Designing a constitution-drafting process: lessons from Kenya.

 
NOTE CONTENTS 
 
INTRODUCTION 
 
I.  KENYA'S CONSTITUTION-WRITING PROCESS 
    A. The Movement for a New Constitution 
    B. The Review Act 
    C. The Constitutional Review Process 
       1. The 2002 Elections and the Review Process 
       2. The Constitutional Conference at Bomas 
       3. Revisions to the Draft 
       4. The Referendum 
II. LESSONS LEARNED 
    A. Lesson 1: The Benefits and Costs of a Participatory Review 
       Process 
       1. The Participation Myt 
       2. The Benefits of Broad Participation in Kenya 
       3. The Costs of Broad Participation in Kenya 
    B. Lesson 2: The Risk of Capture 
    C. Lesson 3: The Need for a Veil of Ignorance 
    D. Lesson 4: Mitigating Ethnic Tensions 
    E. Lesson 5: The Need for a Process That Produces a Coherent 
       Design 
    F. Lesson 6: The Dynamism of Political Environments 
III. DESIGNING CONSTITUTIONAL REVIEW PROCEDURES 
    A. A Limited Number of Drafters with Broad Consultation Duties 
    B. Including Nonnationals as Delegates and Excluding National 
       Politicians 
    C. Adopting the New Constitution After the General Election 
    D. Checks and Balances in the Choice of Delegates 
    E. The Value of a Referendum 
CONCLUSION 

INTRODUCTION

Democratization and constitution-writing are frequent bedfellows, and the rise of democratic reform movements throughout the world has often been accompanied by pressure for constitutional reform. (1) More than half of the national constitutions in existence today were written or rewritten in the last thirty years. (2) This flurry of constitution-writing has generated a vast academic literature on the content of new constitutions. This scholarship recognizes that institutional choices such as separation of powers, the structure of the executive branch, and the centralization or devolution of power have implications for ethnic polarization and the risk of conflict, as well as for democracy promotion and economic development. (3)

But constitution-writing processes matter too. Procedural choices can affect the legitimacy of the final document as well as its content. As Bruce Ackerman has noted, "A workable constitution is worthless unless [the framers] can get it accepted...." (4) Scholars have recognized this principle and have made empirical, (5) historical, (6) and philosophical (7) arguments for how best to design the process of constitution-writing. (8) Most accounts place a premium on public participation and deliberation. (9) Strong versions argue that process is essentially dispositive of the success of the final product, (10) while moderate versions suggest that good process can at least mitigate ethnic tensions and the risk of violence by providing democratic legitimacy and by ensuring the "buy-in" of diverse communities. (11) Other commentators have taken a more pessimistic view, arguing that most constitution-drafting processes are doomed to failure because the compromises that drafters will be forced to make will undermine the final product. (12)

Despite this extensive theoretical literature, there is a dearth of "carefully targeted case studies" (13) on constitution-drafting processes, as well as a surprising lack of scholarly analysis of why particular constitution-writing processes succeed or fail. (14) As political scientist Donald Horowitz has observed, "[T]he spillage rates [between drafting and adoption] are great, but our knowledge of them is thus far so primitive that we can only regard spillage as being close to random in its incidence and configuration." (15) Most case studies of constitution-drafting experiences are purely descriptive, (16) and analyses that do use case studies to glean lessons about process typically focus on only a few examples: the experiences of North American and Western European nations (particularly the United States), (17) of societies emerging from conflict, (18) and of nations with constitutions imposed by outsiders, such as Japan (or, more recently, Iraq). (19) But many emerging democracies do not fit into any of these categories. Rather, many countries undergo popular democratic reform movements that involve lobbying existing governments for constitutional change. (20) These countries face unique challenges. Most seriously, a successful drafting process must avoid capture by the current government yet remain politically palatable to those same leaders. Furthermore, the fact that constitution-drafting is part of a broader democratic reform movement may itself affect what kind of procedural choices are necessary for public legitimacy. Analyses of countries with constitutions imposed by outsiders, or of countries that rebuilt institutions devastated by war or nullified by revolution, can only provide limited guidance. (21)

This Note attempts to fill that gap by examining Kenya's recent constitution-writing experience as a case study for designing constitution-drafting processes in emerging democracies. Kenya is a particularly valuable case study because it represents the relatively unexamined experience of an emerging democracy undergoing a popular constitutional reform movement. Furthermore, no other scholarly work has provided a detailed description of Kenya's recent constitution-drafting experience. (22)

Kenya's constitutional review process grew from a broader democratic reform movement and was designed to be "people-driven," (23) with broad consultation across the country and a representative constitution-drafting conference. Kenya was even cited as a model for the participatory approach to constitution-drafting. (24) Yet in November 2005, eight years after Kenya's drafting process officially began, Kenyans roundly defeated the proposed constitution in a national referendum: the constitution lost in seven of Kenya's eight provinces in an up-or-down vote, with 57% of voters choosing "No" overall. (25)

This Note details the history of Kenya's constitutional reform movement, discusses key lessons from Kenya's experience, and uses Kenya as a lens through which to consider how designers of other constitution-drafting processes might take these lessons into account. Among other things, it argues that Kenya's experience offers insight into the benefits and costs of highly participatory constitution-writing processes and that it illustrates the risk of capture by elites. This Note also makes concrete recommendations about how Kenya might have better structured its constitutional review process, including proposals about the size and composition of a drafting delegation, the role of existing governmental bodies, and the structure of a referendum.

I. KENYA'S CONSTITUTION-WRITING PROCESS

A. The Movement for a New Constitution

For many Kenyans, Kenya's current constitution is a symbol of both British colonialism and domestic political oppression. (26) Negotiated in London, the constitution dates to Kenya's independence from Great Britain in 1963. It is also a product of domestic political influence; Kenya's ruling party amended the constitution over thirty times, (27) for purposes that included centralizing power, strengthening executive authority, and, for a significant portion of Kenya's history, banning opposition parties. (28)

While Kenya has been at peace since achieving independence, it has been a repressive one-party state throughout most of its history. (29) Kenya's first president was Jomo Kenyatta, a hero from Kenya's liberation struggle, who ruled from 1963 until his death in 1978 and created a de facto one-party state in 1969. (30) Daniel arap Moi succeeded Kenyatta and introduced de jure one-party rule in 1982. (31) Both Kenyatta and Moi silenced opposition, sometimes through the use of torture and intimidation. (32) They also used their political power for patronage, often in support of their ethnic groups and home regions, fostering resentment and exacerbating ethnic and regional tensions. (33)

Agitation for constitutional reform in Kenya began in 1990-1991 and was accompanied by calls for multiparty elections, presidential term limits, and expanded political freedom under the highly repressive Moi regime. (34) The primary impetus for reform came from elites in Kenya's civil society, including religious and human rights groups, which mobilized opposition political parties and their supporters and which helped create a popular movement. (35) In 1991 Moi acceded to international and domestic pressure (36) and permitted a constitutional amendment reforming the presidential election process (37) and reinstating a multiparty political structure. (38) These reforms, however, failed to bring opposition leaders the gains they had anticipated, in part due to continued structural disadvantages in the political system. (39) By 1994, democratic agitation in Kenya had become linked with the call for a new constitution (40): "Constitution-making became the sole vehicle through which democratization, promotion and protection of human rights and social justice were robustly agitated." (41) This agitation led to additional moderate constitutional reforms in 1997 and to the enactment of the Constitution of Kenya Review Act ("Review Act"), (42) which was amended in 2001 to provide for a comprehensive review of the constitution and the option to draft a new document.

B. The Review Act

The Review Act outlined a three-step constitutional review process for Kenya: (1) public consultation and initial drafting by a small review commission, (2) revisions to the draft by a national convention, and (3) ratification by Parliament. Strikingly, the Act seemed consistent with many of the preconditions that scholars have argued are necessary for successful constitution-writing: it included several measures to ensure that the document was "home-grown" (43) and would create "a sense of ownership," (44) and it included checks to ensure that "the government [w]ould neither control nor unduly influence" the process. (45) In particular, the Act emphasized consultation with ordinary Kenyans and extensive deliberation among drafters, and it attempted to sidestep interference in the process by the President or by Parliament.

First, the Act created a Review Commission (46) that was empowered to "collect and collate the views of the people of Kenya" on proposals to amend or rewrite the constitution, and to draft a bill to alter the constitution for presentation to Parliament. (47) The Act required the Commission to visit every constituency in Kenya to collect citizens' views and to disseminate the draft constitution widely among the public. (48)

Next, the Act required the Commission to convene a National Constitutional Conference for "discussion, debate, amendment and adoption" of the Commission report and draft constitution. (49) This National Conference consisted of 629 delegates, (50) including the commissioners as nonvoting members, every MP, and representatives from each district and political party in Kenya, as well as from religious, professional, and other civil organizations, (51) The Act mandated that the National Conference agree to the draft constitution by "consensus" and required a two-thirds majority for amendments, (52) Contentious amendments, which were neither supported by two-thirds of delegates nor opposed by a third of delegates or more, could be submitted to a national referendum, (53) Finally, under the Act, the Commission had to submit the revised draft to Parliament, which could accept or reject the proposed constitution, without amendment, in an up-or-down vote. (54)

Thus, the Review Act emphasized broad public participation at every stage of the drafting process and deliberately mitigated the influence of Parliament and the President in constitution-writing. The President could not participate in the Review Commission or in the National Conference, and MPs composed less than a majority of the Conference and could only approve the draft constitution in an up-or-down vote, without subsequent amendment. But while Kenya's process was designed to avoid partisan capture and to reflect the will of the populace, numerous ex post revisions undermined those ambitions.

C. The Constitutional Review Process

1. The 2002 Elections and the Review Process

Kenya's review process began in late 2001, (55) and the Review Commission completed the Act's first stage of information-gathering, public education, and initial drafting by mid-2002. The Commission planned to start the second stage of the process, the National Conference, in October 2002, and to have the entire process completed before December 2002 so that a new constitution could be in place before Kenya's presidential and parliamentary elections, (56) However, the process's interaction with Kenya's December 2002 elections fundamentally altered its terms.

President Moi dealt the first blow to the review process by dissolving Parliament in October 2002 and ending the terms of Kenya's MPs (a power he had as President under the existing constitution). This eliminated any possibility that the review process would be completed before the December elections; by truncating the terms of the MPs, Moi moved them outside the scope of the Review Act, so that no MPs would be represented in the Conference. Holding a Conference without the MPs was arguably inconsistent with the Review Act and was clearly politically untenable. The Chairman of the Review Commission was therefore forced to postpone the Conference until after the elections, (57) Moi, a fierce opponent of constitutional reform, had ensured that the 2002 elections would take place under the old constitution. (58)

The second blow to the review process came from the National Rainbow Coalition (NARC), the opposition coalition to the ruling Kenya African National Union (KANU) (59) during the 2002 elections. In previous multiparty elections, opposition to KANU had fragmented. (60) NARC thus chose to solidify its alliance by drafting a "Memorandum of Understanding" that promised key leaders new positions created by the as-yet-unratified draft constitution. In particular, opposition leader Raila Odinga, whose support was particularly critical for NARC's electoral survival, (61) agreed to join NARC and to throw his weight behind coalition leader Mwai Kibaki rather than to vie independently for the presidency, in exchange for a promise that he would become Prime Minister when the new constitution creating this position was ratified. (62) NARC announced this agreement publicly, and it was widely discussed in the media. (63) At the same time, Kibaki promised to have a new constitution in place within a hundred days of assuming power. (64)

The Memorandum of Understanding succeeded in keeping NARC unified before the elections and thereby facilitated Kenya's peaceful democratic transition from KANU rule. Bucking expectations, NARC won control of Parliament from KANU in a free and fair election, and Kibaki defeated Moi's chosen successor to win the presidency. (65)

Despite its role in Kenya's political transition, however, the Memorandum of Understanding also helped make the subsequent Constitutional Conference highly divisive and acrimonious, prompting ex post alterations to the constitutional review process by Parliament and leading to a number of competing drafts. The Memorandum of Understanding had allowed Kibaki to ride the coattails of the idea of the new constitution without facing divisive debates about its content. When constitutional negotiations recommenced after the election, however, such debates quickly emerged.

2. The Constitutional Conference at Bomas

Almost immediately after his election in late December 2002, Kibaki pushed back his promise of a new constitution to June 2003. The second stage of the Review Act, the Constitutional Conference, began in late April 2003. (66) The Conference took place at the Bomas of Kenya theater facility and was widely referred to simply as "Bomas." Talks quickly became acrimonious, and the Conference ultimately required three rounds of negotiations ending in March 2004. (67) The Bomas process did produce a draft constitution, but it also led to lawsuits, the withdrawal of the Kibaki government from negotiations, parliamentary bills to alter both the Act and the existing constitution, and alternative drafts. Despite almost a year of negotiations, the Bomas draft was never enacted by Parliament or presented to the public for a referendum.

Three major issues dominated Bomas: (1) the structure of the executive (whether there should be a Prime Minister in addition to the President, and if so, what powers the position should enjoy); (2) devolution (whether Kenya should have a federal system with significant lawmaking powers at the local level); and (3) Khadis courts (whether Kenya should codify separate civil courts for Muslims). However, the issue of executive power was by far the most publicized issue and the most divisive among delegates. (68)

During the Bomas negotiations, Odinga led a coalition that called for a powerful executive Prime Minister, while President Kibaki's supporters strongly opposed an executive Prime Minister, arguing that executive power should be concentrated and that checks and balances through other branches could be used to balance the President's power. (69) The Bomas debate made explicit reference to the coalition's preelection agreement and identified Odinga as the would-be Prime Minister. (70)

While the first two rounds of Bomas were tense, the process completely broke down in the third and final round of negotiations. Key NGO representatives pulled out of the Conference in January 2004. (71) Meanwhile, negotiations at Bomas continued, and a consensus team drafted a set of proposals that included a mixed executive system with a strong President. However, when brought to a full vote, Bomas delegates rejected the consensus document and voted for a constitution with a strong Prime Minister (as well as devolution of political powers). (72) President Kibaki's allies proceeded to walk out of Bomas, (73) and the following day the Kibaki government announced its withdrawal altogether from the National Conference. (74)

The remaining members of the Conference continued their work without the government or its representatives and subsequently passed a draft constitution--which included a strong Prime Minister--to be submitted to Parliament under the Review Act. (75)

3. Revisions to the Draft

With the Bomas Conference completed, parliamentary proponents and opponents of the draft constitution entered bitter negotiations over how to move forward. Complicating matters, just as Bomas was completing its work, Kenya's High Court issued a ruling that any new constitution ultimately needed to be ratified through a national referendum. (76) Parliament amended the Review Act in response to the High Court's ruling, adding a provision for a referendum subsequent to Parliament's ratification of the draft constitution. (77) However, Kibaki's supporters further demanded that Parliament amend the Act to enable itself to alter the Bomas draft before presenting it to the public, instead of having to accept or reject the draft in an up-or-down vote. Kibaki argued that given the disputes over the Bomas draft, "consensus" among MPs--particularly on the issue of executive power--was the only way forward. (78)

Critics charged that the proposed constitutional consensus-building was nothing more than dealmaking among MPs, (79) and prominent leaders such as Odinga refused to participate, arguing that any parliamentary alteration to the draft constitution would be illegitimate. (80) The "consensus group" nonetheless agreed to a proposal to give Parliament power to amend the Bomas draft before voting on it. (81) This revision to the Review Act passed Parliament despite strong resistance from Odinga and opposition parties, including the former ruling party KANU. (82) Disputes over the draft constitution also spilled over into other aspects of governing; for example, dissenting MPs left Parliament to prevent a quorum during important votes in order to protest the government's handling of the review. (83)

With new power to make amendments, MPs participated in a series of retreats to rework the Bomas draft in light of the ruling government's concerns, particularly over the scope of executive power. (84) Odinga's followers and the opposition party boycotted these retreats, accusing the government of "mutilating" the Bomas draft. (85) However, the retreats went on, and the drafters finalized a revised constitution in July 2005, (86) promising a referendum for November. (87) The new draft mandated, among other changes, a Prime Minister who was appointed by and reported to the …