In December 2002, the National Rainbow Coalition (NARC) swept to power promising ‘comprehensive political and economic changes in Kenya.’ These reforms included the rapid ‘completion of the current constitutional review process’ (NARC, 2002:vii), which was finally completed in March 2004 after much infighting amongst the coalition. It was not until July 2005 that the draft of the new constitution was finally presented to parliament for debate and amendment. The draft that parliament ultimately put forward for public approval by referendum was very different to that agreed at the National Constitutional Conference. Most contentiously, the redrafted constitution retained the powers of the president and central government (Kenya, 2005). As Raila Odinga, an opponent of the redrafted constitution, argued, ‘The power of the President has not been devolved. Instead, power has been concentrated on the President’ (‘Too Much Control Vested in the President’, Daily Nation, 12 October 2005). Controversially, the draft approved by parliament substantially reduced the responsibilities of the proposed office of prime minister, intended to act as a check on presidential power. Less attention was paid to the proposed fate of the bureaucracy, despite the importance of the prefectural provincial administration to Kenyan politics. While MPs acceded to the demise of the provincial administration, they agreed that the body would be replaced by an elected district administration that would remain answerable to central government (‘Narc Wins Crucial Vote to Change Bomas Draft’, Daily Nation, 22 July 2005). Disagreement over the draft constitution split the already fractious NARC coalition. Ahead of the referendum held in November 2005, cabinet members openly campaigned against one another. Ultimately, the ‘no’ campaign emerged victorious, with 58 per cent of those voting rejecting the new constitution.
With the defeat of the proposed constitution, Kenya's existing ‘top-heavy’ constitution remains in place. The alliance between the bureaucracy and the executive, which has dominated governance for the past century, has survived the most recent assault upon its privileged position. The longevity of this alliance cannot be explained solely by a sequence of historical events. Instead, a deeper structural and theoretical approach is necessary. This paper represents an attempt to carry out this project by locating the Kenyan post-colonial state in the universe of authoritarian regimes and by delineating its distinctive qualities. In so doing, we borrow heavily from the literature on bureaucratic-authoritarianism to develop an understanding of how the post-colonial state was shaped by processes of class formation and institutional development during the colonial era. We argue that the Kenyan post-colonial state represents a particularly strong combination of administrative and executive power underpinned by an alliance of elites. In the immediate post-colonial period the legitimacy of the executive and the coercive capacity of the administration resulted in the consolidation of a highly resilient ‘bureaucratic-executive’ state. This bureaucratic-executive state, we suggest, should be seen as an outlier among sub-Saharan African states because of its strength in both urban and rural areas. In making this argument we are challenging the claim, explicitly stated by Zolberg and often implicitly assumed elsewhere, that African states can control the ‘centre’ but not the ‘periphery’ (Zolberg, 1966 & 1968; Chabal & Daloz, 1999; Jackson & Rosberg, 1982).
In developing an understanding of the bureaucratic-executive state it is imperative to escape the intellectual confines inherent in the adoption of the temporal categories of ‘colonial’ and ‘post-colonial’ (for an example of this approach see Cooper, 2002). Avoiding this constraint provides a necessary corrective to the tendency of some studies of post-colonial states to ‘oddly privilege colonialism’ by describing the irrationality and disorder of post-colonial states without a comparable examination of the colonial period. Such an approach, Ranger argues, tarnishes the post-colonial state through ‘an unintended implied contrast with a period of relative stability and competence’ (Ranger, 1996:273). Of course, we must also beware of any re-evaluation of the state in Africa that simply projects notions of post-colonial instability back into the colonial era. Chaos is fast becoming the normative state of African politics. We do not believe asking ‘Why has Kenya not experienced a full-blown civil war?’ (Kimenyi & Ndung'u, 2005) is a particularly useful analytical approach. In contrast to the prevailing and generalised focus upon disorder and criminality (Bayart, 1993; Bayart, Ellis & Hibou, 1999; Chabal & Daloz, 1993), this paper is a study of order within a particular African state, or at least the political aspiration for order.
We understand order to be ‘the recognition of patterned regularity in social and political life’ (Lieberman, 2002:698), and to be an aspiration of governments during the period in question rather than a description of the lived experience of the various regimes. In 1966 Daniel arap Moi attempted to excuse the political biases of Kenyatta's regime by claiming that ‘even if it's a political Government, it is an orderly Government, it is not a Government of disorder’ (quoted in Gertzel, Goldschmidt & Rothchild, 1969:126). This desire for order encapsulated an intolerance of dissent, the maintenance of profound social inequality and a determination to maintain control for its own sake (Throup & Hornsby, 1998:42-47). Whether or not order actually characterised the political system is another question, as the periodic but carefully targeted recourse to the tool of political assassination suggests. However, while the Kenyatta regime made increasing use of nepotism and the extra-legal powers of the provincial administration (Gathaara, 1982), only after 1978 did the Kenyan political system begin to resemble the informalised models outlined in recent scholarship (Chabal & Daloz, 1999:1).
In understanding post-colonial continuities in the Kenyan case it is important to recognise that states can be Janus-faced - chaotic and ordered at the same time. As Berman has written, ‘The most striking characteristic of the colonial state was the ambiguous, indeed, contradictory character of its structures and processes’ (Berman, 1992:141). For Cooper, in modification to Foucault's capillaries where power is endlessly diffused, colonial power was arterial:disseminated in irregular spurts. In most colonies, the state's influence was most keenly felt close to ideological and spatial centres of colonial governance (Cooper, 1994:1533). The colonial state could thus be simultaneously powerful and weak. Similarly, Chabal and Daloz have argued that the post-colonial state is ‘both strong and powerless, overdeveloped in size and underdeveloped in functional terms’ (Chabal & Daloz, 1999:9). The extent to which post-colonial states can be therefore said to be substantially different from their colonial predecessors, either by being ‘Africanised' or in their capture by society, appears questionable.
We follow Lonsdale and Berman in understanding the state as ‘the historically conditioned set of institutions in any class society which, more or less adequately, secures the social conditions for the reproduction of the dominant mode of production, in this case capitalism’ (Lonsdale and Berman, 1979:489). Here our focus is two specific institutions, namely the bureaucracy and the executive. Whilst we recognise that the formal institutions of government constitute but one actor in the theatre of politics, institutions are particularly useful for our purposes, because they lend themselves well to attempts to theorise the development of the state over time. This is neither the first attempt to develop such an understanding nor the first recognition of the importance of colonial institutions to post-colonial developments in Kenya and beyond (Firmin-Sellers, 2000; Gertzel, 1970:1-31; Mamdani, 1996; Mueller, 1984). However, this paper breaks new ground by developing a theoretical understanding of the Kenyan post-colonial state that supports a more direct comparison between the Kenyan state and other examples of authoritarian rule.
Although beyond the scope of this paper, we do not dismiss the ability of Kenyans to manipulate, subvert or simply ignore the state and its institutions. Recent historical, anthropological and empirical-based political analyses have extensively detailed the practice by Africans of ‘agency in tight corners’ (Lonsdale, 2000). Yet there is a need to periodically reconsider just how tight those corners were. Our intention is to explain the enclosure and asphyxiation of formal political space so as to defeat challenges to an elitist political and economic settlement. It is this process which forced opponents of colonial and post-colonial Kenyan regimes to seek the imaginative and alternative vehicles to express their discontent that have been discussed elsewhere. The Mau Mau insurgency of the 1950s is perhaps the most extreme demonstration of this phenomenon, but informal and less publicised methods of political protest are equally valid examples (Haugerud, 1995).
The post-colonial state
Scholars’ awareness of the post-colonial continuities in the structure of the political-economies of African states first emerged with the dependency theorists of the 1960s (see for the Kenyan example see Leys, 1975). Subsequent studies have been less willing to adopt the vocabulary of neo-colonialism, preferring instead to identify post-colonial discontinuities (Young, 1994:9-10). Nevertheless the case for viewing contemporary African politics in the longue durée remains compelling. We remain unconvinced that ‘From whatever point examined, what we are witnessing in Africa is clearly the establishment of a different political economy and the invention of new systems of coercion and exploitation’ (Mbembe, 2001:93). Kenya presents an example of the manner in which the capacity and knowledge structures of government developed during colonial rule continue to shape the post-colonial state. As Mbembe concedes, ‘post-colonial African regimes have not invented what they know of government from scratch’ (Mbembe, 2001:24).
Theorising the development of the Kenyan state demands a comparative approach to conceptualising patterns of state formation in Africa. Following Allen, we argue that Kenya belongs to a category of ‘centralised-bureaucratic’ regimes along with, at different times, countries such as Tanzania, Zambia and Senegal. Such regimes are marked by the retention of clientelism, the centralisation of power in an executive, the displacement of the party by a bureaucracy answerable to the head of state, and the downgrading of representative institutions (Allen, 1995:305-306). Within this overarching classification there is sufficient space, and indeed the necessity, for nuanced exploration of the specificities of the Kenyan case and consideration of variations within the ‘centralised-bureaucratic’ model. This we do here by drawing upon and modifying the Latin American model of ‘bureaucratic-authoritarianism’ which we discuss below.
Berman has already described the colonial state in Africa as ‘one of the most striking examples of bureaucratic authoritarianism’ (Berman, 1992:144). It is true that the post-colonial Kenyan state shares many features with the ‘bureaucratic-authoritarian’ states of Latin America. However, the Kenyan case, in common with many African states, differs from the Latin American model in several distinctive ways:the emergence of the state out of a nationalist movement that became institutionalised in a dominant political party; the position of the executive as a key source of legitimacy for the regime; and the extent to which issues of ethnicity and nationalism complicated issues of class. While not unique, these features have had two profound consequences for Kenya. First, the ruling elite was forced to confront the widespread popular understanding of nationalism as a movement for self-determination. Second, to borrow from Weber, the strength and legitimacy of the executive and the capacity of the provincial administration meant that the Kenyan post-colonial state was bolstered by semblances of both ‘rational-legal’ and ‘charismatic’ authority (Weber, 1965). This proved to be a powerful combination and has underpinned the stability and longevity of the Kenyan state. The profound tension between the rhetoric of nationalism and the reality of social and economic domination were common to many post-colonial societies. The distinctive features of the Kenyan state lie in its ability to demobilise these forces through the primacy and strength of the executive and the provincial administration. For this reason we argue that the Kenyan state is best thought of as a type of ‘bureaucratic-executive’ state that represents an important sub-set in the spectrum of ‘centralised-bureaucratic’ states. What renders the ‘bureaucratic-executive’ state distinctive is the highly developed capacity of the provincial administration which works to ‘facilitate political control in the countryside as well as in the cities’ (Mueller, 1984:400). As a result, and in contrast to much of the African experience, the picture of limited African governments that have no authority in the ‘periphery’ is misleading in the Kenyan context (Mueller, 1984:427).
By reshaping the ‘bureaucratic-authoritarian’ model, we argue that understanding the nature of the Kenyan ‘bureaucratic-executive’ state requires a careful analysis of the process of institution building and class formation under colonial rule. During this period, Kenya witnessed the emergence of a profoundly unbalanced institutional landscape in which coercive capacity resided with a strong provincial administration and political parties remained underdeveloped. In the process of decolonisation, members of African elites were recruited to man the provincial administration, won seats in the legislature and emerged as large farmers and traders. Together with representatives of transnational capital, it was this group that maintained political and economic control and who were amongst the chief beneficiaries of independence. These elites not only attained control of the commanding heights of the Kenyan economy, they also assumed, through their dominance in the legislature and the bureaucracy, the capacity to reproduce that position. The post-colonial state must therefore be conceptualised as a representation of the interests promoted during the latter years of colonial rule. In the Kenyan case, the post-colonial state represented a ‘pact-of-domination’ (Cardoso, 1979:55) between transnational capital, the Kenyan elite, the provincial administration and the executive (here understood to be the colonial governor, the post-colonial president and their closest advisors, formal or otherwise). The ability of this coalition to reproduce itself over time lay in its capacity to demobilise popular forces, especially those ‘radical’ elements of the nationalist movement that questioned both the social and economic divisions of the post-colonial state. Whilst Kenya may have experienced changes in the executive and the political landscape, the basic structure of the state itself has demonstrated remarkable continuity.
Conceptualising the bureaucratic-executive state
The significance of the provincial administration in Kenyan political life and the closing off of political space suggests parallels to the bureaucratic-authoritarian systems in a range of Latin American countries, including Brazil in the late 1960s, Argentina post-1976, and Chile and Uruguay after 1973 (Collier, 1979; O'Donnell, 1973). Bureaucratic-authoritarianism is characterised by rule by a technocratic elite, the demobilisation and exclusion of the ‘popular sector’ (any mass association of lower income groups) and the undermining of democracy. The bureaucratic-authoritarian model reflects many of the characteristics of one-party states in Africa. For example, Cardoso writes that ‘bureaucratic-authoritarian regimes organise the relations of power in favour of the executive.’ He continues, ‘the strengthening of the executive involves increased centralisation that undermines the federal tradition, where it existed before. It also involves the elimination or sharp reduction in the role of the legislature. Moreover, the judiciary is controlled in practice, if not in theory, by the Executive’ (Cardoso, 1979:41). As in the bureaucratic-authoritarian model, in order to protect their own position the Kenyan ruling elite was forced to demobilise popular forces through absorption. As Cardoso has argued, the ‘links between civil society and the bureaucratic-authoritarian regime are achieved through the cooptation of individuals and private interests into the system’ (Cardoso, 1979:37). In Kenya, the need to co-opt popular forces was particularly acute, as the nationalist movement had effected mass popular mobilisation. Consequently, the post-colonial executive was forced to demobilise its own support base in order to be able to reproduce the power relations inherited from the colonial period.
Despite many shared characteristics, the Kenyan state differs from the Latin American model in three ways. First, Kenyan authoritarian rule was not the product of an alliance between the military and the bureaucratic elite, despite the appointment of Kenyatta's allies as commanders of the various branches of the police and general service unit (Tamarkin, 1978:301). Instead, the roots of Kenyan authoritarianism were to be found indirectly in the democratic election of a highly popular executive in 1963. Second, and consequently, whilst bureaucratic-authoritarianism is seen as rule by an institution (the military), the personal position of Jomo Kenyatta, Kenya's first president, was crucial to the development of the Kenyan political system. Finally, in the Kenyan case the very foundation of the state's existence was fundamentally different. Latin American bureaucratic-authoritarianism came into being because of the perceived need, on behalf of the military, to ‘reorganize the nation in accordance with the ‘national security’ ideology of modern military doctrine’ (Cardoso, 1979:36) and to ‘ensure the restoration of ‘order’ in society by means of the political deactivation of the popular sector’ (O'Donnell, 1979:292). In other words, the bureaucratic-authoritarian state results from the takeover of power by highly selective and coercive interests at the expense of popular forces. This ensures that ‘by the nature of its founding, this state entails an anticipated rejection of the basis for its own legitimation’ with the consequence that it is almost ‘impossible for bureaucratic-authoritarianism to legitimate itself’ (O'Donnell, 1979:286).
In contrast, Kenyan authoritarianism never abandoned popular forces entirely. As elsewhere in Africa and Asia, the Kenyan post-colonial state grew out of a nationalist coalition that embodied majority claims to self-determination:it grew out of the victory, not the defeat, of popular forces. The state therefore emerged into the dawn of the post-colonial period with greater legitimacy and popular support than a typical bureaucratic-authoritarian regime. All states experience a ‘tension between the underlying reality of the state as guarantor and organiser of social domination, on the one hand, and as agent of a general interest which, though particularised and limited, is not fictitious, on the other’ (Cardoso, 1979:28). However, this tension was particularly acute in post-colonial Kenya, and in comparable African states, because independence was supposedly the raison d'etre of the regime. At the same time, due to the popular legitimacy many prominent members of the new political elite derived from their role in achieving independence, the Kenyan state, and particularly the executive, had far greater non-coercive resources with which to overcome the dialectics of its role. Leaders including Kenyatta, Paul Ngei, Bildad Kaggia and Peter Kigondu had experienced detention under colonial rule. Those that did not, such as Tom Mboya and Oginga Odinga, played critical roles in the constitutional negotiations leading to independence. The mantle of national heroes impelled members of the post-colonial regime to maintain avenues of political participation in order to retain legitimacy, but also made it easier for them to do so in a manner suited to their own particular interests. The danger of permitting a degree of political activity further decreased once the dominant elite had been solidified into the de facto one-party state at the end of 1964 by the Kenya African Democratic Union's (KADU) absorption into the ruling Kenya African National Union (KANU). Elections for constituency members of parliament (MP) from a selection of candidates standing on the KANU platform could then be maintained in the knowledge that such competition did not threaten the dominance of Kenyatta and KANU, or the direction of public policy.
These greater avenues of political participation set Kenya aside from many bureaucratic-authoritarian states, with the possible exceptions of the Mexican ‘civil-authoritarian’ state under the Partido Revolucionario Institucional (Institutional Revolutionary Party; see Stevens, 1977; Cotler, 1979:270-272) and the Peruvian ‘experiment’ of 1968-80 (Lowenthal & McClintock, 1983). Whilst much has been written about bureaucratic-authoritarian states, significantly less attention has focussed on variants such as civil-authoritarianism. All too often such cases are assimilated under the broad ‘bureaucratic-authoritarian’ label, which strips the variants of their distinctive feature and the term of much of its accuracy (Cardoso, 1979:39-40). As a result, ‘although the bureaucratic-authoritarian model explains why leaders may seek not to mobilise a populace but to engender apathy’, we must look to the specific characteristics of the Kenyan case for an ‘understanding of the institutions and practices favoured' (Widner, 1992:21). Carrying out such a task requires an analysis of the forces that underpinned the creation of the post-transition state. This has already been done in the case of bureaucratic-authoritarian regimes, where O'Donnell has argued that the nature of the resulting state is related to the fact that it is a phenomenon triggered by processes of industrialisation. The structure of bureaucratic-authoritarianism is therefore seen as being closely related to a specific process of economic development. While economic change and class formation was equally important to the emergence of the bureaucratic-executive state, these developments need to be understood in the distinctive cultural, ideological and institutional context of the colonial period.
The Kenyan ‘pact-of-domination’ was the result of processes intrinsic to the demise of colonial rule. Of particular importance was the creation of a cleavage between Mau Mau supporters and their former opponents, the loyalists. Membership or support of the latter group bound the interests of the executive, the bureaucratic elite (top civil servants, government advisors and the heads of parastatals) and the provincial administration (provincial and district commissioners and their junior officers). Importantly, this elite did not just share the common interests implied by their dominant socio-economic position. Through their control of the provincial administration and the government the elite had the capacity to reproduce their dominance. The supremacy of the provincial administration and the weakness of KANU supported the expansion of the former and the atrophy of the latter. In the absence of dissenting political organisations, the distinctive feature of the bureaucratic-executive state is the effective control of political space by an alliance of the executive and provincial administration. The executive enhanced its position by establishing patron-client relationships with strong local politicians who delivered the support of their ethnically mobilised constituents and clients. This system, which maintained regular elections to create local avenues of accountability, enabled the executive to demobilise opposition forces whilst maintaining a veneer of legitimacy.
Founding the Kenyan bureaucratic-executive state
The political struggles that developed in Kenya were not simply the product of a well-disguised class struggle. As elsewhere, issues of class formation were intertwined with expressions of ethnicity and disputes over who could claim to be heir to the nationalist struggle (Bayart, 1993:58-9; Atieno Odhiambo, 2003). At the centre of the complex relationship between class, ethnicity and nationalism lay the legacy of the Mau Mau rebellion and, consequently, the lasting influence of the colonial government's attempts to create a loyal middle-class as part of the counterinsurgency effort. In its attempts to resolve the conflict and in its later preparations for departure, the colonial government passed control of the state to a multiracial alliance of elites. The manner by which the state reproduced itself during this process subsequently proved to be the keystone of the bureaucratic-executive state. This capacity for reproduction was related to the strength of the prefectural provincial administration, first established in 1900. By independence in 1963, the provincial administration had grown significantly. The demands of two world wars, post-war development policies efforts that ironically initially threatened the influence of administrative officials, the suppression of African political activity and the counter-insurgency campaign of the 1950s all aided its growth in size and influence. The most significant period of expansion occurred in the final decade of colonial rule, in large part because of the role played by the provincial administration in the war against Mau Mau. Between 1951 and 1962 the number of administrative officer posts increased from 184 to 370, greatly enhancing the scope and societal penetration of the provincial administration (Berman, 1990:73-114, 199-218, 347-71).
Under colonial rule, ultimate authority resided with the secretary of state for the colonies in Westminster, but effective power resided with the governor. The governor, as executive, initially ruled in the absence of a strong legislature. As the colonial period progressed, the executive retained much of its power in spite of the emergence of an independent legislative council. The executive was empowered by its control, via the chief native commissioner, of the provincial administration. This enabled the executive to bypass the legislature when necessary. As a result of their role as the instrument of executive agency, the provincial administration quickly developed an extensive portfolio of responsibilities. As Gertzel has summarised ‘before independence it could be said that the Provincial Administration had power, authority and influence. They also had three major functions:control, co-ordination and mobilisation of the public for development’ (Gertzel, 1970:26). From the late 1950s onwards, the provincial administration was entrusted with responsibility for the running of African Courts, the collection of local taxes and the regulation of elections. As the Chief Secretary wrote in 1960, the provincial administration oversaw ‘the general working of the machinery of Government’ (quoted in Gertzel, Goldschmidt & Rothchild, 1969:368). Within this range of duties, the most significant was the role of the provincial administration as the primary body charged with maintaining law and order (Kenya, 1954: para. 6).
The provincial administration's theoretical role as the conduit for local grievances was rendered all the more significant because of the paucity of other avenues for African representation. Africans were not appointed to the legislative council until 1944 and African elected political representatives did not achieve a majority in the legislature until 1961. Whilst some form of African representation was possible earlier via the system of local advisory boards established from 1924, ‘the Provincial Commissioners and District Commissioners suppressed the native authorities with self-confident paternalism’ (Hughes, 1963:26). However, the political role of the provincial administration was not a simple one, as the circle of interests of the metropole, settlers and African subjects could never be squared (Berman & Lonsdale, 1992:77-100; Hyden, 1984:104). As elsewhere (Berman, 1992:152), the colonial provincial administration was critical to the restriction of African political organisations by preventing the raising of funds, refusing permission for public meetings and suppressing protests (Kenya National Archive (KNA) PC/CP.8/5/3, ‘Kikuyu Central Association’; KNA DC/MUR/3/1/10, ‘Kenya African Union’; Throup, 1988:139-70). Throughout the 1950s it was the provincial administration which stood at the forefront of the counter-insurgency effort in Central Province (Berman, 1990:347-76).
Besides implementing the villagisation policy and controlling the process of interrogation and detention of Mau Mau suspects, during the Emergency the provincial administration oversaw the promotion of loyalist political and economic interests. With accelerated class formation deployed as a counter-insurgency tactic (Heyer, 1981:102; Sorrenson, 1967:118), rewards granted to loyalists included the introduction of private land titles that guaranteed security of tenure, access to expanding cash crop production, and preferential access to the labour market (Branch, 2005:162-73; Sorrenson, 1967). The government intended to break the alliance of middle and lower peasantry of Central Province that sustained the Mau Mau movement, whilst providing the foundation for local loyalist domination (Berman, 1990:371; Ng'ang'a, 1977:365-84; Okoth-Ogendo, 1976:163; Sorrenson, 1967:118, 166-7). This pattern continued up to independence (Berman, 1990:371; Ng'ang'a, 1977:365-84; Okoth-Ogendo, 1976:163; Sorrenson, 1967:118, 166-7). The outgoing regime intended to use resettlement programmes such as the Million Acre Scheme to ‘lance the boil of land seizures and growing lawlessness’ (Bienen, 1974:164). Whilst the government wanted to make political capital by resettling poor and landless Africans on former settler property, the Scheme also aimed at strengthening the growing strata of middle-sized farmers and allowing the government to ‘reward its own powerful supporters’ (Bienen, 1974:165). Although the motivations and outcomes of late colonial land policy have proved contentious, it is clear that the period was critical in the development of a privileged land owning elite sympathetic to British interests (for a range of views on land policy see Bienen, 1974; Coldham, 1978; Cowen, 1982; Harbeson, 1971; Heyer, 1981; Leys, 1975; Sorrensen, 1967).
The gains derived by the African elite from agricultural and land reforms meant little without the protection provided by control of the main institutions of the state. Consequently, formal political power was progressively transferred to this same group during the final years of colonial rule. There were three main features of the attempt to promote ‘loyal’ African leaders:bias in their favour during elections to the legislative council in 1957 and 1958, preferential treatment in the course of the Africanisation of the provincial administration and the unequal distribution of new economic opportunities. The colonial government manipulated the electoral system to enhance the likelihood of a strong loyalist and conservative presence in political structures with the intention of accelerating class formation in a manner conducive to British interests. To do this, in 1957 and 1958 the regime enforced a loyalty test in areas that had been directly affected by the Emergency, which effectively excluded the more radical elements of the political community. As a result, in 1957 only 7.4 per cent of the adult population of Central Province were registered to vote (Engholm, 1960:421). Their votes were further diluted by the electoral boundaries that penalised the centres of the Mau Mau rebellion. The restrictive criteria for voter registration ensured that the qualitative franchise became, alongside land consolidation, a means through which the colonial regime attempted to consolidate the position of a conservative political elite across the colony (Anderson, 2005:550; Branch, 2005:194-214; Engholm, 1960:391-461).
Despite the termination of the loyalist monopoly on the franchise, elite African politicians retained their dominant position following the polls in 1961 (for an account of this election see Bennett & Rosberg, 1961). Electoral boundaries now favoured the pastoral communities of the Rift Valley, and thus the candidates of the KADU over those of KANU (Bennett & Rosberg, 1961:47-9). The provincial administration further buttressed KADU by its attempts to ban KANU candidates and leaders from addressing public meetings in Central Province, on the basis that most KANU ‘office bearers are ex-detainees and rather a dangerous lot’ (KNA VP/ 9/102, PC Central, ‘Handing Over Notes’, 8 September 1960:2). In reality, KANU in Central Province was a broad church and contained both loyalists and Mau Mau. First elected in 1957 and 1958 and subsequently re-elected in 1961 and 1963 as a consequence of their conciliatory rhetoric and modernist credentials, the legitimacy of loyalists such as Julius Kiano and Jeremiah Nyagah was further safeguarded by their participation in the negotiations that resulted in Kenyatta's release and independence. The loyalist background of this first generation of elected African leaders was considered less significant than their perceived ability to deliver economic development and political independence.
In order to understand the post-colonial alliance between the legislature, the Executive, and the Administration, it is important to appreciate the process through which elites with similar economic interests came to capture all three institutions. Like the legislature, the provincial administration underwent a significant transformation during the course of the 1950s and early 1960s. The lower ranks had been opened to Africans for the first time in 1947, but they were unable to progress further than the rank of assistant administrative officer. With the onset of the Emergency, the twin necessity to reward loyalists and to cope with extra security demands reinvigorated the growth of African representation within the institution (Branch, 2005:214-27). The new recruits rose through the ranks at great speed and, by early 1965, the entire provincial administration had been Africanised (Gertzel, 1970:36). The context of the Mau Mau war meant recruits from Central Province were disproportionately represented amongst the new cohort. Loyalist appointees were well connected to local elites, highly educated and with prior experience of public service as chiefs or in the lower ranks of the provincial administration (Branch, 2005:222-3). Recruits from elsewhere had a similar background. Andrew Mnyolmo, for example, had been a clerk in native authority courts, a district assistant in the Rift Valley, a member of the Pokot Law Panel and acted as a magistrate in Baringo prior to his appointment (KNA VP/1/102, ‘Confidential Personal Report – A.M. Mnyolmo’, 23 July 1964). With independence approaching, African elites were finding their way into positions of authority. For those from Central Province, recruitment offered them protection against the ultimately unrealised threat of ex-detainees and made the loyalist recruits essential to the everyday functioning of the state. Collectively, Africanisation of the legislature and the provincial administration established the domination of the state by specific class interests which would come to dominate the post-colonial bureaucratic-executive state.
The new recruits carefully protected the functions of the provincial administration as British rule entered its final months, despite concern amongst the outgoing European administrators in Central Province with the institution's inbuilt political biases. It was suggested by some departing officials that the entire provincial administration should be converted into a more accountable model of local government in order to discard the trappings of imperialism (KNA VP/1/111, DC Kiambu to PC Central, 25 February 1963). Of particular importance was the future of the tribal police (later known as the administration police), the provincial administration's own police force, a central actor in the anti-Mau Mau campaign. The most vocal advocate of retaining the tribal police was John Michuki, the newly appointed district commissioner in Nyeri. Eulogising the Tribal Police's ‘unofficial, liberal outlook’, Michuki rejected ‘any suggestions that would amount to total annihilation of the Tribal Police Force’ (KNA VP/1/111, DC Nyeri to PC Central, 7 March 1963). The new generation of Kenya's governors became convinced of the virtues of a force accountable only to the provincial administration that could straddle hazy legal and political boundaries. Indeed, throughout this period the provincial administration and the rest of the civil service came to be highly valued by the emergent African elite due to its capacity to deliver order and stability (Mboya, 1993:162). On the eve of independence, Oginga Odinga, as minister for home affairs, prevented the disbandment of the Tribal Police or its removal from the auspices of central government (KNA WC/CM/1/13, Minister for Home Affairs, ‘Future of the Tribal Police’, 21 November 1963). Odinga further retained central government control over the appointment of chiefs (KNA WC/CM/1/13, Minister for Home Affairs, ‘Future of Chiefs and Sub-Chiefs’, 21 November 1963). Both policies contravened the independence constitution by maintaining a parallel police force and contesting the sole responsibility of the Public Service Commission for the appointment of public servants (Kenya, 1963:111, 135). Consequently, coercive capacity was not decentralised and responsibility for maintaining order remained with the Administration.
The process of decolonisation grafted an African political and administrative elite onto existing European and Asian economic elites. As both African and European elites stood to lose from a shift to a more radical redistributive state, both were concerned to see that the fundamental structures of the economy remained intact throughout the decolonisation process. Despite some tensions, the African elite recognised the advantages of maintaining good relations with Britain and the expansion of foreign investment. The post-colonial executive therefore resisted confiscation of European land and Africanisation of business interests, instead favouring economic continuity. Although there was an influx of African employees into companies during the 1960s, expatriate managers remained in place (Bienen, 1974:142). While many did leave, European settlers were free to remain on their land. Relations with Asian capital were more troubled. However, despite the opposition of several MPs, the government protected the position of Asian capital through its continued commitment to the Kenyanisation, rather than Africanisation, of the economy. In reality, allowing preferential access to economic opportunities to all citizens regardless of race had little immediate effect. In 1967, just 10,000 of the 37,000 Asian middle and higher managers were citizens. By 1970, less than 40 per cent of the Asian population as a whole were thought to hold Kenyan citizenship (Ghai, 1970:126).
The ability of the multiracial ‘pact-of-domination’ to protect its interests depended on the capacities of the political institutions under its control. The powers of those institutions taken over by the newly promoted loyal African elite were therefore extended at the direct expense of more independent institutions, most notably KANU. The difficulty of transforming congress-style anti-colonial movements into the foundations of post-colonial governance was common to many African states (Williams, 2003). In Kenya, however, little effort was went into addressing the fragmented nature of the ruling party. Partly a reflection of the fact that Kenyatta ‘was not a party man’ (Okumu, 1984:50), this condition was also a product of long historical trends. Even prior to Mau Mau, the varied impact of colonialism according to geographical and economic considerations had led to an uneven mobilisation of African political communities, a phenomenon most apparent amongst the peoples of Central and Nyanza Provinces. There, the greater direct impact of colonialism enhanced access to the colonial infrastructure and thus to superior economic development, albeit unevenly distributed (Atieno Odhiambo, 1995; Bennett, 1957; Ogot, 1995; Throup, 1988). The intra- and inter-ethnic competition that Luo and Kikuyu political and economic development aroused helped to establish the position of local leaders. Given this wider context, the restrictions placed on colony-wide African political activity by the provincial administration only served to institutionalise a system of ethnically based ‘boss politics’ (Gertzel, 1970:15).
The influence of these trends upon KANU was highly visible after 1963. Any development of a national party structure was fatally undermined. Within KANU, even such basic requirements as central party funds and a widely accepted manifesto had not been developed (Good, 1968). Kenyatta had little incentive to address such problems. The party, unlike the provincial administration, contained significant radical elements that sought to call into question the policy of continuity. It was, therefore, the provincial administration that shared specific class interests with the executive and which represented the path of least resistance to the extension of executive control. Significantly, the existing capacity of the provincial administration enabled the executive to act without consulting the party or the legislature, replicating the direct control of the colonial Governor. This unchecked institutional capability, married to the popularity of many within the African elite and the class alliance that bound together the ‘pact-of-domination’, created a powerful system which was able to both maintain its dominance and its legitimacy in the immediate post-colonial era.
The Post-colonial bureaucratic-executive state in operation
Recognising the strength of the bureaucratic-executive structure, Kenyatta chose to retain its main components (Bienen, 1974:45; Gertzel, 1970:36; Kyle, 1999:204). In order to effect this smooth transition, the executive had to overcome the independence constitution that created a regional system of government with a regionally elected upper house and regional assemblies intended to control independent revenue streams. The immediate pre- and post-independence periods, therefore, saw a battle over the centralisation of authority within the Kenyan state. Ironically, considering his later role as one of the most prominent critics of the post-colonial state, Oginga Odinga was the principal architect of this process. As minister of home affairs, Odinga added to his protection of the Tribal Police and chiefs (see above) by playing a critical role in the successful attempts to resist the decentralisation of the provincial administration until KANU had an opportunity to introduce a new constitution (Odinga, 1967:243). Within this process, the provincial administration was entrusted by the executive with the task of scuppering any attempt at devolved government (Anderson, 2005:557; Mboya, 1993:167).
KANU's desire for greater centralisation of the political system was institutionalised by the establishment of a republic in December 1964, which necessitated new constitutional arrangements. These changes included the abolition of the short-lived regional system of government and the creation of the post of president in place of the prime minister (Ochieng’, 1995:107). Not only was the provincial administration once again placed under the direct control of the executive, but the positions of provincial and district commissioners were reinstated with the Office of the President given control of appointments. Equally significantly, the demise of devolution saw control of land transfers, held briefly by regional authorities, revert to the centre (Harbeson, 1971:241). The executive not only sought to replicate the colonial system of rule, but to improve upon it. The continuity of the political system ensured that the colonial and post-colonial provincial administrations shared significant characteristics. Both were local instruments of executive power (Gertzel, 1966:201; Gertzel, 1970:24, 36). Despite proclaiming neutrality and impartiality, both were willing to become involved in an explicitly biased manner in local political affairs on their own initiative or at the behest of central government, often to the considerable frustration of the population (Gertzel, 1966:203, 205; Gertzel, 1970:23, 27, 39-30; Onalo, 2003:132-3; Oyugi, 1994; Throup & Hornsby, 1998:3, 14). We therefore dispute the assertion that there is insufficient evidence to support the claim ‘that the civil service in Kenya operates in a fashion similar to its colonial predecessor – exercising, as a cohesive institution, wide powers over the rest of society’ (Hyden, 1984:104).
By the late 1960s, the provincial administration had amassed an impressive list of responsibilities (see Office of the President, ‘The Role of the Provincial Administration’, May 1966, reproduced in Gertzel, Goldschmidt & Rothchild, 1969:365-7). It assessed and collected graduated personal tax, remained the eyes and the ears of the executive and from March 1965 assumed responsibility for the collection of local authority taxes (Ministry of Local Government quoted in Gertzel, Goldschmidt & Rothchild, 1969:428). Moreover, Administrative officers chaired a great number of local committees, including the Land Control Boards and those responsible for the land settlement schemes. Most troubling for opponents of the regime was the provincial administration's position as regulator of local political activity. Effective power drained from local authorities to the provincial administration. The surviving requirement for all political meetings, even those organised by MPs, to be licensed by the district commissioner was particularly controversial. As J.M. Shikuku argued in parliament in February 1966:
‘the issue of licences was instituted by the imperialist Government and that was because they were afraid of us.’ Shikuku then asked Moi, the minister for Home Affairs, ‘Is this Government now, which is an African majority Government, also afraid of us?’ (quoted in Gertzel, Goldschmidt & Rothchild, 1969:126).
by the middle of 1968 the executive in independent Kenya enjoyed the position very similar to that of the executive during the days of colonial rule. The President occupied a position very much akin to that of the Governor, both in the scope of his powers and in the manner in which he could call upon the provincial administration to ensure Central Government control (Gertzel, 1970:171).
Despite KANU's incorporation of large numbers of dissenters, their attempts to influence policy were frustrated by the party's impotence and the personal loyalty of the bureaucracy and a majority of the legislature to Kenyatta. However, the ability of the executive to restrict political space was constrained by popular expectations of uhuru and the strength of local leaders. However, with the creation of a de facto one-party state at the end of 1964 and the further atrophy of KANU, it became easier for the executive to isolate and control more radical political factions within the ruling party. Elections for constituency MPs were maintained, but the executive adapted this necessary reality to its own needs by establishing a patron-client relationship that co-opted significant political groupings. The increasing dependence of MPs on executive patronage reduced the independent power of the legislature whilst centralising political authority. Of crucial importance to the ability of the executive to command loyalty and thus perform the delicate balancing act between participation and control was the initial legitimacy afforded to Kenyatta as Baba Taifa, father of the nation. Kenyatta's personal popularity and influence trumped any configuration of rival leaders, however influential. As one dissident MP recognised in 1966,
If Jomo Kenyatta was not leading KANU, and KPU started working under Kaggia and Odinga, I dare say there would be an awful lot of people moving to it. But Kenyatta leads KANU (quoted in Gertzel, Goldschmidt & Rothchild, 1969:157).
While many African states may be characterised as ‘empirically weak and underdeveloped', such a description is misleading when applied to Kenya (Jackson and Rosberg, 1982:12). In this sense Kenya is an important outlier. While much of the recent literature has focussed on the ‘vacuous’ and ‘ineffectual’ nature of the African state (Chabal and Daloz, 1999:14), the Kenyan state apparatus was relatively extensive and efficient, and gave incumbent regimes a ‘monopoly of key coercive sanctions and economic resources’ (Mueller, 1984:399). This monopoly enabled the elite to restrain those disgruntled with its conservative economic approach and proactively promote an explicitly elitist ideology. The Kenyan elite felt able to present a political platform tailored to their own vested interests as a set of policies intended to benefit all. In so doing, they achieved a relatively high degree of consensus by appearing ‘as agents of a general interest of a community – the nation – that transcends the reproduction of daily life in civil society’ (O'Donnell, 1979:288). The manner by which the state was able to see off the challenge posed between 1966 and 1969 by the Kenya People's Union (KPU) and its demands for land and wealth redistribution, is a prime example of this process (Gertzel, 1970; Mueller, 1984).
During the pre-independence constitutional negotiations, African political leaders agreed to the British demand that European land be purchased by the Kenyan government prior to any large-scale resettlement scheme, rather than repossessed (Wasserman, 1976). These attitudes became entrenched within the modus operandi of the post-colonial state. The promotion by the departed colonial regime of members of the African elite continued to pay dividends for those whose interests the process of Africanisation was intended to safeguard. Members of the African elite were fiercely protective of their newfound influence and wealth and were thus resistant to any demands for significant structural reforms. The MP for Kitale East, chairman of the Maize and Produce Board and a major farmer in Trans-Nzoia, Masinde Muliro rejected the demands by the KPU for land redistribution. The former schoolmaster, first elected to the legislative council in 1957, believed ‘An African socialist is by nature a capitalist’ (quoted in Gertzel, Goldschmidt & Rothchild, 1969:83). Muliro's opinions mirrored official policy. Authored by Mwai Kibaki and Tom Mboya, Sessional Paper No. 10 of 1965, African Socialism and its Application to Planning in Kenya, committed the state to private enterprise, increasing foreign investment, and wealth creation through economic expansion rather than the redistribution of resources (Hyden & Leys, 1972:393; Kenya, 1965). The post-colonial state was a significant investor in Kenyan capitalist estate agriculture and industry and encouraged further private Kenyan and international capital investment (Swainson, 1980:17). This configuration of interests supported a capitalist development programme. Pre-independence nationalist aspirations for major land reforms were sacrificed for economic and political stability. An elitist process of statecraft restricted to those with access to capital had replaced the popular desire for nation building that had emerged in the final years of colonial rule (Harbeson, 1971:250-1; Swainson, 1980:18).
The ‘pact-of-domination’ worked effectively to resist the redistribution of land and wealth, but not without significant dissent. The initial acquiescence to British demands on property rights and the adoption of a development agenda based upon private enterprise angered many radical activists inside and outside the government. Following the departure of some European landowners, the availability of land that could be redistributed to ease the political pressure from below was a great resource. However, that there was not enough land for all of the landless, and that the established elite did not wish to give up its own significant holdings, was a source of great tension. The conflict over land redistribution came to be embodied in a struggle between KANU and the KPU. The KPU formed following by the departure of dissident elements from KANU in 1966. In Oginga Odinga and Bildad Kaggia, the radicals had strong regional leaders with a demonstrable commitment to political nationalism and social equality, primarily in the form of free education and land redistribution (Leys, 1975:214). This was reflected in Kaggia's support base in his home district of Murang'a, which could be ‘fairly accurately defined as poor, landless or owning very little land, and predominantly ex-detainee’ (Lamb, 1974:135). At political rallies, Kaggia told his audience that Kenyatta ‘had taught him that the land belonged to Africans and will never be bought’ (KNA VP/1/98, Anon., ‘Notes on Political KANU Meeting of 21 February [1966] for Provincial Commissioner’). Such sentiments not only went against prevailing policy, they questioned the very economic and ideological basis of the bureaucratic-executive state. Furthermore, Odinga's support base in Nyanza Province and Kaggia's role as the voice of poor Kikuyu in Central Province and along the eastern fringe of the Rift Valley threatened the geographical and ethnic centres of the regime's power.
In the face of the KPU's challenge, the state supported a process of economic continuity and sought to delegitimise its opponents. Kenyatta stood firmly against the redistribution of land to the poor and landless. Free grants of land ran counter to Kenyatta's own principles (Lonsdale, 2002). Wealth through land was to be earned (Bienen, 1974:164). In 1965, Kenyatta dismissed Kaggia's demands for land redistribution as the selfish begging of the lazy for ‘free things’ (Lamb, 1974:36). Beneath such condemnations lay the elitist premise that while labour begot wealth, indolence led inexorably to poverty and criminality (Lonsdale, 1992). In asserting that land was to be earned and not given, the president was appealing to a broader antipathy towards egalitarian policies (Gertzel, 1970:86-7).
The twin powers of the administrative/executive axis supported the defeat of the KPU. The considerable public support for Kenyatta enabled him to outmanoeuvre the personal threat to his position posed by a national leader of Odinga's stature and a local figure of Kaggia's influence. Kenyatta personally legitimised a post-colonial state that fell short of many Kenyans’ expectations. For all the president's strengths, it was his foot-soldiers amongst the provincial administration that underpinned Kenyatta's position through their ability to regulate political activity and their capacity for unmonitored, extra-legal behaviour (Lamb, 1974:137). During the by-election campaigns that followed the creation of the KPU in 1966, known as the Little General Election, a number of KPU candidates complained of interference by the provincial administration, such as the refusal to grant permits for election rallies (Gertzel, 1970:82, 116, Mueller, 1984:413). Accusations were made in parliament that the provincial administration was threatening KPU voters with expulsion from land settlement schemes, detention and the withdrawal of government loans (Gertzel, Goldschmidt & Rothchild, 1969:158-61). Even after the defeat of the KPU in 1966, the provincial administration was employed two years later to ensure the disqualification on various technicalities of all KPU candidates attempting to stand (Bienen, 1974:259; Gertzel, 1970:165; Hyden & Leys, 1972:394). Such practices continued even in the wake of the demise of the KPU and the return to a one-party state in 1969. During elections that year, the provincial administration apparently used its control of an English language proficiency test in order to prevent known opponents of the government standing as candidates (Hyden & Leys, 1972, 393).
Outside periods of electioneering, the provincial administration exploited its position as the principal agent of development in rural areas. The rural scarcity of economic and political opportunities made the capture of local bodies overseeing development imperative for control. The provincial administration's domination of land, agricultural and educational committees allowed the state to manipulate access to key socio-economic resources at the district level (Lamb, 1974:52; Mueller, 1984:407). Supporters of the regime were thus able to consolidate local power-bases through networks of clientage. In this way the influence of the bureaucratic-executive state was extended to the periphery. With the conflict between the regime and KPU occurring outside of Nairobi, the ability of the provincial administration to penetrate the periphery was critical to its ability to control political space. Consequently, Kenya's bureaucratic-executive state represents the opposite balance of power between the centre and the periphery to that which is often assumed in the literature on African states (for the origins of the centre-periphery distinction see Zolberg, 1966 & 1968). The Kenyan state arguably exerted greater political control in the countryside than in the urban areas (Mueller, 1984:414). This is in contrast to the common assumption that dominant party regimes in Africa can police the capital city and other major urban centres, but have insufficient resources to effectively manage the ‘periphery’.
The defeat of the KPU represented a crucial stage in the development of the post-colonial state. Subsequently, dissent was muted and state-level political competition became the personal fiefdom of the elite. Kenyatta's defeat of Kaggia in Murang'a by appealing to Kikuyu values such as self-improvement demonstrated the difficulties of mobilising a popular political force along ‘class’ or ‘egalitarian’ lines. It is telling that J.M. Karuiki, the most significant threat to Kenyatta's hegemony in the decade following the demise of the KPU, appealed to few of that party's stated egalitarian values. A Mau Mau detainee, successful businessman and a junior minister in the Kenyatta government, Kariuki was able to enhance his already significant popular appeal by making large contributions to harambee (community self-help projects). Kariuki rapidly became a well-known and highly popular politician (Widner, 1992:86-91), but did not speak out against wealth accumulation as the KPU had done. He rather attacked the way in which the Kenyan elite neglected their duties as ‘Big Men’, most notably their responsibility to support the advancement of others. Consequently, his potent platform combined Kikuyu moral ethnicity and a broad populist appeal. Kariuki represented a different and possibly greater threat to the internal unity of the regime than the KPU, but had no formal institutional support base. His threat was neutralised by his assassination in 1975, the third major political assassination of the post-colonial era after Pio Pinto (1965) and Tom Mboya (1969). The value of carefully targeted political assassinations was again demon-strated in 1990 with the murder of Robert Ouko, then minister for foreign affairs in Moi's cabinet.
Following Kariuki's murder and consolidating trends visible much earlier, intra-elite conflict came to be focused on direct control of the state and avoided the question of the structure of the state and its resources. The bureaucratic-executive state had been consolidated, and there were few among the elite who did not recognise the value of maintaining it. Intra-elite competition could be fierce and all sections of the elite by no means benefited equally from the system of domination. On the whole, however, all benefited sufficiently to ensure that a dominant majority continued to support the system of domination itself, thus preserving the elite consensus. This was particularly evident during the Kenyatta succession, when different factions within KANU jostled for position in a bid to ensure that their representative inherited the presidency. As Tamarkin has argued,
The struggle for succession was essentially an intra-elite one, the two factions striving to control the regime rather than subvert it. Once the succession was decided, the elite, and the bourgeoisie as a whole, had an overriding interest in stabilizing the regime upon which they thrived (Tamarkin, 1979:33).
Conclusion
The bureaucratic-executive state emerged as the result of a specific process of institution building and class formation during the colonial era. These twin processes placed an elite of wealthy ‘conservatives’ in control of an extensive system of administration and a powerful executive. The resulting ‘pact-of-domination’ was able to maintain its privileged position by protecting and extending the authority of the president and the capacity of the administration. Unlike its bureaucratic-authoritarian counterpart, this bureaucratic-executive state resulted from the victory of popular forces and so could call on both popular legitimacy and coercive measures to contain opposition. Consequently, the ‘conservatives’ could demobilise the challenges their economic and political dominance rendered highly probable. Although clearly an example of the set of African ‘bureaucratic-centralised' states identified by Allen (1995), the bureaucratic-executive state demands to be seen as a distinctive state formation located within that spectrum. What renders the bureaucratic-executive state distinctive is not so much the ‘charismatic’ authority of the executive, but the capacity of the regime to monitor and influence political developments through the provincial administration. The control of the bureau-cratic-executive state did not begin at State House and decrease with every mile travelled away from the ‘centre’. Instead, the provincial administration acted as a conduit for executive power, with the consequence that political space was, at times, as tightly regulated in the ‘periphery’ as it was in the ‘centre’, if not more so. It is this that separates Kenya from Zambia, Tanzania and many of the other ‘weak’ and ‘soft’ African states (Jackson & Rosberg, 1982:4-12; Chabal & Daloz, 1999:8-16).
The role of the provincial administration is fundamental to an understanding of the longevity of the bureaucratic-executive state. While the dual structure of the executive and provincial administration was underpinned by the coalition of the ‘pact of domination’, the axis was intimidating and powerful. It was only with the refusal of the Moi regime to share political power and economic opportunities that the pattern of elite consensus was eroded and the question of the appropriate structure of the state became a source of dissent once again. Despite this, the bureaucratic-executive axis is too important for incumbent governments to dismantle. Even in the context of heated political competition throughout the multi-party era the basic structure of the state has remained intact. Under Moi, KANU continued to make extensive use of the provincial administration to ensure victory in the multi-party elections of 1992 and 1997 (Omukada, 2002:84-5). Despite the success of NARC in 2002, the essential components of the ‘pact-of-domination’ are still in place. As Njoya has commented, ‘We now know what ‘NARC’ means:‘Nothing-Actually-Really-Changed!’ (quoted in Wolf, 2005:16). Despite NARC's promise of a new dawn in Kenyan politics, the legacy of the bureaucratic-executive state remains intact. Whilst the regime may have undergone periodic changes, the state itself has not.