Kenya has for many years been a favourite tourist destination, at least partly because it is the natural locus of white fantasies about Africa. All the elements are there: spectacular landscapes, teeming wildlife, "picturesque" natives, colonial kitsch, Isak Dinesen's Out of Africa and Walt Disney's The Lion King. But there is also a sinister side: memories of Mau-Mau insurgency, the arrogant settlers of White Mischief, Bruce Berman's Unhappy Valley, Robert Ruark's novels, runaway crime - and today this, violence in paradise.

Gérard Prunier is research professor at the University of Paris. He is the author of The Rwanda Crisis: History of a Genocide (C Hurst, 1998), Darfur: The Ambiguous Genocide (C Hurst, revised edition, 2007), and
From Genocide to Continental War: The Congolese Conflict and the Crisis of Contemporary Africa
(C Hurst, 2006)

Also by Gérard Prunier in openDemocracy:

"Darfur's Sudan problem"
(15 September 2006)

"The DR Congo's political opportunity"
(14 March 2007)

"Chad, the CAR and Darfur: dynamics of conflict" (18 April 2007)

"Chad's tragedy"
(7 September 2007)
"Sudan between war and peace"
(1 November 2007)

"Khartoum's calculated fever"
(5 December 2007)

"Kenya: roots of crisis" (7 January 2008)

"Chad: between Sudan's blitzkrieg and Darfur's war"
(19 February 2008)
The crisis that consumed Kenya between the presidential election of 27 December 2007 and the power-sharing agreement of 28 February 2008 (aasuming the latter really does mark its end) - has added a new dimension to this extraordinary duality. The violence that followed the declaration of incumbent Mwai Kibaki's victory over his main challenger Raila Odinga has taken more than 1,000 lives (perhaps as many as 1,500) and displaced around 250,000 people. It has exposed ethnic faultlines and political flaws that will - even with sound leadership, progressive ideas and huge investment of money and time - will take years, even decades, to heal. Yet the hiatus in the attempt by former United Nations secretary-general Kofi Annan to find a resolution to the political impasse indicates that even an interim solution is not yet in sight.

Kenya's trauma continues - albeit in a lower key than in the first, shocking weeks of killing and expulsion in January-February 2008, now rather neglected by a world media caravan endlessly in search of other headline disasters. As it does so, the wounds fester, the weapons are counted, the crimes are unpunished, the people wait. It is an impossible situation which cannot be allowed to continue.

The responsibility to see Kenya clearly and whole is even more important in such a period. But the exotic imagery - benign or malign - with which I began is an obstacle here. For in the portrait of the recent unleashing of "incomprehensible" violence, the old clichés can be seen steadily and unerringly resurfacing. Many of today's reports and analyses depict today's "rampaging ethnic gangs" as tribal fanatics or as manipulated nitwits used by evil politicians for their own dark ends (an image effectively identical to that of the Mau-Mau of the 1950s).

In this kind of language and coverage - and of course there have been contributions of hugh quality too, not least in the series of articles published by openDemocracy - what is missing (even inadmissable) is the notion that what is happening might be a systemic problem rather than one of wrong behaviour. If this insight seems hardly bearable, it is because of what long-term observers of Kenyan reality David W Cohen and ES Atieno Odhiambo have called "presentism".

In the Kenyan context, "presentism" has been a way of explaining the present reality by...the present reality. This involves a tautological concentration on the "latest news", a self-explanatory universe where successive and self-referential paradigms are offered as a miraculous potential solvent of the country's problems: nationalism (1960s), "development versus dependency" (1970s-1980s), "democratisation" (1990s), and - now - the exhaustion of "African pathology". "Presentism" is an intellectual and political pathology which goes way beyond Kenya or even Africa; but it has been strongly influential in Kenya in particular, perhaps precisely because this country has been for so long perceived as a figment of the white man's imaginations, hopes, fears and quick-fix solutions.

Kenya's shadow politics

For the west, Kenya has long been a necessary bulwark of stability both internally and internationally. During a crucial period of the cold war, Somalia under Siad Barre was pro-communist before sinking into civil war; Milton Obote's Uganda was "socialist" and then sank into Idi Amin's folly; Ethiopia went communist in 1974; and Julius Nyerere's Tanzania was suspiciously neutralist. This left Kenya as a very lone bastion of pro-western capitalism in the region - something which was conveniently equated if not with fully-fledged democracy then at least with some kind of semi-democratic dispensation.

A sort of political wishful thinking in the west imaginately translated Kenya's friendly foreign-policy stance into the practice of benign domestic politics. During the Jomo Kenyatta era (1963-78), western powers were so happy to see a man they had deemed to be a dangerous terrorist turn out to be a pro-western statesman that they chose not to look too closely. The result was that they paid very little attention to the land-grabbing that went hand-in-hand with the "land reform" of the mid-1960s; to the frequent imprisonment of political dissidents; and indeed to the political murders that constituted the final resource of the regime when it felt threatened.

A series of precise, targeted political assassinations (Pio Gama Pinto in 1965, Tom Mboya in 1969, JM Kariuki in 1975) enabled Kenyatta to fine-tune political control. The target was always the same: something seen as "radicals", in the American sense of the word, i.e. people who were of a strong democratic persuasion and whose international politics might be inspired by a dangerous domestic idealism. The members of this category could come in several guises: Pio Gama Pinto was a communist sympathiser, Tom Mboya a pro-United States democrat and JM Kariuki a kind of non-denominational populist.

More than any comparable African pattern, their murders resembled those of political activists in Latin America at the time. Somehow, moreover, the Luo/Kikuyu tribal rivalry always found itself at the heart of the controversy: Gama Pinto, an Asian, was a legal advisor to Jaramogi Oginga Odinga, the Luo (and allegedly pro-Russian) rival of Jomo Kenyatta; Tom Mboya was politically on the opposite side but a Luo anyway, and suspected of wanting to challenge the pro-western "good" Kikuyu; and Kariuki, although himself a Kikuyu, was accused by fellow Kikuyu of spoiling the broth and betraying the tribe. It is in this climate that Raila Odinga, Jaramogi Oginga Odinga's son and present leader of the Orange Democratic Movement, grew up.

A system of ethnicised class domination supported by the British and the Americans grew up around the "reformed terrorist" Kenyatta. The Kikuyu grew to be at the centre of this new system - but as an elite, not as a tribe. This is where the temptation to blindly "ethnicise" the present crisis are wrong. Between 1963 and 1970, Kenyatta and his entourage developed a powerful ethno-class in the central province, with a weaker pseudopod in the Rift valley. This was done through the Kenya Coffee Growers' Association which, although legally private (it was a cooperative) was a main recipient of government financing. The beneficiaries were the "good" Kikuyu who - more often than not - were the families of the anti-Mau-Mau "loyalists" who had supported the British during the 1952-56 "emergency".

The former Mau-Mau families, often impoverished as a result of detention during the emergency, lost out to the loyalists in the post-independence competition for land (the "million-acre scheme" and later the "Haraka scheme"). The result was the creation of a Kikuyu ethno-elite, entailing the social, political and economic domination of a segment of the tribe (which had quickly used its coffee money to branch off into transport, light manufacturing and juicy government sub-contracting) over the rest of Kenyan society - including the other Kikuyu.

Non-Kikuyu members of that elite - like Daniel arap Moi, Kenya's president during 1978-2002 - followed a set pattern: first, submit to the Kikuyu elite; second, use members of that elite against the rest (one of the Kikuyu used by Moi to manipulate the others during 1978-85 was his vice-president of the time, Mwai Kibaki, like himself a former "young Kenyatta boy" - but a Kikuyu one); third, once the necessary crowbars had prised open the Kenyatta ethno-elitist system, toss it aside. This cycle left, as before, two sets of opponents pitted against the post-Kenyatta ethno-elitist system: disenfranchised leftist Kikuyu and Luo of all stripes (the more radical, the more opposed). This opposition coalesced in the ethno-political coup of August 1982 which was crushed by Moi with relative ease.

Indeed, August 1982 marked a turning-point: the appearance of violence no longer as targeted individual murders but as a globalised tool. The 1980s still saw mostly "classical" repression, with the so-called Mwakenya "revolutionary movement" whose members (Wanyiri Kihoro, John Khamwina, Paddy Onyango, Mukaru Ng'ang'a, George Anyona, Koigi wa Wamwere, Raila Odinga himself) were a multi-ethnic cross-section of the Kenya intellectual and political elite. This was the period when President Moi called the multiparty democracy supporters "tribalists surviving on borrowed ideas".

Moi had tried to replace the Kikuyu ethno-elite with his own Kalenjin group. But this did not work, for a variety of reasons:

* In power for over fifteen years, the Kikuyu ethno-elite had had time to entrench itself in the civil service and in private business; the Kalenjin elite was a comparative latecomer

* Even if the Kikuyu had been divided among various geographical locations (at times called "mafias") like Kiambu or Murang'a, basically they were one. This was not true of the so-called "Kalenjin", who are in fact a British colonial regrouping of seven different tribes, two of which (the Nandi and the Kipsigi) had not bought into Moi's attempted system of ethnic domination and had been labelled "opponents" (and brutally treated as such)

* The movement for "democratisation" was so widespread and trans-ethnic that trying to both fight it and fight the old Kikuyu ethno-elite at the same time proved to be too much in the long run

* The old Luo rivals were waiting in the wings, ready to pounce, using "democratisation" as the bandwagon.

The patterns of violence

By the early 1990s, Daniel arap Moi was getting desperate. When foreign minister Robert Ouko (a Luo) was murdered in 1990, a Scotland Yard inquiry found that two people were "directly implicated": Hezekiah Oyugi, Moi's secretary, and Nicholas Biwott, Moi's cousin, the energy minister and the president's heir apparent. The days of Gama Pinto and JM Kariuki were back - but this was not enough any more. The system was now bursting at the seams and more direct violence became necessary to keep it operational.

During 1991-92, in preparation for the December 1992 election, organised violence by the General Service Unit in the Rift Valley (the very area where violence is now again prevalent) was developed by the regime against the Luo, the Luhyia and the Kikuyu settlers. Nobody knows exactly how many were killed but a figure of around 1,500 (i.e. higher than that of the violence in December 2007-February 2008) is conservative. Journalists who denounced Moi's ethnic militias at the time were arrested. But the west's protest remained mild, even after an International Monetary Fund (IMF) communiqué in November 1991 stated that 50% of Kenyan-owned assets ($1.31 bn) were in government hands and had been transferred abroad over previous five years.
Also in openDemocracy on Kenya's crisis of 2007-08:

Peter Kimani, "A past of power more than tribe in Kenya's turmoil"
(2 January 2008)

Michael Holman, "Kenya: chaos and responsibility"
(3 January 2008)

Gérard Prunier, "Kenya: roots of crisis"
(7 January 2008)

Roger Southall, "South African lessons for Kenya"
(8 January 2008)

Wanyama Masinde, "Kenya's trauma, and how to end it"
(9 January 2008)

John Lonsdale, "Ethnicity, tribe, and state in Kenya"
(17 January 2008)

Angelique Haugerud, "Kenya: spaces of hope"
(23 January 2008)

Anna Husarska, "Kenya's displaced people: a photo-essay" (5 February 2008)
The December 1997 elections were a remake of 1992, but with a vengeance. Moi lost all restraint and unleashed systematic ethnic violence all over the country:

* Kalenjin elements deemed "treacherous" like the Nandi and Kipsigi were hit

* More violence was organised against the Luo and the Kikuyu settlers in the Rift Valley

* The government incited its old Luo enemies against their Kisii and Kuria neighbours in order to intervene later and "re-establish order"

* On the coast, the Likoni area which was known to be anti-Kanu (the ruling party), people were literally thrown to the dogs with ethnic militias ransacking whole villages and communities; they particularly targeted the Mijikenda tribe which was suspected of sympathy with the "fundamentalist" Islamic Party of Kenya (IPK).

Around 2,000 people were killed. This is why it is surprising to see people today express astonishment at "ethnic violence in the Kenyan paradise" - when 1992 and 1997 witnessed much worse cases of violence than December 2007-February 2008. Why is there such a selective memory? Because the socio-economic patterns of 1992 / 1997 and of 2008 are largely different:

* In 1992 and in 1997 alike the violence was organised - a state-sponsored phenomenon where the government killed whom it wanted, through the agency of its own paramilitary forces, with its own money, for its own purposes, for as long as it needed to kill and for no longer than it had to

* When the government kept killing after the election (such as during the raids against the Kikuyu in Narok and West Pokot in 1998), it was because the regime felt these areas should have voted "well"; that they had not; and therefore they should be properly "punished" until they saw the light

* The centralised and organised nature of what was at the time coyly called "ethnic clashes" drastically limited the economic fallout. This was "clean" violence, only poor people were killed and infrastructure was hardly damaged. On the coast the tourists - some of whom were stranded during bouts of violence - soon left and were replaced by others who knew nothing and were happy to enjoy the sun and the beaches in blissful ignorance. Thus the economic impact was minimised and the foreigners little disturbed

* Everything was blamed on evil old Moi instead of seeing Moi as simply the latest - even if perhaps so far the most repulsive - incarnation of a political and socio-economic system which had been born with independence.

In many ways the December 2002 elections were the peak of political alienation and blindness; because, once that old devil Daniel arap Moi had been sidelined, there was a pervasive assumption that happiness and "good governance" were just around the corner .

A case of false consciousness

There were many disturbing signs that, the "democratisation" phenomenon notwithstanding, the system was alive and well. One of those was the growth of the Mungiki politicised sect, a Kikuyu formation. The Mungiki were, in many ways, the product of the Kikuyu's partial exclusion from the spoils of power during the Moi years. Kenya politics are not ethnic, they are ethno-fragmentary: that is, winners and losers are tribal segments or sub-groups rather than whole tribes.

The Mungiki rank-and-file are the lost boys of the ethno-elitist system. The behavioural cycle of these young and unemployed kids follows a familiar pattern: street gangs, religious radicalism (neo-traditionalist "Kikuyu pagans" to Muslims, and when that still didn't work, back to evangelical Christianity), larger-scale organised crime, political extortion. But the Mungiki made the wrong choice. In trying to ingratiate themselves with the powers that be, they picked as their candidate Moi's last protégé, Uhuru Kenyatta, son of the first president and a lacklustre character whom Moi used to manipulate. Uhuru was trounced in December 2002 and the Mungiki boys found they had landed themselves in the political wilderness. The only thing left was a return to their originating social violence, to which the police reacted by shooting them like dogs.

They were perhaps not the nicest young men around, but the treatment meted out to them was atrocious: police death-squads hunted them down in the slums. This was done under the "democratic" government of President Mwai Kibaki, the man elected in December 2002 as the carrier of the hopes of the democratic, multiparty NARC opposition which had finally managed to oust Moi, the symbol of corruption and oppression.

The fate of the Mungiki boys was hardly a central problem in post-2002 Kenya, but it was sadly emblematic of the fate of such (very large) fringe groups in a society that had begun to experience record economic growth. The average age In Kenya is 18, and in 2002-07, 3 million new voters have been added to the electoral roll. These are the people the World Bank-approved 6% rate of (jobless) economic growth has left discarded on the side of the road; and it is from their ranks that the people engaged in the "ethnic" killings are drawn. The place of prominent members of civil society at the opposite end of the socio-political spectrum should have been another warning signal if anybody had cared to look. NARC and the reform movement had used civil society as a kind of glue with which to bind the various social groups and tribes which had been marginalised by the (now) forty-year-old ethno-elitist system. It was fraying; but no one noticed.

Since the victims of the system were much more numerous than its beneficiaries, it had not been too difficult to get a majority for virtue. Moi had painted himself in a corner and the victims (the Luo, the small tribes, the disenfranchised social groups, the non-establishment Kikuyu) all found themselves on the other side. Raila Odinga - son of the old Luo radical, former "communist" Mwakenya suspect and jailbird - had supported Kibaki. This groundswell of hope found its most visible expression in the appointment of John Githongo - former chairman of the Kenya branch of Transparency International - as the anti-corruption Czar.

In the 2004-05 period, Githongo did a very good job. Too good in fact: he unveiled a series of massive scams - from the Anglo-Leasing "system's" fake security contracts worth hundreds of millions of dollars to the bizarre acquisition of an un-needed and never delivered "oceanographic vessel" costing €52 million from a non-existent Spanish shipbuilding. This is where Kibaki the "great reformer" completely betrayed the trust of the electorate and showed himself to be the direct heir of two presidents, Jomo Kenyatta and Daniel arap Moi, in protecting his corrupt associates working within the framework of the ethno-elitist system and distributing stolen money to his supporters in descending order of ethnic weight and social prestige.

Kibaki had become a (willing?) prisoner of a system run by such men as former defence minister Njenga Karume, finance minister Amos Kimunya, and former security minister Chris Murungaru; businessmen like Nathanel Kangethe, Joseph Kanyogo, and Nicolas Wanjohi; and the menacing interior minister George Saitoti, who had exploited his half-Maasai ancestry during the Moi years when killing Kikuyu was the preferred way of keeping electoral control, and who is now playing his other, half-Kikuyu parentage when the "house of Mumbi" (the Kikuyu's imaginary ancestor) is back in control. John Githongo knew what he was up against. When he realised that Chris Murungaru and a number of others were after him, he prudently fled to Oxford in order not to add his (Kikuyu) name to the long list of Kenya's victims of political assassination.

The end of the road

The system's violence has escaped its usual manipulators. This indeed is a clue to the inner character of the present anarchic situation and of the west's horrified reaction: namely, that the 2007-08 killings, contrary to those of 1992 and 1997-98, are horribly "democratic". Some observers have noted that many of the murderous rampages were also organised, but in fact the two interpretations are not contradictory: ethnic killings are occurring in spontaneous reaction to the frustration of the electorate at having been cheated out of the possibility of replacing its fallen reformist champion by a new one, while at the same time diverse groups of "big men" have instrumentalised the spontaneous violence as a route to power and pillage.

The Mungiki boys have been rescued from the political wilderness and are now moving into a new role as defenders of the Kikuyu community. Various other ethnic militias are being organised, armed and trained. The attitude of the international community, as exemplified by Colin Bruce - the Kenya representative of the World Bank (whom every street kid in Nairobi knows is renting his house from the Kibaki family) - has not helped. Its approach is to consider the violence as a kind of popular madness fuelled by the obstinacy of two men, Mwai Kibaki and Raila Odinga: everything would be solved if only they "really talked" and if their followers could be persuaded to desist from their "crazy ethnic prejudices". In that case, the "legal order" would prevail, the incumbent would be "re-elected" and the "challenger" would accept some juicy face-saving compensation.

This approach is a complete denial of the systemic nature of the problem. But the international community cannot look realistically at the facts because it keeps treating a corrupt, ethnically biased and violent system as a constructive development partner simply because it has been capable of achieving a (jobless) rate of economic growth of around 6%. It is this attitude which fuels the present violence: regular young unemployed boys kill other regular Kenyans out of frustration at not being able to reach those who are really responsible for the situation. The perceptions are tragically reductive, targeting the "ethnic symbols" because nobody can reach the "big men" who are the true agents of the new social war.

This is what makes Kenya's post-election impasse so intractable. In 1992 or 1997, a cynical dictatorship could turn the killing machine off and on. But today a democratic electorate frustrated by the lies of fake reformers is sinking into low-level victimisation. Any "explanation" that relies on "ethnic prejudice" or "irrationality" has no chance of solving anything. What is happening is a social war; but as is so often the case, the victims are not the real perpetrators.