KE-B-02: The 21 November 2005 Constitutional Referendum — The Wako Draft, the Orange–Banana Campaign, and the Reconfiguration of Kenyan Politics (1997–2005)

Status: DRAFTWords: 10,721

1. Key Takeaways

  • The 21 November 2005 referendum was the first national vote in Kenyan history on a constitutional text and the first occasion on which a sitting government's flagship constitutional proposal was defeated at the polls. Voters were asked to approve "The Proposed New Constitution of Kenya, 2005" — the Wako Draft published in the Kenya Gazette on 22 August 2005 by Attorney General S. Amos Wako. The result, declared by the Electoral Commission of Kenya, was No 58.1% to Yes 41.9% on a turnout of approximately 53% of registered voters [TBD-VERIFY: the ECK's Report on the National Referendum gives No 3,548,477 to Yes 2,545,495 with rejected ballots and abstentions accounting for the remainder; the round percentages cited in most secondary literature are No ~58% / Yes ~42%]. Voters rejected the government's draft on a clearly higher turnout than the Moi-era multi-party elections of 1992 and 1997 had achieved in many constituencies, and the rejection was geographically broad — only the Mt Kenya provinces (Central) and parts of the Eastern Embu/Meru sub-region returned solid Banana majorities. The vote was therefore not a narrow technical rebuff but a popular repudiation of the government draft as such.

  • The Orange and Banana symbols were a deliberate response to Kenya's illiteracy distribution and became politically determinative beyond the referendum itself. Under the Constitution of Kenya Review Act 1997 (Amendment) and the Referendum Regulations gazetted in September 2005, the ECK was empowered to allocate symbols for the Yes and No positions to enable voters who could not read the ballot to mark their preference. The Yes position was assigned the banana; the No position the orange. The choice — formally a neutral administrative allocation — became a political instrument: the No coalition rapidly converted the orange into a coalition brand, with orange T-shirts, orange-painted rallies, and orange political iconography that the No leaders carried into the post-2005 period. When the No coalition formalised itself as a political party in 2006, the name "Orange Democratic Movement" (ODM) explicitly anchored the new party to the 2005 mobilisation. The banana, the Yes side's emblem, did not survive politically in the same way; the Kibaki camp's post-referendum reorganisation into the Party of National Unity (PNU) for 2007 did not retain the banana brand. The asymmetric afterlife of the two symbols — the Orange becoming a durable opposition identity, the Banana fading — is itself part of the political legacy of the 2005 vote.

  • The constitutional-review process that produced the contested 2005 draft began with the 1997 Inter-Parties Parliamentary Group (IPPG) reforms and the Constitution of Kenya Review Act 2000, and the Bomas Conference of 2003–2004 was its participatory high point. The IPPG package of August 1997 had been a minimal-reform compromise that secured the 1997 election and committed the government to a comprehensive constitutional review afterwards. The Constitution of Kenya Review Act, first passed in 1997 and substantially reconstituted in 2000 under President Moi's government, established the Constitution of Kenya Review Commission (CKRC). The CKRC, chaired from 2001 by Yash Pal Ghai — a Kenyan-born constitutional lawyer who had returned from a distinguished international career — undertook the most systematic constitutional consultation in Kenyan history, holding public hearings in every constituency in 2001–2002 and producing the CKRC Draft Constitution of 27 September 2002. The National Constitutional Conference at the Bomas of Kenya cultural centre on the outskirts of Nairobi, convened in three sessions between April 2003 and 15 March 2004 with 629 delegates including all MPs, civil society, religious bodies, and women's organisations, produced the Bomas Draft. The Bomas Draft proposed a powerful Prime Minister leading the Executive with a ceremonial President, a strong regional/devolved tier of government, a Bill of Rights with justiciable socio-economic guarantees, and a series of accountability commissions. Whatever its compromises, the Bomas Draft was the participatory product of three years of structured consultation.

  • The Wako Draft was the Kibaki government's revision of the Bomas Draft, and its two most consequential departures were the dropping of the Prime Minister system and the dilution of devolution. Between March 2004 (Bomas Draft) and August 2005 (Wako Draft publication), the Kibaki Cabinet — through Justice Minister Kiraitu Murungi, Attorney General Amos Wako, Vice-President Moody Awori, and Constitutional Affairs Minister Martha Karua, with influential input from the Mt Kenya circle around the President — substantially rewrote the Bomas Draft's executive architecture. The Bomas premier-with-executive-authority was demoted to a presidentially appointed Prime Minister with administrative coordination but no independent constitutional standing; the Bomas regional tier was reduced; the powers of the proposed Senate were trimmed. The August 2003 walk-out of the National Convention Executive Council (NCEC) civil-society caucus from the Bomas Conference, in protest at what they read as government attempts to steer the process, was the early signal that the government and the participatory-conference architecture were on a collision course. By July–August 2005 the government had pushed the revised draft through Parliament's Constitutional Review (Amendment) Act 2004 and on to the Attorney General for publication.

  • The opposition coalition that became the No campaign was built inside the National Rainbow Coalition government itself, around the Liberal Democratic Party (LDP) faction led by Raila Odinga. The Memorandum of Understanding that had bound the Kibaki–Odinga coalition before the 2002 election had committed the government to deliver a new constitution with a Prime Minister post for Odinga; that commitment had been functionally nullified by mid-2003 (KE-B-01 §3 documents the betrayal). By 2004 the LDP, with Odinga at its head and including ministers Kalonzo Musyoka (Vice-President of the LDP and Cabinet Minister for Environment until 2004; subsequently Foreign Affairs), William Ruto (then a KANU MP for Eldoret North aligned with the LDP-led opposition), Najib Balala, Anyang' Nyong'o (the LDP Secretary-General whose institutional coordination held the coalition together), William Ole Ntimama, and several others, was operating as an opposition bloc inside the Cabinet. The September 2005 publication of the Wako Draft and the announcement of a referendum forced the Cabinet rebellion into the open: ministers who had signed Cabinet collective responsibility for the draft now campaigned against it in their constituencies. The Kibaki Cabinet's dilemma — whether to dismiss the dissenters before, during, or after the referendum — was managed by allowing them to remain until 23 November 2005, two days after the vote, when Kibaki dismissed the entire Cabinet and reconstituted it without the LDP rebels.

  • The September–November 2005 campaign was the most polarised national mobilisation Kenya had seen since the multi-party transition, and the No coalition won the air war as well as the ground war. The Banana coalition was led personally by President Kibaki, with Vice-President Awori, Cabinet ministers from the Democratic Party–NARC-Kenya wing (Karua, Murungi, Mwiraria, Murungaru, Michuki), and substantial state-resource deployment in mass rallies across Central Province and parts of the Rift Valley. The Orange coalition was led by Odinga as principal spokesperson, with Musyoka, Ruto, Balala, Ntimama, and FORD-Kenya leader Musikari Kombo, joined by KANU's Uhuru Kenyatta (whose 2002-defeated KANU had repositioned itself into opposition under Kenyatta's leadership). The Orange campaign mobilised on three grounds — the betrayed promise of devolution, the betrayed promise of the Prime Minister system, and the Cabinet's broken-MoU bad faith — and on one mobilisational ground that civil-society allies (the Law Society of Kenya, the National Convention Executive Council, the Kenya National Commission on Human Rights, sections of the National Council of Churches of Kenya) found compelling: the principle that a draft revised by an Attorney General behind closed doors should not be ratified over a participatory conference draft. The Orange campaign drew large crowds in Nyanza, Western, Coast, North-Eastern, and Rift Valley provinces; opinion polling by Steadman Group in the final fortnight showed the No camp consistently ahead by margins of 8–15 percentage points.

  • The 21 November 2005 result was decisive and ECK-administered without significant procedural challenge. Polling was peaceful by Kenyan election-day standards; the ECK under chairman Samuel Kivuitu — operating two years before the disputed 2007 count — administered a process that the international observer missions, including the Commonwealth Secretariat and the European Union, rated as credible. The final tally, declared on 22 November 2005, was Yes 2,545,495 (41.9%) to No 3,548,477 (58.1%) [TBD-VERIFY: precise vote totals vary by source — the ECK's Report on the National Referendum is the authoritative record; the percentage split has been quoted variously as 58.12/41.88 and 57/43 in different summaries]. Turnout was approximately 53% of registered voters, lower than the 2002 general election but high for a single-issue referendum on a complex constitutional text. The geographic pattern was politically consequential: Yes won pluralities in Central Province (Kibaki's home base, where the Mt Kenya political establishment campaigned hard), parts of Eastern (Embu and Meru), and small pockets of the Rift Valley. No won majorities, often by 70–80% margins, in Nyanza (Odinga's Luo heartland), Western (Luhya, where Mudavadi was campaigning No), Coast (a region with long-standing devolution claims), North-Eastern (Somali-Kenyan), and most of the Rift Valley (Kalenjin under Ruto's mobilisation).

  • President Kibaki's response — dismissing his entire Cabinet on 23 November 2005 and reconstituting it on 7 December without the LDP rebels — was decisive, divisive, and consequential. The dismissal was constitutionally permissible (the Kenyan presidency before the 2010 Constitution carried unfettered Cabinet-appointment power), but the rapid timing — two days after a referendum the government had lost decisively — communicated that the President was treating the No vote as a personal political affront rather than as a constitutional mandate for revising the draft. The reconstituted Cabinet, sworn in on 7 December 2005 (some appointments delayed into early 2006 because nominees declined to serve), excluded Odinga, Musyoka, Ruto, Ntimama, Balala, and the other LDP rebels. It included new Mt Kenya allies and some opportunistic crossovers. The dismissal converted the No coalition from a single-issue alliance into a self-conscious opposition bloc with a shared political identity, and it eliminated the last remaining institutional check on the Mt Kenya inner circle's dominance of the Kibaki government. The 2007 election campaign began politically — though not formally — within weeks of the dismissals.

  • The Orange Democratic Movement was launched in 2006 as the institutional successor of the No coalition; it split in late 2006–early 2007 into ODM (Odinga) and ODM-Kenya (Musyoka). The formal launch of ODM as a political party — a process that involved the registration of the Orange Democratic Movement as a distinct entity under the Political Parties Act and the contested capture of an existing party shell to enable rapid contestation in the 2007 cycle — proceeded through 2006. The intra-ODM presidential primary in August–September 2007, contested by Odinga, Musyoka, Ruto, Balala, Mudavadi, and Joseph Nyaga, was won by Odinga; Musyoka rejected the result and broke away with his ODM-Kenya (ODM-K) faction, ultimately running as a third-party candidate in December 2007 and taking the Kamba vote out of the Odinga coalition. The split, narrated by Musyoka's camp as a procedural objection to the primary and by Odinga's camp as a Kibaki-orchestrated splinter, was foundational to the December 2007 electoral arithmetic and to the post-election crisis. The 2005 No coalition's two key non-Luo anchors — Musyoka (Kamba) and Ruto (Kalenjin) — would, after 2007, follow very different trajectories: Musyoka into the post-2008 Grand Coalition as Vice-President under Kibaki; Ruto into a long alliance with Kenyatta culminating in the 2013 Jubilee victory and the 2022 Ruto presidency.

  • The 2005 defeat closed the constitutional-reform path for the remainder of the Kibaki first term and re-opened it only through the post-2008 Agenda Item 4 of the National Accord. Between November 2005 and the December 2007 election, the Kibaki government made no serious attempt to revive the constitutional-review process. The Constitution of Kenya Review Act was allowed to lapse; the CKRC was wound down; the question of devolution was deferred. When the 2007 election dispute precipitated the post-election violence (KE-B-03) and the Kofi Annan mediation produced the 28 February 2008 National Accord (KE-B-04), Agenda Item 4 of the Accord — long-term issues, including constitutional reform — explicitly took up the unfinished business that the 2005 referendum had blocked. The Committee of Experts established under the Constitution of Kenya Review Act 2008 (KE-C-01 §3) was given the task of producing a new draft by triangulating the Bomas Draft, the Wako Draft, and the original CKRC Draft. The Constitution of Kenya 2010, promulgated on 27 August 2010 with 67% Yes on a 72.6% turnout, restored most of what Bomas had proposed and the Wako Draft had stripped out — devolution to 47 counties, a Bill of Rights with justiciable socio-economic provisions, an independent Judiciary with a Supreme Court holding original presidential-petition jurisdiction. The 2010 architecture was, in significant part, the Bomas Draft delivered on the second attempt. The 2005 defeat is therefore best read not as a constitutional dead end but as the moment at which the participatory-conference legitimacy of Bomas survived a government attempt to rewrite it, and waited five years for the political conditions to re-open the path.

  • Three contested accounts of the 2005 referendum's significance, the Kibaki Cabinet's gamble, and the Orange–Banana legacy continue to circulate in the scholarship, and §11 develops each. The first account — pro-No, broadly the dominant Kenyan and international reading — holds the 2005 vote as a legitimate democratic rejection of a flawed government draft that preserved the participatory-conference principle and opened the path to the 2010 Constitution. The second — pro-Yes, defended by some former Cabinet members and by sympathetic biographers — holds that the No coalition was driven by political opportunism (Odinga's frustration with the unfulfilled MoU, Ruto's KANU-rooted opposition, Musyoka's Cabinet ambitions) more than by constitutional concerns, and that the Wako Draft, while imperfect, was the realistically attainable settlement of the period. The third — structural, developed in Ghai and Cottrell Ghai (2011), Mueller (2008), and Kanyinga (2010) — holds that the 2005 defeat reflected Kenya's broader devolution-versus-centralisation cleavage, dating to the majimbo/centralism debate at independence, that no single referendum could resolve and that 2010 only partially settled. Section 11 develops all three.

2. The Constitutional Pre-History: From the IPPG Reforms (1997) to the Constitution of Kenya Review Act (2000)

The constitutional question that the 2005 referendum tried — and failed — to resolve had been on Kenya's political agenda since the early 1990s civil-society reform movement. The 1991 repeal of Section 2A of the 1969 Constitution had restored multipartyism but left the rest of the Moi-era constitutional architecture intact, including a near-imperial presidency, executive control over the Judiciary and the electoral commission, and a unitary state with no constitutionally protected sub-national tier. Willy Mutunga's Constitution-Making from the Middle (1999) documents the civil-society networks — the National Convention Assembly, the Citizens' Coalition for Constitutional Change (4Cs), the Law Society of Kenya, and the church-based platforms — that mobilised through the mid-1990s for a comprehensive review.

The proximate trigger for action came in the run-up to the 1997 general election. Through the first half of 1997, civil-society and opposition mobilisation around the demand for comprehensive constitutional reform before the election produced episodes of state violence — the "Saba Saba" anniversary protests of 7 July 1997 in Nairobi, the killing of demonstrators in several constituencies — that threatened the credibility of the election cycle and the government's external legitimacy. The Moi government's response was the Inter-Parties Parliamentary Group (IPPG) compromise, negotiated through August–November 1997 between KANU and the parliamentary opposition (the Democratic Party of Kibaki, FORD-Kenya, the National Development Party of Odinga, and others). The IPPG package, enacted as a series of statutory amendments before the December 1997 election, delivered minimal-but-significant reforms: the licensing of broadcast media beyond the state-controlled KBC, the lifting of the worst restrictions on public assembly and association, the formal reconstitution of the Electoral Commission with some opposition-nominated commissioners, and the principle that comprehensive constitutional review would proceed after the election. The IPPG was a parliamentary deal that explicitly excluded the civil-society reform networks from the negotiation; it secured the 1997 election (which Moi won amid renewed opposition fragmentation) and bequeathed the post-election review commitment to the new Parliament.

The post-1997 review process moved with halting progress. The Constitution of Kenya Review Act 1997, intended to operationalise the IPPG commitment, was structurally inadequate and was substantially redrafted. Through 1998–2000, parallel review tracks emerged: a Parliamentary track, controlled by KANU, that emphasised technical legal drafting under government supervision; and a Civil Society/Ufungamano track, named after the Ufungamano House in Nairobi where the National Council of Churches of Kenya hosted the civil-society alternative process, that emphasised broad public consultation and the participatory-conference principle. The two tracks were merged through the Safari Park Accord of 28 September 2000 — a negotiated settlement between the parliamentary and Ufungamano sides that produced the consolidated Constitution of Kenya Review (Amendment) Act 2000.

The 2000 Act, enacted by Parliament in November 2000, established the Constitution of Kenya Review Commission (CKRC) as a 27-member body with commissioners nominated by Parliament (15), the Ufungamano process (12), and the President (with deemed appointment for nominees of both processes). The Act gave the CKRC a four-stage mandate: (a) collect public views nationwide; (b) draft a constitution; (c) submit the draft to a National Constitutional Conference; and (d) submit the conference output to Parliament for enactment, with provision (added by later amendment) for a referendum where required. The Commission's first chairperson, the senior advocate Charles Maranga Kibara, was succeeded in early 2001 by Yash Pal Ghai. The Ghai chairmanship — beginning before the 2002 Moi-to-Kibaki transition and continuing through the first Kibaki term — was the institutional locus of the 2001–2004 constitutional-review process.

3. The Constitution of Kenya Review Commission under Yash Pal Ghai (2001–2004) and the CKRC Draft of September 2002

Yash Pal Ghai was, by 2001, one of the world's most experienced comparative constitutional lawyers. Kenyan-born and educated at Oxford and Harvard, he had taught at the University of Warwick, the University of Hong Kong (where he had been a key adviser on the Hong Kong Basic Law in the 1980s), and elsewhere, and he had advised post-conflict constitutional processes in Papua New Guinea, Fiji, Nepal, Cambodia, and East Timor. His return to chair the CKRC was understood, at the time, as a serious institutional commitment: an independent figure with deep technical authority and no personal stake in Kenyan party politics. Kenya's Constitution: An Instrument for Change (Ghai and Cottrell Ghai, 2011) is his and his collaborator-spouse Jill Cottrell Ghai's authoritative reconstruction of the process from inside.

The CKRC's 2001–2002 public consultation was the most systematic constitutional outreach in Kenyan history. The Commission held public hearings in all 210 constituencies through 2001–early 2002, receiving submissions from individuals, churches, women's groups, professional bodies, ethnic associations, and political parties. The methodology — well documented in the CKRC's Final Report (February 2005) — was to record submissions verbatim, code them thematically, and report the results back to the Commission as the empirical input to drafting. The Commission's analysis of submissions identified strong popular support across regions for: (a) significant devolution of power to regional or provincial tiers; (b) a reduced and constitutionally constrained presidency; (c) a justiciable Bill of Rights including socio-economic guarantees; (d) accountability commissions for ethics, human rights, and gender; and (e) a Kadhi courts question that divided Muslim and Christian respondents and was retained as constitutionally anchored.

The CKRC's draft constitution, submitted to the Commission and made public on 27 September 2002 — three months before the December 2002 election that ended KANU rule — proposed a substantial restructuring. The Executive was to be a hybrid system: a President directly elected with reduced powers, and a Prime Minister leading the Government with executive authority and accountability to Parliament. Devolution would be to a tier of regions or provinces (the exact number and naming was an open question for the conference). The Bill of Rights, drafted with reference to the post-apartheid South African Bill of Rights, included justiciable economic and social rights. An array of accountability commissions — for human rights, ethics, gender, land, judicial service — was specified. The Kadhi courts were retained with constitutional anchoring. The September 2002 CKRC Draft was the document the incoming Kibaki administration inherited from the outgoing Moi government as the basis for the National Constitutional Conference.

The 2002 election interrupted the CKRC's timetable in ways that proved consequential. The Conference had been scheduled to begin in October 2002, but was deferred to allow the new government to take office. When the National Rainbow Coalition swept into power on 30 December 2002 with a Memorandum of Understanding (KE-B-01 §2–3) that included a commitment to deliver a new constitution within 100 days, the incoming Kibaki administration's interest in the CKRC process was substantial — at least at first. Justice Minister Kiraitu Murungi, Constitutional Affairs Minister Karua, and Attorney General Wako were all named as senior interlocutors with the CKRC. The 100-day commitment, however, proved impossible to deliver; the conference was rescheduled for April 2003.

4. The Bomas of Kenya Conference (2003–2004) and the Bomas Draft of March 2004

The National Constitutional Conference convened at the Bomas of Kenya — a cultural-and-conference centre on the southern outskirts of Nairobi, originally built in the 1970s as a venue for ethnographic performances and conferences — on 28 April 2003. The Conference comprised 629 delegates: all 222 MPs (under the pre-2010 Parliament), 41 representatives of district consultation processes, 126 representatives of civil-society and religious organisations, and additional delegates representing women, persons with disabilities, and youth. The Conference's chair was Professor Ghai in his capacity as CKRC chairperson; the deputy chairs and the various technical committees drew from across the commissioner ranks. The Conference was scheduled in three sessions across 2003 and into early 2004, allowing delegates to return to constituencies between sessions and report back.

The first Bomas session (April–June 2003) was structurally focused on the principles of the draft — the presidency/premier question, the devolution question, the Bill of Rights structure — and produced extensive debate but no final text. The second session (August–September 2003) was the most contentious and produced the major political ruptures. The Cabinet itself was divided: Murungi and Wako, broadly representing the Kibaki/Mt Kenya wing, pushed for retention of a strong presidency and a weaker premier; LDP members of the Cabinet, including Anyang' Nyong'o and Raila Odinga, pushed for the CKRC Draft's executive prime-ministerial settlement. The National Convention Executive Council (NCEC) — the civil-society coalition that had pushed the 1990s reform movement and that held substantial delegate strength at Bomas — walked out of the Conference in August 2003 in protest at what they characterised as government attempts to direct the outcomes through whip-line control of MP-delegates. The walk-out was symbolically consequential: it revealed that the Conference's participatory legitimacy was already contested before the votes had been taken.

The third Bomas session (January–March 2004) produced the final Bomas Draft after extensive procedural manoeuvring. On the executive question, the Conference voted for a system in which the Prime Minister, nominated from the parliamentary majority and confirmed by Parliament, would be Head of Government with the President as Head of State; the Cabinet would be drawn from MPs and would be accountable to Parliament for the conduct of government business. On devolution, the Conference voted for a regional tier — initially proposed as 14 regions, subsequently modified through the drafting committee — with constitutionally specified functions and revenue. On the Bill of Rights, the Conference adopted the CKRC Draft's justiciable economic and social rights provisions essentially intact. On the Kadhi courts, the Conference confirmed retention with constitutional anchoring. The Bomas Draft of 15 March 2004 was the consensus product of three sessions and approximately eleven months of structured deliberation.

The Bomas Draft, however, was politically vulnerable. The Mt Kenya wing of the Cabinet — including Kibaki personally, by the accounts of contemporaries — was opposed to the demotion of the presidency that the executive premier system implied. The MoU commitment to create a premier post for Odinga had not survived the post-2002 coalition politics (KE-B-01 §3); the prospect of a constitutional system that would deliver that post to Odinga through ordinary majority politics was even less attractive to the Kibaki camp. Through March–December 2004, the government took the Bomas Draft and began the revision process that would produce the Wako Draft. The institutional vehicle was the Constitutional Review (Amendment) Act 2004, which gave Parliament power to amend the Bomas Draft and to schedule a referendum.

5. The 2004–2005 Government Revision: From Bomas to the Wako Draft

The transition from the Bomas Draft of March 2004 to the Wako Draft published in August 2005 was the most consequential constitutional-drafting episode of the Kibaki first term, and the secondary literature is divided on how to characterise it. Branch (Kenya: Between Hope and Despair, 2011) and Hornsby (Kenya: A History Since Independence, 2012) treat it as a government rewrite that stripped from the participatory-conference text the provisions most likely to constrain the Mt Kenya inner circle's hold on executive power. Ghai and Cottrell Ghai (2011) — Ghai writing from the position of the CKRC chair who had presided over the Bomas Conference and whose institutional product was being rewritten — are explicit that the government revision violated the spirit of the participatory process and several of its procedural commitments. Defenders of the revision, including subsequent statements by Wako and by Murungi in published interviews and op-eds, argue that the Bomas Draft contained internal contradictions and drafting weaknesses that required technical revision, and that the executive-architecture changes were a legitimate exercise of constitutional drafting authority by an elected government.

The Constitutional Review (Amendment) Act 2004, enacted by Parliament in December 2004, was the legal vehicle for the revision. The Act amended the Constitution of Kenya Review Act 2000 to give Parliament power to revise the Bomas Draft before publication and to schedule a national referendum on the published draft. Civil-society litigation challenged the Act — the Reverend Timothy Njoya and others filed a constitutional petition arguing that the Bomas Draft, as the product of the National Constitutional Conference, could not be revised by Parliament without returning to the Conference for ratification. The High Court of Kenya, in a March 2004 ruling that pre-empted the December 2004 Act (the Njoya litigation was filed against the procedural framework as it was being constructed), held that the people of Kenya had a constituent power that could not be exercised by Parliament alone and that any new constitution required a referendum. The 2004 amending Act incorporated the referendum requirement; it did not, however, address the question of whether Parliament could rewrite the Bomas text before referendum, which Njoya and the NCEC continued to dispute.

The revisions made between March 2004 and August 2005 fell into four principal categories. First, the executive architecture was rewritten. The Bomas Draft's executive Prime Minister, leading the Government and accountable to Parliament, was replaced by a presidentially appointed Prime Minister whose role was administrative coordination of Cabinet business under the President's direction. The Cabinet remained primarily a presidential instrument; Parliament's confirmation role was reduced. The result was a presidency closer to the pre-2010 status quo than to the Bomas settlement. Second, the devolution architecture was diluted. The Bomas regional tier with constitutionally specified functions was replaced with a weaker district-level structure under more discretionary central-government control; the revenue-sharing provisions were softened. Third, several of the accountability-commission provisions were trimmed in their independence guarantees. Fourth, drafting and technical changes — some genuine improvements, some politically motivated rewordings — accumulated across multiple chapters.

By July 2005 the revised draft had passed through the Parliamentary Select Committee and the National Assembly. On 22 August 2005, Attorney General Wako gazetted The Proposed New Constitution of Kenya, 2005 as Kenya Gazette Supplement No. 63. The publication began the constitutionally-required period before a referendum; the ECK was empowered to set the date, and 21 November 2005 was announced shortly thereafter. The label "Wako Draft" — sometimes also called the "Kilifi Draft" after the Kilifi retreat at which the final Parliamentary revisions had been agreed — fixed the public identification of the document with the Attorney General whose signature appeared at the foot of the gazette publication, even though Wako was a process-author rather than a sole drafter.

6. The Coalition of Refusal: Raila Odinga, the LDP Rebellion within NARC, and the Cabinet Dilemma (2004–2005)

The opposition to the Wako Draft was not built from outside the Kibaki government; it was built from inside it. The Liberal Democratic Party, the NARC component led by Raila Odinga that had carried the Luo vote into the 2002 victory, had operated as an internal opposition since the 2003 MoU betrayal. By 2004, the LDP's institutional identity inside NARC was anomalous: LDP politicians held Cabinet portfolios under Cabinet collective responsibility while simultaneously campaigning publicly on the unfulfilled MoU and the constitutional-review question.

The principal figures of the LDP rebellion were a small group of Cabinet ministers and senior politicians. Raila Odinga, holding the Roads, Public Works and Housing portfolio, was the public face of the rebellion. Kalonzo Musyoka, a Kamba politician who had served under Moi and crossed to NARC for the 2002 cycle and held the Environment and subsequently Foreign Affairs portfolios, was the second figure; he positioned himself as a moderate within the LDP camp with his own presidential ambitions. Professor Anyang' Nyong'o, the LDP Secretary-General and Minister for Planning and National Development, provided the intellectual and institutional coordination — his organisational work through 2004–2005 in convening LDP-faction meetings, briefing media, and managing the political calendar was central to the coalition's coherence. Najib Balala, the Mombasa-based Coast politician, brought Coastal and devolutionist support. William Ole Ntimama, the long-serving Maasai politician and Cabinet veteran, brought Rift Valley pastoralist representation. Linah Kilimo, the East African Cooperation Minister, was the most senior woman in the rebellion.

William Ruto's position was distinct and consequential. In 2005, Ruto was not in the Kibaki Cabinet; he was a KANU MP for Eldoret North, having retained his seat in the 2002 election that had otherwise reduced KANU to a parliamentary rump. As KANU's national organising secretary under Uhuru Kenyatta's post-2002 reconstruction of the party, Ruto led the KANU mobilisation against the Wako Draft. The Kalenjin political establishment — formed during the Moi era and still cohesive in 2005 — followed Ruto's lead. The Orange coalition's Kalenjin component was therefore a KANU-rooted bloc rather than a NARC-defection, which mattered for the post-referendum institutional development of ODM.

The Kibaki Cabinet faced a dilemma through September–November 2005. The LDP ministers were campaigning publicly against the government's own constitutional draft, violating Cabinet collective responsibility on what the President had identified as a flagship policy. Constitutionally, the President had unfettered power to dismiss Cabinet members. Politically, dismissal before the referendum risked converting an internal coalition dispute into a public rupture that could mobilise additional No-side support; allowing the dissent risked the appearance of government weakness. The Cabinet operated through October–November 2005 with this contradiction unresolved. Kibaki himself maintained, at public rallies and at the State House, that ministers had the right to campaign according to conscience, while making clear that he expected Cabinet members to support the official position. The contradiction was managed by the President's selective public silences and by an informal understanding that the post-referendum Cabinet review would address the question whatever the referendum result.

Civil-society allies of the No camp added institutional weight to the LDP rebellion. The Law Society of Kenya, under its 2005 chairman Tom Ojienda, formally opposed the Wako Draft on procedural grounds (the unauthorised revision of the Bomas Draft). The NCEC, the National Council of Churches of Kenya (NCCK), the Kenya National Commission on Human Rights (KNCHR), and the Federation of Women Lawyers (FIDA-Kenya) issued analyses critical of the draft. The Catholic Justice and Peace Commission was more measured but in many parishes participated in No-side civic education. The Anglican Church of Kenya, under Archbishop Benjamin Nzimbi, was split, with some bishops openly Yes-aligned and others Orange-aligned. The composition of the No civil-society coalition mattered because it gave the campaign access to institutional networks — bar associations, church platforms, women's groups — that extended beyond the political parties and that complicated any government attempt to characterise the No side as a narrow Odinga-led political vendetta.

7. The Campaign, September–November 2005: Orange versus Banana

The official campaign period ran from the August 22 Gazette publication of the Wako Draft to the November 20 eve-of-poll. The ECK's referendum regulations, gazetted in September 2005, set the symbol-allocation procedure. The Yes side was assigned the banana; the No side the orange. The choice has been variously explained — some accounts attribute it to ECK administrative random allocation; others to negotiation between the campaign committees. The political effect was immediate: the Orange and the Banana became the dominant iconography of the campaign and, in the Orange's case, of the political coalition that emerged from it.

The Banana campaign was personally led by President Kibaki. State House mobilisation, the deployment of provincial administration (DCs, DOs, chiefs) in supportive roles, the use of presidential rallies and State broadcasting through KBC, and the channelling of Cabinet ministers and Mt Kenya MPs into the campaign created a structurally well-resourced campaign with national reach. The Banana platform, articulated by Kibaki in major rallies at Uhuru Park (Nairobi), Nyeri, Embu, and Meru through September–November, emphasised three themes: that the Wako Draft was a genuine reform document that delivered the Bill of Rights and the modernising changes Kenya needed; that the alternative to approval was a return to the unreformed 1969 constitution that had served KANU's authoritarianism; and that political stability — and continued economic growth, which had accelerated to over 5% — depended on settling the constitutional question. Cabinet ministers Karua, Murungi, Murungaru, Michuki, and Mwiraria led the Banana campaign in their home regions; the Mt Kenya political establishment, including the GEMA (Gikuyu, Embu, Meru Association) networks, mobilised through ethnic and business associations.

The Orange campaign was led tactically by Odinga, with intellectual coordination from Nyong'o and field leadership distributed across the LDP, KANU, and FORD-Kenya networks. Major Orange rallies — at Kamukunji grounds (Nairobi) in September, at Kisumu Stadium, at Mombasa's Tononoka grounds, at Eldoret 64 Stadium, at Bukhungu in Kakamega — drew crowds that observers regularly estimated in the tens of thousands, with the Kasarani rally in October 2005 cited by contemporary press accounts as one of the largest political rallies of the Kibaki first term. The Orange platform articulated four grounds for rejection. First, that the Wako Draft had stripped from the Bomas text the executive premier and the strong devolution that the Conference had endorsed. Second, that the Kibaki government had broken the 2002 MoU and could not be trusted to implement even the modest reforms the Wako Draft did contain. Third, that the constitutional process had been hijacked: that a participatory-conference text had been rewritten behind closed doors and presented as a fait accompli. Fourth, that on the substance of devolution and the structure of the presidency, the Wako Draft preserved rather than reformed the central problem of post-independence Kenyan governance — concentrated executive power without effective accountability.

The campaign included several memorable mobilisational moments. Odinga's "the Banana is bitter" speech, delivered at the Kamukunji rally and replayed widely, framed the Yes campaign's symbol as itself a deception. The Orange-coalition Sunday-newspaper insert campaigns through October 2005 — pamphlets distributed with the Sunday Nation and the Standard, written by Nyong'o and others — provided constituency-by-constituency analyses of what the Wako Draft would and would not deliver. The Banana side's counter-mobilisation included the deployment of donor-funded civic-education programmes, some of which civil-society monitors documented as having been politically captured by Yes-aligned messaging. The Kenya Anti-Corruption Commission's report on Anglo Leasing — which had implicated several Banana-side Cabinet ministers (KE-B-01 §3) — was, on the No side's framing, evidence that the Cabinet asking voters to approve a constitution was a Cabinet that could not be trusted with constitutional discretion.

Opinion polling through October–November 2005 consistently showed the Orange side ahead. The Steadman Group's tracking polls, the Strategic Public Relations and Research surveys, and the Synovate poll published in mid-November 2005 all showed No leading Yes by margins between 8 and 15 percentage points, with the gap widening through the final fortnight. The Kibaki camp's internal polling, according to subsequent media accounts, showed similar trends; the President's continued vigorous campaigning into the final week was reportedly an attempt to close a gap that internal numbers had identified as substantial.

8. The 21 November 2005 Vote: Result, Geography, and Demography

Polling on 21 November 2005 was peaceful by Kenyan election standards. The ECK, under chairman Samuel Kivuitu and operating two years before the disputed 2007 count, deployed across approximately 20,000 polling stations [TBD-VERIFY: ECK Report on the National Referendum gives the precise polling-station figure]. International observer missions, including the Commonwealth Secretariat under former Tanzanian Prime Minister Frederick Sumaye, the European Union election observation team, and the African Union mission, rated the process as credible. Domestic observers from the Kenya Domestic Observation Programme (K-DOP), the Catholic Justice and Peace Commission, and the NCCK deployed thousands of monitors. Isolated incidents of disruption were reported — minor delays in some Northern constituencies, scuffles at a few polling stations in Nairobi — but nothing that affected the overall integrity of the count.

The result, declared by the ECK on 22 November 2005, was:

  • Yes (Banana): 2,545,495 votes — 41.9%
  • No (Orange): 3,548,477 votes — 58.1%
  • Turnout: approximately 53% of registered voters [TBD-VERIFY: the ECK report gives a precise figure; the percentages cited here are the most widely reproduced consensus numbers]

The geographic pattern of the vote, as reported by the ECK constituency tally, mapped the emerging post-2005 political coalitions. Central Province — Kibaki's home province and the Mt Kenya political base — returned the strongest Yes majorities, with constituencies in Nyeri, Kirinyaga, Murang'a, and Kiambu producing Yes margins frequently above 80%. The Embu and Meru sub-regions of Eastern Province (the broader GEMA cultural-political area) returned Yes majorities. The rest of the country returned No majorities, often by large margins. Nyanza Province — Odinga's Luo heartland and the Kisii sub-region — returned No majorities frequently above 85%. Western Province — Luhya-dominant, with Musalia Mudavadi a senior No-side figure — returned No majorities, though with significant Yes minorities in some Bukusu sub-regions. Coast Province — with Balala leading the campaign and with long-standing devolutionist sentiment — returned No majorities. North-Eastern Province — predominantly Somali-Kenyan, with strong devolution and Bill-of-Rights expectations — returned strong No majorities. The Rift Valley, the largest province, returned No majorities under Ruto's Kalenjin-led mobilisation, with some constituencies (in the Kalenjin highlands and Maasai areas) producing No margins above 80%.

Demographically, the survey-based analysis subsequently produced by Whitaker and Giersch (Journal of Contemporary African Studies, 2009) using Afrobarometer data showed several consistent patterns. Voters who identified the constitutional draft's executive structure (presidency versus premier) as their primary concern were disproportionately No voters. Voters in areas with strong devolution claims (Coast, North-Eastern, parts of Rift Valley, parts of Western) were disproportionately No. Voters who identified themselves as supporters of Kibaki/NARC in the 2002 election and as primarily concerned about economic continuity were disproportionately Yes. Voters identifying as Christian Pentecostals — the bloc that would later become important in the 2010 referendum's smaller No campaign on abortion and Kadhi-courts grounds — were in 2005 not yet a coherent referendum constituency; the Christian-conservative voting bloc that emerged for 2010 was an artefact of post-2008 Pentecostal political mobilisation, not a feature of the 2005 cleavage.

The Yes campaign conceded the result on the evening of 22 November 2005. President Kibaki, in a televised address on the evening of 23 November, accepted the result and acknowledged that the people of Kenya had spoken. The speech did not contain a concession on the constitutional question — Kibaki did not commit to returning to the Bomas Draft or to a fresh review — but it accepted the procedural legitimacy of the No vote. Within the same address, and in the Cabinet meeting that immediately followed, Kibaki announced the dismissal of his entire Cabinet.

9. The Immediate Aftermath: November 2005 Cabinet Dismissal and the Formation of ODM

The Cabinet dismissal of 23 November 2005 was a single act with multiple downstream consequences. Constitutionally, Kibaki was within his power: the pre-2010 Constitution gave the President unfettered authority to appoint and dismiss Cabinet ministers, with no parliamentary confirmation requirement. Politically, the dismissal converted the No coalition from a single-issue alliance into a permanent opposition bloc and removed the institutional means by which the LDP-NARC contradiction might have been managed through bargaining and compromise.

The reconstituted Cabinet was announced on 7 December 2005, with further additions and resolutions of declined-appointment seats through January 2006. The new Cabinet excluded Odinga, Musyoka, Nyong'o, Balala, Ntimama, Kilimo, and the other LDP rebels. Kibaki retained the Mt Kenya core (Karua, Murungi, Michuki, Mwiraria — though Mwiraria would shortly resign over Anglo Leasing, KE-B-01 §3) and brought in new figures, including several MPs who had crossed from the LDP to the President's side after the referendum. Vice-President Awori was retained. Several appointments were declined: notably, Mwai Kibaki's offer to FORD-Kenya's Musikari Kombo was reportedly accepted, while offers to some other potential bridge-builders were declined. The post-2005 Cabinet was, in political composition, narrower than the 2003 Cabinet had been — a NARC-Kenya/Mt Kenya government rather than a NARC coalition government.

The dismissal was politically interpreted across two competing frames. The pro-Kibaki framing — articulated in subsequent op-eds by Karua and by some of the academic defenders of Kibaki's record — held that Cabinet ministers who had campaigned publicly against the government's flagship constitutional draft had violated collective responsibility and that their removal was the constitutionally and politically appropriate consequence. The pro-Odinga framing — articulated by LDP leaders in subsequent rallies and by Nyong'o in his post-2005 writings — held that the dismissals confirmed the Mt Kenya inner circle's incapacity for genuine coalition governance and the President's preference for ethnic-network rule over the cross-ethnic NARC mandate of 2002. The two readings have continued to coexist in the secondary literature; Branch (2011) and Hornsby (2012) both note the structural inevitability of some such rupture given the post-2003 NARC dysfunction.

The Orange Democratic Movement was constituted as a political brand through 2006 and as a registered political party through late 2006 and 2007. The institutional history is complex: the No-coalition leadership initially sought to operate as a loose movement, but the requirements of the Political Parties Act and the looming 2007 election cycle pushed toward party formation. Through 2006, the ODM leadership — Odinga, Musyoka, Ruto, Ntimama, Balala, Mudavadi (who had been a senior KANU figure under Moi and was now a No-side ally) — coordinated through joint rallies, press platforms, and parliamentary motions. A registered party shell, the Liberal Democratic Party of Kenya, was repurposed; the Orange Democratic Movement-Kenya party emerged through an internal process whose precise legal form was the subject of significant litigation in 2006–2007.

The ODM presidential primary, held in August–September 2007, was the moment at which the No-coalition broke into its post-2007 constituent parts. Odinga, Musyoka, Ruto, Mudavadi, Balala, and Joseph Nyaga contested the nomination. Odinga won the primary; Musyoka rejected the outcome, alleging procedural irregularities, and broke away to form ODM-Kenya (ODM-K) as a separate party with his own presidential candidacy. The Odinga camp's framing of the split — that Musyoka had been induced to splinter by Mt Kenya interests in order to dilute the ODM vote and re-elect Kibaki — and the Musyoka camp's framing — that the primary had been manipulated by the Odinga camp — are both elements of the contested record. The downstream electoral effect was significant: Musyoka took the Kamba vote out of the Odinga column for December 2007, contributing to the close national result that the post-election crisis fought over.

10. The Long Arc: From the 2005 Defeat to the 2007 Election, the 2008 Accord, and the 2010 Constitution

The two years between the November 2005 referendum and the December 2007 election were structured by the absence of further constitutional progress and the consolidation of two opposing electoral coalitions. The Kibaki government, having lost the referendum, made no serious attempt to revive constitutional review. The CKRC was wound down — Ghai's chairmanship ended; the Commission's operational machinery was dispersed; the institutional memory was preserved by individual commissioners and by the published Final Report but not by an active drafting body. Through 2006 and into 2007, the constitutional question receded from formal government priority; the 2010 Constitution's eventual emergence required the post-2008 crisis to re-open the question.

The 2007 campaign reprised the 2005 referendum coalitional architecture in modified form. The Party of National Unity (PNU), launched by Kibaki in mid-2007 as the umbrella for his re-election campaign, brought together the Banana coalition components — the Mt Kenya core, Awori's wing of NARC-Kenya, and various smaller parties. ODM, under Odinga, brought together the Orange coalition minus Musyoka (now ODM-K) — Luo, Luhya (Mudavadi), Kalenjin (Ruto), Coast (Balala), and parts of the urban informal-settlement vote. The election held on 27 December 2007 produced the disputed count documented in KE-B-03 and KE-B-01 §5. The Kriegler Commission's later finding that the announced result was unverifiable did not adjudicate which candidate had genuinely won; what is certain is that the 2005-built coalitions were the structural fault line along which the disputed count became a political and ethnic crisis.

The post-election violence of 30 December 2007 to February 2008 — approximately 1,300 deaths and 660,000 displaced — was the catastrophic consequence (KE-B-03). The Kofi Annan mediation and the 28 February 2008 National Accord (KE-B-04) produced a Grand Coalition Government that included Odinga as Prime Minister, Musyoka as Vice-President, and Cabinet representation across the post-2005 coalitions. Agenda Item 4 of the Accord — long-term issues, with constitutional reform as the central item — committed the parties to delivering the new constitution that the 2005 referendum had blocked. The Constitution of Kenya Review Act 2008 (KE-C-01 §2–3) established the Committee of Experts under Nzamba Kitonga SC; the Committee's method was textual triangulation of the Bomas Draft, the Wako Draft, and the CKRC Draft, producing a Harmonised Draft Constitution in November 2009 and a Revised Harmonised Draft in February 2010. The 6 May 2010 Gazette publication of the Proposed Constitution began the referendum period; the 4 August 2010 vote returned Yes 67% on a 72.6% turnout (KE-C-01 §4–5). The Constitution was promulgated by Kibaki at Uhuru Park on 27 August 2010.

The comparison between the 2005 Wako Draft and the 2010 Constitution is the simplest way to read the long-arc significance of the 2005 defeat. On devolution, the 2005 Wako Draft offered a weak district-level structure under central-government control; the 2010 Constitution delivers 47 counties with directly elected governors and county assemblies, exclusive and concurrent functions schedules, and an equitable-share floor of 15% of national revenue rising to 20%. On the executive, the 2005 Wako Draft preserved an essentially unreformed presidency with a presidentially appointed coordinating Prime Minister; the 2010 Constitution delivers a presidential system but with significant constitutional constraints — Chapter 6 integrity requirements, two-thirds majority requirements for impeachment, a Supreme Court with original jurisdiction over presidential election petitions, and an Independent Electoral and Boundaries Commission with constitutionally specified independence. On the Bill of Rights, the 2005 Wako Draft included most of the Bomas Draft's provisions; the 2010 Constitution includes them in a Chapter 4 that has produced the most extensive rights jurisprudence in Kenyan history. On the Kadhi courts, both drafts retained them; the 2010 Constitution anchors them in Article 170. On accountability commissions, the 2005 Wako Draft had trimmed independence guarantees; the 2010 Constitution elevates the independent commissions to constitutional-organ status under Article 1(3).

The 2010 Constitution is therefore best read, on the comparison, as the Bomas Draft delivered on the second attempt — a participatory-conference text that survived a government revision attempt in 2005 and was restored, in substantially its original form, through the post-2008 process. The 2005 defeat was the moment at which this survival was secured. Had the Wako Draft passed, the resulting constitution would have entrenched a near-imperial presidency with token devolution; the 2010 referendum, with a 67% Yes mandate, would not have been politically available against a recently ratified constitution. The 2005 No vote, in this reading, kept the constitutional question open.

The political coalitions of 2005 also shaped the longer arc. The Orange Democratic Movement, formed from the No coalition, was Kenya's principal opposition party for the decade between 2007 and 2017; Odinga contested every presidential election from 2007 to 2022 as the ODM or ODM-allied candidate. Ruto's 2005 alignment with the No camp — through KANU rather than through ODM — and his subsequent shifting alliances (ODM in 2007, Jubilee with Kenyatta in 2013, UDA against Kenyatta-Odinga in 2022) traced a trajectory that was structurally enabled by the 2005-built Kalenjin political identity (Lynch, I Say to You, 2011). The Mt Kenya/Kalenjin/Luo three-bloc coalitional arithmetic that has structured Kenyan presidential politics from 2007 through the 2022 cycle was built in the 2005 referendum campaign and the immediate post-referendum political reorganisation.

11. Three Accounts of the 2005 Referendum, the Kibaki Cabinet's Gamble, and the Orange–Banana Legacy

Account 1 — The 2005 referendum's significance. The pro-No framing, broadly the dominant Kenyan and international academic reading, holds that the 2005 vote was a legitimate democratic rejection of a flawed government draft that preserved the participatory-conference principle and opened the path to the 2010 Constitution. On this reading, the No coalition's grounds — defence of the Bomas Draft against the Wako revision, defence of devolution, defence of the participatory process — were intellectually serious and politically vindicated; the 67% Yes vote on the 2010 Constitution, which restored most of what 2005 had defended, is the retrospective validation. The pro-Yes framing, defended by some former Cabinet members and by sympathetic biographers (and developed in defensive commentary by Karua and Murungi in subsequent decades), holds that the No coalition was driven by political opportunism — Odinga's frustration with the unfulfilled MoU, Ruto's KANU-rooted opposition, Musyoka's own Cabinet ambitions — more than by constitutional concerns, and that the Wako Draft, while imperfect, was the realistically attainable settlement of the period. The structural framing, developed in Ghai and Cottrell Ghai (2011), Mueller (2008), and the Kanyinga/Okello volume (2010), holds that the 2005 defeat reflected Kenya's broader devolution-versus-centralisation cleavage, dating to the majimbo/centralism debate at independence and the Kenyatta-era abolition of the regional tier, and that no single referendum could resolve it; 2010 only partially settled the cleavage, with implementation of devolution continuing as a contested live issue through the Kenyatta and Ruto presidencies.

Account 2 — The Kibaki Cabinet's gamble. The government framing, articulated by the Kibaki camp at the time and by sympathetic post-2005 commentary, holds that the November 2005 Cabinet dismissal was a legitimate exercise of executive prerogative under a constitution that gave the President unfettered Cabinet-appointment power; ministers who had publicly campaigned against the government's flagship draft had broken Cabinet collective responsibility and could not credibly continue in office. The opposition framing, developed by Odinga, Nyong'o, and the broader LDP-then-ODM camp, holds that the Cabinet dismissal — combined with the Wako Draft's content and the Anglo Leasing record of the implicated Cabinet members — revealed a Wako-Murungi-Karua-led dilution of the participatory-conference settlement and a betrayal of the 2002 NARC mandate; on this reading, the dismissal was the natural conclusion of a coalition-management failure that had been visible since 2003. The party-system framing, developed in Wanyande (2005) and in subsequent comparative work on African coalition politics, holds that NARC's internal contradictions — built on the 2002 MoU but never institutionalised through formal coalition-management procedures or constitutional change — made some such rupture inevitable regardless of the referendum; the 2005 vote and dismissal were the form the rupture took, not the cause of it.

Account 3 — The Orange/Banana campaign and its post-2005 legacy. The pro-ODM framing, dominant in Odinga-camp and broader opposition commentary, holds that the Orange campaign forged a durable opposition coalition that became Kenya's leading opposition party for the next decade and the principal vehicle of democratic accountability through the Kibaki, Kenyatta, and Ruto eras; ODM's role in the 2008 National Accord, in the 2010 Constitution Yes campaign (where ODM and PNU were on the same side), in the 2013 and 2017 election petitions, and in the 2018 Handshake and subsequent realignments, traces a continuous lineage from 2005. The critical framing, advanced by some PEV-era commentators and by scholars (Mueller, 2008; Lynch, 2011) writing about the deeper political-economic preconditions of 2007–08, holds that the symbolic Orange–Banana campaign deepened ethnic-political alignment patterns — Mt Kenya as a defensive Yes bloc, Nyanza/Western/Coast/Rift Valley as a coalition of grievances — that fed directly into the December 2007 mobilisation and the post-election violence; on this reading, the durable ODM coalition was bought at the price of more rigid ethnic alignment in subsequent electoral cycles. The structural framing holds that Kenya's ethnic-political party-formation patterns predate 2005 and reflect the deeper political-economic structure of post-independence Kenyan governance (land allocation, patronage networks, presidential-system coalition arithmetic); the ODM/PNU split of 2005–2007 would have produced similar coalitions through some other political flashpoint regardless of the referendum. All three readings have evidentiary support; the corpus's discipline is to present them in their own voice rather than to adjudicate.

12. Conclusion and Spiral Index

The 21 November 2005 referendum on the Wako Draft was the first occasion in Kenyan history on which a sitting government's flagship constitutional proposal was defeated at the polls, and it was the moment at which the political coalitions that have structured Kenyan presidential politics for the subsequent two decades were assembled and named. The defeat closed the immediate constitutional-reform path and re-opened it only through the post-2008 emergency; it produced the Orange Democratic Movement as a permanent feature of the political landscape; and it produced the Cabinet rupture that converted the Kibaki–Odinga MoU contradiction into open opposition warfare for the December 2007 election. The post-election violence of 2007–08 (KE-B-03), the National Accord of 2008 (KE-B-04), and the Constitution of Kenya 2010 (KE-C-01) are all downstream consequences in different ways. The constitutional architecture that 2010 delivered — devolution to 47 counties, a justiciable Bill of Rights, an independent Judiciary, the Independent Electoral and Boundaries Commission — is in substantial part the Bomas Draft of 2004 delivered on the second attempt, after the 2005 No vote had preserved its viability against the Wako revision.

Spiral Index:

Key Persons:

  • Mwai Kibaki (1931–2022): Third President of Kenya; led the Banana (Yes) campaign in 2005; dismissed Cabinet 23 November 2005 (KE-B-01; KE-H-PRES-03 forthcoming).
  • Raila Odinga (b. 1945): LDP leader within NARC; led the Orange (No) campaign; principal post-2005 opposition figure; ODM leader from 2006; subsequent Prime Minister 2008–2013 (KE-B-01; KE-H-OPP-01 forthcoming).
  • Kalonzo Musyoka (b. 1953): Vice-President of LDP; senior No-side campaigner; broke from ODM in 2007 to form ODM-Kenya; Vice-President 2008–2013 under the Grand Coalition.
  • William Ruto (b. 1966): KANU MP for Eldoret North in 2005; led Kalenjin No-side mobilisation through KANU; subsequent Jubilee Deputy President (2013–2022) and President (2022–) (KE-E-01).
  • Anyang' Nyong'o (b. 1945): LDP Secretary-General; institutional coordinator of the No coalition; subsequent ODM Secretary-General and Kisumu Governor.
  • William Ole Ntimama (1931–2016): Maasai Cabinet veteran; No-side campaigner; symbol of pastoralist devolutionist support.
  • Najib Balala (b. 1967): Coast-based politician; senior No-side campaigner; subsequent Cabinet Secretary for Tourism in later administrations.
  • Yash Pal Ghai (b. 1938): Chair of the CKRC 2001–2005; principal architect of the participatory-conference review process.
  • S. Amos Wako (b. 1945): Attorney General 1991–2011; signatory of the 2005 Proposed Constitution gazette publication; institutional name on the "Wako Draft".
  • Kiraitu Murungi (b. 1952): Justice and Constitutional Affairs Minister 2003–2005; principal Banana-side defender of the government draft.
  • Martha Karua (b. 1957): Constitutional Affairs Minister 2005; senior Banana-side campaigner; subsequent PNU negotiator at the 2008 National Accord.
  • Samuel Kivuitu (1933–2011): ECK Chairman; administered the 2005 referendum without significant procedural challenge (compare 2007).
  • Reverend Timothy Njoya: Civil-society constitutional-reform litigant; petitioner in the Njoya case on constituent power.

Key Events and Dates:

  • August 1997: IPPG reforms negotiated; minimal package secures 1997 election and commits to comprehensive review.
  • 28 September 2000: Safari Park Accord merges parliamentary and Ufungamano review tracks.
  • November 2000: Constitution of Kenya Review (Amendment) Act 2000 enacted; CKRC established.
  • 2001: Yash Pal Ghai appointed Chair of CKRC.
  • 27 September 2002: CKRC Draft Constitution published.
  • 28 April 2003: National Constitutional Conference (Bomas) convenes.
  • August 2003: NCEC walk-out from Bomas Conference.
  • 15 March 2004: Bomas Draft Constitution adopted by Conference.
  • December 2004: Constitutional Review (Amendment) Act 2004 enacted; Parliament given power to revise Bomas Draft.
  • 22 August 2005: Wako Draft (The Proposed New Constitution of Kenya, 2005) gazetted.
  • September–November 2005: Orange (No) versus Banana (Yes) referendum campaign.
  • 21 November 2005: Referendum vote; No 58.1% / Yes 41.9% on ~53% turnout.
  • 22 November 2005: ECK declares result.
  • 23 November 2005: Kibaki dismisses entire Cabinet.
  • 7 December 2005: Reconstituted Cabinet sworn in (excludes LDP rebels).
  • 2006: ODM coalesces as political brand; party-registration process initiated.
  • August–September 2007: ODM presidential primary; Odinga wins; Musyoka splits to form ODM-K.
  • 27 December 2007: General election (disputed) — see KE-B-03.
  • 28 February 2008: National Accord — see KE-B-04.
  • 27 August 2010: Constitution of Kenya 2010 promulgated — see KE-C-01.

Key Instruments and Documents:

  • Constitution of Kenya Review Act 1997 (initial); Constitution of Kenya Review (Amendment) Act 2000 (CKRC-establishing); Constitutional Review (Amendment) Act 2004 (revision authority).
  • CKRC Draft Constitution, 27 September 2002.
  • Bomas Draft Constitution, 15 March 2004.
  • The Proposed New Constitution of Kenya, 2005 (Wako Draft), Kenya Gazette Supplement No. 63, 22 August 2005.
  • Electoral Commission of Kenya, Report on the National Referendum on the Proposed New Constitution, 21 November 2005, Nairobi, 2006.
  • CKRC, Final Report of the Constitution of Kenya Review Commission, Nairobi, 10 February 2005.

Thematic Connections to Other Kenya Corpus Documents:

  • KE-A-02 (Moi Presidency): pre-1997 constitutional context; IPPG reforms negotiated under Moi.
  • KE-B-01 (Kibaki Presidency): direct concurrent narrative; §4 is the antecedent summary; §3 documents the 2003 MoU betrayal that made 2005 inevitable.
  • KE-B-03 (2007–08 PEV): downstream consequence; the 2005-built coalitions were the structural fault line of the post-2007 crisis.
  • KE-B-04 (2008 National Accord): the resolution that re-opened the constitutional-reform path the 2005 defeat had closed.
  • KE-C-01 (2010 Constitution): the eventual successor; the 2010 architecture is, in substantial part, the Bomas Draft of 2004 delivered on the second attempt after the 2005 No vote had preserved its viability.
  • KE-D-01 (Kenyatta Presidency): the post-2013 implementation of the 2010 settlement that 2005 had made possible.
  • KE-E-01 (Ruto Presidency): Ruto's 2005 No-campaign role through KANU is part of the political biography that produced the 2013 Jubilee alliance and the 2022 Ruto victory.

Sources

  1. Andreassen, Bård-Anders and Tostensen, Arne (eds.). Of Oranges and Bananas: The 2005 Kenyan Referendum on the Constitution. CMI Working Paper, Chr. Michelsen Institute, Bergen, 2006 — the principal academic-record compendium on the referendum.
  2. Ghai, Yash Pal and Cottrell Ghai, Jill. Kenya's Constitution: An Instrument for Change. Nairobi: Katiba Institute, 2011 — by the CKRC chair; the authoritative reconstruction of the 2000–2005 review process from inside.
  3. Ghai, Yash Pal. "Devolution: Restructuring the Kenyan State." Journal of Eastern African Studies 2, no. 2 (2008): 211–226.
  4. Branch, Daniel. Kenya: Between Hope and Despair, 1963–2011. New Haven: Yale University Press, 2011, especially Chapter 9 on the Kibaki first term and the 2005 referendum.
  5. Hornsby, Charles. Kenya: A History Since Independence. London: I.B. Tauris, 2012, especially the chapters on the 2002–2007 cycle, the constitutional review, and the post-referendum realignment.
  6. Kanyinga, Karuti and Okello, Duncan (eds.). Tensions and Reversals in Democratic Transitions: The Kenya 2007 General Elections. Nairobi: Society for International Development and Institute for Development Studies, University of Nairobi, 2010.
  7. Mutunga, Willy. Constitution-Making from the Middle: Civil Society and Transition Politics in Kenya, 1992–1997. Nairobi: MWENGO and SAREAT, 1999 — the foundational text on the civil-society constitutional-reform movement in the pre-CKRC period.
  8. Constitution of Kenya Review Commission (CKRC). The Final Report of the Constitution of Kenya Review Commission. Nairobi, 10 February 2005 — the CKRC's own record of its 2001–2005 work, the Bomas Conference, and the contested handover.
  9. Constitution of Kenya Review Commission. The Draft Constitution of Kenya, 2002 (the "CKRC Draft" of 27 September 2002).
  10. National Constitutional Conference (Bomas). The Draft Constitution of Kenya, 2004 (the "Bomas Draft" of 15 March 2004) — the conference output that the Wako Draft revised.
  11. Republic of Kenya. The Proposed New Constitution of Kenya, 2005 (the "Wako Draft"), Kenya Gazette Supplement, 22 August 2005, signed by Attorney General S. Amos Wako.
  12. Electoral Commission of Kenya (ECK). Report on the National Referendum on the Proposed New Constitution, 21 November 2005. Nairobi, 2006 — the official tally and process record.
  13. Cheeseman, Nic. "The Kenyan Elections of 2007: An Introduction." Journal of Eastern African Studies 2, no. 2 (2008): 166–184 — frames the referendum as the proximate political antecedent of the 2007 election.
  14. Cheeseman, Nic; Lynch, Gabrielle; and Willis, Justin. "Democracy and Its Discontents: Understanding Kenya's 2013 Elections." Journal of Eastern African Studies 8, no. 1 (2014): 2–24 — for the longer-arc coalitional inheritance.
  15. Lynch, Gabrielle. I Say to You: Ethnic Politics and the Kalenjin in Kenya. Chicago: University of Chicago Press, 2011 — on the Ruto-led Kalenjin alignment with the No campaign.
  16. Barkan, Joel D. "Kenya After Moi." Foreign Affairs 83, no. 1 (2004): 87–100; and Barkan, Joel D. "Devolution and the New Politics of Development in Kenya," African Studies Review 57, no. 2 (2014): 167–184.
  17. Mueller, Susanne D. "The Political Economy of Kenya's Crisis." Journal of Eastern African Studies 2, no. 2 (2008): 185–210.
  18. Wanyande, Peter. "The Politics of Alliance Building in Kenya: The Search for an Alternative to the Ruling Party in the 2002 Elections." In Maupeu, Hervé; Katumanga, Musambayi; and Mitullah, Winnie (eds.), The Moi Succession: Elections 2002. Nairobi: Transafrica Press, 2005.
  19. Murunga, Godwin R. and Nasong'o, Shadrack W. (eds.). Kenya: The Struggle for Democracy. Dakar and London: CODESRIA and Zed Books, 2007 — includes essays on the constitutional review and the referendum.
  20. Whitaker, Beth Elise and Giersch, Jason. "Voting on a Constitution: Implications for Democracy in Kenya." Journal of Contemporary African Studies 27, no. 1 (2009): 1–20 — survey-data analysis of the referendum vote.
  • KE-A-02: The Moi Presidency (1978–2002) — for the pre-1997 constitutional context and the IPPG reforms.
  • KE-B-01: The NARC Victory and Kibaki Presidency (2002–2013) — the direct concurrent narrative; §4 of KE-B-01 is the antecedent treatment of the 2005 referendum.
  • KE-B-03: The 2007–2008 Post-Election Violence — the successor crisis to which the 2005 coalition rupture directly contributed.
  • KE-B-04: The 2008 National Accord and Coalition Government — the resolution that re-opened the constitutional-reform path the 2005 defeat had closed.
  • KE-C-01: The Constitution of Kenya 2010 — the eventual successor document; the 2005 defeat is the immediate antecedent referendum.
  • KE-D-01: The Uhuru Kenyatta Presidency (2013–2022) — for the long-arc inheritance of the ODM/PNU coalitional split.
  • KE-E-01: The William Ruto Presidency (2022–) — Ruto's 2005 No-campaign role is part of his political biography.
  • KE-D-02: 2017 Election Annulment and Re-Run
  • KE-D-03: Building Bridges Initiative (2018-2022)
  • KE-R-01: Kenya Governance Books Canon
  • KE-A-04: 2010 constitution and the katiba decade 2010 2025
  • KE-A-03: back-reference added by symmetry sweep
  • KE-G-01: back-reference added by symmetry sweep
  • KE-H-PRES-04: Uhuru Muigai Kenyatta — A Biography
  • KE-H-PRES-05: William Samoei Ruto — A Biography
  • KE-K-01: The 2002 Moi Succession Decision and the KANU Collapse
  • KE-F-01: Kenya-China Relations — The SGR, the Debt, and the Look-East Decades
ArchiveSourcesChat