Volume 8 No 1 71

THE POLITICAL AND INSTITUTIONAL

CONTEXT OF THE 2007 KENYAN ELECTIONS
AND REFORMS NEEDED FOR THE FUTURE
David K Leonard and Felix Odhiambo Owuor
With contributions from Katherine George

Felix Odhiambo Owuor is Senior Programme Manager of the National

Democratic Institute for International Affairs Political Parties Programme
in Nairobi
E-mail: [email protected]
Katherine George received her MA in Governance from the Institute for
Development Studies (UK) in 2008

ABSTRACT

For many Kenyans the outcome of the 2007 presidential election represented
a continuation of the betrayal of the promise made by Mwai Kibaki’s
government, elected in 2002, that a new Constitution would be drafted which
would help to deal with Kenya’s governance problems. The consequence was
a closely contested election, ethnic division, a flawed election process, and
serious post-election violence, which lasted well into 2008.
This article analyses the underlying political features of Kenya that
led to the election failure itself and the fundamental changes to the Kenyan
system, including its Constitution, that are necessary to avert a recurrence
of the 2007 election violence in the future.

INTRODUCTION: KENYA, CONFLICT AND ELECTIONS

A failure of governance
In December 2007 Kenya erupted into violence after the announcement that its
presidential election had been won by incumbent President Mwai Kibaki. The
world was shocked, as Kenya had been thought to have made the transition to
stable democracy with its peaceful elections in 2002 and 2005. The history of
failed constitutional reform, the reopening after 40 years of a particularly toxic
combination of ethnic political conflicts, and electoral fraud resulted in post-
election violence that exceeded even the scale of the 1992 and 1997 clashes.
71
72 Journal of African Elections

In this article we examine the underlying causes of this violence and the
fundamental issues that will need to be addressed if the next national elections,
scheduled for 2012, are to be successful. There have been numerous other analyses
of the details of the elections, with which we broadly agree and which we will
not replicate here (see, the Kriegler Commission Report; the Waki Commission
Report; Sheehy & Maina 2008; Journal of Africa Elections 7.2).

Kenya’s quest for constitutional reform

Failures of governance were at the core of the violence that followed Kenya’s
2007 national elections. The 2002 government of Mwai Kibaki had promised
a new Constitution that would help to deal with Kenya’s many governance
problems – an overly powerful presidency with a weak legislature and judiciary,
a centralised state, mismanagement of and unresolved disputes over land, a
history of impunity for violence and corruption, inequalities between ethnic
groups, and poverty and unemployment. For many Kenyans, the outcome of the
2007 presidential election was a continuation of the betrayal of that promise, as
reflected in the ‘Wako’ draft constitution Kibaki had submitted to a referendum
in 2005, which the people had rejected.
Kenya’s independence Constitution of 1963 was based on federal devolution,
individual rights, and the separation of powers between a symbolic head of state
and an executive prime minister. Its current Constitution evolved through a series
of fundamental changes between 1963 and 1969 that undermined democratic
safeguards (Gathii 1994). These alterations entrenched executive power in a
powerful president by changing from a parliamentary system of government
with an executive prime minister to a presidential system; replacing a bicameral
with a unicameral legislature; converting from federal devolution to a unitary,
centralised state; and moving from a multiparty to a single-party system (de facto
in 1969; de jure in 1982) (Chitere, Chweya, Maya, Tostensen & Waiganjo 2006, p 2).
The amended sections have been the subject of constitutional debate since 1992,
occupied the 2000-2005 constitutional review process, and dominated political
party campaign platforms in the 2007 election.

Breakdown into violence

The violence that followed the 2007 elections, during which an estimated 1500
people died and as many as 350 000 were displaced (IRIN 2008a) occurred in
three discernable waves (OHCHR 2008). The first came after the announcement
of the election results and involved spontaneous looting by youths in the slums
of Nairobi and Kisumu of government buildings and of the shops and houses of
Volume 8 No 1 73

known supporters of Kibaki’s Party of National Unity (PNU) and his co-ethnic
Kikuyu, who were also presumed to have supported him.
The second wave of violence was organised in part before the election by
opposition and ethnic leaders as a response should Kibaki win the election. At
the announcement of the election results groups of youths attacked Kikuyu and
government supporters in the central Rift Valley, with the aim of driving them
out of their constituencies for good.
The third wave comprised reprisal attacks, organised by government
supporters and Kikuyu militias, mainly targeting migrant workers thought to be
opposition supporters in parts of the Rift Valley Province, Central Province, and
the Nairobi slums. Male youths played the main role all three waves of violence.
The police were also responsible for much of the violence, either because they
used excessive force to deal with protestors (potentially accounting for 40% of
deaths) or because they chose not to prevent it.
The violence disrupted crop production and transport, resulted in a sharp
economic downturn, an 80 per cent reduction in tourism revenue, and a rise in
the price of basic goods (Bevan 2008). It also entrenched social fragmentation
between ethnic groups in the areas that were hardest hit.
As the post-election disputes and skirmishes dragged on, the international
community supported a panel of eminent African personalities, led by former
United Nations Secretary General Kofi Annan, to mediate and find a solution.
After several weeks of negotiation, on 28 February 2008 the opposition Orange
Democratic Movement (ODM) and the PNU reached a power-sharing agreement
which created a coalition government, including the position of prime minister,
and a redistribution of Cabinet and other ministerial posts (RoK 2008a).
The most concrete aspects of the agreement were: an internationally-
composed independent review commission to investigate all aspects of the
2007 general elections and recommend areas for electoral reforms (RoK 2008c);
a commission to investigate the causes and effects of the post-election violence
and propose solutions (RoK 2008b); a recommendation that a Truth, Justice and
Reconciliation Commission (TJRC) should be established to facilitate a lasting
healing process and unite the country;1 and the rekindling of the constitutional
review process that aims to provide solutions to the underlying causes of the
violence and produce a new draft to be approved by a public referendum (RoK
2008d; 2008e).

1 The TJRC has yet to be convened, as it would require an Act of Parliament.

74 Journal of African Elections

THE INSTITUTIONAL STRUCTURE OF KENYAN DEMOCRACY

The sociological foundations

In Kenya, as in the rest of Africa, voting is largely determined by ethnicity, kinship
and neighbourhood. In the rural areas, where all three tend to coincide, voting at
polling station level is generally in favour of one particular candidate, with the
decision effectively a collective one, often enforced by violence. Barkan, Densham
& Rushton (2006, p 933), found that the average vote for the winning candidate
for Parliament in 1997 was 65 per cent, even though there was an average of 4,4
candidates per constituency. And if one had looked at the percentages at polling
station level the concentration would have been still higher.) Only in the major
urban areas is the coincidence of family and neighbourhood broken. Although
kinship and rural ties are still influential among all but second-generation elites,
adherence to them will not be as obvious at urban polling stations.
As discussed in the introductory article to this issue these social dynamics
contribute to the dominance of patron-client networks in most of the political
systems of Africa and, as a consequence, those who are able to distribute patronage
goods from their personal wealth dominate politics on the continent. In the 45
years since Kenya’s independence African elites have used political office to gain
economic advantages, which they have ploughed back into political advancement
in an upward spiral.
By now there are wealthy businessmen (and some women) in all but the
very smallest Kenyan ethnic groups, who compete with each other for political
office. Since political clients trade their votes for benefits targeted to themselves
as individuals or to their immediate communities, the patronage-dispensing
businessmen are able to direct public policy and its implementation to their
personal and collective interests. Not all Kenyan businessmen are politicians, and
many (such as the local South Asian community) participate only indirectly in
politics. But all but a very few MPs and Cabinet ministers are businessmen.
The strength and weakness of Kenya is that it is unapologetically friendly
to business: the major political parties compete over whose businesses benefit
from access to the state, not over whether they do (Leonard 1991). As a result, the
collapse of the country’s tourism and export economy in response to the continued
post-election violence eventually prompted powerful businessmen to press for
a negotiated settlement.

A History of political violence

Another critical dimension of Kenyan politics is its violent heritage. In most of
sub-Saharan Africa independence was achieved relatively peacefully. In Kenya,
Volume 8 No 1 75

however, the Kikuyu Mau Mau rebellion had broken out in the late 1940s and
was put down in the 1950s only with substantial military assistance from the
United Kingdom. The presence of those British troops deprived the colony’s white
settlers of the local control that might have enabled them to mount a Rhodesia-
like unilateral declaration of independence.
Thus, even though the Mau Mau were defeated, it is common in Kenya to
credit them with hastening the country’s independence, and their rituals may
sometimes be recreated as an instrument of Kikuyu political struggle – as they
were in the oathing movement of the mid-1970s and by the underclass Mungiki
in the past decade and during the 2008 post-election riots.
The violence of Mau Mau seems to have legitimated other forms of political
violence as well. Kenya has had five or six prominent political assassinations – in
order, Pio Pinto (a radical Asian), Arwings-Kodhek (a Luo Cabinet Minister), Tom
Mboya (a Luo heir apparent to Kenyatta), J M Kariuki (a populist Kikuyu hoping
to succeed Kenyatta), possibly Ronald Ngala (a Cabinet minister from the Coast
who also figured as a possible successor) and Robert Ouko (a Luo Cabinet minister
who was thought to have offended Moi). It was not lost on the Luo ethnic group
that all these men were Luo or were reaching out to them politically. All but the
last of these assassinations took place during the Kenyatta administration.
Extensive violence accompanied the first multiparty elections, in 1992,
and also attended the elections in 1997. Violence targeted mainly the Luo and
Kikuyu communities, which had refused to toe the Moi line. In his efforts to
retain power for the small ethnic groups when he changed the Constitution to
re-permit multiparty competition in 1992, President Moi inserted a requirement
that a victorious presidential candidate had to win both an overall majority and
no less than 25 per cent of the vote in five of Kenya’s eight provinces.
A Kikuyu candidate in an ethnically divided field will easily pass the latter
threshold in Central, Nairobi, and (with their Embu and Meru allies) Eastern
provinces. There are also substantial numbers of Kikuyus in smallholder
settlement schemes and the largest towns in the Rift Valley and Coast provinces.
In the rural areas of the latter two provinces most Kikuyu live in multiethnic
settlements on land to which other groups had historical title (of which more
below).
The strategy for those running against a Kikuyu candidate for president has,
therefore, been to deny him the ‘25 per cent in 5’ threshold by mobilising those
in the central Rift Valley or on the Coast who were indigenous at the start of
colonialism and who are aggrieved about the land issue to attack Kikuyu settlers
there, displacing them and rendering them unable to vote.
Moi used this strategy successfully in both 1992 and 1997 but it became
irrelevant in 2002 when Moi manoeuvred the nomination of Uhuru Kenyatta
76 Journal of African Elections

(the late President’s son) as the candidate for the Kenyan African National Union
(KANU) to stand against the opposition’s Kibaki, so that both candidates were
Kikuyu. The political logic of creating internally displaced persons (IDPs) became
salient again in 2007 when a Kikuyu president (Kibaki) was challenged by a Luo
(Odinga), who had Kalenjin indigenous allies in the central Rift Valley. In this
case, however, the violence broke out only after the election and did not engulf
the Coast. We elaborate on this below.

Competing narratives of grievance

The tragedy of Kenyan politics is not just that it is organised around ethnic
differences and the ‘horizontal’ inequalities between ethnic groups (for more
about this term, see Stewart 2008a), characteristics that are common to most
African countries. The problem is that in Kenya three competing and legitimate
grievance narratives accentuate these features.
Early white settlers in Kenya appropriated most of their land from the Kikuyu
(Central and Nairobi provinces), Miji-Kenda (Coast), Maasai and Kalenjin (Rift
Valley) areas. Land pressures resulted first and most acutely among the Kikuyu,
who, as a result, were, by World War I, taking up disproportionate numbers of
tenant-labourer positions on white farms (as well as availing themselves of the
education the missionaries brought).
The eviction of these tenant-labourers in the late 1930s laid the foundation for
the Mau Mau rebellion (Rosberg & Nottingham 1966). To suppress it, the British
rusticated Kikuyu labourers, moved Kikuyu peasants into fortified villages,
detained large numbers of suspects, and, while they were away, conducted land
registration.
At independence the Kikuyu had not only suffered most during the anti-
colonial struggle, they alone faced serious problems of landlessness. To pacify
them both the British and President Kenyatta favoured them disproportionately
in the smallholder settlement schemes that government created out of the old
white farms.
The Kikuyu also did well in the labour market (as they had been early
entrants, were better educated, and had the country’s capital in their territory).
Furthermore, their education, entrepreneurialism and support from the Kenyatta
presidency enabled them to rise rapidly in the civil service and to do well in
business. The combination of all these factors then meant that when white farms
were sold in the post-settlement era the Kikuyu were more likely to be able to
buy them and divide them up into still more smallholdings.
The Miji-Kenda (Coast), Maasai and Kalenjin (Rift Valley) came to this
competition for land later, with less capital and later political support. They also
Volume 8 No 1 77

suffered from the fact that both Kenyatta and Moi used the presidency’s ultimate
authority over land to reward their most prominent clients (and themselves)
with large farms in these areas. Today the Miji-Kenda, Maasai and Kalenjin are
aggrieved that lands that were theirs at the time of the British incursion are all
occupied and their young adults no longer have access to farmland.
From 1982 on Moi worked aggressively to counter-balance the historical
advantage of the Kikuyu, resulting in their becoming disadvantaged, to some
degree, in education, civil service employment and business contracts and
privileges, while the Kalenjin received special favours.
Despite the removal of state patronage, however, the Kikuyu continued to
do well in education and business; their prominence in trade making them the
target of still more envy today. But they are aggrieved because they were the
most prominent victims of British colonialism and of Moi’s affirmative action; to
them the fact that they have nonetheless done well is a sign of virtue, not of quite
clear favouritism under presidents Kenyatta and Kibaki. The Kikuyu discourse of
grievance focuses on the colonial period and the period between 1982 and 2002;
other Kenyans focus on 1963-1978 and 2002-2007.
Our purpose is not to determine who, on balance, is more right or wrong
but to indicate the ways in which all the discourses have some basis in historical
fact and talk past each other.
The third aggrieved people are the Luo. At independence they were as well
educated as the Kikuyu and more entrenched in the civil service. They were the
victims of Kenyatta’s turn against the left in 1966 (both in politics and in the civil
service), many of their leaders were assassinated and they have remained political
outsiders until this year.
Each of these groups has difficulty appreciating the injustices felt by the
others. The leaders of all three groups also tend to use grievances to mobilise
their non-elites and then to deliver ‘compensating’ benefits to their elites. As
the Kikuyu elite (but not their poor) has prospered despite the difficulties it has
sometimes faced, it is a lightning rod for the problems faced by everyone else
– and its members aggravate the situation by assigning themselves still more
‘compensating’ advantages.
The tragedy of Kenya is that it has no unifying national ideology – only
incompatible narratives of injustice. The country is not divided on economic
ideology – even though the large inequalities it manifests might make such policy
divisions seem natural. The question is not whether the elites will benefit from
a pro-business environment but which ones will benefit. For a brief period after
independence nationalism did provide a unifying political ideology. It is possible
that it could again, but to date no political leader has succeeded with it or any
other option.
78 Journal of African Elections

The strong presidency

African states are weak in the face of very strong societies (Migdal 1988). As
Aristide Zolberg (1966) noted more than 40 years ago, this makes nervous national
politicians construct governmental systems that are as strong as possible – hence
the long period of one-party rule on the continent and the persistence even today
of very strong presidencies. Kenya’s presidents exercise unchecked authority over
virtually every aspect of government. Until recently the ability of Parliament to
initiate legislation has been all but non-existent and neither it nor the courts has
been able to check the executive. Since immediately after independence civil
servants have felt obligated to execute presidential directives even when they are
manifestly illegal. The result is that the stakes for winning or losing State House
become inordinately high.

A Parliament of rural ambassadors

As noted in the introductory article in this issue, the pressures of patronage cause
parliamentarians in Africa to see themselves primarily as supplicants to the centre
for discrete goods that will benefit their constituents. Kenya conforms fully to
this pattern. Nonetheless, many of the country’s MPs do also care deeply about
public policy. When the democratisation movement in the country began in the
early 1990s such MPs had no institutional support. It became the responsibility
of various civil society organisations (CSOs), usually with international funding,
to analyse pending legislation, note policy areas of concern, draft new laws, and
so on. The Constitutional Amendment Bill of 1999 finally created a Parliamentary
Service Commission and staff support, which enabled committee work and
provided some autonomy.
The MPs who were newly elected in 2002 showed increased legislative
independence – but they were most active in relation to matters that enhanced
their role as patrons. They voted themselves high salaries and created constituency
development funds (CDF), both of which enhanced their ability to distribute
patronage.
Ironically, the very fact that they visibly controlled this patronage may have
created as much jealousy as loyalty. When MPs were ‘ambassadors’ seeking
patronage for their clients from an ‘imperial presidency’ they could take full
credit for anything that was provided while blaming the president for any
disappointments – after all, if they did well for some in this round they also might
be able to help in the next round those who had been let down. Now, however,
MPs are clearly responsible for the patronage they distribute and must bear the
brunt of criticism from those who received nothing. This dynamic may help to
explain the fact that two-thirds of sitting MPs were defeated in the 2007 elections.
Volume 8 No 1 79

On the other hand, the CDFs do make the MPs independent of the president in
their quest for patronage resources.

Political parties and partisanship

Political parties in Kenya are organised along ethnic lines. Half a century ago it
was not obvious that this would be the case. The Mau Mau rebellion arose among
the Kikuyu as a result of landlessness and displacement from large white farms
(Rosberg & Nottingham 1966). The movement was never wholly about poverty
but those Kikuyu who supported the British and joined the Home Guards during
the uprising were more likely to come from the better off and/or missionary-
educated sectors of the population.
When the British began to permit African political parties among Kenya’s
ethnic groups in the mid-1950s they restricted them to district-only organisations
– which, given the geographical specificity of ethnicity in most of Kenya, meant
that they represented particular groups. When Jomo Kenyatta emerged from
detention by the British in 1961 he faced a political landscape designed to be
ethnically divided. Although he was a national hero he joined KANU, which was
an alliance of the Kikuyu and Luo (the two largest and best educated groups),
with a streak of Mau Mau-inspired radicalism. It was opposed by an alliance (the
Kenyan African Democratic Union – KADU) of all the other, smaller groups, who
feared dominance by the Kikuyu and Luo. Kenyatta perceived the potential for
class conflict (particularly over land) to fragment political unity among the Kikuyu
– but only among them. If he were to support the radical wing in Kenyan politics
his own ethnic base would split and KADU would be politically dominant. But if
he insisted on class unity (the more conservative line) and provided the landless
Kikuyu privileged with access to land vacated by white settlers he could keep
the Kikuyu together and emerge as president.
In a series of brilliant political moves in the mid-1960s he enticed KADU
to join KANU to create a government of national unity and then destroyed the
radical wing of KANU. Radicalism prevailed only among the Luo, because of
their remaining leader, Oginga Odinga (father of Raila Odinga, the opposition
presidential candidate in 2007), helping to make them the permanent outsiders
in Kenyan politics. In the de facto one-party state that resulted, with the Luo and
radical Kikuyu excluded, the government in effect became a KADU state, with the
addition of conservative Kikuyu in the leadership roles (Leonard 1991, pp 73-81).
When Kenyatta died in 1978 he was succeeded by Daniel arap Moi, his vice-
president, a minority Kalenjin and a former leader of KADU. After the attempted
coup staged by a group of airforce officers in 1982 Moi was able to jettison the
conservative Kikuyus who had supported him (in effect making the government
80 Journal of African Elections

even more a KADU one) and to concentrate on reversing the gains made by the
Kikuyu under Kenyatta (Leonard 1991, pp 168-69; 176-77).
By the late 1970s all ideology had disappeared from Kenyan politics, replaced
by purely ethnic competition over the division of spoils derived from control of
the state, with a deep split between the Kikuyu (joined by the Embu and Meru)
and the small ethnicities/tribes (led by the Kalenjin) and with the electoral balance
decided by the second- through fourth-largest groups – the Luo, the Luhya and
the Kamba. The battle to control State House began in earnest in 1969 after the
assassination of Tom Mboya and the accompanying Kikuyu oathing, whose
explicit objective was to retain the presidency. The non-ideological nature of this
competition began with the banning of the Kenya People’s Union (KPU) (to create
the de facto single-party state) and the assassination in 1975 of the populist, J M
Kariuki. The curtailment of fundamental freedoms and political space, which
accelerated after the 1982 coup attempt, eventually led to a sustained clamour for
the restoration of multiparty democracy, a movement spearheaded by politicians
who had fallen out with President Moi joined by civil society and the international
community. The movement was also aided in part by the collapse of the Cold
War, which allowed Western governments, no longer needing to worry about
‘Communist’ competition, to insist on democracy and good governance as a
prerequisite for donor engagement. This pressure led to the reintroduction of
multiparty politics through the repeal in 1991 of s 2A of the Constitution.
During the first multiparty elections in 1992 it was obvious that Moi and
KANU did not have the support of the majority of the country. The newly
registered pressure group, the Forum for Restoration of Democracy (FORD),
quickly gained and mobilised support across the country. FORD brought together
once again the independence era alliance of Kikuyu and Luo, their unity posing
a direct threat to Moi and KANU. Nonetheless, the opposition to Moi fell into
competing camps, neither group being willing to stand aside and endorse one
of the other’s leaders as president. After letting Moi through in a three-way race
in 1992 and 1997 the opposition finally won in 2002 by uniting around Mwai
Kibaki, a Kikuyu and former vice-president. Raila Odinga, who was central to
bringing the opposition together, was promised the post of prime minister under
a reformed constitution.
Kibaki has been a low-key president and Kikuyu ministers close to him
persuaded him to renege on this deal (which was embodied in the ‘Bomas’2 draft
of a new constitution that was prepared just before the 2002 elections). Instead

2 The ‘Bomas’ draft emanated from the National Constitutional Conference which took place at the
Bomas of Kenya theatre facility, Nairobi, and became known simply as the ‘Bomas Talks’. The draft
was initially formulated by the Constitution of Kenya Review Commission, appointed in 2000 to gauge
the views of Kenyans about what they wished to see incorporated in their new constitution.
Volume 8 No 1 81

Kibaki put to a referendum in 2005 the ‘Wako’ draft constitution,3 which made
no provision for a prime minister (and also omitted other key provisions of the
‘Bomas’ draft) and was defeated by the ODM, led by Odinga. This history helps
to explain why Odinga was at first unwilling to accept Kibaki’s offer of a prime
ministership in a coalition government after the 2008 elections; he had been there
before and felt he had been betrayed. Odinga’s concerns could only be assuaged
by a constitutional amendment creating the position and making it answerable to
Parliament (where the ODM has the largest number of seats – 99). This amendment
was delivered as part of the Annan agreement.
The other major political parties are Kibaki’s PNU, with 43 MPs; the rump
of KANU, which supported Kibaki and has 14 MPs; and the Kamba breakaway,
ODM-Kenya, which supported Kalonzo Musyoka for president and has 16 MPs.
Kibaki appointed Musyoka to the vice-presidency in the midst of the post-election
violence.
Although there is stability in the major ethnic blocks of Kenyan politics
there is considerable factionalism within and between them as the ‘big men’ try
to assemble and hold together a winning coalition. In the past decade Odinga has
continually surprised his opponents by his ability to assemble such coalitions, first
for Kibaki, then against him, and finally for himself. To do so, however, the shape
of the alliances and the names of the winning group have constantly changed.
In order to counter the fragmentation and instability of parties, in October
2007 Parliament passed the Political Parties Act, which provides a framework for
the registration, regulation and financing of political parties. The Act took effect
on 1 July 2008 and all registered political parties had to adhere to it by no later
than 1 January 2009. Thus far 38 of the 168 previously registered political parties
have been certified as complying with the new standards.
Until the old ‘brief case’4 political parties have been definitively deregistered
the intention of the Act to impose discipline within parties by arresting the
defections and factionalisation and the irregularities in internal processes which
so often characterise them will not have been achieved.

The courts
An independent judiciary is critical to ensuring that the executive respects
the laws passed by the legislature. Otherwise the election of a Parliament and
its deliberations are meaningless. The judiciary is also central to ensuring the

3 The Wako draft – the government version of the constitution – was named for Amos Wako, Chief
Justice of Kenya, who was believed to be the key author of this alternative constitutional document.
4 These are parties that exist only on paper and whose rights to legal existence are kept in the ‘brief
cases’ of individual leaders in order to facilitate factional splits.
82 Journal of African Elections

framework within which democratic competition takes places – by protecting

human and minority rights and by enforcing the integrity of the electoral process.
The courts in Kenya, emerging from a long period of authoritarian rule, do not yet
play these roles fully and hence were not trusted during the crisis, contributing
to its escalation.
The judiciary has even less power than Parliament to check the executive. The
current Constitution does not provide guidelines for the exercise of judicial powers,
it merely defines their bare structure. The judiciary has weak infrastructure, biased
courts, incompetent and corrupt personnel, and serious problems upholding
the rule of law (Chitere, Chweya, Masya, Tostensen & Waiganjo 2006). The
president appoints without oversight the High Court, Appeals Court, and other
judges, as well as the Judicial Service Commission, which recommends judicial
appointments (NEPAD 2007). The executive has fiscal control over the courts
(Chitere et al. 2006).
It is true that judicial appointments now can be made only on the
recommendation of the Judicial Service Commission with the approval of the
president (which Kibaki has never denied) and judges can be removed from office
only by a tribunal – both improvements over the Moi era. On the other hand the
current chief justice manages the judicial system in such a way as to prevent its
ever challenging President Kibaki, for example, through his assignment of critical
cases.
When there is no competent and independent judiciary there are few
incentives for the legislature to spend time and political capital writing or
amending laws that will restrict or challenge the presidency – for such laws will
not be enforced.
Even during the one-party era the courts heard petitions contesting the
outcome of primary elections (which determined who would sit in Parliament).
These proceedings, however, could be and continue to be very slow – in some
cases lasting as long as an MP’s term of office. Mwai Kibaki, when he was running
for president in 1997, refused to appeal to the courts for this reason. Thus it is no
surprise that the ODM also refused to test its contentions of vote rigging before
the courts in 2008 (Gathii 2008).

The media
Independent and competent media are generally considered essential to
democracy. Without them citizens and even societal elites have no hope of being
informed about critical issues of public policy or of the performance of the
government they elect. The existence of strong media also gives opposition and
back-bench MPs an incentive to be vigilant in their oversight of the executive;
Volume 8 No 1 83

they are sure to be outvoted on any censure motions but information they obtain
by subpoena and/or by speaking aloud under parliamentary privilege will be
disseminated by the media.
The past decade has seen significant growth in the Kenyan media. Indeed,
that period has been characterised by the mushrooming of print and electronic
media with national coverage. The conduct of the media particularly during the
2002, 2005 and even the start of the 2007 elections was very good. However, in the
final periods of 2007 some vernacular and regional media were used to propagate
ethnic animosity and hate speech. KBC, the national broadcasting corporation, also
failed, by and large, to accord equal coverage to the main presidential candidates
despite agreements reached by the Inter-Parties Parliamentary Group (IPPG)
of 1997.
On the whole, Kenya’s print media are among the best in Africa. They are
professional and represent competing points of view. It is the proliferation of
new, private FM radio stations that has presented a problem. Talk radio shows
and live coverage of political events became vehicles for the propagation and
implicit legitimation of inter-ethnic hatred in 2007. There is no effective regulation
of these inflammatory practices. In fairness, one should acknowledge that some
of the problems stem from a lack of training of radio hosts in handling talk radio
contributions in an inflammatory situation and that some vernacular radios also
played a major role in urging peace after the conflict erupted (BBC 2008).

Civil society
Today civil society organisations are regarded as essential to democracy. Kenya has
a very large number of them, covering virtually every public interest and including
many working on election issues. These groups were extremely important in
the attempts to reform the Constitution, in advocating a return to multiparty
democracy, and in monitoring the end of Kenya’s 35 years of one-party rule.
In the transitional elections they were able to put large numbers of non-
partisan volunteers into the field as observers. However, most of these individuals
are now involved with government or with the new, competing democratic parties
and are therefore no longer non-partisan. Thus, there are now fewer genuinely
non-partisan groups promoting best democratic practice in Kenya and those that
exist are smaller and more dependent on international finance than they were
in 2002. The democratisation CSOs were late in mobilising for voter education
and election observation and found it harder to recruit observers prepared to be
wholly non-partisan. The competing political loyalties of the CSOs may have
contributed to the failure of their basket-fund management to deliver finances
to them in a timely manner (Sheehy & Maina 2008).
84 Journal of African Elections

The management of the electoral process

We turn now from the general architecture of democracy to the governance of its
electoral processes. The Election Commission of Kenya (ECK) managed the key
components of voter registration, election-day mechanics and certification of the
results. It had performed very well in 2002 and 2005, was thought to be robust,
and its chairman, Samuel Kivuitu, was highly respected.
Kibaki’s last-minute appointment of 18 of 21 new commissioners without
consulting the opposition, as he had agreed, during the 1997 IPPG negotiations
would be the case and the refusal of the ECK to carry out a number of technical
improvements recommended by the donor-provided experts assisting it were
important danger signals, but the second was not widely known and their
significance was appreciated by the international community only afterwards.
Some Kenyans, particularly ODM supporters, did appreciate the significance
of these actions, however, and, prior to the elections, began to plan violence in the
event of ODM losing the election – for, given the ECK’s structure, a loss at the polls
would not be accepted as legitimate (Kenya National Commission for Human
Rights 2007). The ECK’s failures were the most obvious and fatal part of the 2007
elections and have been subject to a detailed investigation by the International
Review Commission (IREC) headed by Johann Kriegler, retired South African
Constitutional Court judge and the chair of South Africa’s first Independent
Electoral Commission (hereafter the Kriegler Commission). The commission
found that electoral fraud began at polling-station level and was rampant.
It determined that the errors made in the various stages of the tallying process
were so great and widespread that it is impossible to reconstruct from the formal
record who, in fact, won the presidential contest. The commission focused on the
many computational errors it found and a majority of the its members concluded
that they had ‘not been able to substantiate … accusations [of] … malfeasance [at
the national tallying centre]’ (Kriegler Commission report 2008, ch 6, p 131). The
commission’s findings therefore implicitly legitimised the negotiated construction
of a government of national unity for the country. Our findings concur on all
these points.
We note, however, certain limitations in the Kriegler Commission report.
Most of the members of the commission were lawyers. Their finding that they
could not ‘substantiate accusations [of] malfeasance’ at the national tallying centre
should therefore be taken in a strict legal sense (which raises the threshold for the
admissibility of evidence and the burden of proof) – no witnesses were willing to
come forward and testify under oath. Our own interviews did uncover those who
had received first-hand reports from people who had seen such malfeasance but
were unwilling to give testimony under oath. A verdict of ‘not guilty’ was correct,
Volume 8 No 1 85

but we believe that should not be construed to mean a finding of ‘innocent’. In

any case, from the point of view of the election itself, what did or did not happen
at the tallying centre proper, is less important than the commission’s conclusion
that there was massive fraud in the general process.
The Kriegler Commission also did not seem to appreciate the significance or
evaluate the reliability of the evidence that emerged from the exit poll conducted
by the University of California/International Republican Institute (IRI) (Gibson
& Long 2008). IRI did not make these results available until after the government
of national unity had been negotiated, and polls are not an accepted basis in any
country for deciding an election. Nonetheless, in the absence of reliable data
from the polls themselves, this survey is the best evidence we have of the way
the people of Kenya who actually cast their ballots claimed to have voted. The
results are superior to the flawed election returns in that they do not include any
ballot-box stuffing, were collected under strict controls by well-trained personnel
and are free of computational errors. The poll indicates clearly and with statistical
reliability that Raila Odinga should have won the election with approximately 46
per cent of the vote (vs Kibaki’s 40%).5
Finally, the Kriegler Commission, surprisingly, did not explore whether or
not President Kibaki met the minimum standard of 25 per cent of the vote in the
required five provinces even if he had won a plurality from the national. The
survey evidence indicates that he fell short, in that he received only 17 per cent
of the vote in the North-Eastern Province (the ECK claimed he won 50%). In these
circumstances there should have been a run-off election, which would probably
have produced a decisive result.
None of the foregoing can, or should, provide a legal basis for changing the
negotiated settlement brokered by Kofi Annan. We present these concerns instead
as an indication that the flaws in the election process not only produced a violent
outcome but also one quite different from that which may have been intended
by a majority of Kenyans.

The electoral trustees

The above processes and organisations are formally responsible for the conduct
of free and fair elections and the implementation of the results. But they hardly

5 The question arises as to why the survey results were not released until after the conclusion of the Annan-
mediated agreement. The study had been funded by US Government on the explicit understanding
that it would be held in confidence for a period after the elections, doubtless so as not to be seen as
interfering with them. Also, some individuals in IRI had doubts about the results (for reasons that are
not wholly clear to us) and before they would release them insisted they undergo two US survey audits
(which they passed). It has been both alleged and denied that there was a political motive behind the
delay in releasing the results.
86 Journal of African Elections

acted alone. The support of other actors is key to their success. Among these are
the army and police, who insure that democracy is not subverted by violence, and
the international community, which provides finance and technical assistance.
Save for the UN Development Programme (UNDP) it would be a mistake to
classify these bodies as participating directly in the management of elections,
but their indirect roles are critical. It is helpful to think of them as the trustees of
the electoral process.

Networked donors
The donors who support democratisation work in Kenya, and the UNDP, to
which they supplied joint programme funds (which paid for full-time election
expertise), were well coordinated, held meetings monthly, and had very good
interpersonal relations. The lead donor was the USA, deputised by the Danish
embassy. Nonetheless, early danger signals about the performance of the ECK
were not well communicated within the donor group and therefore neither it nor
the diplomats were mobilised into precautionary activity.
The major political focus was on assuring the reappointment of Kivuitu as
chair. He was highly respected and had performed well in leading the ECK in
2002 and 2005 but he was not in good health and had taken several months off
in 2007. Thus the group over-estimated his and other respected commissioners’
ability to organise a complex process and to stand up to intense pressures in a
very high-stakes election. In effect, at both the international and the ECK level,
the problem was that too much faith was placed in too few people.

The army
Kenya’s armed forces might be likened to the Hound of the Baskervilles, which
did not bark – they were notably absent in the aftermath of the election. There
was the attempted coup by the airforce officers in 1982 and an army mutiny
in the 1960s, but the Kenyan military has never successfully intervened in the
country’s politics. Today it seems determined to be professional and apolitical, a
commitment reinforced by its participation in peace-keeping missions elsewhere
in Africa. During the post-election violence the generals made it clear to the
president that they were not willing to be called out, that they had seen during
their peace-keeping work that military involvement could make domestic conflicts
worse, and that political problems needed political, not military solutions. Such
conclusions bode well for the future of democracy in Kenya.

The police
The assessment of the police is much less positive. Of course their job includes
dealing with domestic violence, but they are not well prepared to handle it,
Volume 8 No 1 87

meeting it with more force rather than containing or de-escalating it. The police
were provided with money from the donor election basket but they used it
to purchase riot gear rather than for training. They are alleged to have been
responsible for much of the violence, either because they chose not to prevent it
or because they used excessive force to deal with protestors (possibly accounting
for 43% of an estimated 1 500 people killed in the election aftermath [IMLU 2008]).
The Waki Commission is highly critical of their role.6

COULD THE VIOLENCE HAVE BEEN PREDICTED?

Clearly some international observers and Kenyans were more sanguine about the
2007 elections than they should have been. They reasoned that there had been no
violence or significant fraud during the 2002 presidential elections, despite these
having led to a change in governing party, or in the 2005 constitutional referendum,
which had been lost by the new president. Furthermore, Samuel Kivuitu would
once again head the ECK.
The weaknesses in the ECK were either unknown or poorly communicated to
those who might have been able to mitigate them (particularly the internationals).
Surely, it was believed, Kenya had now entered the ranks of mature democracies.
This conclusion, however, ignored important differences between the 2002 and
2007 elections. In 2002 both the competing presidential candidates were Kikuyu,
which removed from contention the most toxic ethnic issue in Kenyan politics.
In the 2002 presidential elections Kibaki had promised a new constitution
with a position of prime minister in return for Odinga’s active support, with
the understanding that the post would be given to Odinga. Kibaki kept his
commitment to submit a new constitution to a referendum but it did not include
the prime ministership that had been promised to Odinga. Odinga then led the
fight to defeat this new constitution and won a ‘no’ vote, leaving Kibaki with the
old Constitution, with the full powers of the Kenyatta-Moi ‘imperial’ presidency
fully in place. In important ways he won by losing.
In contrast to the 2002 elections the candidates in the 2007 presidential contest
represented two of the ethnic communities with the bitterest history of conflict
– Kibaki, the Kikuyu, and Odinga, the Luo – with the added elements that their
fall out over the constitution meant the two men did not trust each other. The
presence of the third- party ‘spoiler’ candidate, Musyoka (Kamba), made it just
possible that Kibaki might squeak through despite widespread disappointment
with the corruption of the Kikuyus who surrounded him in the government. A
sitting president, a deep ethnic divide, personal animosity between the contestants,

6 http://www.nation.co.ke/blob/view/-/483378/data/46262/-/4cocav/-/WakiRep.pdf.
88 Journal of African Elections

the extraordinary powers of an ‘imperial’ presidency at stake, and the likelihood

of a very close election all carried the threat of trouble.
Perhaps some internationals did not want to see the danger signals. Despite
his corrupt administration, Kibaki was highly respected as an individual, the
country was undergoing excellent economic growth, and Kenya was a good ally
in dealing with the ‘war on terror’ in the Horn of Africa. If this were the view of
the US, which played a leading role in supporting the elections, it might have
encouraged some members of the Kibaki government to think they could get
away with tampering with the election results.
This was not the case with the European Commission or the UK High
Commission, however, both of which foresaw that the election would be difficult
and expected that there might be violence in the central Rift Valley. As a result,
the EC representatives in Kenya put in a strong, early bid for a European Union
observer mission and, with support from the UK High Commission, was able to
overcome the optimism of Brussels and obtain field observers.
The EC and the UK hoped that a strong international presence would be all
that was necessary to keep the election reasonably fair. Very close to the election
there were signals that the most likely strategy for illegitimately keeping Odinga
from the presidency would be to fraudulently deny him a victory in his Nairobi
parliamentary seat. (In Kenya only an MP can become president.) Thus the UK
high commissioner and the US ambassador personally observed the elections in
Odinga’s Langata constituency.
There were other danger signals the international community failed to see.
For example, the significance of the fact that the ECK declined to implement
important recommendations for its technical operations was not adequately
conveyed to the donors who were not directly supervising the election support
programme.
Nonetheless, it is perplexing that the only attempt to mitigate the dangers
of violence in the powder keg of the central Rift Valley, where ethnic conflicts
over land always boil just below the surface, was to provide election observers.
The Rift Valley had exploded in similar electoral circumstances in 1992 and 1997
and internal and external experts on conflict warned that the same might happen
again (Bayne, Muragu, Newton & Thomas 2007).
The donors had offered to finance district peace committees, which might
have mitigated the danger, or at least provided information about the violence
being planned. The ECK turned down this offer, which mystifies us, since it
was known that the major victims of such violence would be poor Kikuyu, the
ethnic group that controlled the presidency. Were the Kikuyu political elite so
anxious to retain full control over the election machinery that they were willing
to sacrifice their poorer brethren? Perhaps the attitudes of both the ECK and the
Volume 8 No 1 89

police should have sent warning signals to the donors. If, indeed, they did, the
signals were ignored.
The major point we are making here is that if there was a failure by domestic
and international officials to prepare for violence during the 2007 elections it was
not due to the absence of risk analysis but to de facto policy. Although the full scale
of the violence was not anticipated, and although it was not predicted to happen
in the locations where it would damage international and elite interests, there
was enough warning that it should have been a major concern to the government
and development agencies in the country. The loss of life in the central Rift Valley
in 1992 and 1997 was considerable and the large numbers of IDPs that resulted
from it had never been adequately dealt with.

COULD THE VIOLENCE HAVE BEEN AVOIDED?

There were eight main causes of the violence. Starting from the premise that
a trigger factor is required to shift a country with underlying tensions from a
peaceful to a violent state (Kimenyi & Ndung’u 2005), the immediate cause of
the 2008 election violence was the mismanagement and likely rigging of the
elections. Two intermediate causes: the role of the media and hate speech and the
predominance of ethnic politics. Finally, there were five underlying causes: an
overly powerful executive, a centralised state with ethno-regional inequalities,
mismanagement of land, a history of impunity for violence and corruption,
and poverty and youth unemployment. All these causes were intertwined and
mutually reinforcing.
Observers of Kenya’s politics agree that the gross mismanagement of the
elections triggered the violence. If the polling and counting had been conducted
transparently and fairly and if the loser had accepted the results, it is probable that
the country would have experienced only minor violence. If the donors had been
fully aware of the impending technical difficulties in the electoral management
process and had pressed aggressively for their repair and if the international
community had spoken strongly to President Kibaki in a unified voice about the
advantages of conducting a fair election and accepting its outcome, it is possible
that the trigger would not have been pulled and the violence would have been
avoided. In the event, as noted above, the international community missed or
failed to prioritise some of the important signals of impending trouble and it may
not have been united in its message to Kibaki. Finally, it is possible that if donors
had created a conflict mediation and early warning system for the central Rift
Valley early enough and had forcefully signalled its disapproval of any electoral
violence, it might have been possible to persuade the key perpetrators to take a
different course of action.
90 Journal of African Elections

Nonetheless, the very factors that made the consequences of electoral

mismanagement so explosive also made it possible that the Kenyan government
and its opponents, in a close contest, would have been willing to risk manipulating
the results. Too much was at stake and the chances of squeaking through were
too great. In the event, what is past cannot be undone. On the other hand, as
Santayana said, those who ignore history are doomed to repeat it. What is crucially
important is that the underlying causes of Kenya’s recurring electoral violence
now be addressed.

WHAT IS NEEDED NOW?

Agenda 4 of the Annan agreement committed the signatories to address most of

the sources of violent conflict in Kenyan society. The process of doing so has only
just begun and may possibly be stalled. Here we review what we believe are the
most important of these issues and what might be done about them.

Constitutional issues
A number of the most difficult issues facing the country have constitutional roots.
Several attempts have been made in the past decade to reform the Constitution,
good ideas have been deliberated and public views solicited. There is no need
to begin the process afresh. The most thorough constitutional review was led by
a distinguished Kenyan Asian law professor, Yash Ghai, and resulted in what is
called the ‘Bomas’ draft. If, as originally planned, it had been adopted before the
2002 elections it would probably have been easily accepted and created a far less
conflictual foundation for Kenyan politics. For a variety of reasons, however, its
presentation was delayed, and the opposition felt it was unacceptably dangerous
to extend in any way the term of office of then President Daniel arap Moi.
As a consequence President Kibaki was elected under the old Constitution
and, once in power, was unwilling to surrender the powers of the office. Instead,
he submitted to a referendum what is called the ‘Wako’ draft, which did not satisfy
other sections of the reform coalition and was defeated by the Orange Democratic
Movement under the leadership of Raila Odinga.
It is said that Professor Ghai has recently prepared another draft that pulls
together the most acceptable features of all the drafts that are under consideration.
Justice Minister Martha Karua has also been charged with presenting still another
constitutional proposal. Nonetheless, many observers believe that Kenya is
now more likely to proceed by amending the existing constitution rather than
putting a new one to another referendum. Whatever the strategy, a number of
key constitutional issues must be addressed.
Volume 8 No 1 91

Presidential authoritarianism must be reduced

A key dimension of the violence that frequently surrounds presidential elections
in Africa is the extreme concentration of authority in the presidency, with
almost no checks on its power. The result is not only an election in which ‘the
winner takes all for life’ but in which there is almost no limit to the president’s
control. In the political science literature this phenomenon is known as ‘imperial
presidency’, but a Cabinet minister we consulted preferred to speak of ‘presidential
authoritarianism’, which may be a more politic term.
The deconcentration of the powers of the presidency may follow from a
number of coordinate reforms which are under consideration in Kenya – the
creation of a prime minister with specified authority, an enhanced role for
Parliament, an independent and impartial judiciary, legal limits on what civil
servants can be asked to do, and devolution of authority to local government
bodies.

Clarity about the role of the prime minister

Following the Annan agreement, Kenya’s constitution was amended to provide for
a prime minister, answerable to Parliament. The duties and authority associated
with this position are not clear, however, and are contested within the Cabinet.
Currently the issue is not pressing because Kibaki is willing to let Odinga take the
initiative as prime minister. It may, however, become a source of conflict between
any succeeding president and prime minister.
Some analysts express doubts about the long-term feasibility of a hybrid
presidential-parliamentary system, pointing to its inherent potential for
inter-governmental conflict. However, others call attention to the fact that a
strengthened Parliament with a prime minister counterbalances the presidency
and injects accountability into the executive, and that most Kenyans, in fact,
support the concept of an executive president with diminished authority.

Executive responsibility to the electorate or Parliament

There is significant debate over what form of representation would best prevent
ethnic politics and violence. The problem is that direct presidential elections
create incentives for political elites with firm majorities in their constituencies
either to prevent opponents and minorities from voting or to displace them from
their constituencies and provinces. The reason for this is that if a candidate is to
achieve the 25 per cent threshold every vote counts in each constituency and in
each province. A Westminster-style executive selected by MPs according to a
single-member-first-past-the-post (SM-FPTP) system would reduce the incentive
to ethnic violence in a society like Kenya because the votes of ethnic minorities
(who are currently driven from the polls) would become irrelevant. The fact that
92 Journal of African Elections

a significant proportion of the executive (the prime minister and those he selects
to the Cabinet) is responsible to Parliament would therefore somewhat reduce
the incentive to violence.

Method of Representation
The allocation of seats in the Kenyan Parliament is hugely unequal, with the
largest constituency more than 20 times more populous than the smallest. These
inequities result from the merging of the National Assembly and the Senate in
the 1960s and the addition in the Moi years of districts for small ethnic groups. In
both periods Parliament gave symbolic representation but had no real power, so
the disparities in constituency size did not matter. Now that Parliament has re-
emerged as a significant governmental institution and selects the prime minister,
the extent of the inequalities is no longer supportable. As noted above, single-
member constituencies with plurality votes (SM-FPTP) in a purely parliamentary
democracy, as advocated in ‘Bomas’, reduce the incentives for violence because
most constituencies are drawn along ethnic lines and are thus controlled by one
ethnic group. But to achieve equality among SM-FPTP constituencies would
require major changes in the electoral map. At the other end of the spectrum,
Kanyinga (2006) argues that pure proportional representation (PR) using a list
system would require leaders to take a national rather than geographic outlook
during elections. But pure PR might well increase the chances of inter-ethnic
conflict because it would make the country a single multi-ethnic constituency.
A mixed system, with a proportion of MPs selected by FPTP and a
compensating balance voted in by a list PR system, has been advocated as a recipe
for a more ethnically balanced Parliament. While this system would probably
provide a balance in Parliament , it might still produce incentives to use violence
to prevent minority opponents from voting. A final possibility would be to return
to the two-house system created at independence, with the Assembly chosen in
equal population constituencies and the Senate based on distinct ethnic groups
(which would give more effective representation to the smaller groups).
This last option shares with the SM-FPTP system the advantage that it
provides no incentives for violence against minority voters. No electoral reforms
formally proposed to date call for pure PR or MPR. Barkan, Densham & Rushton
(2006, pp 926-39) use 1997 demographic and electoral data to model how the
different systems would affect the distribution of parliamentary seats between
the major parties and find that the effects would be remarkably modest, which
enhances their political feasibility. It is clear that one of these reforms must
be introduced in the interests of the long-term legitimacy of parliamentary
representation and, given Kenya’s history, those that would reduce the incentives
for violence deserve special scrutiny.
Volume 8 No 1 93

Devolution
The ‘Bomas’ draft proposed a return to the system, provided for in the inde­
pendence Constitution, of regional governments placed between the centre
and the district and municipal councils. The Swahili term for regions is majimbo,
which is the word used when proposals for such devolution are debated in Kenya.
On the one hand, devolution would decrease the powers of the centre, make the
attraction of capturing the presidency a bit less overwhelming, and provide more
alternative bases of power for ethnic politicians who were out of favour at the
centre – all of which would seem to reduce electoral violence. On the other hand,
majimbo has become a code word for returning control of land to local communities
and driving out the ‘foreign’ ethnic groups. As such, majimbo is advocated as an
instrument of autarchy by the aggrieved Kalenjin and Miji-Kenda and feared as
a mechanism for discrimination by the Kikuyu, who are land scarce and widely
settled throughout the country.
Mahmood Mamdani (1996) has argued that such ethnically exclusive control
over land is fundamentally regressive, both socially and politically, as it gives the
local chiefs who control the land despotic powers over their subjects. Independent
of the question of what is historically progressive, for a number of reasons it is
probably not an opportune moment to move forward on majimbo.

• The topic itself has become deeply divisive and greater spur to
violence than the other constitutional issues.
• Devolution, or some greater local voice on land issues, could just as
easily be addressed by expanding the powers of the district councils,
which are already in place.
• The need for greater equity and transparency in the distribution of
central resources could as easily be achieved by means of legislated
allocation formulae.7
• International evidence suggests that there would be fewer governance
gains from regional governments than from the centre and few
democratisation gains over the district councils. Politicians with
frustrated national aspirations would the main beneficiaries.
• The majimbo debate is really about conflict over land, an issue that
would be better addressed directly, as we suggest below.

7 Note, however, that Kenyatta (through harambee) decentralised to communities the responsibility for
initiating service expansion in order to legitimate an unequal distribution to Kikuyus (who were better
placed to raise funds locally). Also note that redistribution was a major agenda for Moi and that central
reallocation is unlikely to go further than the extent he achieved.
94 Journal of African Elections

Electoral Management
Re-establishment of the Electoral Commission of Kenya
The 1997 Inter-Party Parliamentary Group (IPPG) agreement made the ECK a
representative body. There are two downsides to this reform.
Firstly, there is no effective oversight mechanism to ensure that opposition
parties nominate commissioners. Instead of relying on legislation (which may
never be enacted) the ‘Bomas’ draft entrenched an oversight mechanism in the
Constitution that hindered the president’s ability to appoint commissioners
without consultation. It required that Parliament approve presidential nominees
to the renamed Electoral and Boundaries Commission (EBC) (Chitere, Chweya,
Masya, Testensen & Waiganjo 2006).
Secondly, most commissioners in a representative ECK are not sufficiently
qualified to carry out their tasks (Muite 2008), leaving the commission open to
incompetence and abuse. The 2006 Constitutional Minimum Reforms Package,
which called for detailed legislation providing for a professional ECK comprising
nine commissioners (instead of the current 21) who would formulate electoral
policy and supervise a professional secretariat, would have professionalised the
ECK. Paul Muite (2008), the MP who spearheaded the package, argues that the
professionalisation of the ECK prior to the elections would have resulted in free
and fair elections in which the results, even if the victor won by only a small
margin, would have been accepted. In terms of the Minimum Reforms Package
commissioners would not be appointed but would be openly recruited with regard
to merit, regional diversity, gender and approval by Parliament (Muite 2008).
This is one area where there has been some progress. Parliament has
passed a constitutional amendment creating the Interim Independent Electoral
Commission and the Interim Boundary Commission and has disbanded the
disgraced ECK.

Voter Registration
The Kriegler report (2008) details problems with the voter registration process
and suggests ways of improving it. It finds that it is highly likely that there are
significant numbers of deceased voters still on the voters’ rolls (in some provinces
more than others) and that the voting procedures make it easy for voters to cast
their ballot in one electoral contest and not in another (making it impossible to
use the roll to determine whether ballot boxes have been stuffed).
During the course of our interviews a senior person in a CSO concerned
with voter education reported an incident he had personally witnessed in his
home rural constituency on which he has never given evidence. At the end of
the election day a voting official announced to those present at the counting
Volume 8 No 1 95

that X number of presidential ballots had been ‘unused’ (presumably because the
voters had died or because they had cast their ballots in one of the other elections
and not in the presidential election).
The group agreed that those ballots would be cast for the presidential
candidate who was overwhelmingly favoured at that polling station and that
the surplus parliamentary ballots would be distributed evenly between all the
candidates so that they would not affect those results and would not produce
a discrepancy between the two races. The agent for the party that was hurt by
this procedure announced that he had no objection as his sole concern was for
his party’s parliamentary candidate; he did not disagree with the others about
the presidency.
If even someone who is professionally engaged in voter education will not
act against such fraud it is clear that the temptations and social pressures are
overwhelming in (rural) constituencies where the community is united in its
wishes. The only thing that can be done, realistically, is to reduce the temptation
and this demands both that the names of deceased voters be removed from the
rolls well in advance and that voters be issued with all their ballot papers when
they arrive at the polling station.

The courts
Both the ‘Bomas’ Draft and the ‘Wako’ Bill strengthen the judiciary by making it
more independent and able to check the executive. ‘Bomas’ and ‘Wako’ describe
principles that guide the exercise of judicial power and create a Supreme Court
that exercises jurisdiction over presidential impeachment, election petitions, and
constitutional matters.
Both documents recommend a Judicial Service Commission composed
of appointees who are independent of the executive. The commission’s
recommendations for the appointment of judges must be approved by Parliament,
and the judiciary is granted fiscal autonomy from the executive (NCC 2004,
pp 85-95; RoK 2005, pp 93-102). If they had been implemented these reforms might
have restored some faith in the judiciary as a credible institution independent of
the executive. One article which is especially salient to the prevention of election
violence is that the Supreme Court would replace the current ad hoc system of
resolving election disputes. If such a system had been in place the ODM and its
supporters would have been able to choose a legal rather than a political – and
violent – route to resolve the election dispute.
On the other hand, it is unrealistic to think that the courts can stand alone
against extreme pressure from the presidency. Such changes in the Constitution
may well make the courts effective in resolving disputes over parliamentary
96 Journal of African Elections

seats, but for them to decide on disputes with a sitting president the current
concentration of powers in the presidency would have to be reduced or the stakes
would be too high.

The police
The ‘Commission of Inquiry into Post-Election Violence’ chaired by Justice Philip
Nyamu Waki (2008) submitted a hard-hitting report suggesting a number of
reforms in relation to the police. The report as a whole initially met with a hostile
reception from all the major parties but it is not clear at this stage whether or not
this also applies to the recommendations about the police.

Prosecutions
Political violence in Kenya has been conducted with impunity. Prosecutions
have been extremely rare, which has only encouraged further violence and
built resentment and calls for revenge. The Waki Commission called for an
independent tribunal to prosecute those against whom it is believed there is a
prima facie case of organising electoral violence. It gave Kofi Annan a confidential
list of those whom it felt should be charged and asked that it be turned over to
the International Criminal Court if the government of Kenya did not prosecute
them (Waki 2008).8 Despite the hostility of the major political parties to the Waki
report a local tribunal seemed less threatening than the prospect of international
indictments so the government moved swiftly to propose a mechanism for
establishing such a tribunal and Parliament was recalled on Tuesday 20 January
2008 to pass the necessary legislation. Only time will tell how vigorously the
cases will be prosecuted.

National Land Commission

We noted above that the government’s failure to address land conflicts and
the manipulation by political elites of grievances over land have been the
underlying cause of election violence since the introduction of multiparty politics.
Furthermore, the earlier rounds of electoral violence inspired by land disputes
(dating back to 1992) have left substantial numbers of internally-displaced families
whom it is inhumane to ignore any longer and whose own inclination to violent
revenge is undeterred.
It is vital that the land-inspired conflicts be addressed. But it is also important
to recognise that they are unlikely to be ‘solved’. First, the competing discourses

8 http://www.nation.co.ke/News/politics/-/1064/485606/-/yw8abjz/-/index.html.
Volume 8 No 1 97

over land are not only incompatible; they are also legitimate. It is true both that
the central Rift Valley was not occupied by Kikuyu when the colonial incursion
occurred but also that they have now been living there in significant numbers for
nearly a century and that they moved there in good faith.
The land was stolen, but by whites. Kikuyu smallholders who live there
now bought the land and are as poor as the Kalenjin who want the land ‘back’.
Secondly, many of the large landholders did acquire their land under questionable
conditions and, in many cases, justice would be served by their being made to
surrender it – either by order or by the pressure of progressive land taxation.
However, the amount of such land that is viable for smallholder crop production
is inadequate to meet the demands of those with historical claims to be in the
area, much less of those who are landless.
These questionable large landowners are also among the most politically
powerful people in Kenya and are therefore unlikely to be easily or entirely
dislodged. Thus, no matter what is done, there will be a ‘land problem’. According
to experts in this field the crucial step is to create a process of deliberation,
negotiation and mediation so that all parties come to realise the limits of what
is possible and reach a compromise with which they can live. The first steps for
such deliberative and participatory processes are contained in a Bill proposed by
Lands Minister James Orengo.
The current Constitution concentrates control over land in the hands of the
president, who holds it in trust for the state. The president uses these powers
without oversight to make land grants to individuals and corporations and to
arbitrate land disputes. One recommendation for resolving the land problems that
is contained in both the ‘Bomas’ Draft and ‘Wako’ Bill is to remove powers over
land allocation from the hands of the executive by creating a central regulatory
agency, the National Land Commission (NCC 2004, p 38; RoK 2005, pp 40-1). The
Land Commission would regulate the ownership and use of land on behalf of the
national and devolved governments.
One drawback is that neither draft constitutionally guarantees the Land
Com­mission’s independence. Both are silent on its funding, regulation, and
appointment of commissioners.
The ‘Bomas’ Draft goes a necessary step further than the ‘Wako’ Bill by
outlining legislation on land for Parliament to enact, which includes key provisions
for revising land laws and policy, affirming spousal land rights, investigating the
legality of government grants and dispositions of public land, settling the landless,
redistributing land from individuals back to the appropriate communities, and
setting limits on maximum land holdings (NCC 2004, pp 38-9).
If these articles had been implemented they would have started the much
needed process of establishing a tribunal to investigate and settle land disputes
98 Journal of African Elections

by peaceful means and would have contributed significantly to preventing the

2007 election violence.
The Draft Land Policy 2006, which closely mirrors the ‘Bomas’ recom­
mendations (Wambua 2008), was to have been introduced in Parliament by Cabinet
Minister Orengo, and the passage and implementation of this policy may reduce
the likelihood of further violent conflicts over land. The debate over land reform
seems, in early 2009, to be subsumed once again in the whole discussion about a
comprehensive review process for the Constitution (of which more below).

Employment-led development
Kenya is poor; 50 per cent of the population lives in absolute poverty (UNDP 2005).
Empirical studies find that violence is much more likely to occur in a low-income
country (Fearon & Laitin 2003; Collier, Hoeffler & Sambanis 2005; Collier 2007).
Kenya is currently experiencing a ‘youth bulge’, where the median age is 18
years and 42 per cent of the population is under 15 (UNICEF 2006). Forty per cent
or more of these youths are unemployed and most are under-educated; only 22 per
cent go on to secondary education (SID 2004). Being young, poor, uneducated, and
with limited prospects of employment or access to land makes people more likely
to engage in violence (Collier 2007). A recent analysis of the role of the youth in the
election violence (Youth Agenda 2008) confirms that they are easily manipulated
to violent ends when the bulk of them are unemployed and disillusioned. The
Kenyan government’s strategy for reconciliation acknowledged that youth violence
was due to a ‘lack of effective opportunities to integrate the majority of Kenyan
youths into the mainstream economic activities’ (RoK 2008b, p viii).
Youth vigilante groups have emerged in Kenya in the past decade offering
an attractive route for socialisation, income, and informal employment. Many
of these groups were activated during the election violence (OHCHR 2008). For
other youths recruitment was based on promises of a new leadership that would
redirect state assets to their ethnic group, direct payment (through a locally levied
‘fighting’ tax [HRW 2008]), land, jobs, and goods derived from looting. Grievances
about inequality and marginalisation were used to justify their actions.
Patterns of violence in some districts of Central and Western provinces
corroborate that unemployment is a motivating factor, for violence was directed
against the homes and food reserves of (competing) commercial and tea-farm
workers but left the commercial infrastructure intact (Mozersky 2008). While the
youth perpetrated 73 per cent of the attacks they were involved in planning and
financing only 7 per cent. Youth Agenda (2008) concludes that the youth were at
the tail end of the perpetrator chain and constituted the majority of the victims
of violence.
Volume 8 No 1 99

Young men are likely to remain available for violent mobilisation until they
find reasonable employment, accumulate some capital, and get married. An
‘employment-led development strategy’, as suggested in the International Labour
Office’s 1972 report on Kenya, would address both poverty and violence.

Conflict: resolution and management

Youth violence has been a persistent feature of political activity in Africa for the
past 60 years, particularly in relation to elections. It would seem that other, less
violent tools for conducting conflict are weakly developed, so the repertoire of
mass action is confined to the alternatives of passive acceptance of repression or
riot, with nothing in between. Instruments that other societies have discovered
for pursuing personal interests in the face of resistance or repression – petitions,
demonstrations, strikes, civil disobedience – are weakly developed in most of
Africa. The flowering of civil society in Kenya has provided some tools of protest
and change for the middle classes but far fewer for the poor.
Tools used to mediate inter-communal violent conflict and bring about
reconciliation have fallen into disrepair. International non-governmental
organisations and the Conflict Early Warning project based in Addis Ababa are
involved in innovative mediation work to reduce cattle raiding in the ‘Karamajong
cluster’, which includes Kenya but applies only to pastoralist areas.
In terms of reconciliation, the African Great Lakes Initiative of Friends Peace
Teams has made creative use of ‘alternatives to violence’ techniques in Rwanda
and Burundi and modest efforts are underway in Kenya (Peace Ways 2008). Much
more could be done on all these fronts. It would be unwise to allow Kenya to
slip back into silent resentment and unresolved wishes for revenge. Similarly, the
internally displaced persons created during episodes of election violence should
not be forgotten. There is an urgent need, on both humanitarian and conflict
prevention grounds, for programmes which will either reintegrate them in the
places from which they were chased or find them suitable alternative land on
which to settle.

WHAT ARE THE PROSPECTS FOR REFORM?

Agenda item 4 of the agreement Kofi Annan negotiated between the PNU and
the ODM nominally committed the parties to addressing the above issues. The
question is whether they will do so. In the past Kenya’s leaders have simply
ignored the causes and consequences of electoral violence and pretended they
were no longer relevant. This time around, the response has been more positive,
although still mixed.
100 Journal of African Elections

As noted, the Kriegler Commission’s call for the members of the ECK to
resign went unheeded. Ultimately, however, Parliament took the matter out of
the commissioners’ hands and passed a constitutional amendment creating
the Interim Independent Electoral Commission and the Interim Boundary
Commission. Appointments to the two commissions will be made by the
Parliamentary Select Committee on Constitutional Review (PSC), and confirmed
by Parliament as a whole. The term of office of the two commissions will be two
years, after which they will be dissolved.
The selection of the commissioners to serve in the IIEC was not without
drama. Parliament voted to reject the individual nominated to serve as chairman,
forcing the PSC to begin the recruitment procedures afresh.
Similarly, both the ODM and the PNU voted against the implementation
of the criminal prosecution recommendations of the Waki Commission.9 The
government drafted a Bill to establish a local tribunal to try suspects (some of
them Cabinet Ministers) rather than have them indicted by the International
Criminal Court, but Parliament voted against the Bill, preferring to have the
alleged perpetrators tried in the Hague.
Now that Odinga and the ODM have succeeded in achieving an amendment
to the Constitution to provide for the post of prime minister do they have
an incentive to attempt to reduce the constitutional powers of the ‘imperial’
presidency? As Kibaki is not an activist president Odinga, as prime minister, is,
in practice, already exercising most of the powers he might claim and he is in a
favourable position to win the presidency in 2012.
Those in the PNU who will challenge Odinga in his attempt to succeed Kibaki
have a better chance of winning the presidency than the prime ministership if
parliamentary constituencies continue to be constructed as they are at present,
so their motivation to press for constitutional change might not be as strong as
it would seem.
Informed observers of Kenyan politics believed it was more likely that
selective amendments would be made to the existing Constitution than that a
completely new one would be drafted and that even those changes would have
to be motivated by dedicated reformers. Opinion polls showed that there had not
been strong pressure from the public for a new constitution.
Contrary to these assessments, however, towards the end of 2008 Parliament
passed the Constitution of Kenya Review Act to pave the way for the completion
of the constitutional review process. The Act provides for a committee of seven
experts, of which three are international. The committee has been nominated

9 http://www.nation.co.ke/News/politics/-/1064/485606/-/yw8abjz/-/index.html.
Volume 8 No 1 101

by the PSC and thereafter appointed by the president. Thus, after a slow start,
progress is now being made.
The fact that Kenyans were deeply shaken by their country’s descent into
violence after the December 2007 elections, as was the international community,
has created an atmosphere in which the democratic changes for which the
country’s citizens have battled since 1990 can now be fully achieved. This is the
moment when Kenya’s democratic future can be secured. There is resistance to
fundamental change, but there is hope that that can be overcome.

– REFERENCES AND MATERIALS CONSULTED –

Abdi, J & J Deane. 2008. The Kenyan 2007 Elections and their aftermath: the role of media
and communication. An Analysis. Interim Draft for Presentation to Nairobi
Meeting of Donors, 13 March.
Anderson, D. 2008. ‘Kenya’s crisis and Africa’s fragile democracy’. Prospect Magazine,
February.
_____. & E Lochery. 2008. ‘Violence and Exodus in Kenya’s Rift Valley, 2008:
Predictable and Preventable?’ Journal of Eastern African Studies Special Issue:
Election Fever: Kenya’s Crisis 2(2).
Barkan, J D. 2008a. ‘Progress and Retreat in Africa. Legislatures on the rise?’ Journal
of Democracy 19(2).
_____. 2008b. Breaking the Stalemate in Kenya. Washington, DC: Center for Strategic
and International Studies Africa Program.
_____. 2008c. Too Close to Call: Why Kibaki Might Lose the 2007 Kenyan Election.
Washington, DC: Center for Strategic and International Studies Africa
Program.
_____. 2008d. ‘Will the Kenyan Settlement Hold?’ Current History 107(708).
_____. 2008e. ‘Testimony “Hearing on the Immediate and Underlying Causes and
Consequences of Flawed Democracy in Kenya”'. Prepared for the Senate
Foreign Relations Committee's Subcommittee on African Affairs.
_____, P J Densham & G Rushton. 2006. ‘Space Matters: Designing Better Electoral
Systems for Emerging Democracies’. American Journal of Political Science 50(4),
October.
Barton, M & M S Kimenyi. 2008. ‘Voting in Kenya: Putting Ethnicity in Perspective’.
Journal of Eastern African Studies 2(2).
Bayne, S, M Muragu, D Newton & B Thomas. 2007. ‘Strategic Conflict Assessment:
Kenya’. Nairobi: DFID Kenya.
Bayne, S. 2008. Kenya Conflict Assessment – Review of post-election violence.
102 Journal of African Elections

BBC. 2008. ‘The Kenyan 2007 elections and their aftermath: the role of media and
communication’. Policy Briefing 1. London: BBC World Service Trust.
Bevan, T. 2008. The Aftermath of the Kenyan Election – Focus on the Economy.
Consultancy Africa Intelligence. Available at: www.consultancyafrica.com/
elections/kenya_2007_followup
Bratton, M & M S Kimenyi. 2008. ‘Voting in Kenya: Putting Ethnicity in Perspective’.
Journal of Eastern African Studies. Special Issue: Election Fever: Kenya’s Crisis
2(2).
Carothers, T. 2008. Current Questions about the roles of elections and International Elections
Assistance in Developing and Transitional Countries. Carnegie Endowment
for International Peace, Prepared for the Politics and The State Team, UK
Department for International Development.
Cheeseman, N. 2008. ‘The Kenyan Elections of 2007: An Introduction’. Journal of Eastern
African Studies. Special Issue: Election Fever: Kenya’s Crisis 2(2).
Chitere, P, L Chweya, J Masya, A Tostensen & K Waiganjo. 2006. Kenya’s Constitutional
Documents: A Comparative Analysis. CMI Report: CHR Michelson Institute.
Coalition for Accountable Political Financing (CAPF). 2008. The Money Factor in Poll
Race. A Monitoring Report of the 2007 Elections.
Cook, M. 2007. ‘Final Report on the Activities of the 2007 Elections Assistance
Programme’.
_____. 2008. ‘Final report on the activities of the 2007 elections assistance programme’.
Unedited version. UNDP.
Collier, P, A Hoeffler & N Sambanis. 2005. ‘The Collier-Hoeffler Model of Civil War
Onset and the Case Study Project Research Design’. In P Collier & N Sambanis
(eds). Understanding Civil War Vol. I: Africa. World Bank.
Collier, P. 2007. The Bottom Billion. New York: Oxford University Press.
Democratic Governance Donor Group Review of Post Election Crisis Damage and
Needs Assessments.
DFID Kenya 2007a. Welcome to Nairobi, DFID Kenya Visitor’s Guide.
_____. 2007b. ‘Kenya Country Governance Analysis. Executive Summary’.
_____. 2007c. ‘Country Governance Assessment, Kenya. Governance strategy
December 10 2007’.
_____. 2008a. ‘Kenya CAP 2008 Statistics Annex’.
_____. 2008b. ‘Press Lines Post Election Violence in Kenya. Issue: UK response to post
election violence in Kenya’, 29 February.
_____. 2008d. ‘Kenya. Country Governance Analysis – Update June 2008. Joint analysis
agreed by FCO and MoD at post’.
_____. CAP. 2008. ‘Kenya Country Assistance Plan’. Framing Paper.
European Commission. 2008. ‘Updating the Governance Profile’. Presentation to EU
Member States, 3 April.
Volume 8 No 1 103

European Union Election Observation Mission. 2008. ‘Kenya Final Report General
Elections 27 December 2007’.
Fearon, J D & D D Laitin. 2003. ‘Ethnicity, Insurgency, and Civil War’. American
Political Science Review 97(1).
Gathii, J. 1994. ‘Kenya’s Legislative Culture and the Evolution of the Kenya
Constitution’. In Y Vyas, O Kibwana & S Wanjala (eds). Law and Development
in the Third World. Nairobi: Faculty of Law, University of Nairobi.
George, K. 2008. ‘Constitutional Reform and Conflict in Kenya’. Brighton: Institute
of Development Studies.
Ghai, Y. 2008. ‘Devolution: Restructuring the Kenyan State’. Journal of Eastern African
Studies. Special Issue: Election Fever: Kenya’s Crisis 2(2).
Gibson, C C & J D Long. 2008a. ‘Governance in Kenya: Ethnicity and Government
Performance in the 2007 Elections’. Department of Political Science, University
of California, San Diego. Paper prepared for presentation at the Center for
Strategic and International Studies, Washington, DC, 8 July.
_____. 2008b. ‘What Explains the 2007 Kenyan Vote? Using Exit Poll Data to Explore
Ethnicity and Government Performance in Voters’ Choice’. Department of
Political Science, UC San Diego. Paper presented at the SAIS, Johns Hopkins
University, July.
_____. 2008. ‘The 2007 Kenyan Exit Poll: Presentation to Independent Review
Commission’. San Diego: Department of Political Science, University of
California.
Githinji, M W & F Holmquist. 2008. ‘Kenya's Hopes and Impediments: The Anatomy
of a Crisis of Exclusion’. Journal of Eastern African Studies. Special Issue: Election
Fever: Kenya’s Crisis 2(2).
Githongo, J. 2008. ‘Kenya – Riding the Tiger’. Journal of Eastern African Studies. Special
Issue: Election Fever: Kenya’s Crisis 2(2).
Gona, G. 2008. ‘Changing Political Faces on Kenya’s Coast, 1992-2007’. Journal of
Eastern African Studies. Special Issue: Election Fever: Kenya’s Crisis 2(2).
Human Rights Watch. 2003. ‘Kenya: Human Rights Developments’. In Human
Rights Watch World Report 2003, 14 January. New York.
Human Rights Watch. 2008. Ballots to Bullets: Organised Political Violence and Kenya’s
Crisis of Governance, 20(1)(A)
Humanitarian Policy Group (HPG). 2008. ‘Crisis in Kenya: land, displacement and
the search for “durable solutions”’. HPG Discussion paper. Draft.
Humanitarian Services Committee. nd. ‘Transformative Leadership Values &
Ethics for Peace & Nation Building’. Humanitarian Services Committee
Subcommittee: Peace and Nation Building.
Independent Review Commission. 2008. ‘Report of the Independent Review
Commission on the General Elections held in Kenya on 27th December,
104 Journal of African Elections

2007’. Presented to His Excellency Hon Mwai Kibaki, C G H, MP, President

and Commander-in-chief of the Armed Forces of the Republic of Kenya.
Nairobi.
Initiative of Friends Peace Teams. 2008. Peace Ways: AGLI, III, 2. Kipkarren River,
Kenya: African Great Lakes, Initiative of Friends Peace Teams.
Institute for Education in Democracy and Constitution and Reform Education
Consortium. 2005. ‘Kenya Civil Society Observation Programme (KCSOP).
Report on the Referendum on the Proposed New Constitution of Kenya. 21st
November 2005’.
International Crisis Group. 2008. ‘Kenya in Crisis’. Africa Report 137.
International Labour Office. 1972. Employment, Incomes and Equality: A Strategy for
Increasing Productive Employment in Kenya. Geneva: International Labour
Office.
Internews. 2008. ‘Kenya: Media in the Post-Election Conflict’. Internews
Network.
IRIN 2008a. Republic of Kenya Humanitarian Country Profile. UN Office for the
Coordination of Humanitarian Affairs (UNOCHA).
_____. 2008b. ‘Kenya: It’s the economy, stupid (not just “tribalism”)’. IRIN Africa,
13 August 2008, UNOCHA.
Jensen, B. 2008 ‘Letter. To the Co-Chairs of the National Accord Implementation
Committee (Hon Prof George Saitoti and Hon Peter Anyang Nyongo)’.
Kanyinga, K. 2006. ‘Governance Institutions and Inequality in Kenya’. In Readings
on Inequality in Kenya: Sectoral Dynamics and Perspectives. Nairobi: Society for
International Development (SID), East Africa.
Kavulla, T R. 2008. ‘ “Our Enemies are God’s enemies”: The Religion and Politics
of Bishop Margaret Wanjiru, MP’. Journal of Eastern African Studies. Special
Issue: Election Fever: Kenya’s Crisis 2(2).
Kenya Joint Assistance Strategy. 2007-2012. Improving Aid and Effectiveness.
Kenya National Commission for Human Rights (KNCHR). 2007. Still Behaving Badly:
Second periodic report of the Election-Monitoring Project. Nairobi.
_____. 2008a. ‘Documentation and Investigation of Serious Violence of Human
Rights in the Post Election Violence in Kenya – KNCHR's Progress Report’.
Press Release. Nairobi, 27 March.
_____. 2008b. ‘KNCHR Progress Report on Investigating and Documenting Post-
Election Violence’. 27 March.
Kimenyi, M S & R G Romero. 2008a. ‘Identity, Grievance, and Economic Deter­
minants of Voting in the 2007 Kenyan Elections. Improving Institutions for
Pro-Poor Growth (iiG). iiG Workshop for CSOs Programme. The Kenyatta
International Conference Centre, 17 September.
_____. 2008b. ‘Tribalism as a Minimax-Regret Strategy: Evidence from Voting in the
Volume 8 No 1 105

2007 Kenyan Elections’. Improving Institutions for Pro-Poor Growth (iiG).

iiG Workshop for CSOs Programme. The Kenyatta International Conference
Centre. 17 September.
_____ & S Dercon. 2008. ‘Key Findings from Surveys’.
Kimenyi, M S & N S Ndung’u. 2005. ‘Sporadic Ethnic Violence: Why Has Kenya
Not Experienced a Full-Blown Civil War?’. In P Collier and N Sambanis (eds).
Understanding Civil War Vol. I: Africa. World Bank
Konrad Adenauer Stiftung. 2007. Parties and Democracy. The KAS Democracy Report
2007. Bonn: Bouvier.
Kriegler, J J et al. 2008. ‘Report of the Independent Review Commission on
the General Elections held in Kenya on 27th December, 2007’. Nairobi:
Government Printer, Government of Kenya, 17 September.
Leonard, D K. 1991. African Successes. Berkeley: University of California Press.
Lonsdale, J. 2008. ‘Soil, Work, Civilization and Citizenship in Kenya’. Journal of
Eastern African Studies. Special Issue: Election Fever: Kenya’s Crisis 2(2).
MacArthur, J. 2008. ‘How the West was Won: Regional Politics and Prophetic
Promises in the 2007 Kenyan Elections’. Journal of Eastern African Studies.
Special Issue: Election Fever: Kenya’s Crisis 2(2).
Manby, B. 2008. Was the APRM process in Kenya a waste of time? Lessons that should be
learned for the future. Open Society Institute, Africa Governance Monitoring &
Advocacy Project (AfriMAP).
Mamdani, M. 1996. Citizen and Subject: Contemporary Africa and the Legacy of Late
Colonialism. Princeton: Princeton University Press.
Mathenge, O. 2008. ‘Insecurity on the rise, says report’. Daily Nation, Tuesday 16
September.
Migdal, J. 1988. Strong Societies and Weak States: State-Society Relations and State Capabilities
in the Third World. Princeton: Princeton University Press.
Mozersky, D. 2008. ‘Testimony of David Mozersky, Horn of Africa Project Director.
International Crisis Group’. Hearing ‘The Immediate and Underlying Causes and
Consequences of Flawed Democracy in Kenya’. United States Committee on Foreign
Relations, Sub-Committee on African Affairs, 7 February
Mueller, S D. 2008. ‘The Political Economy of Kenya’s Crisis’. Journal of Eastern African
Studies. Special Issue: Election Fever: Kenya’s Crisis 2(2).
Muite, P. 2008. ‘Blatant disregard of agreements contributed to post-election chaos’.
Comment: Post-Election Violence, The Standard, 20 July.
Myrum, J, J Ward & M Marsh. 2008. A Rapid Assessment of Gender-Based Violence
during the Post-Election Violence in Kenya. UNFPA, UNICEF, UNIFEM, Christian
Children’s Fund.
National Accord Implementation Committee. 2008. ‘Report on National
Reconciliation and Emergency Social and Economic Recovery Strategy’.
106 Journal of African Elections

National Constitutional Conference. 2004. The Draft Constitution of Kenya 2004,

20 October 2004 (adopted by the National Constitutional Conference on 15
March 2004). Nairobi.
New Partnership for Africa’s Development. 2007. African Peer Review Mission Report
on Kenya. Nairobi: NEPAD Kenya Secretariat.
News, B. 2008. ‘Timeline: Sierra Leone’. http://newsvote.bbc.co.uk/mpapps/
pagetools/print/news.bbc.co.uk/1/hi/world/africa/country_
profiles/1065898.stm. Accessed: 02/09/2008, 09.51pm.’
Nowrojee, P. 2006. The November 2005 Referendum and Kenyan Politics Today.
Directions in Constitutional Reform. Nairobi: African Research and Resource
Forum.
_____. 2008. ‘Chief Justice: Is the Issue Hypocrisy or Control?’.
Osborn, M. 2008. ‘Fuelling the Flames: Rumour and Politics in Kibera’. Journal of
Eastern African Studies. Special Issue: Election Fever: Kenya’s Crisis 2(2).
Oyaro, K. 2008. ‘Kenya: The Media is not Innocent’. http://www.ipsnews.net/news.
asp?idnews=41049.
Peace Ways. 2008. Peace Ways: AGLI, III, 2. Kipkarren River, Kenya: African Great
Lakes Initiative of Friends Peace Teams, Fall.
Republic of Kenya (RoK). 2005. ‘The Proposed New Constitution of Kenya’. Kenya
Gazette Supplement, Nairobi.
_____.  2008a. ‘Agreement on the Principles of Partnership of the Coalition
Government and the National Reconciliation and Accord Act 2008’, 28
February.
_____.  2008b. ‘Report of the National Accord Implementation Committee on
National Reconciliation and Emergency Social and Economic Recovery
Strategy’, March.
_____.  2008c. ‘Kenyan National Dialogue and Reconciliation Independent Review
Committee Terms of Reference’, 4 March.
_____.  2008d. ‘Kenyan National Dialogue and Reconciliation Longer-Term Issues
and Solutions: Constitutional Review’.
_____. 2008e. The Constitution of Kenya Review Bill 2008. Kenya Gazette Supplement
46 (Bills No 14), Nairobi, 20 June.
Rosberg, C G & J Nottingham. 1966. The Myth of Mau Mau: Nationalism in Kenya.
Nairobi: East African Publishing House.
Sheehy, T & W Maina. 2008. ‘Final Evaluation Report: The 2007 Joint Elections
Assistance Programme (EAP)’. Nairobi: UNDP.
Society for International Development. 2004. Pulling Apart: Facts and Figures on
Inequality in Kenya. Nairobi: SID
Stewart, F. 2008a. Horizontal Inequalities and Conflict: Understanding Group Violence
in Multiethnic Societies. Hampshire: Palgrave Macmillan.
Volume 8 No 1 107

_____. 2008b. ‘Note for Discussion: Kenya, Horizontal Inequalities and the Political
Disturbances of 2008’. Oxford: Centre for Research on Inequality, Human
Security and Ethnicity, CRISE, Department of International Development,
University of Oxford.
The Kenya Gazette CX(23). 2008. Special Issue. Published by Authority of the Republic
of Kenya.
Throup, D W. 2008. ‘The Count’. Journal of Eastern African Studies. Special Issue:
Election Fever: Kenya’s Crisis 2(2).
Tomoko Nishimoto (Country Director). 2008. ‘Project progress review report
– Support to the African Union (AU) Panel of Eminent Person’s facilitation
efforts on peace building and reconciliation’, February-July.
United Nations Development Programme. 2005. Kenya Human Development Report.
hdr.undp.org/en/reports/nationalreports/Africa/Kenya/Kenya_2005_
en.pdf
UNDP Electoral Assistance Division, D. o. P. A. 2006. ‘Report of electoral needs
assessment mission, Kenya (mission dates 10-18 July 2006)’.
UNICEF. 2006. Statistics Kenya. Available at: www.unicef.org/infobycountry/
kenya_statistics.html (accessed 14 August 2008)
United Nations High Commissioner for Human Rights. 2008. ‘Report from OHCHR
Fact-finding Mission to Kenya, 6-28 February’.
Wachira, G. 2007. ‘Peace Agenda for faith-based bodies’. Wajibu. A Journal of Social
and Ethical Concern. Peacebuilding-Gaining or Losing? 22(1), April/May.
Waki, Justice P N. 2008. Waki Commission of Inquiry into Post-Election Violence: Final
Report. Nairobi, Government Printer, Government of Kenya, 15 Oct. Available
at: http://www.nation.co.ke/blob/view/-/483378/data/46262/-/4cocav/-
/WakiRep.pdf.
Willis, J. 2008. ‘What Has He Got Up His Sleeve? Advertising the Kenya Presidential
Candidates in 2007’. Journal of Eastern African Studies. Special Issue: Election
Fever: Kenya’s Crisis 2(2).
Youth Agenda. 2008. Who is Guilty? Youth Perspectives on the 2007 Pre-Polls Electoral
Violence in Kenya. Nairobi: Youth Agenda. Available at: youthagenda.net/
pdf/YAA_Guilty.pdf)
Zolberg, A. 1966 Creating Political Order: The Party-States of West Africa. Chicago:
Rand McNally.