Abstract: The study of Africa's anti-colonial nationalisms flourished in the 1960s.  In the 1970s it faded out, as nationhood itself appeared to wither away in many of Africa's post-colonial states.  Nationalism seemed to have become a past without a future.  Its rueful memory was part of 'the disturbing discourse of failing modernity' (Rathbone 2000: 164).  Since modern historians tend to regard all history as contemporary history, the story of how the past became the present, there seemed little point in digging up the corpse of movements which turned out to be less than they had proclaimed themselves to be.  Embarrassment at our own credulity, moreover, was best covered up in scholarly silence and a turn to other themes.  Other subjects had a continuing rationale of study or became new objects of attention.  Gender, peasantries, poverty, ethnicity, urbanisation, the environment-these would all persist, despite all change.  Unlike nationalism they could not disappear in a puff of deflated purple prose about the masses pulling (or hoeing) together in unity, on the road to freedom from poverty, ignorance, and disease..

Scholarly interest in Africa's nationalisms-if not the first then perhaps the second of Terence Ranger's many fields of research-is now reviving, nonetheless.  The moving frontier of archival access now permits study of the 1960s.  That is one reason.  There are helpful new insights too, offered meanwhile by students of culture, identity, and political thought.  With the failure of other grand narratives of the past, not the sorry tale of African nationalism alone, historians are more alert to the little imaginations and memories that go to make up the larger gambles on contingency that are undertaken, in hope or despair, by humankind (Lonsdale above).  We are turning generally, with good analytical reason, to study the intellectual energies that tie together the personal, the social, and the political in any search for purposeful authority-in this particular case 'national' authority.  Without discussion of their intellectual or moral discourse political movements become unintelligible, whether in their rise or, especially, in their decline.  All authority has to be socially (which means, first, imaginatively) constructed.  Not all construction proves sound.  At a time when all nationalisms are again in question, not least in the still-United Kingdom, the study of African nationalisms too may have a future if we take a different, less credulous, look at the past..

Ranger's own approach to the history of nationalism-mainly in Rhodesia-Zimbabwe but with glances at eastern Africa from his chair of History at Dar es Salaam-has undergone some change, unsurprising in a span of forty years, but perhaps surprisingly little change all the same.  In the 1960s Terry pursued the connections between African resistance or rebellions against colonial conquest at the turn of the twentieth century and the mass nationalisms of two generations, or half a century, later.  This interest grew out of his classic work on Zimbabwe's first chimurenga war in the mid 1890s (Ranger 1967).  Even in these heady, early, days of his enquiry, searching for the springs of mass commitment, he was careful to stress the complex and contradictory nature of African thought and action.  Nationalism was never 'a monolithic movement of common interests' (1970: 231).  Terry dwelt on the uncomfortable relations between secular and religious leaderships, between educated elites and ordinary people, urban or rural; on the fragile balance between traditional values and desire for the Europeans' 'nice things' (1968: 453); on odd alliances between spirit mediums and ex-policemen.  There were difficult successions from one sort of mass movement to another, from millenarianism, say, to popular nationalism, not only an heir but also a rival.  One can also see, as early as 1968 (1968: 641), Ranger's liberal reservations about the unhappy-if in Rhodesian conditions inevitable-conflict between the facilitating politics of a Thompson Samkange in the 1940s and the all-devouring mobilisation for liberation war which overtook it in the 1980s.  Moderate nationalism, of its nature, tolerated the autonomy of other civil associations; in war associational freedom can look like treachery (1995).  Perhaps the dominant thread of Terry's writings on nationalist complexity was the African search for authority, some link between social status or political (or prophetic) utterance, and the ability to carry a meeting, to control a movement, to co-opt or silence rivals.  Other studies of nationalism were less penetratingly concerned with this problem.  In the spirit of the rest of this collection, this essay is a belated attempt to repair that deficiency with respect to Kenya, by discussing the narrative sources of political agency in the colony's first nationalist age.  It finds that Kenyan nationalism lived more stories than the one we thought we knew.  Nor did it altogether fail. .

The Kenya African Union, KAU or 'cow', was founded in October 1944 and banned in June 1953.  The party's founders put their hopes in a reforming post-war colonialism.  Their successors were jailed not ten years later, convicted of managing a terrorist movement called Mau Mau.  The first studies of Kenya's post-war nationalism, in this short KAU decade, were dominated by the question of Mau Mau.  What did the rise of this secret organisation, confined largely to one ethnic group, the Kikuyu and their nearest linguistic neighbours, 'the highland Bantu', say about the previously pan-ethnic, open and constitutional, congress-like umbrella, KAU?  For Mau Mau, despite its problematic claims to be called 'nationalist'-an issue which troubles Kenyan debate to this day-forced the issue of power in a way that KAU had never done, from which indeed it seemed to shy away (Buijtenhuijs 1973; Atieno Odhiambo 1991; Sabar-Friedman 1995; Clough 1998: 248-58).  It was not that Mau Mau won its war against the British; guerilla movements rarely win in military terms; and militarily Mau Mau was defeated (Heather 1994; Percox 1998).  But in order to crown peace with sustainable civil governance-and thus reopen a prospect of controlled decolonisation-the British had to abandon 'multiracialism' and adopt African rule as their vision of Kenya's future.  London's attempts to combine the devolution of imperial power with the preservation of privilege for white settlers and Indian businessmen had neither convinced the immigrant minorities nor attracted the African majority.  KAU had constantly petitioned for African rule.  The British had taken no notice.  Mau Mau could not be similarly ignored.  The blood of Mau Mau, no matter how peculiarly ethnic in source and aim, was the seed of Kenya's all-African sovereignty.  Scholarship in the violent, controversial, shadow of Mau Mau has hitherto condemned KAU-and with it, the post-war British administration of Kenya-to various failures, depending on one's expectations of historical process in the twilight of empire.  In brief, KAU is said to have failed at the level of high politics, organisation, class representation, and leadership.  That has been the KAU story.  Its sad epilogue was the future it was presumed to have lost.  What KAU moderation and British blindness lost to Mau Mau's violence, it has been said, was all prospect of a liberal, facilitating, politics for the Kenya that gained independence in 1963.  The violence of the Mau Mau war had been too sectarian, destroying inter-ethnic trust.  This essay is not so certain of that cause and effect. .

At the high-political level, KAU certainly failed to persuade the British of the justice of its aims, which were to unseat the white settlers from their economic throne of political influence and to re-open two African frontiers.  The immediate African need was for the return of at least some of the 'stolen lands' of white settlement, to make room for a rising black population and draw the sting of looming landlessness.  The further frontier was freedom under African, not white minority, rule.  KAU was scarcely listened to on either score.  British officials on the spot, even some Labour politicians in London, despised African nationalism as an artifice designed to grant to the educated few the freedom to exploit the ignorant many (Mitchell 1947).  Colonial administrators were bound to take this view-by their British upbringing, from professional self-interest and, not least, by the intensely local terms of their alliances with African 'chiefs'.  Officials were also buoyed up by the optimistic spirit of the post-war age.  The energies of 'the second colonial occupation' gave them a new sense of purpose.  Imperialism had two new pursuits to follow, now it was released from the miserly constraints the Treasury had imposed between the wars.  There were the dollar-earning agricultural exports, in whose production Africans were beginning to share, that promised to prop up the war-shattered pound sterling.  More upliftingly, socially engineered domesticities, civic communities, producer co-operatives and trades unions might, in the longer run, turn instinctively local tribesfolk into critically national citizens, fit for the rigours of self-government (Low and Lonsdale 1976: 12-16; Lewis 2000).[1]  All this ambition deafened the British ear to the African voice (Berman 1990: 327-34).  Nonetheless, KAU proved helpless to make local officials and British politicians change their mind.  From this lack of any political success its other failures can well be argued to follow.  The old KAU story is not wrong.  It was not, however, the only story.  Perhaps it was not the most important. .

In its organisation, next, KAU was undeniably uneven.  The party generated little support outside the highland Bantu peoples, Kenya's central ethnic cluster, one quarter of the population.  KAU was also ineffective.  It soon 'lost control' of those peoples' more desperate members, the young and the poor in land, often the underemployed in town (Rosberg & Nottingham 1966; Spencer 1985).  Having lost, indeed alienated, its more desperately determined spirits, KAU could not even defend itself against the British charge of acting as legal cover for the banned Mau Mau movement.  While their lack of success meant that the party's moderates had no means to quash the ardour of the Mau Mau militants, it was questionable for the British to treat their political failure as treasonous subversion.  In early 1953 a trial which raised many legal eyebrows at the time, and left many unpersuaded, convicted KAU's leading officials, Jomo Kenyatta at their head, on a charge of managing Mau Mau (Lonsdale 2000 a).  KAU was itself banned shortly after.  The uniform quality of the new activists who had meanwhile stepped, at short notice, into Kenyatta and his colleagues' shoes, showed just how impressive a more united KAU might have been.  Intellectually crisp, socially urbane, they gave diehard settlers the first shocking hint of how agreeable a multi-racial society, or at least its elite, might be (KAU 1952; Anon 1953; Awori 1953; Mumford 1953; Goldsworthy 1982: 19-20; Murumbi n.d.).  This thread of African political failure, British blindness, might-have-beens, is part of the KAU story.  It is the most visible, the central, plot.  There were provincial stories too.  They were different. .

If nationalism is understood as a distillate of class struggle, then in this context, too, KAU seemed hard to place.  The Union failed to promote in any publicly acknowledged way the interests of the rural capitalists who had most to invest in its moderate politics and who thronged its rural branches (cf Rathbone 1973).  The party seemed no better as the people's champion at the bottom of the social scale, among farmworkers, peasants, and towndwellers.  Some of its officials in the Rift Valley province failed to understand, let alone support, the resistance which the resident, 'squatter', African labourers mounted against their white landlords' new demands (Furedi 1989: 86).[2]  Pursuing the lure of post-war profit with the logic of mechanisation, anxious to prevent a common-law accretion of squatters' rights to their land, settler farmers strove to restrict their resident labourers' private liberties to graze livestock, and cultivate, on the broad-acred farms of the 'white highlands'.  The farmers were too late.  Squatters had often worked for two generations in white employment.  In their indigenous labour theory of value (Lonsdale 1992), their sweat had already made them joint heirs to white property.  KAU was also slow to respond to the angry fears of peasant households in the African Reserves.  Poorer families, especially in Kikuyuland, were facing landlessness, moral extinction.  They were also subject to the compulsory toil of soil conservation, laboriously terracing their hillsides against its erosion by tropical rain.  The government could see no other way to save African lands from further degradation, peasants from pauperisation, and white farmland, therefore, from an African clamour for restitution (Throup 1987: 139-70).  Finally, KAU at first distanced itself from the anger of Kenya's townsfolk, self-employed and migrant workers.  Then in 1951 the party was, in effect, taken over by Nairobi's impatient trades unionists.  The coup destroyed any last KAU claim to be pan-ethnically 'national' (Spencer 1985: 221-5; Throup 1987: 9-11, 188-96, 243-4).  .

This essay takes another view of the relationship between KAU and class.  Party officials knew the issues.  Their headquarters argued them, on paper-when the office existed, as sometimes it did not.  In 1946 KAU analysed Kenya's ills, in fifty-three typescript pages.  Its four authors, teachers and journalists (one of them an accomplished violinist), representing Kenya from the Victoria Nyanza in the west to the Indian Ocean in the east, put their finger on the nub of Kenya's problem.  The colony had not one government but two.  At the centre, the state claimed to be trustee for African interests.  In the white farming districts a second regime  perpetuated a 'modern serfdom' (KAU 1946: 7).  The new squatter contracts intended a 'new slavery' (1946: 5).  At its annual conference a few months earlier the party's president, James Gichuru, had recognised squatters as 'our unfortunate kinsfolk', suffering under 'a process of systematic exploitation unparalleled in history' (EAS 1946). The memorandum went on to analyse the plight of African traders, hindered by lack of capital, discriminated against in the issue of transport, trade and shop-building licences, which seemed to favour established Indian businesses.  KAU was equally informed on landlessness and the squalid conditions of urban labour; its office-bearers had read government reports.  The Colonial Office was impressed.  The Under-Secretary of State, soon to be Secretary of State, Arthur Creech Jones, thought KAU's case 'important and interesting'; it called for 'a reasoned reply'.[3]  There is no indication on file that it ever received one.  The CO's neglect shows what KAU was up against.  But African Kenya's political culture also needs to be understood.  To expect KAU to act more decisively on the part of traders, squatters or wage-workers is to to take the all-inclusive, all-mobilising view of nationalism, common among political scientists in the 1960s.  This takes no heed of how each of these interests, in their own localities, in their own search for authority, insisted that freedom could come only from arguing out how to help themselves. .

To return to, and conclude, my sketch of the received view: in sum, KAU's failure was seen as a failure of successive leaderships.  The Union's early, pan-ethnic moderates are said to have lacked drive and stamina; its later militant extremists were in ethnic terms too sectarian.  In the view of Rosberg and Nottingham (1966), the party's liberal modernizers 'lost control' of more impatient spirits; in Spencer's revision (1985) these latter, the urban radicals, while defeating the liberals, 'lost control' in their turn.  But this established historiography never specified a basis for political leadership, with its supposed capacity for control.  How far a vision of past and future might be a necessary part of leadership went undiscussed and with it, too, large questions about control.

What was missing from these earlier analyses was any enquiry into what I take to be the intimately related issues of political imagination and authority, the ability to sway a following without the use of force.  Did KAU ever have, did it ever create, the authority it is said to have lost?  That has been a seriously unasked question.  Rosberg and Nottingham said more than other historians on political purpose and imagination, but in truth it was very little.  It now seems that they may also have been largely wrong.  Kenyatta, they said, conceived of a 'Kenya people'.  This may be right, but they did not expand, so one cannot tell.  His vision was reinforced by 'cumulative social changes which in many ways challenged the meaning of the tribe as the highest political ideal' (1966: 217).  At the same time-it might be thought in contradiction-they thought urban tribal welfare associations became 'part of the infrastructure of the new nationalism' (210).  That Kikuyu should take the political lead was because they enjoyed greater 'social communication' than other ethnic groups (219).  The substance of this discursive activity was left undisclosed.  Kenyatta's vision of the future was something our two authors called 'territorial integration' (351)  Again, they did not explain.  The problem of how to imagine community is, by contrast, central to this present essay. .

Kenyatta's vision failed to create a common citizenry, a nationhood of political trust, at independence in 1963.  In Rosberg and Nottingham's view, this was the fault of British obduracy, not of the African political imagination (348-54).  In face of colonial resistance to reform, some Africans had perforce abandoned a legal for a violent pursuit of power.  Desperation narrowed a previously broad and tolerant political front, in which a Samkange would have felt at home, into a nationalist passion he would equally have deplored.  Militant fire coursed down the tribally specific channels which colonial rule's 'uneven social mobilization' had already scoured out.  Some ethnic groups had gained and suffered more than others and were therefore more politically determined, Kikuyu most of all.  Social change, it seems, had not 'challenged the meaning of tribe' quite enough.  Spencer (1985) had less interest in thought than in action; and for Throup, Kenya's Namier, the key conflict was between rival African elites, ins and outs, incorporated and spurned, chiefs and their KAU rivals.  Each dealt in the earthy coin of peasant clientage, not in airy political vision.  He even characterised the Mau Mau militants, who transformed the political landscape, as 'urban nihilists' (1987: 241).

More recently the Kenyan historian Atieno Odhiambo has argued, in his usual provocative way, that the historian's attempt to 'tie up' the 'different communities of thought, action and discourse into a narrative on Kenya's nationalism' in the formative KAU era is misguided, a waste of scholarly time.  This was an age of matatizo, mgogoro and matata-of complexity, tumult and commotion (1995: 26, 44).  KAU never had a control to lose; it was 'restlessness [that] served as an agency to action'.  In an unexpected echo of Dame Margery Perham, University aunt to the colonial service, writing in her Colonial Reckoning of forty years ago (1961: ch 2), Atieno thought nationalism sprang principally from the countless personal humiliations imposed by the kalaba, the colour bar (1995: 32-3).  What imposed order and narrative on this inchoate sentiment was neither KAU nor any imagination of a Kenyan people but the British counter-insurgent war.  This cleared the field politically for the forgotten moderates; and fuelled their competition with agrarian reform (1995: 42-3).  This essay arrives at the same conclusion, if by another route. .

What Atieno Odhiambo implicitly denied-and to which the earlier accounts paid little heed-is a thesis to which I remain obstinately committed.  To understand political history, that is, we need a concept I call 'deep' politics, the negotiated relationship between politics 'high' and 'low'.  In their high politics politicians handle that fissile material, power.  Low politics happens on the factory floor, in the fields, on the street, even the household (Peterson 2000: 318-74) when men and/or women are moved to act together, and make themselves a public, alert, demanding.  Deep politics is the field of tension between.  In it imaginative leaders may conjure up self-aware political communities.  By loss of vision, material failure, or by excess of greed, leaders can lose their followers too.  Publics, conversely, may invest in or withdraw from their leaders' reserves of loyalty.  The balance in their banks of deep allegiance determines how boldly leaders dare to engage in the high-rolling gamble of high politics.  Deep politics is the domain of moral narrative.  It is where people imagine, and dispute, the reasons for honouring or breaking their reciprocal demands upon each other, from on high or low.  It is the sphere of public memory, in which there was once honour, and could be again, or where injustice was inflicted that must now be undone. .

Atieno did however emphasise what others tended to forget, that Mau Mau was only one, extreme, example of mgogoro, tumult.  Other peoples, not only Kikuyu, confronted in their own way the divisive threats and promises of 'the second colonial occupation', in which British governments throughout eastern Africa tried to turn the colonies to better account.  That is the central point of what follows.  This essay suggests-it can do no more-how people sustained in their narrative imagination what they struggled to achieve in practice, that is, deep political agreements across the widening divisions of social status, high and low, middle class, peasant, and landless prole. .

Was all this provincial commotion connected?  Does it have to have been linked for Kenyan politics in this era to be construed as a unit of study?  Ranger had not insisted on a 'monolithic movement of common interests' before he called it nationalism.  If not in Southern Rhodesia, no more need it be demanded of Kenyans.  But Terry did write of his manifold Zimbabwean leaders as people with authority, however contradictory it may have been.  Moreover, their authority had had little help from the state.  Rhodesian Africans had notionally been entitled to vote on a common electoral roll since 1898, but in 1951 there were barely 400 on the register (Palley 1966: 135-6, 243-6).  They had never mobilised in hope of electoral victory.  The franchise had not delivered whatever coherence and power they possessed.  The same was true of KAU.  It was founded to support African members of the Kenya Legislative Council.  Of these there was one in 1944 on the party's formation, six by 1953 on the party's demise-in a total of 28 'unofficial' members, 14 of whom were European.  No African member was directly elected.  All were nominated by the governor, from preference-lists chosen by elected African district councils.  To Rosberg and Nottingham this was Kenya's great contrast with British West Africa.  In the Gold Coast (Ghana) or Nigeria, British flexibility had allowed African parties to play a constructive, moderate, role in the transfer of power (1966: 188-9).  Their contemporary, the political scientist Zolberg went so far as to say that it was to fight the elections conceded by the French and British, and for that purpose alone, that West African parties had stirred themselves.  Otherwise an 'organizational vacuum' characterised societies which did not take kindly to outside direction (1966: 14, 33-6).  This thesis from a previous historiographical age has received renewed support in Iliffe's outstanding history of the continent: 'Nationalism (as distinct from anti-colonialism) in West Africa was chiefly a response to elections' (1995: 234-5). .

This essay takes a different view of the KAU years in Kenya.  The party would of course have been stimulated to greater efforts had it been permitted to fight national elections-but Kenya would have to have been a very different colony.  The imaginative and political battles party members fought in the short KAU decade were mostly local (cf Iliffe 1995: 250).  But they were battles and they did, as already suggested, have a common theme.  Their provincial protagonists understood their counterparts from elsewhere when they later met-and confronted-each other, once the British conceded that legal African politics must return, after the defeat of Mau Mau.  The deep bases of authority differed in local geography but in narrative achievement were much the same.  Nationalism in Kenya, the imagination of community-as distinct from anti-colonial resistance to 'the second occupation'-was chiefly a response to gender conflict and class formation.  Local in sentiment it may have been.  The tensions of gender, class and generation which spurred on the search for community, were however felt all over Kenya. .

To understand the common issues in Kenya's different localities it is helpful to recall three events in 1946.  While unrelated, they were intimately linked.  In that year, released from the army, Gakaara Wanjau published Kenya's first vernacular novel.  Many young veterans, of whom there were over 90, 000, had received an army education; they were used to reading the news (Parsons 1999: 115-17, 189-92).  Gakaara knew he had an audience.  A vernacular press, if not quite mass literacy, was at hand.  Gakaara also knew his audience.  His novel was a tragedy, a love story, Uhoro wa Ugurani or Marriage Procedures.  An educated young woman, the heroine, was driven to suicide.  Her equally educated lover could never meet her father's ever more extortionate bridewealth demands.  Modern love and its consummated Christian respectability were destroyed by an ignorant elder's greed (Pugliese 1995: 150-62).  1946 also saw the collapse of the returning soldiers' Anti-Dowry Association.  As they waited for the boat home from Suez some had decided to organise a 'boycott' of girls whose fathers insisted on bridewealth (Kaggia 1975: 57-8).  But elders, future fathers-in-law, had more power than young men over women; Gakaara knew that.  The young men's fathers, too, controlled the distribution of property and with it access to adulthood, the right to marry.  Young men had to submit.  The African press carried as many letters on bridewealth as on politics.[4]   Gakaara's novel has never been out of print.  The inequality of propertied power was as sharp between Africans themselves as between Africans and settlers. .

1946 was also the year that Jomo Kenyatta became, very deliberately, a senior elder.  He returned in September from Britain, after sixteen years away.  His first actions were to marry into one of the most powerful lineages of Kikuyuland, that of senior chief Koinange, and, with the aid of his age-grade, to buy land and build an imposing house, with a large library (Murray-Brown 1972: 229-30; Kershaw 1997: 199-200).  Within a few months he was evasively denying that he had no authority over the young (Farson 1950: 113-14).  How many of the latter, one wonders, had read Gakaara Wanjau?  How many had given up hope in the Anti-Dowry Association?  Intergenerational conflict between men disputed the elders' power to decide the rules of honour.  It was precisely this battle over authority-where it lay, how it could be constructed, or challenged-which joined the high politics of leadership to the low politics of household formation.  A local deep politics of uncertainty and fear for the future emerged, a narrative domain dominated by the determination to work for an ordered, but scarcely equal, freedom. .

Among all the major agricultural peoples of Kenya, men and women were aware of growing conflict between the genders, between generations.[5]  Poverty was at the root.  Poor families, and they were many, suffered the worst fears.  Male migrant labour for increasing lengths of time, unaided women's responsibility for household land that was subject to more pressing litigation by senior neighbours, these were the key stresses on marital relations.  Kikuyu households were now plagued by internal charges of sorcery, at a time when the established churches had little to offer to the poor (Peterson 2000: 318-74; Lonsdale 2000 b).  Among their Kamba neighbours half the able-bodied men were absent from home in 1945 (Grignon 1997: 94).  Gusii marital relations would surely not have been so brutal were sons not becoming more dependent on fathers for access to land (LeVines 1966: 27-8, 40-54).  Luo women were attracted to a 'spirit' church partly in hope of healing from marital strife (Hoehler-Fatton 1996)-although Luo men were also grateful that the Hut Tax, which bore hardly on widows and abandoned wives, had in 1943 been replaced by a Poll Tax on males alone.[6]  What caused men most alarm was the way in which women from the best watered areas, and with easiest access to urban markets (here Kikuyu women were doubly advantaged) could gain financial independence, and some solidarity, by trading foodstuffs.  It was at this juncture of moral panic that African and British males showed common cause, to restrict women's immoral traffic, to impose rational male knowledge on peasant, which now tended to mean female, farming (Robertson 1997: 102-45; Mackenzie 1998).  These women were 'straddling' between productive sectors.  To supplement the returns on smallholder farming they earned an off-farm income too, just like their men.  But while men toiled in the contested order of wage-labour, women gambled on the dangerous enterprise of trade.[7]  Men found the contrast disturbing.  Women, it was widely supposed, traded more than vegetables, charcoal or grain.  Among the main purposes of the urban tribal welfare associations, was to repatriate 'loose women' to their rural homes.  As White has it, the associations imagined 'an authoritarian world in which reliable men entered into responsible relationships that were backed up by laws that would control the movements of wives and children' (White 1990: 190-94; also Mathu 1952). .

While all Kenya's agrarian peoples knew this tension, 'those without previous dealings', as Kikuyu say, 'have no cause to quarrel.'  These were internal strains, rarely inter-ethnic troubles.  Authority was contested, a more equal sociability could be constructed, only within a deep politics of mutual comprehension.  The discursive crucibles of changing social obligation were bound to be local, and vernacular.  'Moral ethnicity' is the term I have adopted for the consciousness they aroused (Lonsdale 1992: 461-8; 1996 a & b).  It could as well be called ethnic patriotism, a sense of how hard it is to be 'ourselves' rather than of a competition with 'others' (cf Parkin 1978: 291-2).  Ethnic groups were, above all, story-telling groups, the domain of deep politics, of public memory, however recently that may have been constructed.  Attention to public debate on the changing moral economy of reciprocity reverses the 1960s orthodoxy that social change 'challenged the meaning of tribe'.  To the contrary, colonial rule and social change was what first made 'tribe' meaningful in Kenya's formerly chiefless small societies.  .

Ethnic consciousness arose from an enlargement of moral community, not its contraction.  Two processes were particularly significant here.  First, the administrative geography which built state control became filled out with patronage and market networks, as social ambition formed around the state.  Administrative districts were filled with a legal and political culture, elements of ethnicity.  Vernacular Christian literacy, secondly, provided another element, when individually alienated converts read biblical images of ethnic destiny which shaped their local civilising missions (Lonsdale 1996 a).  Districts became not merely political arenas but imaginative domains.  Mau Mau's own internal moral economy, for instance, was argued out in face of the agonising question how far relations between generations-peculiar to Kikuyu and rooted in highly specific relations to land-must be rethought in order to promote effective political action (Lonsdale 1992, 1995; Kershaw 1997).  No other ethnic group faced precisely the same question; none could entirely enter Kikuyu moral ethnicity.  In the earlier literature, Mau Mau was the central question of Kenya's nationalism and its failure to build pan-ethnic trust.  If one first considers the need to establish trust within ethnic groups then Mau Mau becomes part of, an acute example in, the story of how all Kenyans tried to imagine wider communities than they had known before.  The question this essay now asks is how far the idea of moral (or civic) ethnicity-a disputed domain of honour and duty in changing times-might reopen the study of nationalism, without transgressing Atieno's proper insistence that good history must not in retrospect impose on any era a story line that none who lived through it would recognise. .

Happily, moral ethnicity is a Kenyan concept, if not so named.  Eliud Mutonyi, for one, would have recognised it.  Chairman of Muhimu , the Mau Mau central committee, he dictated his memoirs soon after independence (1968).  He asked why one who had fought for Kenya's national freedom should extol ethnic loyalty.  His answer was that nepotism was the problem, not tribalism-the term that Kenyans unashamedly use.  'Tribe', in his view, enlarged one's moral community, or sense of obligation, beyond family or clan.  To deplore tribe was to abandon the social disciplines, instilled since childhood, that gave people 'human dignity and moral strength'.  If 'healthy tribes' taught a wider citizenship than clannishness, they could also build a healthy nation.  Moreover, Mutonyi could see no reason why the discursive bounds of ethnicity should not expand-where experience taught that neighbouring others shared the same principles of social justice, dignity and public morality.  His was a cultural politics of cooperative self-determination, not of cultural or moral relativism.  Indeed, to Mutonyi-as to Kenyatta (1938: 251; and in Muoria 1947)-the latter, which they called 'detribalisation', was a 'process of cultural emasculation'.  It turned free people into slaves.  What destroyed a nation was not ethnicity but the sycophantic modernity of its detribalised elite, their sense of racial inferiority that fed a disdain for the shenzi  [savage] backwardness of their people and a love of money that laid them open to foreign purchase.  Mutonyi did not say so explicitly, but he could well have been criticising the KAU moderates, widely known as the teachers' party. .

KAU's most consistent doctrine was self-help.  Despite Mutonyi's disdain, it was the moral imperative that tied Kenya's few African secondary-school graduates, local boys made good, to the service of the rural localities which had so often clubbed together to fund their education, a ladder which in the 1940s could culminate at Uganda's Makerere College or South Africa's Fort Hare (Kipkorir 1969, 1972, 1974).  Kenyatta had gone further afield, to Moscow's University of the Toilers of the East-where he was put off Communism for life-and Britain's London School of Economics, where he learned the pleasures of bourgeois freedom.[8]   But he enunciated no 'national' cultural policy as KAU's president.  He prefered to adhere to the Kikuyu maxim that 'an ornament you find on the path may make you lose the one that is truly yours.'  Educated people, he warned, could be like a parrot 'with no idea of its own, but keeps on repeating what somebody else has said.'  Addressing a Kamba audience soon after his return from England he defined his multi-culturalism as a matter of mutual respect; nor did political unity presuppose cultural unity.  Like young men and women 'we need to decorate ourselves as well, we need to respect each other [as ethnic groups] and then speak in one voice' (Muoria 1947).  Eliud Mathu, first African member of Kenya's colonial legislature, not formally a member of KAU but a close associate of Kenyatta's, maintained in 1950 that differences in language and custom had not prevented India from becoming a republic.  India had 'made a nation out of a conglomeration of races'; Kenya could do the same (Roelker 1976: 76-80).  But it would have to be a conglomeration of self-respect.  At a public meeting in Taita, near the coast, Mathu had stressed that with even greater force than Kenyatta.  'No people in history has ever risen from the dust to the height of a respected nation without struggle.  No people has ever achieved freedom without working, fighting or dying for it.  No people has ever been given this freedom, real civic responsibility by another nation, just like that!'  Kenya's peoples must be 'busy as bees'.  Government should support their efforts, but if Africans allowed themselves to be 'spoon-fed' they would not only be poor but morally poor, slaves (Mathu 1947). .

Much of KAU's reluctance to follow a national cultural policy was founded on a long-standing male African distrust of towns.  Eliud Mutonyi had started his public career as an official of the Kikuyu General Union in the 1940s, part of 'the infrastructure of the new nationalism'.  His moral ethnicity was decidedly rural, thanks to his male elder's definition of ethnic virtue.  To him, towns meant wanton women and feeble men.  Kikuyu men had long thought so.  Their moral panic about urbanization had shaped the 'female circumcision crisis' of 1929 (Lonsdale 1992: 386-95).  Rising numbers of young, claimants on the elders' resources, now added fresh urgency to concern for generational discipline.  Moral ethnicity, at bottom, was a debate on how to link different rural localities in one (ethnic) system of rights and duties, despite the temptations of town.  Replicated in one rural locality after another, it was an effort to give order to mgogoro, tumult.  Kenyatta's world (reported in Muoria 1947: 15) was sharply divided between Kikuyuini, where parents could expect respect, and Gecombaini, Nairobi town, a place of strangers, whose young showed him none at all (Henderson 1952).  Moral ethnicity, to repeat, is not an historian's alien imposition but an actor's passionate view.  In his first known publication after returning from Britain, Kenyatta entrusted to Kikuyu the task of making 'our country [of Kikuyu] go ahead with stability'-na mbere wina ugariru (preface to Muoria 1947: 2). .

To an extent that has rarely been recognised, late-colonial Kenya did go ahead with remarkable stability.  The state had not lost its capacity to absorb and coopt African ambition.  As for KAU, so for the British, Mau Mau was not the only story.  Government revenues more than doubled in the war.  African local governments did still better; their income tripled.  Local 'establishments' emerged, composed of now-educated chiefs and diploma-trained African vets and agricultural officers.  African maize prices doubled from 1939 to 1945; the state-controlled profits were credited to the growers' district councils as agricultural betterment funds.  The women hawkers' vegetable prices rose still faster.  Two African districts had grown coffee since the 1930s.  The depression had made it clear that the state could no longer afford to protect settlers by denying coffee-growing to Africans; more districts were to join them by 1950.  In the richer districts, officials used their discretion to put more retail trade, especially in rationed clothing, in the hands of African traders, at the expense of Indians.  By no means all Africans benefited from wartime prosperity, however.  Instead, the gap between the comfortable and the poor widened.  In rural areas, as farm profits increased, so the worth of clients and dependents living on the land declined.  African big men were no different to white farmers in that respect.  Property was beginning to be worth more than patronage.  And the wages of urban workers lagged far behind the profits of peasant farmers (Lonsdale 1986: 124-36).  This distinction alone is almost enough to explain the political history of KAU. .

The difference between town and country is illustrated in the contrast between their constructions of leadership.  Urban leadership came from quick thinking in low-political crisis.  Rural leadership had to work hard to win the respect of public memory, the domain of deep politics.  Towns were not then the cradle of the Kenya nation (as they may be becoming today).  Mombasa, Kenya's port city, and Nairobi, its upcountry capital, were the towns that counted.  Mombasa had a long history, the stuff of local, competitive, ethnic myth (Willis 1993).  There is no evidence that the dockworkers who struck in early 1947, most of them from upcountry, knew anything of it.  Nor did the man who emerged as their leader, Chege Kibachia.  He had just arrived from Nairobi to represent his Kikuyu trading company.  This was his chief recommendation to the dockers, that he could put them in touch with KAU, whose newspaper he had helped to edit.  Conversely his motive for helping them was to improve urban living standards, for fear that their present squalor would render migrant workers morally unfit to return to their rural homes (Cooper 1987: 105).  Up in Nairobi, urban militants deliberately suppressed the memory of the nearest the Kikuyu had to a resistance hero, Waiyaki wa Hinga, because his dynastic authority rested on rural deference (Bildad Kaggia interview, 7 December 1984; Lonsdale 1995).  What made Nairobi's Mau Mau activists revolutionary was their determination to use the urban interests of artisans, workers, taxi-drivers, as the raw material of leadership.  They denounced rural leaders as a self-interested class, antagonistic to workers; workers, like others, must rely on themselves (Lonsdale 1992: 425, 430).  Yet the Mau Mau chairman, Mutonyi, owed his position in Nairobi to his rural transport business and sponsorship of dance troupes that he brought in at the weekends (1968).  Town leadership could not entirely dispense with its rural ethnic hinterlands. .

Kenya's countrysides, then, its terroirs as the French would say, were the domains of deep politics, the roots of KAU's different stories.  Rural leadership demanded imaginative social labour over time, to measure up to a deep politics of responsibility imagined within a local tradition.  One can point to at least six contrasting stories here.  At one extreme rural leaders could consciously repudiate their links with KAU.  At the other they might simply ignore the Union.  Most districts were somewhere in between.

The 'white highland' squatters were the community, largely Kikuyu in origin, who hoped most of KAU and were most cast down.  KAU's rural core of patronage, southern Kikuyuland, was for many their ancestral home.  Ties of reciprocity were worth testing.  But as settler pressure to sign the new, restrictive, contracts mounted; as government resettlement measures began to look more like oppression than relief; as Kenyatta and his southern Kikuyu establishment achieved nothing; so squatters turned to self-help.  They had reason to distrust all patrons.  Many of their forebears had been tenants of Kikuyu lineages which had not helped out when their land was first taken, forty years earlier, for whites.  KAU's tie-wearing teachers, the taitai, seemed to have no comprehension.  Settler landlords were repudiating client rights.  Squatters had nothing to trust but the sweat they had themselves invested in white farmlands, in whose private corners they had circumcised their young and buried their dead, equally hallowing commitments to the soil.  Their self-help implied a distrust of superior African outsiders as fierce as some of the worker militants in town.  Squatters, it seems, were the first to convert Kikuyu oaths of solidarity against envious-and politically treacherous-sorcery into a commitment to kill, led by men whose trade in squatter produce enabled them to imagine a domain of conflict wider than each single settler farm (Kanogo 1987: 96-149; Throup 1987: 109-38; Furedi 1989: 75-145; Lonsdale 1992: 417-20, 426-7, 438-9). .

Two peoples who more or less ignored KAU were the Kamba, not forty dusty miles from Nairobi, and the Gusii, in the far west of the colony, whose well-watered hills overlooked the Victoria Nyanza.  Their stories could not be more different.  Kamba neglect of KAU had to do with gender, Gusii self-absorption to do with property, not that the two can ever be disentangled.  But Kamba energies were led by their myriad clan elders and disciplined by their juniors, soldiers returned from the war, both of whom aimed to control the labours of women.  Restoration of Kamba patriarchy and the toil of soil conservation fitted together under the aegis of the colonial state, which called the alliance a 'miracle'.  It was indeed astonishing how little the Kamba, culturally so close to Kikuyu, had to do with either KAU or Mau Mau in building their local politics of clannish male authority (Grignon 1997).  Kamba country was as much pastoral as agricultural; the Gusii hills, untouched by white settlement, were among the richest farmlands of Kenya.  Gusii had done well out of the war; thereafter their coffee began to boom; their rounded hills did not need to be terraced against erosion.  Cooperation with the improving aims of the second colonial occupation was easier than agreement between Gusii themselves.  In 1945 John Kebaso, among KAU's founder members in Nairobi, returned home to start the Kisii Union.  It represented the owners of water-powered maize mills, often the harbingers of rural capitalism.  One of the Union's aims was to abolish laziness; as Mathu had urged on the Taita, Gusii must become busy as bees.  Indeed they were, but on their own account.  There was no local KAU branch.  Instead, there was massive litigation between Gusii, as grain and coffee land soared in value.  Gusii were aware of the vulnerability of the poor.  Proverbially, 'the poor man's property belong[ed] to the rich man.'  But there was little solidarity among the rich: 'rich men fear each other.'  The local politics of responsibility was, unusually but uncontroversially, aimed outwards, against the 'other', the economically less favoured but more numerous southern Luo, with whom Gusii shared a district until it was subdivided in the late 1950s (Maxon 1986; 1989: 114-48; proverbs in LeVines 1966: 10-12).  The internal politics of moral ethnicity was never easy, least of all among Gusii. .

The districts in the middle of the KAU spectrum of rural connection proved that well enough.  KAU knew it had to gain a following in western Kenya, the then Nyanza province, land of the Gusii, Luo and Luyia.  These peoples were numerous, educated, comparatively wealthy.  They had political resources, their allegiance would counter accusations of Kikuyu domination.  For a time it looked as if the Bantu-speaking Luyia of North Nyanza might provide valuable support.  But KAU was dragged in to the local geopolitical quarrels which arose when different histories, different rates of economic change, tore the recently imagined unity of the Luyia apart.  The KAU connection bounced from the north, to the middle, to the south of the district in quick succession.  In the north, the Bukusu congeries of clans drew strength from their large farms and memories of fighting the British fifty years before.[9]  Then the KAU branch was taken over by the main allies of British conquest, the Wanga people, whose rulers were the nearest any Kenya group had to a chief.  Their ex-chief Joseph Mulama, one of several rival brothers, had considered seceding to Ethiopia in the 1930s in search of a local kingship and still harboured hopes of Uganda, while keeping his options open with Kenyatta.  But he was still keener to keep in with the British.[10]  So the KAU link shifted to the crowded and impoverished south of the district, the area with the strongest migrant labour ties to Nairobi.[11]  Here was a story of how a conservative local politics could, with much argument, become more radical in mood, more central in focus. .

The Luo story was more equivocal.  Luo historical discourse was not about resistance to the British but about their own south-eastwards colonisation of the land from their ancestral home on the upper Nile.  All Luo carried an historical atlas in their head of how their continuous migration of tribes and clans nested together in an expanding network of seniority and affinity.  The journeys of the JoRamogi-the people of Ramogi, the Luo progenitor-continued profitably into the twentieth century in a competitive game of colonising the employment of the railway and the state (Whisson 1963, 1964; Cohen and Atieno Odhiambo 1989).  Luo competition between themselves coloured their memory of how they had personally pioneered the building of the British state.  Regimental sergeant major 'the warrior' Jotham Simon Petero Okeyo Ogara, for instance, fought the Germans in the Great War as a Luo warrior, behind a British colonel it is true, but one known as 'son of the witchdoctor', wuod ajuoga; his victories were not British but Luo triumphs (Okeyo 1968).  John Paul Olola, agricultural instructor since 1913, angrily recited to the governor in 1953 what he had done for the British.  He had introduced the first milk separators to Nyanza, taught how to drain swamps for planting of rice and cotton, helped drive the Germans out of Tanganyika in the Great War, like the sergeant major; donated two cows and a son to the Second, instigated local government and legal reform.[12]  Olola, to come to the point, was right-hand man to Oginga Odinga, founder of the Luo Thrift and Trading Corporation (LTTC) in 1945, rather silent member of KAU and admirer of Kenyatta, soon to be the chair of the Luo Union and at independence, if briefly, Kenyatta's vice-president.  Odinga himself had as progressive a past.  As a boy, he had been bag-carrier for a peripatetic Luo Anglican priest and then a Makerere-trained schoolmaster.  Like Kenyatta, he married into a chiefly family.  After the war he put on his wellington boots to itinerate throughout Luoland in order to drum up commercial support.  The LTTC would demonstrate that by economic effort the Luo deserved political emancipation.  Its busiest line, and source of Odinga's local political wisdom, was as building contractor for African local authority schools, health centres and markets.  The LTTC also got the official sugar-transport business (Olola diaries; Okutu Bala interview, 8 April 1965; Odinga 1967; Atieno Odhiambo 1975 a & b).  Luoland represented the best, perhaps the only, example of an emergent local bourgeoisie, friendly to but not effusive in its commitment to KAU, which also enjoyed a close alliance with the district establishment of chiefs and local councillors.  Luo moral ethnicity encouraged that.  Odinga knew his clan connections.  His definition of freedom embraced, as it should, the accumulation of socially responsible wealth (Odera Oruka 1991, 1992).  Luo migration continued on the land, not to the 'white highlands' as vulnerable squatters but into Tanganyika as free settlers.  Good missionary education opened up the white-collared employment opportunities of the state.  Luo tradition made no great thing of the opposition between the generations. .

This was what made Kikuyu different.  They might have rebelled in any event, they certainly had good reason.  But what made their internal division between moderation and militancy so violent, what made Kenyatta the elder's leadership so enigmatic, was the Kikuyu institution of graded age-sets and cyclical generations.  Opposition between young and old was expected; conventionally, it renewed the health of the land.  Authorised Kikuyu tradition, unlike that of the Luo, said nothing at all about Kikuyu migration; they had 'always' been there-unlike the white settlers (Kenyatta 1938).  Their modern myth was not the Kikuyu colonisation of the state but the white settlement of Kikuyu land.  The nearest they had to ethnic heroes, Waiyaki wa Hinga the dynast and Chege wa Kibiru the prophet, were enlisted on opposite sides-one the champion of of elder-controlled lineages, the other of the generational hopes of the young-as Kikuyu argued how to renew themselves, in order to master the British (Lonsdale 1992, 1995; Kershaw 1997).  Kenyatta, and KAU, tried to straddle this argument of moral ethnicity, and failed.  Mau Mau could scarcely do better and plunged not so much Kenya as Kikuyuland into civil war. .

The discourses of moral ethnicity, then, had many narratives.  The fate of the moderate KAU showed not how difficult they were to link together, for that was scarcely tried, but how unforgiving they were when ignored.  Half the KAU's first executive came from one school, the Protestant missions' Alliance High School.  Most had a high sense of local duty; they had been sponsored by their clan, their local mission and their local education board.  Among its central officials, KAU before 1947 was largely an embodiment of the old school tie (Kipkorir 1969, 1972, 1974).  They were however, rather rarely countrymen after the manner of Odinga or Kenyatta.  Their strengths and weaknesses were exaggerated in the persons of Francis Khamisi and Tom Mbotela.  Among the most committed of KAU's officials and the best educated, both were descended from freed slaves of Central African origin.  Without a local ethnicity they would on some assessments be the only true 'Kenyans' KAU possessed.  But certainly not in Mutonyi's view, perhaps not Kenyatta's either.  Mbotela in particular, for a time KAU's vice-president, seemed to show the ethnic patriots the worst defects of the educated, an apparent sense of cultural inferiority.  When he became one of Nairobi's first African city councillors in 1952 he observed 'We [Africans] must free ourselves from primitive and other out-of-date ideologies like tribalism, racialism, hooliganism and other shenzi things which used to exist among us before Livingstone and Stanley came to Africa' (Harris 1987: 133).  It is not surprising that he was murdered, presumably on the orders of Mutonyi's Muhimu, not two months later, or that he was on the police list of possible crown witnesses for Jomo Kenyatta's trial (Lonsdale 2000 a).  But Mbotela had no sense of cultural inferiority.  To the contrary, his bookish liberalism made him highly critical of the unBritish authoritarianism of colonial rule.  Kenyatta and Odinga, in an idiom totally different from that of Mbotela, were more critical of Africans themselves.  It was not, to their mind, that Africans were shenzi, but that modern life was undermining the inherited disciplines that moral ethnicity must restore..

One must end with contradictions.  Politics is like that, as KAU knew.  Mbotela expected much of the state; Kenyatta and Odinga expected more from their people.  Territorial nationalism, like moral ethnicity, was a forum for argument.  The Kikuyu journalist Henry Muoria and the Luo businessman Odinga both saw the connections between indigenous capitalist competition and the need for democracy.  Kenyatta too, (1938: 188, 198) had extolled the democracy of universal circumcision in the Kikuyu past, in order to condemn the colonial suppression of 'the spirit of manhood'.  But in 1947 (Muoria 1947) he also sighed that such democracy made leadership impossible.  Since all Kikuyu were circumcised they thought all should have a voice.  Moral ethnicity was not conducive to the concentration of authority that alone might make KAU effective.  Luo made a different sort of distinction between moral ethnicity and the craftiness of power.  By the 1950s they had two archetypally different leaders.  There was Odinga, countryman, collectively minded, obedient to deep-political tradition, genealogically well connected-but in national high politics rather uncertain of his footing.  There was also the arrogant, individualist, Tom Mboya, too clever by half but whose wiles secured power (Parkin 1978: 215, 219-21).  .

KAU never attained power; but faced its contradictions.  The oppositions between the deep politics of critical moral ethnicity and the wiles of high politics have since independence in 1963 continued to defend Kenyans from dictatorship.  Kenyans retain the critical mind of citizens partly because they are not detribalised.  Odinga, Mutonyi and Kenyatta can, on that score, rest in peace.  There is still good reason to study the history of KAU.  Its different stories did not all fail. .

This paper was first published in the Journal of African Cultural Studies 13 (2000), 107-24, a special issue in honour of Oxford Professor Terence Ranger.

[1]           The concept of the 'second colonial occupation' has now been fruitfully transported to southeast Asia by Harper 1999.

[2]           And the KAU is scarcely to be seen in Kanogo 1987.

[3]           Creech Jones, minute, 23 September 1946.  Public Record Office: C0. 533/537/38672.

[4]           Kenya National Archives (KNA): PC/RVP. 2/27/34 (1946-49) contains Kenya Information Office summaries of African opinion in the English language, Swahili, and vernacular press.

[5]           For reasons of space I have omitted consideration of the major pastoral peoples but similar remarks would apply to them.

[6]           Central Nyanza Local Native Council to Governor, 11 October 1944.  KNA: PC/NZA. 4/1/2/1.

[7]           See Kitching 1980 for a generalisation of the late Michael Cowen's illuminating concept of 'straddling'.

[8]           Points to be developed further in Bruce Berman and John Lonsdale, The House of Custom: Jomo Kenyatta, Louis Leakey, and the Making of the Modern Kikuyu, a work in progress.

[9]           John Victor Khatete, secretary to the Bukusu Union (a KAU branch) to various government officers, 19 February 1946.  KNA: PC/NZA. 3/1943/16.

[10]         Ken Hunter (Nyanza provincial commissioner) to KAU North Nyanza branch, 14 December 1948; same to Superintendent of Police, Nyanza (enclosing Mulama to Kenyatta), 1 April 1948.  KNA PC/NZA. 3/1943/86 & 74.  See also Osogo n.d.

[11]         J. N. Katithi (KAU general secretary) to Chief Secretary, with resolutions of KAU Maragoli branch, 11 December 1950.  KNA: PC/NZA. 3/1943/113.

[12]         Olola to Baring, 10 January 1953.  Olola papers, seen by courtesy of the late John Paul Olola at Kibos Farm, Kisumu, in 1963.  Olola's diaries recorded still more variegated supports for civilisation.