The ongoing protests in Kenya against the 2024 Finance Bill have triggered important litigation around the scope and limits of State power, especially in the context of popular mobilisation. One of the most crucial of these cases is Law Society of Kenya vs The Attorney-General and Ors, which challenges the deployment of the Kenyan Defence Forces [the KDF] alongside civilian authorities, for the purposes of policing the protests. In essence, the Law Society of Kenya argues that under the Constitution, the army can be deployed to aid civilian authorities only with the prior permission of the National Assembly. Admittedly, in this case, there was no prior permission, although the National Assembly did approve the deployment post facto.
The case was filed on 26th June 2024, and the Law Society sought urgent conservatory (i.e., interim) orders from the Court to suspend the Gazette Notice that authorised the deployment of the KDF. Justice Mugambi at the High Court heard arguments on interim relief the same day, and issued a ruling on the 27th of June. In the ruling, Mugambi J upheld the deployment of the KDF, subject to certain terms and conditions. Following this judgment, on 29th June 2024, a Gazette Notice was issued re-authorising the deployment, and purportedly in compliance with the terms and conditions of the Court’s order.
In this blog post I will, first, examine the reasoning underlying Mugambi J.’s order; and, secondly, analyse the framing of the order itself.
Interpreting “Emergencies”
Article 241(3)(b) of the Constitution requires the defence forces to “assist and cooperate with other authorities in situations of emergency or disaster, and report to the National Assembly whenever deployed in such circumstances.” Article 241(3)(c) stipulates that the defence forces “may be deployed to restore peace in any part of Kenya affected by unrest or instability only with the approval of the National Assembly.”
The scheme of Article 241, therefore, indicates that in a case of “unrest or instability”, the defence forces may be deployed only with the approval of the National Assembly, while in a case of “emergency or disaster”, no such prior approval is needed – there is only a reporting requirement.
The key point of contention between the parties was whether the word “emergency” in Article 241(3)(b) referred only to formally declared emergencies under the Constitution (in which case, 241(3)(b) would not apply here, as there was no declared Emergency), or whether it was to be given a “dictionary” meaning (in which case, it could be argued that 241(3)(b) applied). The Court held in favour of the latter interpretation, on the basis that giving “emergency” its technical meaning would render the second part of the provision – “or disaster” – superfluous:
The fact that the framers of the Constitution used a different word alongside the word emergency to describe the situation that may call for military deployment shows that they did not intend to give that word any special meaning such as is used elsewhere in the Constitution where certain specific procedures are associated with it. (para 42)
Giving the word “emergency” its ordinary meaning, the Court ruled that the intervention of the defence forces was justified in the circumstances (para 52).
I suggest that for three reasons – textual, structural, and philosophical – Mugambi J.’s interpretation of the Constitution requires reconsideration.
Textually, there is the cardinal principle that the same word within the same legal instrument should not be given different meanings, unless there are overwhelmingly persuasive reasons for doing so. Part 4 of the Kenyan Constitution specifically deals with a “state of emergency”, defines it narrowly, and imposes onerous procedural safeguards both for triggering an emergency, and with respect to conduct during an emergency (see Article 58). The Court’s holding that “emergency” means one thing in Part 4 of the Constitution, but another thing elsewhere defeats the purpose of Part 4 categorically defining “emergency” in the first place, and confining it within strict contours.
Structurally, let us look at the text of Article 58(1). It authorises the declaration of a “state of emergency” where “the State is threatened by war, invasion, general insurrection, disorder, natural disaster or other public emergency.” In other words, therefore, where the Constitution intends to depart from using the word “emergency” as a term of art, it deliberately uses the term “public emergency,” as supposed to simply “emergency.” Indeed, that “the emergency” is a shorthand for a “state of emergency” is clear from Article 58(6), which restricts the limitations upon rights during a state of emergency to those “strictly required by the emergency.”
The Constitution mentions emergencies at one other place: Article 192(1)(a), which authorises the President to suspend a country government in case of an “emergency arising out of internal conflict or war.” Note, first, that 192(1)(a) itself definitionally qualifies “emergency” with the terms “internal conflict” and “war.” Furthermore, it is evident that Article 192 is a sui generis article which carries within it its own set of inbuilt safeguards which mirror those of Article 58 (such as Senate authorisation, a defined time period, and so on). While the entire scheme of Article 192 may therefore justify giving its terms an autonomous meaning, the same is evidently not true for Article 241(3)(b).
Philosophically, one of the most crucial tasks of a Constitution – especially a transformative Constitution that is, in substantial part, built upon the repudiation of a police State – is to clearly distinguish between “states of exception” and “states of normalcy”, and ensure that the norms that apply to the former do not bleed into the latter. It does this by carefully and precisely defining the state of exception, and stipulating when it can apply, how it can apply, and for how long. The deployment of armed forces within a country’s borders – especially for policing purposes – evidently fall within the “state of exception.” If, then, by according a “non-technical” meaning to “emergency” in Article 241, a state of exception is allowed to exist without compelling the State to go through the formal constitutional mechanisms for declaring a state of exception, this essentially turns the constitutional order in upon itself, and sets the stage for the “exception” to swallow up the rest of that constitutional order. Any interpretation that does this – i.e., that effectively collapses the distinction between the “normal” and the “exception” within the constitutional order – should be immediately rejected.
What of Mugambi J.’s argument that the words “or disaster” after “emergency” require interpreting “emergency” in a non-technical sense, otherwise the former words would become superfluous? I suggest that the arguments advanced above should make it clear that any ambiguity in the provision should be interpreted in favour of the “technical” meaning of emergency. Even otherwise, however, I submit that if this ambiguity had to be resolved, the correct way of resolving it is not by according “non-technical” meanings to both “emergency” and “disaster”, but by according technical meanings to both. The Constitution’s Fourth Schedule, which distributes functions between the national government and the county governments, has a specific entry titled “disaster management” under the national government part of the Schedule. Indeed, in 2023, the National Disaster Risk Management Bill was introduced in the Kenyan National Assembly. Like its counterpart Disaster Management Acts in other countries, the Bill specifically defines what constitutes a “disaster”, and has specificied procedures in place both for localised disasters, as well as a “national state of disaster.” This Bill is an example of what I call a constitutional statute – that is, a law that is explicitly enacted to fulfil an (express or implied) constitutional obligation.
The correct way of reading Article 241, I suggest, is to accord to “emergency’ the technical meaning that is there in Article 58 of the Constitution, and for the word “disaster” to become operable as a basis for armed forces deployment as and when a statute akin to the National Disaster Risk Management Bill is enacted, whose basic purpose is to define – and provide a regulatory mechanism for the mitigation of – “disasters.” Such a reading, I suggest, is not only textually consistent, but also, truer to the purposes of a transformative Constitution which must treat the internal deployment of armed forces always as an exception, and therefore subject to strict constitutional safeguards.
The Terms of the Order
In paragraph 54, Mugambi J. ordered that “the terms of military engagement, duration of engagement be clearly defined and gazetted” within the next two days – ostensibly to prevent an open-ended deployment of armed forces, that would become the basis of state impunity. If that was the objective, however, then it evidently failed, as on 29th June, the Cabinet Secretary for Defence issued another Gazette Notice, ostensibly “further to the order” of the High Court. Paragraph 2 of the Gazette Notice stated that “the terms of engagement shall be the constitutional standards enshrined in the Bill of Rights”, as well as the statutory requirements of the Kenya Defence Forces Act. Paragraph 3 stated that “the deployment shall continue until normalcy is restored.”
It is evident that, even as this Gazette Notice purports to comply with the High Court’s order, in reality, it cocks a snook at it. The Executive does not fulfil its constitutional obligations simply by restating the provisions of the Constitution as terms of engagement. For example, primary or secondary legislation that simply restates constitutional provisions as grounds for restriction of rights would be struck down as unconstitutional. The Terms of Engagement have to be demonstrably compliant with the Constitution, not restate is provisions. Ans secondly, “the deployment shall continue until normalcy is restored” is the very definition of an open-ended assumption of power (exactly unlike states of emergencies, which have an expiry date), to be exercised at the unbounded discretion of the executive (who else, after all, whill define “normalcy”?). It effectively places no restrictions whatsoever on State power: something like a constitutional law equivalent of “the beatings will continue until morale improves”!
One must note with regret, however, that this Gazette Notice has been enabled by the vagueness in Mugambi J.’s order: an order requiring that X or Y be “clearly defined” by the executive does not really have any teeth, as the question of what constitutes “clear definition” is up to the executive. One can, of course, go back to the Court for another round of clarifications, but in the context of something as fluid and changeable as a nationwide public protest, the case might already have become infructuous by then.
It is here that Vincent Blasi’s classic exposition of the pathological approach is helpful. While addressing the question of how free speech laws ought to be drafted and interpreted, Blasi argues that “the overriding objective at all times should be to equip [free speech law] to do maximum service in those historical periods when intolerance of unorthodox ideas is most prevalent and when governments are most able and most likely to stifle dissent systematically.” How is one to accomplish this? Among other things, by “confining the range of discretion left to future decision makers who will be called upon to make judgments when pathological pressures are most intense.”
I suggest that Blasi’s approach is equally applicable to judicial orders, especially those that are of an interim nature, and whose goal is to canalise how the executive ought to act. When crafting such an order, the judge should adopt the pathological perspective: in this case, that would mean that they should equip their orders to do maximum service in the context of an executive determined to push its powers to the hilt; and one way to accomplish that is to confine the range of discretion left to the executive, including in interpreting the order.
Adopting the pathological perspective, thus, would not allow the Court to simply ask the executive to state its terms of engagement and a time period, all of which is open to substantial interpretation (as we have just seen). Rather, the Court would to specify with far greater detail the contours within which the terms of engagement would have to be spelt out, and either stipulate a time period for the deployment to end (or at least, for it to have to pass through onerous procedural safeguards before it can be continued), or hold the executive to a much stricter standard of what constitutes “normalcy.” As the Court has, at the time of writing, kept the matter alive for further rulings, one hopes that this might still happen.
Conclusion
This post has argued that Mugambi J. was wrong to authorise the deployment of the defence forces for the purposes of policing the protests, on the basis that the word “emergency” under Article 241 has to be given its “dictionary” meaning. I have argued that textually, structurally, and philosophically, an interpretation that treats the word “emergency” as it is treated in the Constitution – along with the rigorous procedural safeguards that follow – is a better interpretation of the 2010 Constitution. And secondly, in crafting interim relief, the pathological perspective might serve as a valuable guiding light for judges to ensure that the executive is unable to effectively defeat their orders by simply according them the widest possible reading consistent with its own untrammelled sphere of action. This, I suggest, would be truer to the goals and purposes of the 2010 Constitution.
Constitutional Law and Philosophy 
Gautam may we republish with credits?
Warm Regards Teesta Setalvad http://www.cjp.org.in