—Richard Albert, Professor of World Constitutions and Director of Constitutional Studies, The University of Texas at Austin
The Court of Appeal of Kenya recently concluded four days of oral argument on the constitutionality of the Building Bridges Initiative Constitutional Amendment Bill (BBI), a proposed mega-constitutional amendment bill that would quite substantially reform the Constitution of Kenya.
This post is the first in a short series of reflections on the BBI as we approach the date of the Court’s judgment, expected on August 20.
I begin with a question that is central to constitutional amendment in Kenya and to the future of the BBI, but that has not received nearly as much attention as the constitutionality of the BBI itself:
Does Chapter 16 of the Constitution of Kenya require putting the BBI to voters in a national referendum as (1) an omnibus Constitutional Amendment Bill or (2) as a series of separate and individual amendment proposals?
The BBI is a multi-subject Constitutional Amendment Bill. It combines multiple proposals to amend different and unrelated parts of the Constitution. And it combines all of those amendment proposals into one single enormous Bill.
The BBI is 45 pages. It contains 74 clauses. And it includes two Schedules as appendices. The Bill proposes to amend Chapters Two, Three, Four, Seven, Eight, Nine, Ten, Eleven, Twelve, Thirteen, Fourteen, Fifteen and Sixteen of the Kenyan Constitution, as well as the Third Schedule to the Constitution. The dozens of amendment proposals in the BBI touch virtually every part of the Constitution.
If the BBI is upheld, it will have to be ratified in a national referendum before it takes effect, according to Chapter 16 of the Constitution.
But the text of the Constitution is unclear on how to put the referendum question to voters.
The conventional wisdom suggests only two options for voters in a BBI referendum: (1) voters must either approve or reject the entire package of amendments presented to them in the BBI; or (2) voters must separately approve or reject every single individual amendment proposal in the BBI.
This conventional wisdom misses a third option for a multi-subject constitutional referendum like the BBI.
The third option is for the Court to review the multi-subject Constitutional Amendment Bill after it has been proposed but before it is sent to the voters in a referendum. The purpose of the Court’s review is to group together those amendment proposals that are “related” to each other.
This third option spares voters the potential Hobson’s choice of having to approve or reject the entirety of the amendment proposals in an omnibus Constitutional Amendment Bill. And it also spares voters from the impracticality of having to cast a vote on potentially dozens of separate and individual amendment proposals.
If the Court determines in its review that all amendment proposals in a multi-subject Constitutional Amendment Bill are sufficiently “related” to each other so as to touch upon the same constitutional “subject-matter,” then the multi-subject Constitutional Amendment Bill may be put to a single referendum question in its entirety, as one mega-package, in order for voters to either approve or reject the multi-subject Constitutional Amendment Bill all at once.
However, if the Court determines that the multi-subject Constitutional Amendment Bill consists of several different “subject-matter” groups of “related” amendment proposals, the Court may instruct the voting management body to put these different sets of “related” amendments to voters as separate self-standing referendum questions. These separate but smaller amendment packages would be put to voters as distinct questions for their approval or rejection in the referendum–instead of requiring voters to approve or reject the entirety of the multi-subject Constitutional Amendment Bill and instead also of requiring voters to approve or reject each and every single item in the multi-subject amendment package.
For example, imagine an multi-subject Constitutional Amendment Bill contains 100 amendment proposals. Imagine further that 25 of the amendment proposals concern the powers of the executive branch, 25 concern the powers of the legislative branch, 25 concern the powers of the judicial branch, and 25 concern matters of local government. Rather than requiring voters to approve or reject each of the 100 amendment proposals separately and individually, and rather than requiring voters to approve or reject the entirety of the omnibus Constitutional Amendment Bill, the Court could apply “the rule of relatedness” to require voters to approve or reject each of the four groups of amendment proposals, since each of the groups of amendment proposals would have been determined by the Court to be sufficiently “related” according to its own “subject matter.”
We may call this interpretive rule the “rule of subject-matter relatedness.” This “rule of subject-matter relatedness” requires the Court to strike an appropriate balance between specificity and generality. The Court might wish to avoid requiring voters to approve or reject too many separate amendment proposals so as to recreate the impracticality of voting of an endless list of amendment proposals. Likewise, the Court might wish to avoid requiring voters to approve or reject too few separate amendment proposals so as to recreate the risks posed by an omnibus Constitutional Amendment Bill.
Applying the “rule of subject-matter relatedness” may attract criticisms that the Court is interfering with the constitutional amendment process. But the Court’s intervention here may be warranted by the reality that there are significant problems with requiring voters either (1) to approve or reject the entirety of an multi-subject Constitutional Amendment Bill and or alternatively (2) to approve or reject each separate and individual amendment proposal, especially where the number of separate and individual amendment proposals is quite large and could span multiple dozens of items or pages.
The overriding purpose of this “rule of subject-matter relatedness” is to make the civic duty of voting in a constitutional referendum more coherent, more comprehensible, and more accessible to eligible voters.
The Court may consider applying this “rule of subject-matter relatedness” when it is asked for advice on how a multi-subject constitutional referendum should be put to voters.
Suggested Citation: Richard Albert, The BBI at the Kenyan Court of Appeal | Part I: The Role of the Court in Referendums, Int’l J. Const. L. Blog, July 10, 2021, at http://www.iconnectblog.com/2021/07/bbi-kenya-appeal-hearing-role-of-court-in-referendums.