The Kenyan Supreme Court’s BBI Judgment – III: On Referendum Questions, Other Implications and Untidy Endnotes

This is the third and concluding blog post about the Kenyan Supreme Court’s BBI judgment. Parts One and Part Two can be accessed here. In this post, I will end by examining some of the possible implications of the judgment of the Court, going forward.

Distinct and Separate Referendum Questions

Recall that one of the grounds on which the High Court had invalidated the BBI Bill was that all seventy-four amendments had been lumped together as a “package”. The High Court had held that under Article 257, potential amendments would have to be placed before the People as distinct and separate referendum questions. The Court of Appeal was split on the point, but arguably, a majority of the bench held that at the very least, a “unity of theme” approach would have to be followed: that is, potential amendments that were thematically unrelated could not be lumped together in a package. The one exception was Tuiyott J, who held that the issue was not yet ripe for adjudication, as the IEBC was yet to frame the referendum question – or questions.

A majority of the Supreme Court agreed with Tuiyott J on this point. Thus, while the judgments of the High Court and Court of Appeal were set aside, the question still remains open for adjudication.

In my submission, however, while the Supreme Court did not explicitly decide the question, the overarching logic of its judgment(s) strongly implies that when the question does become ripe at some point in the future, the unity of content approach is to be followed.

The reason for this brings us back to our discussion in the previous post: going forward, any interpretation of Article 257 of the Kenyan Constitution must be informed by the Supreme Court’s finding that the purpose of the tiered amendment process is to provide internal constitutional safeguards against abusive amendments, and – specifically – against the culture of hyper-amendment. Indeed, it is particularly interesting to note that for more than one judge, the fact that no constitutional amendment had been successfully pushed through in the twelve years of the existence of the 2010 Constitution was evidence that the internal safeguards were working.

But now consider the consequences had the High Court’s judgment in May 2021 not stopped the (somewhat advanced) Article 257 in its tracks. Had the process been completed successfully, in one fell swoop, the Kenyan Constitution would have gone from having never been amended in twelve years, to having been amended seventy-four times in twelve years – and if anything can be called a “culture of hyper-amendment”, seventy-four amendments in twelve years would surely fit the bill!

It is therefore not enough to say that the tiered amendment process provides an adequate internal safeguard against hyper-amendments. The tiered amendment process – as set out under Articles 255 – 257 – still leaves a range of interpretive questions open; and precisely how effective it is against hyper-amendments depends on how the courts answer those questions. It is easy to see that lumping all potential amendments into one referendum question is an enabler of hyper-amendments: as Musinga (P) rightly pointed out in the Court of Appeal, this enabled a culture where, in order to push through a potentially unpopular amendment, its proponents will include a range of “sweeteners” to make the Bill as a whole palatable – or, alternatively, raise the cost of not voting for it. One can see a direct link between this kind of constitutional jockeying and the culture of hyper-amendment. It is therefore my submission that the constitutional silence in Article 257 on the question of distinct and separate referendum questions ought to be resolved in favour of the unity of content approach, as that is the interpretation that would further the purposes of Article 257 in checking hyper-amendments. Indeed, this interpretive approach matches precisely the Supreme Court’s approach to the popular initiative question. Article 257 was silent on whether the President could or could not initiate a PI. The Supreme Court engaged in a purpose interpretation of Article 257 to hold that he could not, because the contrary interpretation would defeat the objective of the PI. The same considerations apply to the issue of distinct and separate referendum questions.

Constitutional Gaps

On at least two crucial issues, the Supreme Court’s judgment was informed by a gap in the Constitution that was meant to be filled in by statute, but hadn’t yet been. The first was the issue of public participation. The second was the issue of the initiation of a popular initiative.

The first issue had also been discussed by the judges in the superior courts below: in the absence of a statute setting out the scope and content of public participation under the Article 257 process, the Courts were forced to stumble around a bit and search for the light, although the judges did eventually – relying upon the constitutional standard of public participation – return findings either way on the subject. Assuming, however, that at some point a law is passed that sets out its details, it will be interesting to see how the courts scrutinise its adequacy; any such scrutiny will now need to be judged against the standard of whether or not the statute can serve as a strong enough bulwark against abusive amendments and hyper-amendments; thus, issues such as time to scrutinise bills, language, accessibility, and so on, will need to be considered from this rubric.

The second issue finds mention in Mwilu DCJ’s judgment, although its echoes are present from the High Court, to the Court of Appeal, and to the Supreme Court. This is the issue of the popular initiative: eighteen out of nineteen judges who heard this case agreed that the President cannot initiate a popular initiative under Article 257. The devil, however, is in the detail: in the present case, the President’s involvement – through proxies – was too overt and too categorical for most of the judges to ignore. One can easily imagine, however, that stung by this reversal in all the Courts, a future President might just decide to be a lot more subtle about this, and put in substantially greater distance between themselves and their proxies. At the Court of Appeal, Tuiyott J, and at the Supreme Court, Koome CJ, both exhibited a keen awareness of this problem, but at the end of the day, beyond applying good judicial common sense, there is only so much that Courts can directly do to prevent executive “hardball”. This is why Mwilu DCJ probably had it right when she listed out a range of issues – such as, for example, whether promoters could be members of political parties, or political parties themselves – that might arise in the future; and the fine-grained character of these issues indicates that they are better off addressed by the legislative scalpel rather than the judicial sledgehammer. Of course, the risk here is, given that Article 257 is meant to be a constitutional amendment route that serves as an alternative to Parliament, Parliament itself legislating on the scope of who can activate Article 257 will raise potential conflicts of interest. That is perhaps inevitable, and once again, it might just be the case that the issue will ultimately find its way back to the judiciary, and that the courts will need to consider at what point the indirect involvement of State actors reaches a threshold where it starts to threaten the fundamental purpose of Article 257.

Indeed, there is good reason to think that the BBI litigation marks the beginning and not the end of the story. Coming away from the judgment, we find that there is a window open for judicial intervention to stop constitutionally destructive “amendments” (although it is no longer being called “the basic structure doctrine”), but the length, breadth, and design of this window is also … open (pardon the pun). We also find that it has now been firmly established that the purpose of Chapter XVI – and, specifically Article 257 – is to constrain the imperial Presidency, check abusive amendments, and safeguard against hyper-amendments. But as history shows, the imperial Presidency is not so easy to contain: its “taming” will need more than one set of judgments, but rather, it is a constitutional commitment that will need to be renewed and renewed again. Stopping subtle and indirect hijackings of Article 257, package deal referendums, and inadequate public participation (to name just a few threats) will all be part of that renewal.

Conclusion: Shadow and Light

It remains to end with a disclaimer (or two). As one of the amici before the Supreme Court of Kenya in the present appeal, my analysis is naturally situated within that broader context, and the arguments I have made in these three blog posts reflect some of the arguments in my amicus brief (I am particularly grateful to the Court for having admitted the brief, and then – across multiple judgments – engaged with the arguments closely and in depth). Indeed, these arguments reflect a broader set of intellectual commitments I bring to interpreting Constitutions: I believe that Constitutions are fundamentally about power relations, about deciding who has power and who doesn’t, who gets to wield power and upon whom it is wielded, and how power (State power, in particular) is to be confronted, mitigated, and contained. Our task as interpreters is to try and ensure that Constitutions live up to their own goal (often stated in the Preamble) of democratising power, and of checking abuse and impunity.

Having had the opportunity to engage so deeply with these questions in the context of the Kenyan Constitution over the last one year has been a privilege. As an outsider who has tried to approach the subject with respect and humility, but who – no doubt – has often put his foot in it, it has been particularly wonderful to experience the openness and generosity with which the Kenyan interpretive community has treated me; for that, I am deeply grateful. After all, as Yvonne Owuor once wrote, there is a “cartography not of possession, but of – how odd – belonging.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s