, Kenya Kwanza coalition leaders framed Kenyatta’s demand that the Kikuyu community back Odinga as an illegitimate imposition. During a rally at the Asian Quarter grounds, Nyeri town, UDA politicians asked the crowd whether they were planning to make their own minds up, or have them determined by a retiring politician. At various points during the rally, the crowd responded Hatupangwingi (our choices will not be dictated). Kenya Kwanza leaders also went to great lengths to create the impression that it was the “handshake” deal that saw Kenyatta and Odinga bury their difference and join political forces after the disputed 2013 elections – rather than rising prices due to the war in Ukraine – that was the source of Kenyans’ economic woes. In one speech we observed, for example, Ruto consistently quoted the price of key items before and after the handshake.

Ruto’s ability to communicate in this way owed much to his greater physical presence in Central Kenya during the campaign. Appearing numerous times, both at major rallies and in smaller “drive by” events – where he would speak by standing up through a car sunroof – Ruto’s presence backed up one of his main promises to the people of Central – namely, that he would work hard for their vote and work hard for them once elected. By contrast, Odinga made far fewer personal appearances, and a major rally scheduled for Nyeri town was cancelled at the last minute. In his absence, the campaign relied heavily on other leaders such as Karua. It is unclear whether this reflected Odinga’s advanced age, or was a deliberate strategy to allow Karua to make the most of her regional appeal, but it seems to have created the unhelpful impression that Odinga believed he could “inherit” the Kikuyu vote rather than seeking to earn it. As Portia Roelofs has written, voters in African countries with first-past-the-post electoral systems tend to place a high premium on the ‘accessibility’ of candidates, and Odinga did little to make himself accessible in 2022.³³

In this way, questions of honesty, morality and appropriate leadership were interwoven throughout the campaign, reflecting the complex moral economy of elections recently described by Cheeseman et al.³⁴ One manifestation of this was the attempt of both leaders to curry support with religious leaders and, through them, their supporters. Kenya Kwanza rallies, for example, regularly began with Christian chants. Ruto frequently quoted Psalms 113: 7–8, which speaks of the Lord’s elevation of the poor to sit ‘on the same table as the rich and the rulers’, to legitimize his “bottom up” economic model.³⁵ He backed up these words with large donations, which enabled him to appear as a strident ally of the church. Azimio responded by claiming that Ruto’s donations were immoral and should not be accepted, but this appears to have largely fallen on deaf ears. One reason for this was that Ruto’s greater visibility in church settings leant credulity to his positioning. Another was that previous efforts by leaders in Central to damage Odinga’s credibility had called into question his own religious faith, undermining Azimio’s counter-narrative. As we explain in greater detail below, this highlights the way in which past “decampaigning” of Odinga paved the way for Ruto's success, and made it easier for him to persuade voters to buy into his campaign messages, whether religious or economic.

The social contract and insurgent citizenship

A combination of economic data, survey responses, media coverage and recent literature suggest that Kenya’s economic difficulties shaped political behaviour in 2022 – and that the effect in Central was particularly strong due to higher levels of urbanization, a greater dependence on wage labour, and a history of inequality and internal competition. Ahead of the elections, economic conditions deteriorated due to the combined impact of COVID-19 lockdowns and rising food and fuel prices. In 2020 the economy contracted by 1–1.5%, while the unemployment rate increased significantly.³⁶ A national survey by the Kenya National Commission on Human Rights of 2400 members of vulnerable groups – including youth and the urban poor – found that the proportion of individuals reporting that they were employed fell from 80% before March 2020 to just 34% a year later, and that 38% had been evicted from their homes.³⁷ Very few of these individuals believed they had been helped by the government, with 96% stating they had received no cash benefits.

There are also a number of strong indicators that these trends had a profound impact on Kenyan voters. According to the nationally representative survey data collected by the Afrobarometer,³⁸ there has been a dramatic increase in the proportion of citizens going without basic essentials – food, medicine, a cash income. Nowhere was this trend more pronounced than in Central Kenya, which witnessed the biggest decline in the proportion of respondents who said that neither they nor a family member had been forced to go without a cash income in the previous 12 months: from 58% in 2014 to just 8% in November 2021. Although this trend was reflected elsewhere, the rate of decline was far greater in Central than in the Nyanza (4%) or Rift Valley (8%) regions. One explanation of this distinctive shift is that economic conditions were generally better in Central Kenya ahead of the pandemic, and so the decline has been felt more keenly. Another, as Opalo has pointed out, is that the combination of high population density and comparatively limited dependence on agriculture meant that residents of Central were particularly vulnerable to the loss of jobs in the service and entertainment sectors.³⁹ In 2019, the population density of the counties in Central Kenya was 417 people per square kilometer, far above the national average of 82 people per square kilometer, making it one of the most urbanized parts of the country.

There is also strong evidence that changing economic conditions had a direct impact on perceptions of the government and in particular Kenyatta. The Afrobarometer data reveals that in 2021 a remarkable 83% of Kenyans believed that the government was handling the economy “badly” or “very badly”.⁴⁰ The same year, the share of Kenyans who said that they trusted the president “a lot” hit a record low, falling from 43% in 2014, shortly after Kenyatta was first elected, to just 22%. This reflects a history in which Kenyan voters have regularly held elected representatives accountable for poor performance. While presidential elections have delivered more change than continuity, at the sub-national level there has been a notoriously high rate of turnover,⁴¹ with constituency-based elections understood as referenda on the development performance of the sitting MP.⁴² Two decades of multi-party politics, and the introduction of free primary education in 2003, has strengthened this impulse, despite regular election controversies.⁴³

These processes have been particularly pronounced in Central Kenya, where access to information is especially high, in part due to its proximity to the capital. According to the Kenya National Bureau of Statistics (KNBS), Kiambu (88%), Kirinyaga (87%), and Murang’a (86%) have some of the highest literacy rates in the country.⁴⁴ Along with the history of internal political contestation described above, this led to a particularly fierce critique of government. Perhaps most strikingly, Kikuyu respondents to the latest Afrobarometer survey were 6% more likely than the national average (59% against 53%) to say that the government was performing “very badly”. This is particularly notable given that at this point the Kikuyu community had “one of their own” in State House, and survey analysis has consistently found that both co-ethnicity and affinity with the party in power shape perceptions of government performance.⁴⁵ In the same survey, Kikuyus were also the second most likely ethnic group to say that they would vote for UDA (68%), behind only the Kalenjin (83%).

Differentiating the specific attitudes of the residents of Central Kenya from the broader Kikuyu community is challenging because nationally representative surveys have relatively small sample sizes at the regional level. It is nonetheless notable that the drop in trust in Central Kenya seems to have been greater than among the wider Kikuyu community, which includes Kikuyus normally resident in other parts of the country. While 18% of all Kikuyu respondents reported trusting Kenyatta “not at all” in 2022, this figure was 25% among those living in the counties of Central. As Lockwood has shown,⁴⁶ because Kenyatta’s co-ethnics expected more of him than individuals from other communities, it was in Central that economic hardship was most likely to turn into moral discontent. Perhaps the best example of this is attitudes towards government corruption. When Kenyatta was elected back in 2013, 21% of Afrobarometer respondents in Central Kenya said that none of the people in the president’s office were involved in corruption, twice the national average. By 2021, this figure had fallen to just 7%  – a lower figure than among other communities such as the Luo of Nyanza and the Kalenjin of Rift Valley.

In order to exploit these nascent tensions, Kenya Kwanza targeted lower-income voters and attacked the wealth of the Kenyatta family. It did the former by holding several “economic forums” in counties such as Murang’a, which served to give a voice to economically disaffected Kikuyu. In turn, Ruto-aligned politicians incorporated some of these complaints in their campaigns, alongside powerful critiques of economic exploitation. MPs Kimani Ichungwa, Mohamed Ali and Ndindi Nyoro, for example, attributed the plight of the landless Mau Mau fighters after independence to the Kenyatta family, castigating the president for enriching himself at the expense of others. The power of such statements was often amplified through references to issues known to resonate with voters. This included the Kenyatta family’s Brookside dairy business, which was accused of exploiting farmers.

In making this critique, figures such as Ali and Nyoro drew on the history of “class consciousness” in the region,⁴⁷ and the long-running tensions identified above. As Ndungu Gachane has argued, the early antagonism between landless and landed Kikuyus ‘set the scene for future dissent within the ethnic group’.⁴⁸ In 2023 the tensions between “haves” and “have nots” was further exacerbated by a Kenya Revenue Authority operation against counterfeit goods, which targeted the Nyamakima area in Nairobi’s Central Business District. Although no precise figures exist, Dauti Kahura argues that the confiscation of billions of shillings worth of goods was especially damaging to Kikuyu traders, who had previously helped to mobilize finances and support for Kenyatta’s election campaign.⁴⁹ A particular cause of chagrin was a public event at which Kenyatta personally oversaw the destruction of KSh 1.5 billion worth of goods – after which it was impossible for him to distance himself from the policy.

The sense of betrayal this gave rise to was epitomized by the song Mbari ya Kimendero⁵⁰ (a clan of oppressors) by Muigai wa Njoroge, a Kikuyu musician. In the song, the diminishing fortunes of traders and farmers are juxtaposed with the great wealth of Kenyatta and Odinga, who are said to be too ‘busy shaking hands’ to prevent the suffering of the workers. The extent to which different forms of economic exploitation were recognized and politicized in 2022 suggests a similar process to one identified by James Holston in Brazil, in which the twin processes of democratization and urbanization led to ‘productive encounters’,⁵¹ as citizens mobilized to assert and defend their rights in highly unequal and predatory settings. Holston saw these clashes as productive because they ‘entail conflicts of alternative formulations of citizenship’, and so ‘sites of metropolitan innovation often emerge at the very sites of metropolitan degradation’.⁵² The growing grass-roots anger at a broad range of forms of economic exploitation – from land alienation through to plutocratic economics – suggest that a similar process may be underway. As Maupeu has argued, the combination of these conditions enabled ‘The very conservative William Ruto’ to present himself as acting in ‘defense of the oppressed’. Meanwhile, the rise of a number of ‘social underdogs’ as viable candidates in sub-national contests ‘reveals affective dynamics for challenging a political culture that resembles a patriarchal gerontocracy based on neo-traditional beliefs’.⁵³

Yet it is also important not to exaggerate the extent to which the 2022 general elections saw new subjectivities come to the fore. The extent to which urban identities or class solidarities were explicitly identified as a reason for mobilizing against Kenyatta or for Ruto was limited. Having initially played heavily on his “anti-dynasty” critique, Ruto subsequently pulled back, making it clear that he was no socialist – a term that has been tarnished in Kenya since the sustained attack on left-leaning politicians in the late 1960s.⁵⁴ Kanyinga’s caveat in his discussion of class in Kiambu in the 1980s is therefore still relevant: while issues of economic status were a continual source of tension, ‘they were complicated by issues of sub-ethnic identity and never expressed in terms of a class project’ (emphasis in the original).⁵⁵

Moreover, as Lockwood has argued, there are also other reasons to be skeptical about the idea that “ethnic” logics were replaced by “civic” ones in 2022.⁵⁶ One reason popular frustration with Kenyatta was so high in Central was his failure to comply with a widely held conception of good leadership within the Kikuyu community. As Lonsdale has argued, the sense of what it means to be a good leader ‘springs from below’,⁵⁷ rooted in a distinctive moral ethnicity that evolved out of years of debates about virtue and leadership within the Kikuyu community.⁵⁸ The particular moral ethnicity that Londsale describes is not underpinned by egalitarian notions that help should be given to the poor in general, but rather that political leaders and those with wealth are responsible to the ‘poor to whom one is related or whose service one needs’.⁵⁹ This highlights both the strengths and limitations of moral ethnicity in terms of democratic consolidation. The strength is that its organic emergence within a community over a long period of time gives it considerable resilience, and the potential to constrain the worst excesses of patrimonialism. The weakness – at least from the point of view of the emergence of a form of “non ethnic” politics that might be better places to sustain development and democracy – is that it is rooted in a political subjectivity that assumes that leaders have more of a responsibility to deliver to their own ethnic group than to others.⁶⁰

While recognizing the way the in which moral economies continue to evolve, we should therefore be careful not to lose sight of what Cheeseman has called the ‘dark side’ of moral ethnicity.⁶¹ As Lonsdale notes, because they both depend on ethnically rooted allegiances, moral ethnicity represents the other side of the coin to the ‘political tribalism of invented nationality which may be manipulated from above’.⁶² So long as moral ethnicity sustains the moral power and political significance of communal identities, any new forms of politics it gives rise to will remain vulnerable to subversion from leaders willing to exacerbate inter-ethnic tensions for their own advantage.⁶³ This is particularly significant because, as the next two sections document, the 2022 demonstrates that, in Noah Nathan’s words, urbanization and the expansion of education do not always lead to a ‘transition away from ethnic competition and towards more programmatic elections’.⁶⁴

Clientelism and ethnicity

There are three main ways in which intra- and inter-ethnic considerations played into the 2022 election in Central: the co-option of influential Kikuyu leaders; the way that historic “decampaigning” of Odinga made him an unpalatable leader for many Central Kenyans; and the fact that a significant portion of the critique of Kenyatta was rooted in patrimonial, rather than civic, concerns.⁶⁵

Where alliance formation was concerned, Ruto avoided knitting together a set of “Big Men” in the way that, for example, Odinga had done when putting together the “Pentagon” of regional leaders in 2007. This was intentional, as it would have made it harder for Ruto to present himself as a change candidate, and left him vulnerable to defections. Instead, Ruto focused on forging alliances with lower-level figures.⁶⁶ In Central Kenya, this meant sponsoring the campaigns of a slate of candidates ahead of the 2017 general elections. Although the precise timing that allegiances were formed is hard to track, this is widely believed to have included figures at multiple levels. Geoffrey Gachagua, elected as Member of Parliament for Mathira Constituency in 2017, and Kiharu MP Ndindi Nyoro, an avowed beneficiary of Ruto’s patronage, were early allies.⁶⁷ He was also careful to reach out to a host of lower profile figures, such as Gathoni Wambuchomba, women representative for Kiambu, and Kiruga Thuku, ‘a local assembly representative from Nyeri County’, who subsequently ‘performed a cover song (in native Kikuyu) in support of Ruto at a meeting of county assemblies of Mount Kenya’.⁶⁸ The breadth and depth of this network enabled Ruto to maintain his access – and channel funds and gifts in an election marked by widespread ‘vote buying’⁶⁹ – to the region after he was marginalized within the ruling party. Moreover, Ruto’s capacity to maintain these alliances into 2022 ensured that the face of his campaign in Central on a day-to-day basis was as much Kikuyu as it was Kalenjin.

In this regard, Ruto’s selection of Rigathi Gachagua as his running mate was particularly significant. Gachagua appears to have been chosen – despite numerous corruption accusations that threatened to taint the UDA campaign – for three main reasons. First, his personal wealth meant he could help to fund UDA during a particularly expensive campaign.⁷⁰ Second, his profile as an effective populist mobilizer with deep roots in the region was the perfect foil for Ruto’s campaign, echoing his own approach on the stump while enabling Ruto to tap into the influential networks of the Gachagua family. While Nderitu Gachagua, his late brother, served as the first governor of Nyeri country, his wife, Dorcas Wanjiku Rigathi, is a well-known pastor in the area. Finally, Gachagua’s personal history, having previously served as the personal assistant to Kenyatta, suggested he could be trusted to look after Kikuyu interests.

The significance of Gachagua’s support to Ruto’s victory in Central Kenya can be seen from the way that impromptu “roadside” rallies were structured. At one such rally we observed, Ruto spoke first, allowing Gachagua to appear as the “headline” act. While regional power brokers will often be given one of the most prominent roles at rallies in line with their greater local popularity, it is rare for a presidential candidate to position themselves as the warm-up act to someone who had only held the role of MP. This act of deference was a carefully calculated piece of political theatre of course, designed to fit with the theme of Ruto’s speeches in the region, which emphasized his humility as an outsider asking for the community’s support. It also reflected a wider recognition within UDA that Gachagua, as the most effective Kikuyu mobilizer in Kenya Kwanza, was essential to Ruto’s hopes in Central – and that Kikuyu voters were more likely to back UDA if it was clear that Gachagua would play a prominent role in any future government. Indeed, it is likely that many voters thought of themselves as voting for a Kalenjin/Kikuyu alliance, rather than simply casting a ballot for a Kalenjin president.

Political “decampaigning” and ethnic distrust

Ethnicity also underpinned Ruto’s campaign in other more problematic ways. Over the previous thirty years Kikuyu leaders had sought to ward off the challenge represented by Odinga – the most prominent rival of both Kibaki and Kenyatta – by demonizing him as a dangerous threat. This included claims that Odinga was an irresponsible leader who would trigger chaos and conflict. During the 2007 election campaign, for example, one of the authors was handed a leaflet by a Kibaki supporter during the Party of National Unity’s final rally at Uhuru Park which featured an image of Odinga’s head superimposed onto Idi Amin’s body. Kikuyu voters were also regularly warned that electing Odinga would disadvantage their community. Those elections saw a sustained questioning of Odinga’s competence, with slogans such as “Stop Raila, Save Kenya”.⁷¹ One reason these messages cut through was that they were often laced with, or foregrounded, anti-Luo sentiment. This ranged from claims that Luos were thugs who could not be trusted⁷² to the frequently repeated slur that as a member of a community that does not practice the rite of passage of circumcision, Odinga could not be considered a man, and hence had no right to lead others. As noted above, Odinga was also maligned as having either no religious faith or being a devil worshipper. In 2007, for example, his “pentagon” of regional leaders was depicted in pamphlets as a satanic sign.

It is difficult to ascertain how much of a role this campaign, and the ethnic prejudice that it was infused with, played in the election in Central Kenya. It is clear that prominent political leaders from the community believed such “attack ads” were effective, having previously invested heavily in them.⁷³ When interviewed, however, many Kikuyu voters were quick to express their distaste for ethnic politics and their willingness to consider leaders from any community – and some said they had considered voting for Odinga in previous elections. This may have naturally encouraged some researchers to overlook the significance of anti-Odinga chauvinism in the 2022 context.⁷⁴ Yet our experience on the campaign trail suggests a different story. When conversing with Kamencu and not speaking English – so that Cheeseman was effectively excluded from the discussion – a number of individuals at rallies in different parts of the region expressed their rejection of Odinga in ethnic or chauvinistic terms. The most plausible explanation of this is that these individuals (incorrectly) took Kamencu for a co-ethnic, and so felt comfortable sharing such opinions with him on the assumption that he would be inherently sympathetic. Such sentiments were never expressed to Cheeseman, who, as an outsider, was consistently given justifications that would not risk embarrassing the individual or their group, such as that Ruto would provide more employment. If this interpretation is correct, it seems likely that ethnic prejudices and historical “decampaigning” played a much greater role in Odinga’s defeat in Central Kenya than has hitherto been recognized.

A nationally representative survey conducted by IPSOS Kenya for South Consulting in August 2021 provides further evidence of the impact of decampaigning, although it is challenging to use this data to differentiate the impact of general attacks on Odinga’s character from those rooted in ethnic stereotypes. When asked to say how much they trusted Odinga, an absolute majority (51%) of those living in Central Kenya said “not at all” while only 9% said that they trusted him “a lot”. These figures were significantly different to the general pattern; for example, 20% of Nairobians and 17% of those in Ruto’s Rift Valley base reported “a lot” of trust in Odinga. In the same survey, respondents from Central Kenya identified Luos (15%) as the external group that “most people in your community do not trust”, compared to just 2% who identified the Kalenjin and 2% who identified Luhyas.⁷⁵ These results are far from definitive, but are consistent with the idea that decades of intense competition between Kikuyu and Luo leaders had generated particularly high barriers to Odinga's ability to mobilize support.

One reason that these issues remained pertinent in 2022 was that Kenya Kwanza leaders exploited the tropes previously used by Kenyatta and his allies to demonize Odinga. At a rally in Dada’s grounds in Limuru witnessed by one of the authors, for example, Ruto’s ally Kimani wa Ichungwa drew on historical tropes depicting Odinga as godless – while quoting him out of context – to claim that popular churches in the area would be shut down if the Luo leader was elected. Thus, while we agree with Lockwood’s argument that Kenyatta lost support in Central because he contravened local expectations of what it means to be a good leader,⁷⁶ too heavy a focus on this explanation risks underplaying the extent to which distrust of Odinga also undermined Azimio’s efforts. Indeed, it may well be the case that it was in part because he campaigned for Odinga that Kenyatta came to be seen to have betrayed his community. As a politically well-connected member of the Kikuyu community explained to us,⁷⁷ from whatever angle one looked at Kenyatta’s change of heart, there were reasons to be critical. Either Odinga was a good candidate, in which case Kenyatta – and other Kikuyu leaders – had been misleading the community for years. Or Odinga was a dangerous threat, in which case Kenyatta was selling out his own people by asking them to back him. Kenyatta’s political choices in 2022 thus called into question his moral authority as a community leader in a way that may not have occurred if he had identified a different successor.

Conclusion: moral ethnicity and the future of Kenyan politics

The evidence presented in this article paints a complex picture of the state of Kenyan politics. On the one hand, it provides some support for the argument that major structural changes to Kenya’s socio-economic landscape including urbanization, the expansion of education, and growing experience with multiparty politics, have generated profound changes to the political landscape. Kenyans have long valued the right to select their political representatives,⁷⁸ and there has been growing critique of leaders who fail to match expectations.⁷⁹ Against this backdrop, a combination of severe economic difficulties, the failure of the government to reduce corruption, and the perception that the political elite was colluding against citizen’s interests left the government vulnerable to defeat. Ruto cleverly played on these concerns by exaggerating his role as a political outsider, emphasizing his hustler narrative to question the legitimacy of political “dynasties”, and avoiding the kind of Big Men coalition building that had characterized previous campaigns.

These developments played out particularly strongly in Central Kenya, where high levels of education and population density, combined with low levels of subsistence agriculture, meant that voters were particularly responsive to Ruto’s critique. Although we caution against the idea that the 2022 polls saw a class-based form of mobilization, the campaign, and the greater role played by “cadets sociaux”,⁸⁰ suggest the emergence of nascent forms of the kind of insurgent citizenship documented by Holston in Brazil. In turn, this finding suggests that political leaders found it harder to use ‘political tribalism’ to mobilize support in 2022, and faced stronger demands for political accountability. This conclusion echoes prior scholarship, such as Dominic Burbidge’s account of the distinctive political subjectivities of middle-class Kikuyu voters during the 2013 elections.⁸¹

The article has also demonstrated, however, that it is far too early to conclude that the political power of ethnicity has entered a period of inevitable decline. Ruto’s ability to penetrate Central Kenya was underpinned by a process of alliance formation with influential Kikuyu leaders and ensured that Kikuyu voters felt strongly represented within his campaign. Criticism of Kenyatta was driven by economic difficulties and the emergence of stronger claims on citizenship, but in some cases was also rooted in a moral ethnicity that reifies, rather than challenges, the idea that leaders have a greater commitment to deliver to co-ethnics than to other citizens. At the same time, Odinga’s struggles in the region can be traced, in part, to the “decampaigning” he suffered over many years, which regularly invoked negative ethnic stereotypes. As a result, Karua and Kenyatta struggled to sell Odinga in Central. This story is not unique to Kenya, and similar processes have played out in a number of other African states where a leader of one ethnicity has been succeeded by a candidate of another.⁸² More broadly, it is also clear that many other communities – including the Kalenjin and Luo – did vote overwhelmingly for a co-ethnic candidate. These findings of continuity alongside change also echo recent scholarship. Cheeseman and Miles Larmer, for example, have argued that a particularly pronounced history of ethnic political mobilization, combined with the lack of a strong trade union movement to bring together voters across communal lines, has militated against the emergence of more inclusive kinds of populist leadership and instead give rise to a distinctive form of ‘ethno-populism’.⁸³

Given this complex picture, it is worth reflecting on what happened the last time that researchers began to predict that Kenya was moving beyond ‘political tribalism’.⁸⁴ The country’s first democratic transfer of power in 2002, combined with a spate of popular protests against abuses of power such as land grabbing, led to a vibrant debate about whether poorer Kenyans would increasingly recognize their common cause, undermining ties of ethnicity and patronage.⁸⁵ There were good reasons to expect this following the rise to power of a multi-ethnic government that promised constitutional reform and an end to corruption. Yet less than two years later the refusal of leaders to genuinely share power led to the fragmentation of that government along ethnic lines, setting the scene for one of the country’s most divisive elections in 2007. In turn, the controversy surrounding the flawed electoral process triggered a prolonged period of violence so destabilizing that some commentators feared it might fatally undermine the democratic project.⁸⁶

It is therefore important to keep in mind that now, as then, contingent factors can both exacerbate and reduce the salience ethnicity – and that this will depend as much on the strategies adopted by political leaders as it does on the structural changes that came to the fore during the 2022 campaign. Had Kenyatta supported Ruto’s candidacy, the elections of 2022 may well have mirrored those of 2017, and would have triggered little discussion of the declining power of ethnicity. Had Ruto not developed the idea of the “hustler nation” during a period of severe economic difficulties, and instead fallen back on past political languages such as “majimboism”,⁸⁷ issues of economic inequality would not have come so powerfully to the fore. Moreover, had Kenyatta backed a leader other than Odinga against Ruto, the Kalenjin leader’s victory among the Kikuyu of Central Kenya would most likely have been considerably less pronounced.

The significance of these contingent factors in the outcome of the 2022 polls should warn against teleological interpretations of Kenya’s political trajectory. We therefore agree with Opalo that Kenya’s political future is not certain, but will depend on ‘whether elite pacts result in strong inter-generational non-ethnic parties that offer opportunities for upward mobility regardless of ethnicity’. Where we differ with Opalo is perhaps on the likelihood of this happening, at least in the short to medium-term. If Ruto struggles to retain popular support amidst a challenging economic context, and opposition to the president coalesces around a Kikuyu candidate, his ability to control the vote in Central Kenya will be considerably reduced. Even if this does not happen, the fragmentation of the government that often occurs when presidents lose elections or stand down due to term limits could in time lead to a political realignment that would see a return to the bloc politics of 2017.⁸⁸ This suggests that ethnic sentiments will continue to influence electoral behaviour, and the manipulation of political tribalism could yet reverse not only the emergence of insurgent citizenship, but also the process of democratization itself.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Lynch, “Negotiating Ethnicity”; Mueller, “The Political Economy”.

2 Lucas Barasa, “Is This the End of Ethnic Groupings in Determining Who Becomes King?” The East African, 14 August 2022, https://www.theeastafrican.co.ke/tea/news/east-africa/is-this-the-end-of-ethnic-groupings-in-determining-president-3913988.

3 Kinoti and Mwita, “Elections and Stability,” 241.

4 Maupeu, “Les élections générales de 2022”.

5 Lockwood, “Hustler Populism,” 2; Opalo, “Hustlers vs. Dynasties,” 1.

6 Cheeseman et al., The Moral Economy of Elections; Lonsdale, “Moral and Political Argument”.

7 Cheeseman et al., The Moral Economy of Elections.

8 Opalo, “Hustlers vs. Dynasties,” 1, 26.

9 Holston, “Insurgent Citizenship”.

10 Lonsdale, “Moral and Political Argument”.

11 Klopp, “Can Moral Ethnicity Trump Political Tribalism?”; Orvis, “Moral Ethnicity”.

12 See: Agbiboa, They Eat Our Sweat; Nathan, Electoral Politics; Paller, Democracy in Ghana.

13 Agbiboa, They Eat Our Sweat; Cheeseman, “(Mis)Understanding Urban Africa”.

14 Kanyinga, “Ethnicity, Patronage and Class”.

15 Throup, “Elections and Political Legitimacy”.

16 The Standard, “Central Female’s Life Expectancy Highest,” 5 May 2022, https://www.standardmedia.co.ke/national/article/2001444743/if-you-are-male-in-kenya-you-are-likely-to-die-by-your-60th-birthday.

17 Branch, Defeating Mau Mau; Odhiambo and Lonsdale, Mau Mau & Nationhood.

18 Imperato, “Differing Perspectives on Mau Mau”.

19 Lamb, Peasant Politics.

20 Ogot, “Transitions from Single Party to Multiparty Political System.

21 Throup and Hornsby, Multi-Party Politics.

22 Electoral Geography, “Kenya: Presidential Election 2013,” n.d., https://www.electoralgeography.com/new/en/countries/k/kenya/kenya-presidential-election-2013.html.

23 Ndegwa, “Kenya: Third Time Lucky?”.

24 Lynch, “Electing the Alliance”.

25 Opalo, “Hustlers Versus Dynasties?” 25.

26 Mwakuni, “Kenya’s Gubernatorial Impeachments”.

27 Burbidge and Raji, “Central”.

28 Opalo, “Hustlers Versus Dynasties?”.

29 Ibid.

30 Al Jazeera, ‘Who Do Kenyans Want for Their Next President?” 25 April 2022, https://www.aljazeera.com/program/the-stream/2022/4/24/who-do-kenyans-want-for-their-next-president.

31 Oxford Analytica, “Running-Mate Choice Boosts Odinga’s Kenya Poll Hopes,” 23 May 2022, https://www.emerald.com/insight/content/doi/10.1108/OXAN-DB270406/full/html.

32 Reuters, “Kenya Suspends Ministers Over Graft Allegations,” 14 February 2020, https://www.reuters.com/article/us-kenya-corruption-idUSTRE61D24U20100214.

33 Roelofs, “Beyond Programmatic Versus Patrimonial Politics”.

34 Cheeseman et al., The Moral Economy.

35 Citizen TV Kenya, https://web.facebook.com/CitizenTVKe/videos/dp-ruto-bottom-up-ni-maneno-ya-bibilia-ukisoma-zaburi-1137-8-humwinua-fukara-kut/682589316319959/?_rdc=1&_rdr.

36 BTI, “BTI 2022 Country Report Kenya”.

37 Scovian Lillian, “‘The Pandemic Hit Kenya’s Economy Hard,” The Continent, 23 May 2021, https://m.facebook.com/democracyinafrica/photos/a.2113373265365412/4001305499905503/?type=3&paipv=0&eav=AfZcDxsuZnE4dqCyqmRkseZ03e1jORyqUGdI3wXjS_2bI4v7-Yq90kcAdRV61w21MM4&_rdr.

38 Afrobarometer, “Kenya Data,” 2021, https://www.afrobarometer.org/online-data-analysis/.

39 Opalo, “Hustlers Versus Dynasties?”

40 Afrobarometer, “Kenya Data”.

41 Opalo, Legislative Development in Africa.

42 Barkan, Legislative Power.

43 Throup, “Elections and Political Legitimacy”.

44 The Standard, “Turkana and Wajir Counties Have Highest Levels of Illiteracy,” n.d., https://www.standardmedia.co.ke/lifestyle/article/2000098680/turkana-and-wajir-counties-have-highest-levels-of-illiteracy.

45 Bratton and Kimenyi, “Voting in Kenya”.

46 Lockwood, “Hustler Populism”; we thank an anonymous reviewer for helping us with this point.

47 Ndungu Gachane,”Nyamakima Kingmakers: How Traders Stopped Uhuru Kenyatta’s Plan,” 12 September 2022, https://nation.africa/kenya/news/nyamakima-kingmakers-how-traders-stopped-uhuru-plan-3944954?fbclid=IwAR2uen0gYemlNxRYnImSCnfH2TmOpWfRyQlqWh0Sfj-cGxXfN-_lUUDNdxw.

48 Ndungu Gachane, “Nyamakima Kingmakers: How Traders Stopped Uhuru Kenyatta’s Plan,” 12 September 2022, https://nation.africa/kenya/news/nyamakima-kingmakers-how-traders-stopped-uhuru-plan-3944954?fbclid=IwAR2uen0gYemlNxRYnImSCnfH2TmOpWfRyQlqWh0Sfj-cGxXfN-_lUUDNdxw.

49 Dauti Kahura, “A Judas Moment: Betrayal in Nyamakima,” The Elephant, 31 May 2019, https://www.theelephant.info/features/2019/05/31/a-judas-moment-betrayal-in-nyamakima/.

50 Muigai wa Njoroge, “MUIGAI WA NJOROGE – MBARI YA KIMENDERO,” 10 August 2018, https://www.youtube.com/watch?v=s93-DDXnvQw.

51 Holston, Insurgent Citizenship, 245.

52 Ibid. “Insurgent Citizenship,” 245.

53 Maupeu, “Les élections générales de 2022”.

54 Attwood, The Reds and the Blacks.

55 Kanyinga, “Ethnicity, Patronage and Class,” 85.

56 Ibid., 1.

57 Lonsdale, “Moral and Political Argument,” 132.

58 Lonsdale suggests this shifted from an “economical ethnicity” to a more clearly defined “moral economy” during colonial rule.

59 Ibid., 140.

60 Cheeseman, “Kenya Since 2002”.

61 Ibid., 100.

62 Lonsdale, “Moral and Political Argument,” 132.

63 Lynch, “Electing the Alliance”.

64 Nathan, Electoral Politics, 5.

65 Cheeseman et al., The Moral Economy of Elections.

66 Willis et al., “The Ruto Roadshow: A Radical Attempt to Transform Kenyan Politics?” The Elephant, 1 November 2021, https://www.theelephant.info/features/2021/11/01/the-ruto-roadshow-a-radical-attempt-to-transform-kenyan-politics/.

67 Ndungu Gachane, “Ndindi Nyoro reveals why he chose Ruto over Uhuru,” Nation, 29 December 2019, https://nation.africa/kenya/news/politics/Why-Ndindi-Nyoro-sticking-with-Ruto/1064-5400916-apmujpz/index.html?fbclid=IwAR25hgPiwaXkoZC_P1LFj6FullpxEyC-8G5XAMnoAwpEuwYuUps6umiAwTM.

68 Son Gatitu, “Kenya: Can William Ruto Secure Support from Mount Kenya in Time for 2022?” The Africa Report, 27 July 20223, https://www.theafricareport.com/108603/kenya-can-william-ruto-secure-support-from-mount-kenya-in-time-for-2022/.

69 IRI, Preliminary Statement of Initial Findings, 11 August 2022, IRI, https://www.iri.org/resources/preliminary-statement-of-initial-findings-and-recommendations-of-the-iri-ndi-international-election-observer-mission-to-kenyas-august-9-2022-general-elections/.

70 Jeff Otieno, “Kenya 2022: The Money and Faces Behind Raila’s and Ruto’s Expensive Campaigns,” The Africa Report, 2 March 2022, https://www.theafricareport.com/180801/kenya-2022-the-money-and-faces-behind-railas-and-rutos-expensive-campaigns/.

71 NTV Kenya, “Cambridge Analytica Unmasked as Claims of Smear Campaign Against Raila are Revealed,” 20 March 2018, https://www.youtube.com/watch?v=ebKEM6Dz4SY.

72 Ibid.

73 Larry Madowo, “How Cambridge Analytica Poisoned Kenya’s Democracy,” Washington Post, 20 March 2018, https://www.washingtonpost.com/news/global-opinions/wp/2018/03/20/how-cambridge-analytica-poisoned-kenyas-democracy/.

74 Lockwood, “Hustler Populism”.

75 It is important to note that 43% of respondents said “none”, the most popular answer.

76 Lockwood, “Hustler Populism”.

77 Anonymous Interview, Nairobi, 4 August 2022.

78 Throup, “Elections and Political Legitimacy”.

79 Cheeseman et al., The Moral Economy of Elections.

80 Maupeu, “Les élections générales de 2022”.

81 Burbidge, “Can Someone Get Me”.

82 Cheeseman, “African Elections”.

83 Cheeseman and Larmer, ‘Ethnopopulism in Africa’.

84 Lonsdale, “Moral and Political Argument”.

85 Klopp, “Can Moral Ethnicity Trump Political Tribalism?”; Orvis, “Moral Ethnicity”.

86 Cheeseman et al., “Democracy and Its Discontents”.

87 Anderson, “‘Yours in Struggle for Majimbo’”.

88 Cheeseman, “African Elections”.