‘Before there is power, there is the country’: civic nationalism and political mobilisation amongst Kenya's opposition coalitions, 2013–2018

Abstract

This paper provides an ethnographic perspective on the street-level deliberations of Kenya's opposition supporters between the 2013 and 2017 elections, arguing that rather than appeals to ethnicity what defines its discourse are broader, inclusive notions of political membership. A civic nationalism is enunciated by opposition supporters that congeals support between multiple ethnic groups through its emphasis on universal values – democracy, due process, equality, adherence to the constitution. However, when such civic ideas are used in political campaigning and mobilising rhetoric, describing a resurgent Kenyan ‘people’ that has been systematically disenfranchised, they take on an exclusionary character. As ‘good constitutionalists’, opposition supporters contrast themselves with ‘bad nationalists’ associated with the government, portrayed as mobilising particularistic ethnic loyalties at the expense of a majority of Kenyans. In practice, their civic ideas remain only potentially inclusive.

INTRODUCTION

In May and early June 2016 supporters of Kenya's alliance of opposition parties, the Coalition for Reforms and Democracy (CORD), took to the streets to demand the resignations of several members of the Independent Electoral and Boundaries Commission (IEBC), the official body responsible for administering Kenya's general elections in 2013. In the middle of May, the country's former Prime Minister and leader of CORD Raila Odinga had publicly faced down media accusations that he and his party were illegitimately causing trouble through the protests when alternatively, they could, detractors suggested, follow parliamentary procedures to try and reform the IEBC. ‘[The] trust bestowed upon institutions has been betrayed times over’, Raila wrote in a Facebook post on 16 May 16. ‘We went to the streets to tell the Commissioners of IEBC that they must resign as they have failed in their duty and we have no faith whatsoever in their ability to be impartial in a political contest.’ According to popular opinion amongst CORD supporters, the IEBC had been complicit in rigging the 2013 election in favour of Raila's opponent, Uhuru Kenyatta (Opalo 2016). Anticipating the same outcome in the forthcoming 2017 election and after a failure to generate change through parliamentary channels, CORD followers went to the streets in an attempt to force the government to engage with their leaders' demands for reforms to the IEBC. In Nairobi, protesters confronted police armed with tear gas outside the headquarters of the electoral commission. In Kisumu, the headquarters of what was once Nyanza Province (Raila's birthplace and the heartland of the Luo people who comprise much of CORD's support), demonstrations turned particularly violent and BBC News (6.6.2016) reported internationally of live bullets being fired and two deaths amongst the protesters. The front of the Taifa Leo newspaper (18.5.2016) depicted a man in a green hooded sweatshirt being stamped upon by a police officer in fatigues.

At least in part, the demonstrations embodied what scholars have typically viewed as the ethno-nationalist character of Kenyan multi-party politics. After an at best ambiguous and at worst outright untrustworthy election result in 2013 (Cheeseman et al. 2014: 3), the protests could be viewed as a continuing confrontation between CORD's then predominantly Luo, Luhya and Kamba supporters and the governing Jubilee Alliance dominated by elite Kikuyu and Kalenjin figures. Much like the 2007 contest between Raila and Mwai Kibaki, the 2013 presidential poll had seen a Kikuyu candidate – the victor, now president, Uhuru – face-off against the veteran Luo politician, Raila. As in 2007, Raila had assembled a multi-ethnic coalition drawn from regions historically marginalised in terms of political representation at the highest level – the former provinces of Coast, Nyanza, Western and Eastern (where, incidentally, the Jubilee Alliance failed to win many seats; Cheeseman et al. 2014: 16). The IEBC protests could be said to have reflected the ethno-nationalist fault-lines that had characterised the 2013 election.

That politics continued to take an ethno-nationalist form in 2016 should hardly have come as a surprise. As Kenyan friends of mine remark, ethnicity runs ‘deep’, an allusion to its recent history in orientating political life. After independence political power came to rest in the hands of a largely Kikuyu oligarchy, owing in part to the strategy adopted by Kenya's first President (and Uhuru's father) Jomo Kenyatta to actively shore-up support amongst his Kikuyu, Meru and Embu allies (Branch 2011: 85). It was during the course of his presidency that disputes in Kenya's nationalist movement gradually shifted from a set of arguments about the future Kenyan state towards a growing tendency for ethno-national political constituencies to vie for a stake in the government machinery itself (Atieno Odhiambo 1987: 177–8; Branch 2011: 81). In the competition for power at the centre, there have been notable winners and losers. Three of Kenya's four post-independence heads of state have been Kikuyu. One cannot fail to notice the exclusion of the Luo from central government positions from Kenyatta's presidency onward (Atieno Odhiambo 2004: 168–9; Branch & Cheeseman 2008: 7). Scholars have argued that the transition from Daniel arap Moi's highly personalised single-party regime to multi-party democracy in 1991 simply created the conditions for ‘winner takes all’ elections in which ethnic blocs began competing for the ultimate prize of the presidency and the control over resources the position brings (Mueller 2008: 186). Accordingly, opposition parties representing large Luo constituencies such as the Orange Democratic Movement (ODM) have argued at recent elections that it is ‘their turn’ to govern (Jonyo 2003).

Yet when one views the rhetoric used by Raila in his defence of the anti-IEBC protests, one does not observe an overt appeal to ethno-nationalist loyalty from his Luo compatriots but rather the language of constitutional democracy and popular sovereignty. Raila's Facebook post on the aforementioned IEBC demonstrations serves as a case in point: ‘[W]e went to the streets to tell all Kenyans that they must not sit back and watch their sovereignty abused and their aspirations betrayed by an irresponsible and unaccountable regime.’ Raila's words and the very fact of the anti-IEBC protests – that Raila argued took place in the name of free and fair elections – suggests that at the very least political mobilisation in contemporary Kenya is a more complex process in its discursive manifestations than evoking only ethnic forms of self–other identification. Raila's constitutionalist rationale for the protests provides an initial point of departure for this paper and its guiding question, namely: what is the role of civic ideas in mobilising political support in contemporary Kenya?

Scholars of Kenya and Africa more broadly have done much to historicise contemporary ethnicity-based politics and its ‘uncivil nationalism’ (Berman 1998; see also Lonsdale 1992; Lynch 2011), arguing that the tension between sub-national ethnic communities and universal citizenship remains the central dilemma for African states post-independence (Berman 2004: 9). This paper suggests that the second half of this formula – the possibility of a civic nationalism in Kenya – warrants more attention from scholars given its prevalence in political discourse. By civic nationalism, I refer to Kenyans’ ongoing attempts to imagine a mode of political governance premised upon general human political equality, subsuming ‘pre-political ties’ (Ingham 1996: 2) within a post-ethnic national community defined by commitment to constitutional principles (see e.g. Klopp 2002: 270). Civic nationalism in Kenya celebrates ‘rational’ thought and decries the ‘backwardness’ of ethnicity and its grip on political subjectivities.

Whilst recognising that such attempts to imagine a more civic Kenya exist (see e.g. Cheeseman 2015), in general scholars have continued to emphasise the strong role that ethnicity has undoubtedly played in recent elections (Bratton and Kimenyi 2008; Mueller 2008: 199; 2011: 108; Cheeseman 2009; Cheeseman et al. 2014, 2019). However, one wonders if Kenya's politics appears thoroughly ethno-nationalist in character because of a focus on election results. Dominic Burbidge's (2014) recent research, which involved soliciting diaries from middle-class Kikuyu interlocutors in Nairobi in the build-up to the 2013 election, provides an interesting counter-point to perspectives gleaned from the ballot box. The diaries revealed in detail how middle-class voters thought through their political loyalties and voting decisions, revealing that ethno-nationalist voting patterns were hardly automatic decisions (Burbidge 2014: 222). The raised political stakes at election time ultimately meant that ethno-nationalist loyalties triumphed for these voters, though not unreflexively or uncritically. Burbidge's account suggests that scholars’ ethnicity-focused accounts of Kenya's politics may be a product of reducing politics to elections at the expense of studying broader events and happenings in which they are situated where alternative political identifications might be observed (see also Hunter 2016: 2).

In contrast to the focus on elections, a recent wave of scholarship has begun to explore ‘everyday’ political debate and knowledge production at sites such as bunge la mwananchi, people's parliaments that exist in a variety of forms across urban Kenya (Diepeveen 2016; Kimari & Rasmussen 2010; Rasmussen & Omanga 2012; Gachihi 2014; Sørensen 2018; cf. Kresse 2009). Studying such spaces, anthropologists have discovered forms of citizenship based on cosmopolitan principles of debate, rather than ethnic identity, even when such parliaments take place beyond Nairobi (Rasmussen & Omanga 2012). Whilst such sites are deliberately cosmopolitan in ethos, and thus very specific manifestations of public debate in Kenya, they provide an alternative perspective on Kenya's politics, one in which membership in their spaces is concertedly trans-ethnic.

At the same time, recent events in Kenya's politics also suggest there might be reasons to expand scholars’ focus on ethnicity towards a renewed emphasis on the ideas and notions that transcend exclusionary and bounded ethnic identification, particularly during times of campaigning. In Kenya, the introduction of the ‘50 plus 1’ stipulation for the presidential ballot in the 2010 Constitution – that the winner must secure over 50% of votes cast – has led to the emergence of new and surprising trans-ethnic alliances (see e.g. Lynch 2014).¹ And as Jacqueline Klopp once noted, scholarship on Kenyan politics that has homed in on the meta-narrative of ethnicity has sometimes hidden the fact that ‘wheeling and dealing across fuzzy ethnic boundaries has been an essential part of patrimonial politics in a polyethnic society’ (Klopp 2002: 275; cf. MacArthur 2008). Given that Kenya has a long history of ethnic groups aligning together to compete for power at the centre, particularly under the recent conditions of a majoritarian democracy, this paper asks what histories, ideas and vocabularies might facilitate relations across the boundaries of ethnicity for contemporary multi-ethnic coalitions.

This paper builds on research into deliberately cosmopolitan spaces, extending its insights into the realm of political mobilisation. The possibility of a civic nationalism is certainly being mooted in Kenya's political sites like bunge la mwananchi. However, such ideas also provide potent forms of political expression and mobilisation for multi-ethnic political coalitions like CORD. A language genre of civic nationalism provides a pool of ideas to be turned into the very stuff of political rhetoric that can encompass a range of ethnic groups based on a shared struggle for democracy itself, one that would simply represent them politically in the first place.

To illustrate the political efficacy of civic ideas amongst Kenya's opposition supporters, I draw upon research carried out at a people's parliament associated with Kenya's opposition coalition between 2015 and 2018, CORD (mentioned above) and later the National Super Alliance (NASA). The fieldwork for this paper involved a series of observations at the CORD bunge then located in Nairobi's CBD between June and August 2015, with follow-up visits in January 2017 and February 2018.² As I go on to discuss, what I discovered there was a multi-ethnic political constituency reflecting upon and discussing its cohesiveness and political viability. The paper's central observation is that civic language observed at the CORD bunge, like Raila's justification for the IEBC protests, mobilises notions of popular sovereignty to open up support for Kenya's opposition coalition, transcending ethno-nationalist particularism via a professed commitment to a higher cause – the values of constitutional democracy.

Whether this language is capable of binding together a multi-ethnic coalition in a lasting sense cannot be argued with the data presented here, which is primarily discursive, drawn from instances of political speech. What I argue instead is that the study of political language amongst a multi-ethnic political coalition such as CORD gives scholars a sense of how such constituencies talk and mobilise themselves during and after elections, how a language of common grievances is formed that harnesses legacies of exclusion from government experienced by a range of ethnic groups. As Klopp notes (2002: 275), such is the fluidity of Kenya's politics that inclusive notions are necessary to encompass the shifting of party alliances. That alliances can be articulated via an inclusive civic nationalism is what this paper observes ethnographically. This is what scholars of ‘constitutional patriotism’ writing after Jurgen Habermas have noted: that through a commitment to democracy itself and a generalised human equality and equal opportunities to political representation, ‘minorities can try to tell stories about ever-widening circles of inclusion’ (Müller 2007: 80). Cosmopolitan CORD supporters at bunge la mwananchi emphasise their collective investment in Kenya as a country, their identity as a ‘people’ who have been disenfranchised, and who are committed to democracy both as a means and an end in their politics. Histories of marginalisation and attributions of electoral fraud allow the opposition to construct itself as ‘the subject of wrongs’ that animate claims towards sovereignty and representation (Samet 2019: 274). However, the civic language I study here renders support for opposition coalitions only potentially open.³ When the conditions and style of political rule itself become the basis of inclusion in a political movement – a campaign tool, so to speak – then opponents to this project are defined as those that appear to block civic identities from coming to the fore: those that practice ‘bad nationalism’ (Calhoun 2002: 150) instead of ‘good’ multi-ethnic cosmopolitanism. Civic nationalism remains exclusionary in practice, though a potent medium of political mobilisation across ethnic boundaries nonetheless.

CONTEXT AND SETTING

Civic nationalism as a moral horizon

A central premise of this paper is that civic ideas in Kenya have been ‘vernacularized’, in other words that ‘values and practices of democracy’ have become ‘embedded in particular cultural and social practices’ and ‘entrenched in the consciousness of ordinary people’ (Michelutti 2007: 639). Practically wherever one looks in Kenya today, one observes the normative force of the concept of democracy in political discourse, so too the value placed on the constitution. One should desire its full realisation and adhere to its logics – these principles are tacit in much of the discussion that takes place within the country's lively public sphere; across its newspapers, radio and television stations, in its politicians' speeches and street-corner debates. Politicians and political pundits trained as lawyers play a prominent role in the media, valorising the neutrality of the constitutional process against the partisan politics of ‘tribalism’.

This positive valuation of democratic constitutionalism in public life has its roots in the era of Daniel arap Moi's single-party rule as leader of the Kenya Africa National Union (KANU) between 1979 and 2002. Opposition to Moi's leadership was regularly couched in terms of ‘multi-party’ democracy, precisely what Moi denied for much of his authoritarian rule. Although Moi had held elections in 1997, these were widely viewed both then and now as an attempt by the President to control growing calls at home and abroad for political pluralism (Branch 2011: 228). It was only in 2002 when Moi's chosen successor Uhuru Kenyatta lost to Mwai Kibaki, who led the National Rainbow Coalition (NARC), that Kenya's ‘second liberation’ appeared at hand (Branch 2011: 249–50). The anti-Moi slogan ‘everything is possible without Moi’ (yote yawezekana bila Moi) spoke to the expectations pinned on the post-Moi era of democracy (Daily Nation 18.11.2002). It was Moi's deliberate use of ethnic violence via gangs as an instrument of oppression (see Haugerud 1995: 39; Branch & Cheeseman 2008) that informed expectations that multi-party democracy would pave the way for a fairer Kenya, defined by the equal political representation regardless of ethnicity.

The era of Kenya's ostensible second liberation and the introduction of multi-party democracy did little to alter the political fortunes of Raila and the struggles of ethnic groups excluded by Moi.⁴ Most notably, the failure of Raila to become President in 2007, coupled with his subsequent loss in 2013 – both amidst widespread claims of rigging (Cheeseman et al. 2014) – has given democracy a distant character to his followers. Despite Raila becoming Prime Minister in 2008 through a power-sharing deal with Kibaki's government, electoral losses to the Kikuyu presidential candidates in 2007, 2013 and now 2017 have given opposition supporters a strong impression that Kenya has been captured by a Kikuyu ethnic monopoly under the leadership of President Uhuru Kenyatta and supported by their Kalenjin allies embodied by the figure of Deputy President William Ruto.

But far from giving up due to discontent (Cheeseman et al. 2014), democracy's status as a vector of hope springs eternal to opposition supporters and is central to its role in what I call the ‘genre’ of civic nationalism that animates their political activity (cf. Werbner 2004: 15). I follow Karin Barber (2007: 32–3) in deploying the notion of genre as a ‘kind’, a category of texts which accrue meaning in its locally situated specificity. This genre draws upon this long history of incomplete democratisation (from the perspective of opposition supporters, at least) along with universalistic notions of fairness, political equality and proper political representation for Kenya's minorities that evokes histories of regionalism (majimboism) that I go on to describe in more detail below. For now, I note that it is precisely democracy's invitation to imagine a better future in terms of fair political representation that makes it so amenable to use by authority in political mobilisation (see also Haugerud 1995: 17–18). Before exploring such instances of street-level political speech, I turn to a more general description of the CORD bunge and its attempt to cultivate a cosmopolitan character.

Bunge la mwananchi

Amongst other sites, it is in Nairobi's central business district (CBD) where debate over the state of Kenya's democracy takes place, where unemployed men, students, off-the-clock taxi drivers and aspiring leaders all with a keen interest in high politics meet on a daily basis to argue on the street corners close to the statue of Mau Mau leader Dedan Kimathi. Popularly, these gatherings are known as bunge la mwananchi (the people's parliament). Much has already been written about bunge la mwananchi in Kenya, particularly its ‘transgressive’ style mimicking Kenya's official parliament (Kimari & Rasmussen 2010) and its aspirations towards performing a cosmopolitan post-ethnic citizenship (Rasmussen & Omanga 2012). However, it should be noted that this account concerns research carried out amongst a particular congregation of bunge – one attended (at the time of fieldwork in 2015) almost entirely by supporters of the aforementioned Coalition for Reforms and Democracy (CORD), the alliance of political parties that contested the 2013 election.

What I found to be at stake at the ‘CORD bunge’ in 2015 were the fortunes and internal politics of CORD. CORD supporters I spoke with claimed they had been upset by electoral malpractice and were searching for guarantees that a fair election would take place in 2017. A level-playing field, they argued, would guarantee them power since they represented ‘all Kenyans’. In late 2017, this bunge moved to Jevanjee Gardens, and there was probably cross-over in terms of members with the Jevanjee Gardens bunge (Gachihi 2014). By then, CORD had become the National Super Alliance (NASA), the more recent iteration of the coalition which had contested the 2017 presidential election, now constituted of four leaders: third-time presidential candidate Raila Odinga of the Orange Democratic Movement (ODM), Kalonzo Musyoka (who ran as deputy) of the Wiper Democratic Movement (WDM), Musailia Mudavadi of the Amani National Congress (ANC) and Moses Wetang'ula, leader of Forum for the Restoration of Democracy-Kenya (FORD-Kenya). After Raila Odinga's boycott of the re-run presidential ballot in October 2017 (the Supreme Court annulled the original poll held in August, citing inconsistencies in the information provided by local polling stations; BBC News 1.9.17), NASA was re-branded as the ‘National Resistance Movement’, and the discourse of ‘revolution’ and ‘resistance’ (kukataa) could be observed at bunge and in other spheres where Luo youth and NASA supporters participated in political debate.

Whilst displaying the character of a coalition meeting, the CORD bunge was a place of deliberation comparable to other manifestations of bunge la mwananchi. At the CORD bunge I discovered members of a multi-ethnic political coalition reflecting on itself, its ethnically diverse make-up, recent history and future fortunes. What its leaders ‘should do’ was regularly discussed, not simply as a naïve attempt to direct a distant and elite CORD leadership, but to perform political opinions and evoke sentiments amongst other members. After all, many of the men present (and it was, almost unfailingly, solely men who participated) were ardent supporters of Raila Odinga. To debate politics and argue one's position was to iterate and perform political viewpoints, to inspire others to maintain their commitment to the cause (as we shall see below). There is much in common here with Stephanie Diepeveen's Arendtian analysis of bunge spaces in Mombasa, where she argues that politics need not be confined to ‘high politics’, but that it takes place through ‘the very act of people speaking and acting together’ (Diepeveen 2016: 268; cf. Kimari & Rasmussen 2010: 150). To follow Karin Barber (1997: 354), such speech-acts are both productive and constitutive of ‘audiences’, collectives wherein common identifications can be sparked.

The content of bunge debates often emphasised fundamental political equalities as a normative good. Consider, for instance, the words of one participant in February 2018 at the height of NASA's opposition to the result of the 2017 election and the prospect of embarking upon a National Resistance Movement that, to some of NASA's coalition members, was seen as Raila Odinga's political project.

I'm seeing Luhya on social media, saying that not being elected is Raila's problem. It's not just Raila's problem. If Raila cannot go to State House, no one can. Not Kalonzo [Musyoka], not Musailia [Mudavadi]. No one outside the Kikuyu-Kalenjin bracket [can be President]. (NASA supporter speaking at bunge la mwananchi. Fieldnotes, February 2018)

The appeal to fairness made by the speaker, to equality of representation and indeed a shared membership in the struggle for democracy and sovereignty, is characteristic of the CORD bunge’s emphasis on a broader civic identity. Raila's struggle to become president, he emphasised, was a litmus test of Kenya's constitutional principles. Notions encoded in democracy's conceptual repertoire – such as equality and the right to political representation and sovereignty – were tools to think with and think through politics and inequality in Kenya.

Habermas might have described such ideas as evidence of a civic nationalism or a ‘constitutional patriotism’ (Cronin & De Greiff 1998). But, as Craig Calhoun notes (2002: 150), such civic identities contrast themselves with ‘the negative force of the nationalist imaginary’ that ultimately cause it to fail. In other words, a cosmopolitan ethic is pursued by a political community to distance itself from those who practice ‘bad nationalism’. This is evident in the spatial and linguistic logics of the CORD bunge where debate took place in both English and Kiswahili, and frequently a mixture of the two. Kiswahili was preferred as a lingua franca capable of being understood by the largest possible audience (though there is also an element of prestige that speakers can hope to project through the command of the English language; Brisset-Foucault 2013: 232). On the other side of the street, Jubilee supporters engaged in an entirely separate debate that often spoke to concerns in central Kenya and regularly took place in the Kikuyu language. The use of Kiswahili and English was therefore a source of pride for attendees of the CORD bunge, who point to the linguistically closed debates amongst Jubilee supporters – further evidence, they argue, that Jubilee does not truly possess the support of all Kenyans. Members of the CORD bunge hail from a range of places – including those that have been historically associated with the cause of majimboism (regionalism) such as coastal Mombasa or the predominantly Luhya province of Kakamega.

Florence Brisset-Foucault has written of similar though notably more formal debating societies in Kampala that many participants are, in a sense, people of the state who wish to distinguish themselves as modern citizens through adherence to an ideal cosmopolitanism via constitutionalism, one that exists against the backdrop of politicised ethnic ties (Brisset-Foucault 2019: 245). In a similar respect, the basis of the opposition glimpsed at the CORD bunge relies on an attempt to ‘rise above’ ethnic politics associated with the government and its supporters. Crucially, an inclusive civic nationalism locates its political other through an implicit elitism that reinforces a distinction between educated cosmopolitan Kenyans and gullible followers of ethno-nationalist politics. In other words, the opposition must not simply compete for power but be moral and right in doing so. This places a limit on the inclusiveness of its civic nationalism that I return to at the close of this paper.

It is as a source of pride and moral strength that we see civic nationalism in the rest of this paper – as a rhetorical call to action as much as an implicit ethos. By turning now to an instance of political oratory that fundamentally reshaped debates at the CORD bunge one afternoon in 2015, I show how such broader notions of citizenship in the Kenyan nation encompass multiple ethnic groups, keeping the alliance together even through events that challenge its cohesion.

A SOCIAL SITUATION: AN AFTERNOON AT THE ‘CORD BUNGE’

An mkubwa arrives

Unusually for Nairobi, where wazungu (sing. mzungu, white foreigner) are routinely mocked for being where they should not be, no one seemed particularly bothered about my presence amongst a group of CORD supporters debating the loss of the 2013 presidential election on a sunny July afternoon in Nairobi's Central Business District in 2015. Apart from the occasional glance or comment about my presence from an audience member, the crowd was primarily focused on the back and forth of the debate in front of them. At the centre of the predominantly middle-aged and elderly male onlookers (dressed in the shirts and suits typical of their age-group) two younger men were arguing over the failure of opposition leader Raila Odinga to win either of the elections he contested in 2007 and 2013, having been defeated in both instances by Kikuyu opponents (Kibaki and Uhuru respectively). Taking it in turn to press home their points, and weigh-up each other's arguments, the temporary adversaries appeared to be at an impasse over the ‘true’ result of the 2013 election. For the most part, the crowd remained quiet, weighing up the points made on either side, though one could hardly miss the smiles when either one would go on the offensive.

Before long, a new participant arrived. Dressed in a flamboyant patterned shirt, a silver watching glinting on his wrist in the sunlight, a middle-aged mkubwa (boss, literally ‘big man’) approached the crowd and was immediately given the floor. This was a man understood by the others present as a person who possessed intimate knowledge of CORD's internal affairs, and therefore entitled to hijack the debate. The mkubwa postured opposite one of the young men at the centre of the circle and began questioning the latter's arguments, hands behind his back and head slightly lowered in deferential expectation. The young man, a student from a local Nairobi university, was put on the spot. ‘Why do you think Raila was not declared in 2013?’, the mkubwa asked him as the onlookers grinned, somewhat surprised by the intervention of a CORD notable in a debate between the two young men on a Nairobi street corner. The student began to marshal legal and constitutional reasons to explain why Raila had lost in 2013. From the crowd a voice shouted ‘50 plus 1!’, indicating the widespread belief amongst the opposition that Raila had indeed rightfully won the first round of the poll, and that the 2013 election had been rigged in favour of the Jubilee Alliance. The crowd laughed in response. The student continued, unsuccessfully attempting to suggest that Raila had lost in 2013 for legitimate reasons, and that the Supreme Court (which ultimately declared the result) had considered the need to ensure the peaceful ‘state of the nation’.

The mkubwa disagreed, firing rhetorical questions at the youth and then the crowd: ‘Was there a power vacuum?’ He looked around as though waiting for an answer. There was no reply from the student, nor the audience. ‘Was there a power vacuum? [Mwai] Kibaki [the incumbent between 2007 and 2013] was still in power! In 2013 Kenya was at peace!’ In other words, it could not have been because of concerns over the state of the nation that Raila had not been declared winner. The mkubwa argued that there was no good reason for him not to have won. He then insisted on having the final word, stressing the ambiguity of the outcome of the 2013 election, and, by implication, that the result had been manipulated:

[Chief Justice Willy] Mutunga, the most educated man in Kenya, the first African to get LLM at Harvard. Why could he not [announce the result]? Even he said it was a non-verifiable result. Let me tell you: After that defeat Raila was very disappointed. He wanted to retire! But the people would not let him!

The mkubwa's intervention immediately changed bunge's character from an arena of multiple voices arguing to a scene of political oratory where he was able to use the recognition of his status in the eyes of others (a status encoded in his dress, demeanour and awareness of his privileged knowledge of CORD) as the basis of a situated authority, one that allowed him to begin addressing the crowd. A single voice was suddenly valued above all others as one with the capacity to mediate the recent history of CORD's electoral fate, and that of Raila in particular.

In this opening scene, one can observe the central premises to be revisited below: the illegitimacy of the previous electoral results, ‘the people’ Raila is imagined as faithfully representing, emphasis on adherence to constitutional procedure. Not only do these initial excerpts suggest that political mobilisation amongst the opposition relies on an understanding of popular sovereignty consistently denied and is therefore a more complex process than directly pandering to ethnic identities, but it warrants further discussion about exactly how multi-ethnic constituencies narrativise and come to understand their mutual political struggle across the boundaries of ethnicity.

Articulating inter-ethnic alliance

Precisely because of CORD's multi-ethnic membership, there is an onus on authority figures like the mkubwa to articulate the basis for this alliance. The importance of overcoming difference becomes clear by observing the mkubwa's reaction to the prospect of conflict within CORD ranks.

After his initial interjection regarding the outcome of the 2013 election, the mkubwa was prompted by the crowd to comment on a separate issue that had taken place the previous day at the ruling of a court case brought by prominent Luhya politician, then Senator of Kakamega County, Boni Khalwale against Luo politician (then Governor of Nairobi) Evans Kidero. Khalwale had attempted to lift a gagging order that would allow him to link Kidero to the collapse of a prominent Kenyan sugar company, suggesting it was Kidero's mismanagement of funds that caused the collapse of the firm. However, at the hearing young male supporters from Kidero and Khalwale's respective camps had brawled (The Standard 15.7.2015). Both sides accused each other of having escalated the confrontation towards physical violence. One of the crowd members at the bunge mentioned to the mkubwa reports that two of Khalwale's supporters had been killed in the melée. The mkubwa was quick to condemn the violence. ‘This is very bad … this is most unfortunate … Blood has been spilt in a court of law!’

The student explained that Khawale supporters had arrived at the courthouse the day before and ‘caused trouble’ prompting Kidero to arrive the following day with his own entourage. He elaborated the logic: ‘If you slap me, I come back with my brothers. He was slapped: he had to come with his brothers.’ The mkubwa was less certain of this wisdom, repeating the words ‘Kidero went to court with his supporters’ in an alarmed tone several times:

MKUBWA: Kidero was being asked about finances when he was in that house [the sugar company].

CROWD: [Simultaneously, in agreement] In that house.

MKUBWA: So he can come and defend himself. But we cannot talk about changing Article 6 [of the Kenyan constitution that sets out the responsibilities of leadership]! This was the constitution sworn on when you enter office! We must maintain that constitution! … If this is true [the reports of two dead at the hands of Kidero's supporters] then Kidero can be sacrificed. Because there has been a loss of life! This is very serious.

CROWD MEMBER: It's our enemies! [A reference to the Jubilee government sewing discord between the Luo and Luhya]

[The mkubwa is repeatedly interrupted by crowd members laughing, agreeing with the outcry]

MKUBWA: Wait your turn! [More commotion] A word is about to come out of my mouth that I don't want to! [Laughter] … Raila has been on this path since before your mothers were born, but Kidero he can be sacrificed. We must be very sober. We must not politicise this. Raila must call a meeting with CORD. [Tuts several times] This is very unfortunate. This is very unfortunate. We can sacrifice Kidero. We cannot sacrifice confraternity. [Audience applause] We have always had good relations with the Luhya, we have intermarried with them. I am a Luo, my wife is a Luhya. I am a Luo and my wife is a Luhya as I stand here today.

Visibly concerned by the implications of internal conflict, the mkubwa lamented the potential for it to damage the CORD alliance and, by extension, Raila's chances of becoming president. Notably, the mkubwa's persuasion took place through the mediation of a history of affinal alliance with the Luhya through marriage, and he deliberately presents himself as a product of that history (MacArthur 2008). If the alleged brawl between Khalwale and Kidero's followers had exposed the fragility of such inter-ethnic alliances at the level of multi-party politics, the mkubwa's response exemplifies the role of moderate figures like himself that appear periodically ‘on the ground’, as it were, in the maintenance and care for such political relations (Neumark 2017: 754). In contrast to the political deals that lead to ‘power-sharing’ across ethno-nationalist lines that are observable at a higher political level (between party leaders, for instance), by evoking such a history of ‘cousin-like’ bonds (Ben-Yehoyada 2014), the mkubwa asked the audience to value relations across difference requiring trust and tolerance. The struggle for democracy as part of the opposition, he suggested, was the product of a kinship that Luo and Luhya already shared and not simply a convenient political alliance between distinct groups.⁵

In this excerpt we observe the mobilisation of political constituencies taking place across ethnic boundaries through the articulation of notions of relatedness. Instances like this prompt further consideration of the relationship between CORD's heterogeneous support-base, not least since ethnic groups are often objectified in such political language as blocs that possess political agency. The objectification of CORD as being comprised of various ethno-national or ethno-regional constituencies relies on the knowledge that bunge participants possess about the shifting political entities that comprise Kenyan political life and how they interact. ODM is often understood as having a predominantly Luo following, just as the Wiper Democratic Movement led by Kalonzo Musyoka is seen (at this larger, national scale) as the party of Akamba. The ethnic make-up of such parties is tacitly understood at such debates, and the participation and activities of leaders in political alliances (as we see with Kidero and Khalwale above) is regularly read back into the broader political milieu as evidence of a given ethnic group's participation in the fortunes of the coalition. That bunge participants like the mkubwa deploy this knowledge of how politics works as a means of building larger political constituencies recalls Rogers Brubaker's important observation about the invocation of ethnicity in political language used both by scholars and interlocutors (2004: 11–13). To speak of groups gives form to politics, creating an impression of the entities that drive its workings. What Kenyans do with recourse to ethnic categories is exemplified in the mkubwa's speech above – ethnic heterogeneity is objectified as the basis of a ‘larger’ struggle for democratic representation that would be achieved were Raila to become President.

No more was this evident than in 2018 when tensions between the government and NASA – which by then had proclaimed itself the National Resistance Movement (NRM) – had reached its nadir. After boycotting a re-run of the Presidential election held in October 2018 (after the Supreme Court had annulled the original held in August), Raila continued to claim that he had been the rightful winner of the original poll. Plans were made to symbolically swear-in (kuapisha) Raila as the ‘people's president’ at Uhuru Park in late 2017. There were widespread fears amongst Kenyans that such an act would lead to a government crackdown and violence at the scene, but the event went ahead without the expected reprisals from law enforcement. After Raila Odinga's swearing in as People's President at Uhuru Park in January 2018, NASA leaders turned their attention to the status of Kalonzo Musyoka's Wiper Democratic Movement within the broader alliance. Kalonzo, who had been NASA's candidate for Deputy President, failed to attend the swearing-in ceremony, possibly fearing arrest. Wiper could not pick and choose when it was part of NASA, argued the coalition's notables (Daily Nation 14.2.2018).

The same day these comments appeared in newspapers, under the trees of Jevanjee Gardens, bunge participants argued about whether Kalonzo, and by extension ‘the Kamba’, were committed to NASA fortunes. The back and forth that day highlighted the tensions that Kalonzo's absence had caused. One Kamba participant argued forcefully that he was not ‘for Kalonzo’ but ‘for Raila’, joking that in Ukambani (‘Kambaland’, to the east of Nairobi) Kamba were walking around with bibles in the event they encountered Kalonzo to swear him in as People's Deputy-President on the spot. Allegiance to Raila indicated allegiance not to an ethnic community – but to the ‘larger’, if you like, ideal of a Kenya rejuvenated by the leadership of Raila the democrat. ‘The Kambas were there [at Uhuru Park]’, another participant recalled. Evidently, the status of particular ethnic groups within the coalition of opposition parties requires maintenance in such instances. It may be that Kamba at bunge felt that day they had to argue for their commitment to NASA in quite forceful terms because of the rather uncomfortable alliance between Kalonzo and Raila. Kalonzo himself was said to fear arrest if he was sworn in since it was thought he planned to vie for the presidency in 2022.

It is precisely because of the fluidity of Kenya's politics that political language and oratory appears significant in achieving temporary unities amongst groups that may have had recent histories of electoral opposition. Where discussion of a history of social relations is possible in the case of the Luo and the Luhya, in a quite different context, Kamba participants argued that their commitment to Raila's cause aligned them with NASA. This brief snapshot of intra-NASA relations in early 2018 foregrounds the next episode of the situation I continue to unfold. It suggests that whilst ethnicity continues to be an important strand of political identification, the particularity of various ethnic experiences of political marginalisation is amalgamated under the umbrella term of ‘the people’ who, according to the mkubwa's ongoing address, will vote Raila into office.

‘Raila must be voted in by the people’

As though trying to end on a positive note, the mkubwa launched into an optimistic speech about the future of Kenya and the emergence of a politics free from ethnic allegiances, a horizon that he argued was embodied by the university student present and his generation. But his speech was quickly interrupted by the young man himself.

MKUBWA: Let me tell you, Kenya is different now. There is more education, like this young man here [gesturing to the student], they do not know their tribe. They think they are Kenyan. Now it's the Kenya of the internet, of Facebook and Twitter.

STUDENT [interrupting]: Can I ask one question? [the mkubwa pauses and looks at him] You tell us to be sober, but are we sure that the other side, are they being sober? [A reference to the Jubilee Party, and the perception that their followers will do whatever it takes to win the upcoming 2017 elections. Followed by raucous laughter from the crowd]

When Raila lost the election in 2007, riots broke out across urban Kenya, at first initiated by his supporters.⁶ The young man appeared to be suggesting that CORD supporters could not be expected to ‘remain sober’ – that is, to accept the result of the 2017 election whatever the outcome, even if they perceived that the government had (as they suspected) rigged the election. The young man evoked the possibility of political violence.

The comment encouraged the mkubwa to rant at length about the qualities and virtues embodied by Raila and that ought to be adopted by all CORD supporters:

MKUBWA: Raila has always been about democracy. We must embrace the constitution! We must follow democracy! Raila does not want to act like a thug. He wants to be voted in by the people. So that the people can say ‘We want Raila’. Otherwise Raila would be president by now. Raila, wakamba [i.e. the Kamba people], some Kikuyu, the Luhya, the Maasai – they don't want him [Uhuru Kenyatta]! Raila must be voted in by the people! Voted in by the people! [To Uhuru Kenyatta] The people together they all say [raised voice] WE WANT YOU … [gesturing towards the government buildings]

AUDIENCE [together]: … OUT!!

MKUBWA [lower tone, calming the crowd to listen to his story]: Let me tell you a story about an uncle of mine in Luoland. His wife was preparing a chicken and you know that a chicken is a very special meal in Luoland [smiling, purposefully playing up to stereotypes about Luo practices to laughter from the audience]. Well, his relatives visit, but they arrive home before him. And because in African culture you must prepare a meal for the guests, the wife makes them a chicken and they eat it before he gets home. So the husband arrives, and there's no chicken – it's been eaten! What does he do? He strikes the wife and burns down the home. The next day people see him sleeping under a mango tree and they wonder if they should help him … What I am saying is: now there is no chicken and no home. We cannot act in those ways … Before there is power there is the country. You cannot have the … [waits for crowd to follow his lead]

AUDIENCE [simultaneously]: POWER!

MKUBWA: without the …

AUDIENCE [simultaneously]: COUNTRY!

The audience roared the final words, and quickly the performance came to an end, the CORD figure departing after shaking hands with one or two audience members.

In this final excerpt, we observe the rhetorical iteration of a higher goal or ideal – that of attaining ‘power’ and realising ‘democracy’ by voting Raila into the State House. Behind this notion of ‘the people’ that seek that goal is the belief that if previous elections had not been rigged then Raila would have won. The implication is that an elite minority of governing Kikuyu and Kalenjin have consistently denied the realisation of the true will of most Kenyans, even ‘some Kikuyu’. By claiming that the electoral system has failed to guarantee the realisation of the popular will, an alternative language of political mobilisation is afforded. Through the experience of consistent electoral defeat articulated by the mkubwa and other CORD supporters this broader constituency of ‘the people’ is expressed, a category that transcends ethnicity and indexes popular sovereignty denied, and yet to be realised. CORD supporters are portrayed as good constitutionalists in their commitment to Kenya and broader civic values of due process – putting the country first – in contrast to partisan nationalists who they oppose.

If suspicions of electoral malpractice generated a sense of marginalisation amongst CORD supporters, such constituencies also have historical roots. As a language of dissent to which bunge participants such as the mkubwa have recourse, a notion of ‘the people’ that encompasses various disillusioned groups and persons irrespective of ethnic affiliation reflects the legacy of political agitation for majimbosim (regionalism) in Kenya. Attempts to draw together a ‘non-Kikuyu’ ethnic alliance from outside Kenya's geographic and political centre were central to Raila's recent opposition campaigns in 2007, 2013 and 2017. In this respect, CORD and, more lately, NASA have their organisational roots in the post-independence period when a policy of devolution was promoted by the Kenya African Democratic Union (KADU) as a means of protecting smaller communities and their regions from the influence of the state. KANU's nationalist orientation, it was thought, was a convenient cloak for Kenya's larger ethnic groups to assert economic and political dominance (Branch 2011: 85). CORD, and more recently, NASA, inherited this legacy of regionalism. The 2010 Constitution and the creation of the counties – a cause for which Raila was a chief advocate – carries its legacy. And as Julie MacArthur notes, at the 2007 election his ODM party ‘relied more on the regional issues of resource distribution and equitable development’ encoded in notions of majimboism than it did on ethnic particularism (MacArthur 2008: 228). If Daniel arap Moi's attempt to create a coalition of minorities during the early 1960s was an outright rejection of national or nationalist aspirations, as David Anderson argues (Anderson 2005: 551), the language of ‘the people’ invoked by the mkubwa evokes less a demand for decentralised power than the struggle of a politically and economically disenfranchised ‘people’ towards central, national power.

It is here that I wish to return to the matter of ethnicity with which I opened the paper. The possibility of concepts that ‘open up’ support not on the basis of ethnicity but through the horizon of democracy and civic values recalls Richard Werbner's (2004) critique of John Lonsdale's historical argument about what he called the ‘Janus-faced’ relationship between ‘moral ethnicity’ and ‘political tribalism’ (Lonsdale 2004) – concepts that Lonsdale introduced into the descriptive language of analyses of Kenyan politics and which are still widely used (Cheeseman 2009: 99–101). The former indexed the internal principles of and conditions for the acquisition of personal virtue and reputation within a loose community, and the latter described the later stages of the colonial era when these loosely defined moral communities were turned into outward-facing and competitive political blocs by the political entrepreneurs of the late colonial and then post-independence eras.

Whilst Lonsdale's work continues to inspire for its commitment to the study of the moral principles and notions through which Kenyans live their political lives, Werbner usefully notes that the story told by Lonsdale about the transformation of Kenya's ethnic groups from moral communities to instrumentalised political blocs has a unidirectionality ‘apparently obviating the precolonial alternative’ (Werbner 2004: 69). The possibility of a ‘moral inter-ethnicity’ appears to be out of the question in some contemporary scholarship that represents Kenyan politics as a terrain of inter-ethnic competition, consigned to a pre-colonial past. And whilst Werbner's work takes another route in describing the ‘postcolonial super-tribalism’ (Werbner 2004: 67) that he tells us has emerged in Botswana, his critique has relevance for the appreciation of Kenya's contemporary politics. The question raised by Werbner is whether the unidirectionality of this historical trend (the social construction of ethnicity, no less) has directed scholars' attention away from the possibilities latent in other political languages to stretch moral community (of any type) beyond the ethnic group. In the above excerpts, we observe a vocabulary that offers an alternative imagining of political community, not contingent upon ethnic membership but on a broader (and often implicit) story of ‘the rest’, if you like, ‘the people’ described by the mkubwa. Though banal, these terms have a useful political neutrality in a multi-ethnic context, providing a language of common grievance that draws on shared histories of political marginalisation. Circulating within multi-ethnic coalitions, then, are more inclusive forms of political identity evoking ‘a wider and more inclusive national consciousness’ (Klopp 2002: 270), even if these inclusive concepts might be limited to those who can argue they have experienced political defeat and marginalisation in post-independence Kenya, though even references to ‘poor Kikuyu’ allow opposition supporters at bunge to claim to be pan-ethnic in the broadest way possible. In terms of political mobilisation, the claim to be ‘the people’ has an important emotional value, allowing supporters to both reason and feel that they are on the right side of history, renewing their commitment to opposition via the normative morality of popular sovereignty.

Though inclusive, these claims are certainly not unfailingly universal. The reference to ‘our enemies’ who might not be ‘sober’ reminds us of the limits of a civic nationalism when used as a rhetoric for political mobilisation. CORD and NASA supporters narrativised themselves as good constitutional patriots but contrasted themselves with ‘bad nationalists’ that the Jubilee Party's Kikuyu and Kalenjin leadership is seen to be (see Calhoun 2002: 150). The aforementioned moral denigration of ‘tribalism’ in Kenya's public sphere more generally finds its way into opposition rhetoric, castigating government supporters as gullible followers of an elite that manipulates them. The civic nationalism practiced by opposition supporters, in so far as it is used in campaigning and mobilisation as in the mkubwa’s speech, remains exclusive, only potentially universal, though nonetheless successful in the temporary fixing of its diverse multi-ethnic constituency.

CONCLUSION

This paper presented a case in which the rhetoric of Kenyan political opposition transcends ethnic particularism through the deployment of alternative concepts of political membership, concepts that draw on a historical experience of exclusion felt by a range of ethnic groups, and which ‘open up’ support for the opposition to groups disillusioned with the current state of affairs in Kenya. The work of producing this political constituency at street-level involves convincing distinct groups that they share a common goal – that they are related in their common democratic struggle. Whilst at times organising across ethnic boundaries involves elaborating notions of kinship, an important vehicle for this political mobilisation draws upon concepts capable of congealing distinct groups together to constitute a ‘people’ or ‘masses’ that have hitherto been denied the experience of representation.

An attentiveness to political language during moments of campaigning complicates and enhances impressions of Kenyan politics gleaned purely from the ballot box. An emphasis on ethnic-based loyalties alone cannot provide an adequate explanation for how Kenyan politics works ‘on the ground’, if you like, particularly in cosmopolitan spaces such as Nairobi where local actors recognise the need to combine and establish a mutual struggle. This is not to say that a seamless and perfect trans-ethnic universal citizenship is established. Far from it, exclusion remains a central aspect of opposition struggle though not purely because of ethnicity itself but via the contrast made – by a multi-ethnic opposition – between themselves (positioned as true civic nationalists) and their opponents that practice a partisan ethno-nationalism. Commitment to the grander scheme of a legitimate political rule in principle becomes a vital tool through which the opposition experiences and narrativises its struggle not merely as a partisan one, but as a moral struggle.

What this research suggests is that more than ethnic and potentially universal political identifications can be a fruitful ground for study, particularly in multi-ethnic contexts where cross-ethnic social relations permeate everyday life and have the potential to be mobilised via the right sort of rhetoric. As Jacob Rasmussen and Duncan Omanga (2012) show, the opening of political space in Kenya since the end of Moi's regime has given rise to several similar people's parliaments where cosmopolitan identities thrive, producing new notions of political subjectivity that complicate an expectation that ethnicity will always trump other political identifications that Kenyans possess. Whether civic identities, deliberately cultivated as they are, will prove to be any less exclusionary remains an open question, one that Kenyans will undoubtedly continue to debate over the years to come.