In 1958, Britain's Colonial Office initiated a policy discussion on the role of elections in the end of empire. The discussion ran over the next two years – even as the pace of decolonization quickened across Africa.1 Elections had become an occasional aspect of the negotiated transfer of power to British-ruled territories decades earlier; by the early 1960s, they were routine. But when a document – tellingly titled ‘Democracy in backward countries’ – finally emerged from the discussion in early 1960, it concluded that ‘Westminster democracy cannot be expected to work well in countries with low standards of living and education.’ Nevertheless, the document argued that, since international expectation and African nationalists demanded elections before independence, they must be held.2
The doubts expressed in ‘Democracy in backward countries’ were not unusual, as we shall show below. Yet British planning, as much as international opinion or nationalist demands, had come to focus on elections by universal adult suffrage and the secret ballot. Critically, these were relative novelties in Europe at the time; and in India, universal suffrage had followed several years after the end of empire, rather than preceding it. So why did they become the norm? Why did African politicians demand such elections? And why did the restive subjects of late-colonial states queue up to register and vote? What part, in short, did elections play in the entangled processes of decolonization and nationalist mobilization?
Prominent as they were in the end of empire, late-colonial elections attracted considerable academic attention at the time,3 but have rarely been studied by historians. That neglect is perhaps unsurprising: the rapid disappearance of multi-party politics across most of the continent within a few years of independence made these polls seem irrelevant, no more than a curious ritual precursor to the lowering of the flag. But universal suffrage and a (nominally) secret ballot never entirely went away.4 And, since 1990, the return of multi-party politics across much of Africa has drawn academic interest back to the ballot box. While the great bulk of the consequent literature has been explicitly concerned with the promotion and measurement of democracy, we take our inspiration from a more critical historical and ethnographic work – on Africa, and other parts of the world – that has asked how the secret ballot and universal suffrage have been imagined, and used, as ways to shape political subjectivities; that is, people's sense of political possibilities, and their own place in political action.5 We use those comparative insights to argue that late-colonial elections – which were vast administrative ‘operations’, as well as vivid political events – were not irrelevant. Nor were they – as a minority of studies have suggested – either pre-eminent tools of a British manipulation which shaped the whole process, or simply the innovation that unintentionally ‘dug the grave’ of empire (though the latter is perhaps closer to our view).6
We argue this through a study of three parts of British-ruled Africa – Ghana (formerly Gold Coast) in West Africa, and Uganda and Kenya in East Africa. Chosen to exemplify the variation of local circumstance, these case-studies also reveal commonality, showing how in each case imperial contradictions and nationalist ambitions combined to make elections seem a powerful tool with which to do different kinds of political work. The performance of the ballot narrowed down the multiple possible political futures that beckoned at the end of empire. They were able to do so because that effect was not immediately apparent to all those who were drawn into a vivid, emergent, electoral culture. Elections asserted the primacy of development as the purpose of legitimate government but simultaneously encouraged clientelism. They ushered voters into an apparently national citizenship but facilitated parochial politics. They created the impression of political equality but gendered political representation. In this particular moment of promise and uncertainty, elections captured the imagination and energy of administrators, politicians, and a wider public, who all looked to elections as they pursued multiple visions of citizenship and nationhood.
We begin with a brief narrative of elections in the three case-studies, which shows how British administrators and politicians first turned to elections in a bid to prolong empire by offering African allies the prize of privileged inclusion in a liberal project. However, African nationalists seized the opportunity to press more dramatic claims. As multiple forces unravelled empire, both British and nationalists committed themselves to a logic that foregrounded elections by adult suffrage as the expression of a powerful linkage of adulthood, citizenship, and sovereignty. However, while colonial administrators were drawn to elections as a way to discipline nationalist leaders, African politicians came to see these events as a means to legitimize their leadership, and ordinary people as a means to select those who could secure benefits for their community. These dynamics are important not only for understanding politics at this critical historical juncture, but also for understanding the longer-lasting appeal of elections during the one-party state and, more recently, into the multi-party era.7
All three territories were colonial creations, and their creation had been rooted in violence and in ideas of racial difference. Over those profound similarities, each was distinctive. Trade in gold, slaves, plant oils, and then cocoa had moved wealth, ideas, and people between what became Ghana and Europe and America since the sixteenth century; one consequence was the presence of an unusually large African elite, of chiefs, businessmen, and professionals (including lawyers). There were tensions within that elite: between chiefs and the educated, between those from the southernmost part of the territory and those from the kingdom of Ashanti.8 But self-confidence and a sense of entitlement united this group – and divided them from the great majority of the population, including those who had found opportunity, but little security, as petty traders, teachers, and clerks in the developing colonial economy and bureaucracy.9 Meanwhile, the north of Ghana lay largely beyond the economic, social, and educational networks, which bound the southern half of the territory together.
Ghana had no white settlers; Kenya, by contrast, was a settler colony, created in the rush of colonial expansion at the end of the nineteenth century. Alongside white settlers, there had developed a large Asian population, involved in commerce and in the colonial bureaucracy. There was little space here for the kind of African elite that had emerged in Ghana: Kenya's late-colonial African politicians mostly began their careers as teachers or as low-ranking employees of government. The settlers shaped the politics of the end of empire; clamorous and insecure, they had built a position of economic and political privilege which rested entirely on the state and its coercive power.10 The crucial moment came when, after bloodily containing an insurgency, the British government decided that it could no longer afford – economically, or politically – the cost of supporting that privilege.11 Yet that decision left open a further question: what would be the shape of an independent Kenya, and how would it be governed?12 Neighbouring Uganda had few white settlers, but many Asians, and its politics were often conducted in the shadow of debates over Kenya. Yet another factor was more important: the kingdom of Buganda. The establishment of British rule had rested on a tense, and often violent, relationship with this kingdom, and the kabaka, or king, had an accordingly privileged position.13 The hierarchy of chiefs under the kabaka, while sometimes in conflict with him, had developed both their status and their wealth through the kingdom's distinctive position within what was formally the Protectorate of Uganda; while a wider population in Buganda might resent the impostures of chiefs, the kingdom had a powerful affective hold, and commanded widespread support.14 The desire of both kabaka and chiefs to secure Bugandan privilege was a constant feature of the politics of the end of empire, in tension as it was with the aspirations of Uganda's emergent political parties, led by a group of teachers, clerks, and local traders from across the territory whose idea of the future offered no space for Buganda's distinct ambition.15
The colonial state in all three had revolved around the production and maintenance of racial difference; and in all three it had launched on late-colonial projects which looked to development for legitimacy, but had made available a language of progress that allowed African trade unionists and politicians to shake the authority of the state, opening up a sense of political possibilities that was both enabling and alarming.16 Those similarities underlay a pattern of late-colonial electoral history; apparent first in Ghana, then repeated a few years later in Kenya and Uganda. In each case – in Ghana in 1951, Kenya in 1957, and Uganda in 1958 – first elections were held in an attempt to breathe new life into the late-colonial state, on the way to a future self-government whose timetable still remained unsettled. In each, carefully devised conditions to control the outcome of those elections failed.17 The African politicians who won elected office refused to accept their assigned role as a privileged but subordinate elite. Instead, they called for rapid progress to independence. The door at which they knocked was already opening rapidly, pushed by an international context that was depriving empire of legitimacy and by a wider process of claims-making enabled by the late-colonial language of development.18 Empire was becoming unaffordable, politically as well as economically: for imperial powers, the pressing need now was to assert the legitimacy of the institutions of stateness that they had created – administration, police, judiciary – so that these might be passed complete into African hands.
Former radicals were now moderates in British eyes, their electoral ambitions suddenly desirable; and in each territory second elections were held – in Ghana as a step to self-government, in Kenya and Uganda to produce what were called ‘responsible’ ministers, now with the expectation that independence was imminent. In each case, the outcome of those second elections proved problematic. In Kenya, continued British equivocation over the future privileges of the settlers combined with – and encouraged – competing African views of the future to produce a ‘frustrated election’ in 1961, which increased rivalries and produced no clear result.19 The apparently conclusive results of Ghana's 1954 elections were quickly clouded by demands from Ashanti, and the north, for distinctive political status, articulated most vociferously by the ambiguously named National Liberation Movement (NLM).20 Uganda's 1961 elections were almost entirely boycotted in Buganda. That boycott delivered victory to one of two competing parties, but at the same time robbed the elections of legitimacy.21
In each case, third elections were held, as British politicians looked increasingly frantically for a way to pass imperial responsibilities to Africans. The precise balance of dynamics varied: in Ghana, it was the British who really insisted on these last elections – confident as they (rightly) were that Kwame Nkrumah's Convention People's Party (CPP), which had won in 1951 and 1954, would secure a clear victory and ensure a unitary constitution for independent Ghana.22 In Uganda, the British found common ground with the Uganda People's Congress (UPC), the party defeated in 1961, when the latter made a deal with the kingdom of Buganda. While both British and the UPC had previously insisted on direct elections across Uganda, it was now agreed that Buganda would hold elections to its own parliament, the lukiiko, which would then choose Buganda's representatives to the new ‘national assembly’. With the support of those indirectly elected members, the UPC emerged with a clear majority from the 1962 elections, and formed independent Uganda's first government.23 In Kenya, particular British concerns (over the settlers, military bases, and political complexion of a future government) focused on the perceived danger posed by radicals in what was evidently the most powerful party, the Kenya African National Union (KANU) – a party that, like the Colonial Office in London, favoured a unitary and centralized political future for Kenya. The British briefly hesitated over elections, hoping that a split in KANU would allow ‘moderates’ to join with the rival Kenya African Democratic Union (KADU) – whose ideas on economics and property rights the British favoured, even as they doubted the viability of KADU's plans for decentralized administration.24 But then the urgent sense of the need to hand over responsibility impelled a sudden acceleration; polls in May 1963 were followed by independence under a KANU government at the end of that year.
There was much that was rushed and opportunistic about these events, and it is hard to see a consistent plan on the part of any involved. Yet the repeating pattern of elections – in these, and other, territories – makes more insistent the query: how and why did multiple different actors, across Africa, pursuing different projects, turn to the ballot?
As Karuna Mantena has argued, the British presented both a universalist justification and a culturalist alibi to explain their empire.25 These were mutually contradictory. The first was avowedly liberal: empire, it suggested, was a tutelary project that sought to spread British institutions, and the goal was to turn subjects into citizens, and colonial possessions into nations. The second was rooted in difference and racial categorization: empire was necessary because its subjects were inherently incapable of ruling themselves justly, and required the British to govern them.26 The tension between these visions ran through an imperial history which can be written in terms of the episodic dominance of one or the other; after 1940, the ‘universalistic, progressive project’ was in the ascendant, with development – constitutional, as well as economic and social – as the key term.27 Yet the universalist vision entailed its own contradiction: the imperial claim to authority required the constant reproduction of the difference that it claimed to eradicate.28 After 1945, that contradiction manifested itself repeatedly, as British advocates of development and modernization expressed their uncertainty over African capacities in what Uday Mehta has called the language of ‘presumed infantilism’; Africans were children, not yet ready for a future for which they required more education.29 The Kenyan trade unionist and politician Tom Mboya ventriloquized that strategy of exclusion with bitter accuracy: ‘“You cannot be free because you are not sufficiently educated.”’30
Adult suffrage and the secret ballot exposed this contradiction most fully. British commentators could – and did – point to metropolitan history for evidence that the slow extension of the franchise was the root of political stability.31 The cautious schemes to develop representative African local government mooted in the 1940s were predicated on the belief that few Africans were ready for the vote: they implied the possibility of a universalist vision of adulthood and citizenship, yet withheld it from most Africans in a language that linked education, responsibility, and experience.32 British observers repeatedly insisted both that irresponsibility made most Africans unfit for suffrage and that illiteracy made them practically unable to cast a secret vote; as The Times put it, ‘[t]here are mechanical means whereby primitive tribes can be made to go through the motions of the ballot, but such a process is neither dignified nor meaningful’.33 Ghana's 1951 elections chose only thirty-eight of the eighty-four members of the assembly; the rest were nominated by councils of chiefs, by business and industry, or by the governor. Of those thirty-eight, only five were directly elected; the rest were chosen by ‘electoral colleges’. Even this degree of popular choice was regarded by many Britons with incredulity: this was ‘handing a native a pistol and carefully explaining its mechanism’, wrote one journalist, while others wrote multiple, mocking, stories of African incomprehension in the face of the ‘white man's new magic’ of the ballot box.34
The CPP's success against this loaded system drove an immediate British reappraisal of Nkrumah. Previously denounced, he was now embraced as a moderate. He used that position to demand direct and secret voting by adult suffrage in the next elections. Alarmed as they were by the idea, the British accepted this as the price of moderation.35 But at same time in Kenya, it was still assumed that adult suffrage ‘will not be possible for many years to come’.36 The presence of white settlers in East Africa made the ‘ignorant, vote-endowed savage’ an unacceptable prospect.37 The aim, instead – as it had been in Ghana only a few years before – was that a managed franchise would reward a modest cohort of responsible subjects and ensure their loyal co-operation in continued British rule: ‘the vote must be reserved at present for those who, by reason of their education, achievements or experience can be expected to make sound use of it’.38 In 1957, a British parliamentary delegation to Kenya reasserted the point: universal suffrage was impossible where the population was ‘immature and undeveloped’.39 When, in 1959, the electoral expert William Mackenzie argued for a rapid extension of the franchise in Kenya, on the grounds that the only way to develop responsibility was to allow Africans to exercise it, British officials and politicians insisted that Africans were unready for the individual responsibility of the vote – and therefore unready for collective self-government.40
But in East Africa, as in Ghana a few years earlier, adult suffrage was rapidly conceded when it became clear that British rule was unsustainable, and the urgent need was to find an ‘inheritance elite’ to whom power could be handed.41 At that point, the urgent need to rationalize and confer respectability on this handover made the universal franchise suddenly attractive as a way to conjure a citizenry, and a nation.42 The crucial moment in East Africa was between late 1959 and early 1960; even as the report on ‘Democracy in backward countries’ was in preparation, a statement by the Commonwealth Relations Office affirming the principle of the qualified franchise was drafted, and then abruptly suppressed.43 Elections now offered a way for the British to reassure themselves – and persuade others – that they were handing power to men who had been moulded into the ‘norms of modernity’.44
For many Africans too, the contradictions of late-colonial thought confirmed that voting was the badge of maturity. Against the infantilizing rhetoric of late colonialism, ‘adult suffrage’ had double significance: to have the vote was to be accepted as a responsible adult, as well as to have the right to participate. Long before the Universal Declaration of Human Rights claimed adult suffrage and the secret ballot for all, the franchise had marked status: in Kenya, white settlers, Asians, and Arabs could cast individual ballots from the 1920s, and the political voice of white privilege by the 1940s was the ‘Electors’ Union’.45 In 1945, the exile intellectuals of the Pan-African Congress had demanded adult suffrage.46 Mackenzie's writing, intended for colonial audiences, made a further point explicit: elections by adult suffrage and secret ballot were the ultimate mark of legitimate government.47 Since the 1920s, international precedent had, haltingly, suggested that elections expressed a collective claim to sovereignty: to hold elections by adult suffrage and secret ballot was to prove national maturity and readiness for self-rule.48 Colonial reluctance underscored the value of the vote for African nationalists, as a practical tool of politics, and as a claim to a future in which Africans would be equal. African nationalism was a particular kind of challenge to colonial rule. It defied the infantilizing logic of difference, and looked to a future of equality, of progress, of science – of change.49 Election was not in itself a novel idea in late-colonial Africa; customary practice and the associational forms of colonialism – school, local government, co-operatives – had made familiar the techniques of ‘acclamation’ at open meetings, public voting by queuing or by show of hands, and indirect voting by delegates. But by the late 1940s, nationalists focused political claims on the adult franchise and the secret ballot; in Ghana, it was at least plausible to claim that there was a ‘popular cry throughout the country for universal adult suffrage’; by 1952 in Uganda, the demand for freedom and equality ‘in this fast moving civilisation of the Atomic Age’ was seen as bound up with adult suffrage.50
In Kenya, the logic of liberal universalism became nationalists’ weapon against settler claims to privilege; their demands for adult suffrage expressed this in terms of the ‘equality’ and ‘civic responsibility’ of the ‘individual citizen’.51 Nationalists gleefully appropriated the logic of electoral progression to adulthood, arguing that in 1957 ‘newly franchised citizens’ in Kenya had ‘displayed an enviably high degree of political maturity’.52 The call for ‘undiluted democracy’ became ubiquitous, and inseparable from the ideas of ‘full nationhood’ and equality. In this vein, Mboya wrote an open letter to Kenya's white and Asian political leaders in 1960, demanding electoral equality: ‘The African, for a long-time a third-class citizen in his own country is conscious of the rising status of fellow Africans elsewhere…He simply wants to be accepted as a fellow and equal human being.’53 When the government in Uganda refused to accept adult suffrage for Africans, Obote denounced this as a ‘complete degradation of the African mentality’; in plangent tone, another Kenyan politician explicitly linked adult suffrage, independence, and adulthood: ‘we know that only in independence shall we be respectable men and women of dignity’.54
Initially hostile, the British came to welcome this once the early transfer of power became the goal: if colonial territories were ready for self-government, then their people must be citizens, ready for the responsibility of the vote. In 1954, Frederick Crawford, then deputy governor of Kenya, had mocked African enthusiasm for the secret ballot and adult suffrage in terms that echoed the British press coverage of Ghana's 1951 elections: ‘they believe there is some juju in it’, he sneered. By the end of the decade, as governor of Uganda, he declared that adult suffrage was ‘desirable both politically and practically’.55 The secret ballot was the unquestioned corollary of adult suffrage; debate over how to make this possible for illiterate voters led to experimentation and variety of practice which quickly gave way to a uniform system within each territory. In Ghana, where voters deemed to be educated had marked their ballot papers in 1951, the need for a ‘common electoral system throughout the country’ led to a system of multiple ballot boxes – with voters placing an unmarked paper in the box of their preferred candidate – from 1954; the ballot boxes themselves were screened from view to make the vote secret.56 The same procedure came to be used in Uganda; in Kenya, by contrast, all voters were expected to mark a choice on their ballot paper, and all papers went into a single box.
These abrupt changes in British policy were driven by imperial expedience. But they were possible partly because elections had – as elsewhere in Africa – become a means to pursue the disciplinary ambitions of late colonialism.57 Colonial officials had long understood their role as one of bringing order to Africa, and had habitually combined the open violence of the bullet and baton charge with the insidious power of the list and the rubber stamp. ‘Operation elections’ – as one official called the 1951 polls in Ghana – offered a new field of endeavour in both authorizing and euphemizing coercion.58 As the electorate grew rapidly, elections – especially the final pre-independence elections – became massive events involving the mobilization and display of state power: 126,000 voters registered in Kenya for the 1957 elections; more than two million for those in 1963. Continuing the military metaphor, Kenya's governor called the 1961 elections ‘an outstandingly successful operation’.59 These were powerful performances, which cast the population as responsible citizens – the focus of an educational campaign of leaflets, posters, and instructional films which taught them how to register, queue, and cast their ballots under the watchful eye of police.60 These were, as British officials and commentators repeatedly asserted, ‘orderly’ events.61 Their cast included an emerging cadre of African administrators, who worked with their British mentors to run the electoral process, requisitioning vehicles, typewriters, and staff from every government department to do so.62 It also drew in thousands of – mostly male – extra polling station staff.63 These were often teachers and senior school students, who had the necessary skills of literacy and a familiarity with lists and colonial spatial order, and whose participation in elections flattered their own sense of distinctive status – particularly since polling stations were often in schools.64 The audience for this performance was, in part, the cast itself; through their involvement, they could see themselves anew, as citizens involved in an event that manifested the ‘triumph of bureaucratic discourse’.65 But there was a wider audience too: elections would ‘demonstrate to the world’ that a territory was ready for independence, as the secretary of state assured Nkrumah in 1956.66 Public, as well as politicians, were reminded of the global gaze: ‘the eyes of the world are upon us’, announced Kenya's new, avowedly national, newspaper on the eve of the 1963 elections.67 As empire threatened to tumble about their ears, the British turned to this vision as validation – this, after all, was what they had meant to do all along.
British politicians and administrators pretended – to themselves, as much as to others – that adult suffrage and the secret ballot marked the culmination of the imperial project, rather than its collapse. They were able to do so because nationalist politicians were willing partners in these electoral projects: ‘expatriate administration and local politics walk[ed] in the same direction’.68 As argued above, nationalists seized on the contradictions in British alibis of empire to lay claim to the adult suffrage as a symbol of individual and collective maturity. But late-colonial elections did other work for nationalists. They too were concerned with discipline, and elections affirmed their distinctive status and claims to power.
The major nationalist leaders of Ghana, Kenya, and Uganda were all men; all with some education – some acquired locally, some the result of educational pilgrimages elsewhere in the world. All were Anglophone, and familiar with the forms and demands of bureaucracy. Nkrumah had studied in the US and spent time in the UK; Kenyatta had spent many years in the UK; Obote, after a brief period of higher education, had worked in Nairobi as a clerk. The local networks that they relied on to mobilize voters were also dominated by men with some limited school education: clerks, primary school teachers, and small businessmen. This was a time of hope, as well as discontent, and nationalists were well aware that popular discontent, and ambition, took many forms.69 Nationalists – and others – had encouraged defiance of colonial authority as they harnessed multiple resentments and ambitions in their cause: Nkrumah and Kenyatta had both been imprisoned. With independence imminent, nationalist politicians in Ghana, Kenya, and Uganda, like others across the continent, faced the task of reasserting authority: persuading the populace to become responsible citizens of a developmental state, and to forget the other ‘futures that might have been’.70
Those other futures seemed alarmingly possible to those who considered themselves nationalists. It was not simply that colonial subjects had become unwilling to pay their taxes or obey officials; some actively sought to dismantle the state. In Kenya, Maasai dreamed of an ethnonational state that would break the borders of Kenya and Tanganyika; ethnic Somalis supported the irredentist claims of the Republic of Somalia; coastal secessionists denied the very possibility of Kenyan sovereignty.71 In Uganda, the ambitions of Buganda threatened the very heart of the state; melodramatically announcing their plans for a separate independence, the members of the lukiiko insisted that ‘Buganda cannot sell her heritage for the purchase of Uganda's independence.’72 In Ghana, dreams of a separate independence for east, or north, or for Ashanti, led to talk of civil war.73 For late-colonial politicians, the survival and authority of the colonial state were by no means assured.
So it was that in Ghana's first elections, administrators were delighted to find that activists of Nkrumah's CPP, who had established themselves as constant critics and opponents of government, were now urging people to pay tax and register as voters.74 In Uganda, where the abuse of power by chiefs had been a constant popular complaint, nationalist politicians showed little interest in challenging the effectively forced registration of voters in 1958 (despite local grumblings).75 In all three countries, over the cycle of pre-independence elections, politicians whom the British had previously regarded as dangerous radicals increasingly joined colonial administrators in urging the importance of law and order.76 They took up the same martial metaphors favoured by colonial officials: Nkrumah announced that the 1954 elections were ‘Operation 104’, as the CPP would win every one of the 104 seats.77 Like colonial officials, they evoked an international gaze to demand that the public ‘demonstrate to the rest of the world that Uganda people were mature and capable of governing themselves’.78 And they used the election campaign to make multiple disciplinary demands: ‘You must be peaceful and law-abiding citizens to expect a good return’, one KANU leader told voters; both CPP and KANU used their manifestos to call for ‘hard work’.79
Elections privileged the educated, and gendered participation, in another even more direct way. To be nominated as a candidate required the payment of a cash deposit, the completion of forms, and the collection of signatures, all to be presented at a set time and place for approval.80 To be a successful candidate required not just popularity, but the ability to mobilize voters to register, helping or encouraging them to fill in forms. Formal educational qualifications for candidates were expressed in terms of a requirement for literacy and fluency in English – a restrictive demand at the time. In Ghana, the requirement was not policed, but in both Uganda and Kenya, local language boards were formed which administered oral and written tests, which significant numbers of would-be candidates failed.81 This combination of practical and regulatory demands excluded most men, and almost all women, from elected office. Among hundreds of aspirant candidates in these late colonial elections, women were only a handful: three in Ghana in 1954; one in Uganda in 1961 and again in 1962; one in Kenya in 1963. Just one of these candidates was successful, in Ghana in 1954. Those elected were, very largely, men in their thirties and forties; teachers were the single largest group.82 Women and older men could be, and were, voters and activists; but the electoral performance and socio-economic realities excluded them from other roles.
Meanwhile, to question the value of adult suffrage and the secret ballot became almost impossible. In the late 1940s in Ghana, the committee of African professionals and chiefs who recommended universal suffrage had balanced their endorsement of the educational value of this with a recommendation for indirect voting in much of the territory, as a safeguard against an unpredictable popular vote.83 But after Nkrumah's victory in 1951, none challenged either adult suffrage or the direct, secret ballot. In 1956, the opposition to the CPP sought local autonomy and an upper house of parliament dominated by chiefs: the constitution would be the curb on the consequences of adult suffrage.84 Accused by the CPP of showing ‘unreadiness for self-government’, they lost the election.85 In Kenya – where the African elite was much smaller, and even more fractured – some reportedly expressed doubts about adult suffrage in the early 1950s to the inquiry on ‘possible methods of African representation’ undertaken by a white colonial official, Walter Coutts.86 But by the early 1960s, no African politician called publicly for a limited franchise. Those who feared the consequences of direct adult suffrage instead sought protection in the constitutional arrangements for an upper house and regionalism.87
Only in Uganda was there a persistent – and successful – challenge to the very principle of direct adult suffrage for the nascent nation. A tiny educated elite had suggested a qualified franchise in the early 1950s; in 1958, the district council in Ankole had rejected even a qualified direct franchise, as the council was already composed of ‘the people who speak English best and represent Ankole well’.88 But it was the attitude of Buganda's elite that proved most important. For the kabaka’s subjects to vote in a way that cast them as citizens of a putative Ugandan nation was unacceptable.89 However, voting itself could be acceptable: Buganda's 1962 lukiiko elections successfully adopted all the forms of pre-independence ‘national’ polls: registration, polling stations, ballot boxes, and papers.90 The kingdom borrowed the services of the Uganda's supervisor of elections – a British official – and sent the bills for the costs of the elections to the Protectorate government.91 Yet the Buganda elections were run as a kingdom event, not a national one. Managed largely by the chiefs, they allegedly saw widespread intimidation and other malpractice.92 This was Bugandan ‘neo-traditionalism’ at its most creative, seizing the forms and processes devised to assert a unitary Ugandan state, and using them to subvert it: 89 per cent of registered voters turned out to cast ballots under the watchful eyes of their chiefs.93
Nationalist politicians and activists, then, seized upon the ballot both as a demonstration of collective political maturity and as a tool to induct a restive public into citizenship. Through elections, they sought to remind the public of the state's ambitions to manage the passage to the future and to prevent the fabric of that state from crumbling just as they took possession of it. Yet that success entailed the abandonment of other possible futures too, though none quite realized this at the time. The essence of African nationalism was futurity, and its rhetoric was often ambitiously pan-African; while vehemently opposed to breaking down the colonial state, nationalist leaders readily expressed enthusiasm for the project of regional and continental unity. But as some foresaw at the time, electoral reliance on the administrative architecture of the colonial territory – the basis of the alliance of interests between colonial officials and emergent nationalists – helped undermine both regional integration and pan-Africanism.94 Pinned to electoral legitimacy, the colonial territory became indistinguishable from the aspiring nation.
Why did significant numbers of people register, and vote – subjecting themselves to bureaucratic process, and queuing at the polling station under the watchful eye of the police to have their names ticked off and their fingers inked? What kind of citizenship, or adulthood, did they hope for? Coercion played a role. The colonial state was violent – most dramatically in Kenya, but in the other cases too, people had become habituated to the physical enforcement of state demands. When chiefs in Ghana ‘beat gong-gong’ to call their subjects to register as voters, they evoked a history of occasional violence and intimidation. In Kenya and Uganda, the summons from tribal police or chiefs’ messengers was the everyday reality of a coercive state. The processes of elections, reliant as they were on that administrative machinery, constantly reminded the public of the state's potential for force. But the success of the electoral process lay in its ability to do more than justify or euphemize violence. While nationalists saw the elections as a disciplinary performance, voters may have understood them rather differently: as Peter Pels has observed of late-colonial elections in Tanganyika, this kind of event ‘always carries a potential for multiple interpretations’.95 The ballot offered a contract: if people voted, their chosen representatives would bring them the prize of future prosperity.96
Formally, parties dominated the final elections before independence in all three countries. A feature of electoral politics in Ghana from the 1951 election, parties developed a little more slowly in Uganda – where they played an uncertain role in the 1958 elections – and in Kenya, where national parties were formally banned until the end of the 1950s. Organizationally, these parties varied significantly in their abilities, but even the CPP – perhaps the most effective – was chronically unable to enforce internal discipline and was dogged by disputes over the choice of local candidates.97 All parties rested on a larger or smaller coalition of influential individuals spread across the country, who lent their local support in the expectation that this would bring them personal benefits and further enhance their local status: the fundamental assumption was that control of the government, and therefore of the state, was the key to prosperity.98 In Uganda, this took religious as well as individual, local, and ethnic form; the popularity of the main rival to the UPC, the Democratic Party (DP) was driven partly by the sense among Catholics that the colonial state had systematically favoured Protestants.99 The state sprawled across the late colonial formal economy: business licences, government contracts, access to market plots, public employment were all – in one way or another – in the hands of government. So too, of course, were the improving projects to which late colonialism was devoted – clinics, schools, water supplies, roads. The party networks, such as they were, were held together by an optimistic sense of the possible rewards of office. Across Ghana, Kenya, and Uganda, campaigns rested not on the manifestos – which all parties produced, in English, for the tiny group who could or would read these – but on the spoken word. Though it was delivered partly by radio – in broadcasts that were subject to a degree of censorship that diminished over time, and which in Kenya were made only by explicitly ‘national’ parties – the real medium for this spoken message was a combination of rallies, smaller public meetings, and canvassing.100
In Ghana, the rallies were from the start noisy and ebullient.101 In Uganda and especially in Kenya, they began as much more constrained affairs, closely scrutinized and managed.102 But they soon became more noisy and theatrical, and occasionally violent.103 In Kenya, in particular, the police remained a prominent presence, ready and very willing to intervene with batons and tear gas.104 Against the bureaucratic order of the register and the polling station, rallies – when not curtailed by the police – performed an order of personal ties and affective loyalty. Women were prominent in them, not usually as speakers but as singers, dancers, and as part of the audience which prayed, and sang and shouted, and sometimes jeered.105 For the parties that emphasized rapid independence, rather than constitutional safeguards – CPP, UPC, KANU – the promise of the rallies was simple: to vote was to secure the future. As the UPC's John Kakonge told a crowd in 1961, immediate independence was necessary because ‘“[w]e are in a terrible hurry to bridge the gap between our poverty-stricken country and the prosperous nations of the western world”’.106 By contrast, those parties – like KADU, or NLM – that made an explicit appeal to regional loyalties mixed the promise of rewards to a local following with warnings of the danger of domination by others.107 Rallies involved the promise of more specific, often mundane, material rewards: from clinics and piped water to jerseys for fishermen.108 So it was that in the wake of the 1963 elections, KANU's central officials were bombarded with complaints from branch officials and activists who had lent vehicles, given fuel, danced and campaigned for the party, and now demanded jobs, contracts, and – quite simply – payment.109
Colonial populations had experienced the state as unpredictable and wilful. They had been told repeatedly that development would transform their lives, yet suspected that the real benefits of prosperity were being withheld from them. Elections embodied an exhilarating offer. This was the chance, at last, to make their own demands on the state. Yet voters were not convinced that the ballot itself was sufficient to ensure accountability; they wanted candidates on whom they had other moral claims, of kinship or shared ethnicity, or at least locality. Such claims were reinforced by the everyday clientelism which vivified social relationships in all three countries: attendance and donation at funerals, petty financial help, and favours all embedded the personal nature of political relationships.110 Nkrumah's repeated attempts to control the selection of CPP constituency candidates were always undercut by patterns of local preference and patronage; official KANU candidates faced ‘KANU independents’ who claimed local endorsement.111 From the outset, campaign costs included displays of wealth, and the buying of food and drink for voters, intended to demonstrate the candidate's acceptance of this responsibility to share wealth with one's community.112 While administrators imagined the ballot as an assertion of the primacy of bureaucratic rationality and orderly citizenship, voters sought to use it to root the new political dispensation in a more familiar logic of moral accountability – a citizenship rooted in clientship, ethnicity, and locality, rather than the nation: as one observer noted of Uganda's 1958 elections, ‘The voting public appeared to be insistent upon learning with some precision from the candidates what each one thought he could do…to benefit the people of his district.’113 Aspiring leaders, drawn as they were to the vision of the bureaucratic state, were required to join in that local negotiation of accountability. Against this backdrop, it was not long before voters were using elections not to perform obedience, but to accuse their representatives of aiming to enrich themselves rather than reward their communities: ‘they say we are seeking big cars’, as one candidate mournfully told a newspaper reporter.114
British politicians and officials made sense of the collapse of their authority through elections which pretended to resolve the contradictions of empire, and looked to the possibility of an imperial afterlife rooted in institutional continuity. African politicians and activists, their gaze fixed on a prosperous future, had seized upon the ballot as a means to demand independence by demonstrating the maturity of the new nation, and to discipline its new citizens. Yet despite the hopeful imaginings of political theorists – and the cheerful pretence of politicians and officials – adult suffrage and the secret ballot did not immediately remake the dissatisfied subjects of colonialism into the citizens of the new nation. Faced by a state that had created itself through violence, and had existed beyond moral accountability, yet which promised development, people across Ghana, Uganda, and Kenya willingly voted. But when it came to making claims on those whom they sent to wrest the prizes of development from the state, they looked for accountability not to the bureaucratic rationale of the self-consciously modern state, but to personal ties, to kinship, ethnicity, and locality.
In all three cases, the glum prophecy of ‘Democracy in backward countries’ initially appeared to be accurate – at least on one level. Elections and government on the Westminster model were not maintained, though the path away from them was different in each case. In Ghana, Nkrumah's pride in his success at the ballot was mixed with an uneasy awareness that parochial preference, rather than CPP edict, determined the success of parliamentary candidates. The combination led him to increasingly authoritarian post-independence electoral experiments which culminated in the farce of the 1965 elections, in which no ballots were cast and the CPP selected all the members of parliament. Less than a year later, the military overthrew him, to public celebrations. In Kenya, the administrative hypertrophy and habits of violence consequent on the counter-insurgency campaign of the 1950s meant that elections were particularly closely organized, and heavily policed. Here, more than elsewhere, a new cadre of African administrators and security officers could see elections as an exercise in order and an expression of their power, not a challenge to it; and independent Kenya continued to hold regular elections, though under an effective single-party system.115 In Uganda, the opportunistic deal agreed by Obote and the British to secure the 1962 elections effectively endorsed Buganda's claims to a distinct status: those who lived in the kingdom were the kabaka’s subjects before they were citizens of Uganda. The result was a particularly weak independent state, lacking legitimacy at its geographical and economic core, in which every subsequent attempt at an election has been problematic.
Yet while neither imperial nor nationalist plans were to be quite what they claimed, the entanglement of the end of empire with the electoral process was of lasting significance. In these three territories, as elsewhere in Africa, elections helped mobilize popular endorsement – or at least acceptance – of a decolonization process that turned multiple possibilities into a single future. Elections narrowed the field of nationalist action to the colonial territory. They exalted literacy and formal education, gendered access to political office, and conferred primacy on a group of relatively young men who were to dominate politics for the next two decades. At the same time, the nature of political competition circulated and normalized a modernist late-colonial rhetoric which insisted that legitimate authority rested on development.
Late-colonial elections – and the politics around them – also embedded a set of linkages: between adult suffrage by secret ballot and national and individual adulthood. The idea that adults had the right to vote, and that government should derive its authority from the ballot, proved far more enduring than the idea that elections should involve popular choice over policy.116 At the same time, the behaviour of voters and politicians had foregrounded the idea that, in the pursuit of development, the task of elected candidates was to secure benefits for their community – local, or ethnic. ‘Operation elections’ may have been conceived as a national project; but the parochial logics of clientship which underwrote its apparent success were to persist, and continue to shape electoral politics in Africa to the present.