Kenya: roots of crisis – by Gerard Prunier

No Comments » January 9th, 2008 posted by // Categories: African Affairs



Published on openDemocracy (

Kenya: roots of crisis

By Gérard Prunier

Created 2008-01-07 16:29

Gérard Prunier is research professor at the University of Paris and
director of the French Centre for Ethiopian Studies in Addis Ababa.

He is the author of The Rwanda Crisis: History of a Genocide [1] (C
Hurst, 1998), Darfur: The Ambiguous Genocide [2] (C Hurst, revised
edition, 2007), and From Genocide to Continental War: The Congolese
Conflict and the Crisis of Contemporary Africa

To many people in the world – and even to many Kenyans themselves
itself – the violence which followed the elections in Kenya on 27
December 2007 [9]has come as a surprise. Unfortunately, it shouldn’t
have. The combination of economic and ethno-political factors in Kenya
had created an explosive mix which was just waiting for the right – or
rather “wrong” – circumstances to explode. The 2002 elections had been
a lucky near-miss; this time, the favourable configuration that
operated then did not repeat itself.

Kenya’s “democratic” politics

To understand the Kenyan crisis [10] in the context of its national,
regional and global situation, it is necessary to examine the regime
which followed independence in 1963. Britain’s withdrawal from the
country had taken place amidst a considerable fear that the Mau Mau
[11]anti-colonial insurrection of 1952-1960 might impinge upon the
politics of the new state and lead to further violence. Nothing of the
sort happened – partly because of the elevation to the presidency of
the leader of the nationalist movement Jomo Kenyatta, who once in
power swerved from radical nationalism to conservative bourgeois

Kenyatta was a Kikuyu (or Gikuyu) and the enigmatic Mau Mau movement
had largely been a Kikuyu phenomenon (most of the 12,000 rebels or
“suspects” killed by colonial forces in a brutal campaign [12]were
Kikuyu). This had caused the British wrongly to conclude that Kenyatta
was the leader of the Mau Mau. But in any case, on becoming president
Kenyatta – head of the Kenya African National Union (Kanu) in an
effectively one-party state – embraced extreme tribalistic politics
and packed the new “Kenyan” bourgeoisie he promoted with Kikuyu and
members of related tribes such as the Embu and the Meru. At the time
of his death in 1978 most of the country’s wealth and power was in the
hands of the organisation which grouped these three tribes: the
Gikuyu-Embu-Meru Association (GEMA).

Kenya [13] has forty-eight tribes, with three – the Kikuyu, the Luo
and the Luhyia – together representing almost 65% of the population.
Meanwhile, the GEMA tribes during Kenyatta’s time (1963-78) composed
perhaps 30% of Kenyans, almost all concentrated in the highlands of
the central province. These figures meant that in order to square the
ethno-political circle in Kenya, power-brokers had to forge deals
between the three big groups and somehow relate to the shifting gaggle
occupying the fourth corner.

In Kenyatta’s time the deal was simple: the Kikuyu and their smaller
relatives, after making an agreement with the minority tribes, ran
everything. The Luo, who eventually tried to challenge this ordering,
were forcefully marginalised as the prudent Luhyia looked on. After
Kenyatta died in 1978, his vice-president Daniel arap Moi – who was
from the Kalenjin minority tribe – inherited the mantle of power on
the understanding that he would not upset the arrangement designed to
keep the two other large tribes (and particularly the Luo) out of
power [14].

But Daniel arap Moi proceeded to use his new status to cleverly divide
his Kikuyu allies (amongst them the man who would be his successor as
president, Mwai Kibaki), so as progressively to sideline them. By
1986, Moi had concentrated all the power – and most of its attendant
economic benefits – into the hands of his Kalenjin tribe and of a
handful of allies from minority groups (see Peter Kimani, “A past of
power more than tribe in Kenya’s turmoil [14]”, 2 January 2008).

But Kikuyu ascendancy had been reined in only, not destroyed. Under
Jomo Kenyatta, the Kikuyu – claiming martyr status for their
sufferings during the Mau-Mau “emergency”, and relying on tacit
government support – had spread beyond their traditional territorial
homelands and “repossessed lands stolen by the whites” – even when
these had previously belonged to other tribes. Thus Kikuyu “colonists”
had fanned out all over Kenya, often creating strong rural

Kenyatta’s successor, Daniel arap Moi [15], used a consummate
juggler’s skill to keep the ethno-political balance working in his
favour. At the same time, the first two multi-party [16] elections
after other movements emerged to challenge Kanu (in 1992 and 1997)
were occasions for carefully state-managed ethnic violence designed to
achieve two objectives: keep the dangerous Kikuyu underfoot, and pit
the Kalenjin’s minority allies against each other in order better to
control them.

By the time of the 2002 election, however, the system had run its
course: foreign donors were alienated, President Moi (having ruled for
twenty-four years) was getting old, and a “democratic” opposition was
gaining momentum. But if everybody agreed on the principle of ridding
Kenya of its Kalenjin-based authoritarian state, the question of who
and what would be the replacement remained open.
Moi had a brainwave: he thought that the best way for him to maintain
his influence over politics after leaving the presidency would be to
pick as the governing party candidate Kenyatta’s own son, Uhuru. This
artful move, Moi calculated, would rally the Kikuyu behind a
prestigious but empty symbol (Uhuru was not overly bright and his name
spoke louder than his personality). But the stratagem backfired
completely and the opposition united behind the veteran Kikuyu
politician, Mwai Kibaki, thus creating a unique situation in which
both leading candidates were Kikuyu.

In other ways, however, they were very different: one embodied the
ghost of yesterday’s near-dictatorship while the other was seen as
offering the hope of a democratic opening. This contrast felicitously
de-ethnicised the election, turning it into a contest between the old
and the new. At the time Raila Odinga [17], the leading Luo
politician, tirelessly campaigned for Kibaki and deployed his tribal
followers behind a man who – albeit a Kikuyu and a Kikuyu with a past
– was seen as the candidate for change. The economic stagnation of
previous years meant that many of the expectations that were invested
in Kibaki were of an economic nature: Kibaki, it was hoped, would
restart the economy and then proceed to share out its benefits more

The Kibaki administration

Mwai Kibaki [18]was elected president in December 2002 with over 62%
of the vote. The country’s foreign backers were only too quick to
salute the polls as “a triumph for democracy”. In a way they were
right – the polls had been free and fair [19], and the candidate for
change had been elected. But in another way this was a hasty form of
wishful thinking because the ostensible “de-tribalisation” of the
election had been due more to a series of fortuitous coincidences than
to a real decline in the appeal [20] of ethnic politics.

The key words in the campaign, however, had been “hope” and “change”,
and to some extent the new Kibaki administration managed to deliver
the goods. The economy did pick up and Kenya witnessed a spectacular
economic recovery, largely based on Keynesian economic recipes and
helped by a favourable international environment.

This can be illustrated by the annual rate of growth in 2002-07, which
reveals a gradual improvement from -1.6 % in 2002 to 2.6% by 2004, 3.4
in 2005, and an estimated 5.5% in 2007. But this was only one side of
the economic coin. Social inequalities also increased; the fruits of
economic growth went disproportionately to the already well-off (and,
among those, to the Kikuyu well-off); and corruption reached new
heights, matching some of the excesses of the Moi years. When John
Githongo, the man appointed by President Kibaki to fight corruption,
blew the whistle in January 2005, he had to flee to Britain in fear of
his life (see Michael Holman, “Kenya: chaos and responsibility [20]”,
3 January 2007). Githongo is himself a Kikuyu, and his denunciation of
a massive series of financial scandals in which hundreds of millions
of dollars had vanished was seen as a betrayal of his tribe as well as
of the government he served.

Moreover, the security situation in Kenya deteriorated steadily in
these years, with the ordinary people bearing the brunt of a triple

* a growing wave of routine crime in urban areas

* rival agrarian claims leading to pitched battles between ethnic
groups fighting for land, particularly around Mount Elgon and in Kisii

* a running feud between the police and the Mungiki sect, which left
over 120 people dead in May-November 2007 alone.

Mungiki is a bizarre cross between pre-Christian Kikuyu
neo-traditionalism and an extortionist gang. The sect ran protection
rackets on the matatu (collective taxi) routes, helping it to prosper
among the poorest urban neighbourhoods and among the landless-peasant
squatters in central province; it also has a tradition of hiring its
muscle-boys to political candidates during election campaigns. In
2002, the Mungiki had backed the losing Uhuru Kenyatta camp. This cost
it dearly in terms of political clout, and it had desperately tried to
recover the lost ground by intensifying its terroristic hold on the
slum population and on the matatu owners.

The accumulating result of these various processes was a feeling of
deep dissatisfaction – not so much with President Kibaki as a person
[21] but with his entourage, with his robbing cronies, and with his
incapacity to sympathise and do something about the plight of poor
Kenyans (made all the more shocking by the level of economic growth
the country was enjoying). Raila Odinga, the candidate [22] of the
Orange Democratic Movement (ODM), was then able to capitalise on that
frustration in a way that fused various types of motivation:

* ethnic (the Kikuyu have grabbed everything and all the other tribes have lost)

* political (Kibaki betrayed his promise for change)

* social (crime and violence are out of control)

* economic (what is the point of economic growth when it does not
bring any benefits to the ordinary citizen).

As the electoral campaign neared its climax in December 2007, the ODM
opposition enjoyed a widespread lead in opinion polls and seemed ready
to sweep Kibaki’s Party of National Unity (PNU) out of power.

The December 2007 election

The election on 27 December 2007 was both a parliamentary and a
presidential one. At the legislative level, 2,548 candidates from 108
parties were vying for 210 seats; at the presidential level, three
candidates – the incumbent Mwai Kibaki , ODM leader Raila Odinga and
former foreign minister Kalonzo Musyoka (who had split from the ODM) –
were competing.

Everybody (including himself) knew that Kalonzo Musyoka had no chance
of winning and that he was simply angling for the position of a
strategic post-election ally who could sell his support to a probable
minority victor in need of additional backing. Kalonzo Musyoka is a
Kamba, and the Kamba – although closely related to the Kikuyu – had
chosen the British camp during the Mau Mau emergency. This gives them
a hybrid status in the Kenyan ethno-political landscape, in which they
hold the capacity to swing either with the Kikuyu or against them.

The polls were a messy business for a number of reasons. The voters’
rolls had been poorly updated or at times not updated at all. Some
dead people were still on the rolls and electors who had changed
residence had not been properly struck off in one place and
re-registered at their new address. The rules governing the help which
could be given to illiterate voters (up to 80% of the electoral body
in some remote constituencies) were poorly enforced. Foreign and
national observers were not always given free access to the polling
stations, and later to the ballots.

But all in all, the parliamentary segment of the election proceeded
smoothly. The definitive results have not at the time of writing been
officially posted, but a provisional tally (based on 181 out of 210
seats) is possible. Twenty-two parties won seats, although only four
can be considered as “serious” (the eighteen others have between one
and three MPs, sharing twenty-eight seats between them): :

* Raila Odinga’s ODM, which won ninety-two seats

* Mwai Kibaki’s PNU, which won thirty-four seats

* Kalonzo Musyoka’s splinter ODM-K, which won sixteen seats

* Uhuru Kenyatta’s Kanu, which won eleven seats.

The results speak for themselves: with 45% of the MPs, the opposition
has a clear majority over the incumbent administration .

This is what makes the results of the presidential election [23]
definitely suspect. Kenya’s electoral commission (ECK) declared on 30
December that Kibaki had garnered 4,584,721 votes against 4,352,993
for his rival Raila Odinga, and immediately proceeded to inaugurate
the incumbent president as the winner. This tight margin (little more
than 230,000 votes, about 2.5% of those cast) is very fragile in view
of the following facts.

In seventy-two of the constituencies, the figures on the ballot forms
signed by the ECK returning officers and the agents of the candidates
differ from the figures released by the national counting centre. At
Ole Kalou constituency, for example, local ECK figures gave Mwai
Kibaki 72,000 and Raila Odinga 5,000 out of 102,000 registered votes.
But by the time the figures for that same constituency were released
at the central level, Kibaki’s winning tally had jumped to 100,980
votes (i.e. 99% of the registered voters).

The pattern was repeated elsewhere. In Elmolo constituency, Kibaki was
said by local ECK officials to have won by 50,145 votes, which then
translated itself into 75,261 votes at the national level. In Kieni
the discrepancy was between 54,337 (local level) and 72,054 (national
tally). In various other constituencies (Lari, Kandara, Kerugoya)
thousands more had “voted” in the presidential election than in the
legislative one, even though the two ballots had been held
concurrently .

All this points to a limited but widespread form of rigging which
would not have had such catastrophic consequences had not the race
been so closely contested. (After all, if several constituencies have
probable rigging levels of 10,000-30,000 votes, there is no way a
victory by 230,000 votes be considered solid.) On 1 January, Samuel
Kivuitu – the respected chairman of the ECK – admitted : “I don’t know
who won the election and I won’t know till I see the original records,
which I can’t for now until the courts authorise it”.

It seems that what happened was that the Mwai Kibaki vote was
artificially inflated rather than that Raila Odinga’s vote was
tampered with. The evidence seems clear: even if gerrymandering had
distorted the legislative vote vis-à-vis the presidential one (during
the Moi years, the “enemy” Kikuyu constituencies had seen their
demographic weight systematically eroded in this way), how could the
pro-ODM trend at the parliamentary level turn itself into a
contradictory support for the anti-ODM president? The possibility of
such a split-personality vote is remote, as it requires that almost
all those voting for minority parties would also have voted for

The bloody aftermath

The results of this manipulation [24] have been disastrous. Almost as
soon as the ECK hastily proclaimed Kibaki to be the winner, both the
Nairobi slums and the western province [25] exploded – the violence of
the slum-dwellers reflecting their social frustration and the
westerners’ arson-cum-machete attacks stemming from their hatred of
the Kikuyu “colonists”. The political violence should thus be seen as
both tribal and socio-economic; because, even if far from all Kikuyu
are rich beneficiaries of the regime, many rich beneficiaries of the
regime are Kikuyu. Such a situation recalls – especially for the Luo –
the frustrations of the 1960s and 1970s.

The vote itself was primarily anti-establishment rather than crudely
anti-Kikuyu, however: only six members of the cabinet survived the
landslide, and many of the victims – including vice-president Moody
Awori, planning minister Henry Obwocha, roads minister Simeon Nyachae,
and tourism minister Moses Dzoro – were not Kikuyu. Even the few Luo
or other westerners who were also PNU members lost their seats.
Several Moi administration survivors – such as former minister
Nicholas Biwott or Moi’s own son Gideon Moi – were also axed, often by
nearly unknown candidates who took their seats with ease. This is one
reason why the minority parties won so many seats: incumbency was a
distinct liability and voters appeared ready to elect anybody who
seemed ready to promote change.

It is when that trend towards long-awaited change appeared about to be
blocked once more by the man who had already betrayed it after 2002
that violence exploded. The configuration of two relationships –
Luo-Kikuyu, and Kikuyu with power – meant in the circumstances that it
could not but be anti-Kikuyu. At the time of writing there have been
at least 600 “official” deaths (as registered in hospitals and by
other reliable sources); but this total is almost certainly an
underestimate, especially if information from all the isolated rural
areas where old scores [32]are being settled were available.

While Luo have slaughtered Kikuyu settlers in their midst in the west,
Mungiki thugs have rallied to the tribe and have been busy killing Luo
in the Nairobi slums, hoping to ingratiate themselves with the big
bosses of Kiambu, Nyeri and Murang’a. There are already as many as
250,000 internally-displaced persons [33] (IDPs) and refugees (into
Uganda). Factories are idle, many roads are closed, and food and
humanitarian [34]crises loom. In Uganda, Rwanda and the eastern DR
Congo, the interruption of fuel supplies coming from Mombasa is
threatening transport. Even Tanzania is beginning to feel the economic
aftershocks of the disturbances. By a conservative estimate, the
Kenyan economy is losing [35]$30 million a day and the loss for the
whole region – though anybody’s guess – must be far greater.

On 2 January 2008, President Kibaki announced that he was “ready to
have a dialogue with the concerned parties”. This was a good start
but, once more, the 76-year-old president seemed to be a prisoner of
his past (and, perhaps, of his entourage). He stalled Desmond Tutu on
the bishop’s arrival from South Africa in the effort to mediate (in
contrast to Raila Odinga, who had immediately met Tutu); and when on 3
January attorney-general Amos Wako announced the creation of three
committees designed to find a solution to the crisis (on peace and
reconciliation, on the media aspects of the situation and on legal
affairs), they were packed with burned-out politicians like Simeon
Nyachae, Njenga Karume or George Saitoti, most of whom had just lost
their seats in the election.

On 7 January, it is reported [36] that Kibaki has invited Ghana’s
president, John Kufuor, to re-engage in the mediation effort that was
proposed as the violence first escalated; and that he has offered [37]
to create a government of national unity with the opposition which (an
official statement says) “would not only unite Kenyans but would also
help in the healing and reconciliation process”.

It is an artful departure from the boast of his precipitous acceptance
speech [38] of 30 December, when President Kibaki had declared:
“Fellow Kenyans, you have given us a vote of confidence in the values
and principles…that we began five years ago. You have chosen the
leaders you wish to serve you during the next five years”.

In the circumstances, the claim was neither truthful nor realistic. It
is unclear whether Mwai Kibaki’s latest manoeuvres represent a genuine
shift of position or a tactical adjustment [39] to desperate
conditions. In any case, the creation of a government of national
unity is now the sole, albeit painful compromise available if Kenya’s
violence is to be contained and some sort of progress beyond this
nightmare made. After that, a just and truthful reckoning with what
has happened in Kenya must be attempted.

Average rating

(5 votes)
Rate this:
Trackback URL for this post:
Source URL:


This article is published by Gérard Prunier , and
under a Creative Commons licence. You may republish it free of charge
with attribution for non-commercial purposes following these
guidelines. If you teach at a university we ask that your department
make a donation. Commercial media must contact us for permission and
fees. Some articles on this site are published under different terms.


Opt In Image
Send Me Free Email Updates

(enter your email address below)

Leave a Reply


Home | About | Contact | Login