[This is a guest post by Kenson Mutethia.]
Introduction
I rarely disagree with Joshua unless it’s a serious matter of principle. Part A of Joshua’s recent blog, “A Constitutional Travesty-II: Unpacking the Kenyan Supreme Court’s Finance Act Judgment,” is one of those rare instances where I fundamentally differ with him. Specifically, I respectfully disagree with Joshua’s criticism of the Supreme Court’s Finance Act Judgment for permitting suspended declarations of invalidity.
In this post, I present a response to Joshua’s criticisms of suspended declarations. I first outline Joshua’s three claims against suspended declarations, discussing each in turn. After each claim, I offer my rebuttal. I hope to persuade the reader that while Joshua’s three main arguments—based on Articles 2(4), 22 & 23, and 10—may seem superficially sound (and I must admit they do), a closer look reveals they are deeply flawed. I argue that these constitutional provisions do not prohibit the use of suspended declarations; rather, when read together with Articles 27(1), 48, 50(1), and 159(2) (a) and (e), they actually support their deployment but in very rare cases, such as those described by the Supreme Court in paragraph 236 of the Finance Act Judgment. I conclude by observing that Joshua’s concluding remarks, in which he concedes that if suspended declarations are issued, they should be combined with the remedy of reading in, are incompatible with his overarching argument that suspended declarations are unconstitutional because that claim rests on the assumption that suspended declarations are incompatible with Articles 2(4), 22 & 23 and 10 of the Constitution.
Joshua’s Overarching Claim(s) and the Rebuttals
There’s little difficulty in understanding Joshua’s claim. He finds it shocking, possibly even disturbing, that in the Finance Act Judgment, the Supreme Court approved a view that contradicts Joshua’s accepted logic, that is, that a law declared unconstitutional can still temporarily apply by suspending its invalidity. Joshua does not find comfort in the Supreme Court’s limitation on the rare circumstances under which suspended declarations can be issued. It is less important to him that the Court clarified in paragraph 236 that a suspended declaration should only be granted when public interest requires it, based on very unique facts that justify such an action, such as preventing a legal vacuum, a threat to the rule of law, lawlessness, chaos, or anarchy. To Joshua, since, after all, under the doctrine of objective constitutional invalidity (see Ferreira v Levin), finding a law to be “in conflict with the Constitution” does not invalidate the law; it “merely declares/establishes it to be invalid,” a declaration of unconstitutionality should be the end of the story. Once declared unconstitutional, no matter the effects, that law cannot have another temporary lease of life through judicial fiat. Joshua’s difficulty with the suspended declarations is based on three main arguments, which I shall address in turn.
Joshua’s Article 2(4) Claim
First, Joshua claims that suspended declarations are totally incompatible and unsupported by Article 2(4) of the Constitution, which provision he claims automatically brings in the immediate invalidity of a law declared unconstitutional. In sum, Joshua’s first argument is that suspended declarations erode the supremacy of the Constitution because they allow the Constitution to be frozen in time, which he considers an anomaly. More pertinently, Joshua points out that, unlike other constitutional texts like section 172 of the South African Constitution, which allows Judges to suspend declarations of unconstitutionality, Article 2(4) of the Kenyan Constitution does not give courts any jurisdiction beyond the declaration of unconstitutionality.
Rebuttal to the Article 2(4) Claim
There are three arguments against Joshua’s claim on Article 2(4).
A Holistic Interpretation of Article 2(4)
First, it is not true that a declaration of unconstitutionality by any court automatically renders the impugned law invalid unless the declaration is final, because an appellate court can either stay the declaration pending appeal or even set aside the declaration on appeal. While an appeal does not amount to an automatic stay even in typical civil cases, the mere existence of a right of appeal defeats, and especially so in deserving cases, the argument that a declaration of unconstitutionality cannot be suspended or stayed pending appeal. The right of appeal, as F. Gikonyo eruditely puts it in James Wangalwa & Another v Agnes Naliaka Cheseto [2012] KEHC 1094 (KLR), “ is a constitutional right that actualises the right to access to justice (Article 48), protection and benefit of the law (Article 27)” and even the right to a fair trial (Article 50). The essential substance of the right of appeal encapsulates that the appeal should not be rendered nugatory, for anything that renders the appeal nugatory impinges on the very right of appeal.
There is no fundamental difference between suspending a declaration of invalidity and staying a declaration of invalidity. One cannot claim that Article 2(4) forbids the deployment of suspended declarations and, at the same time, concede that it does not prohibit granting stays of declarations of unconstitutionality. The Court of Appeal clarified this in the recent RMLF Appeal case, where it allowed, among others, the National Assembly’s request for “a conservatory order to suspend a declaration of unconstitutionality.” In fact, while the Rule 5(2)(b) Application before the Court sought a stay pending appeal, the Court primarily relied on the principles of granting suspended declarations.
The Court of Appeal’s approach effectively confirms that the basis for granting either remedy is generally similar, even if the specific tests differ. Specifically, while the test for suspended declarations is the existence of exceptional circumstances, and that of a stay pending appeal requires an arguable appeal that would become nugatory without a stay, to fulfil either tests, a party would have to demonstrate the same facts, i.e.,: (i) striking down legislation without enacting a replacement could endanger the public or pose a threat to the rule of law; or (iii) legislation was only found unconstitutional due to under-inclusiveness rather than overbreadth and that such relief may be granted to: (i) prevent a constitutional crisis; (ii) avoid harm from immediate invalidation of a law; and (iii) give Parliament time to pass suitable legislation (see LSK v KRA).
Kahinga is a good illustration of the essence of suspended declarations. In Kahinga, the High Court found Sections 295, 296(1), 296(2), 297(1), and 297(2) of the Penal Code unconstitutional for ambiguity related to crimes of robbery and robbery with violence. It suspended the declaration for 18 months, allowing Parliament to address the unconstitutionality. Unfortunately, Parliament did not take corrective action within the period or seek an extension. After the 18 months expired, the Court could not revive the invalidated laws (see, for example, Zondi), meaning there was no law criminalising robbery with violence. Constitutionally, no person should face charges for robbery or robbery with violence after this period, as the provisions effectively ceased to exist. But alas, we are in Kenya, and no one really cares about some constitutional aphorisms.
Joshua’s interpretation of Article 2(4) implying uncheckered immediacy effect of invalidation is, therefore, isolated from other relevant provisions, such as Article 27 (1) on equal benefit and protection under the law, Article 48 on access to Justice, Article 50 on fair hearing from which the right of appeal is situated. Yet, the Constitution must be interpreted as an integrated whole. It is also an unbridled extrapolation of Article 2(4) to arrive at a desired outcome to the extent that it strains it to bear a meaning that it is incapable of bearing, contrary to the holistic interpretation approach extolled by the Supreme Court. In effect, Joshua’s interpretation of Article 2(4) yields a constitutional absurdity and must be resisted on that account alone.
Resolving the Constitutional Silence in Article 2(4)
Secondly, the mere fact that Article 2(4) is silent does not automatically mean that courts are prohibited from issuing suspended declarations. Joshua’s argument fundamentally ignores the serious hermeneutical controversies surrounding constitutional silences. The silence of the Constitution in Article 2(4) reminds me of one wry joke: “In England, everything is permitted except what is forbidden. In Germany, everything is forbidden except what is permitted. In France, everything is allowed, even what is prohibited. In the USSR, everything is prohibited, even what is permitted.” Joshua’s suggestion that the silence of Article 2(4) prohibits suspended declarations in principle amounts to judicial populism to the extent that it insists there is a clear answer to resolving an otherwise controversial question or that the silence of Article 2(4) was more of a design than a default. I may make this argument more profoundly in the future, but for now, it is enough to establish that Article 2(4) does not readily concede that suspended declarations are unconstitutional.
Sophistry in Drawing Comparative Lessons
Lastly, Joshua’s argument that Article 2(4) prohibits suspended declarations because it does not like other comparable tests (section 172 of the South African Constitution) expressly allow them impliedly suggests that that there is no basis for borrowing the practice in other jurisdictions given the textual differences in the Constitution, a logic that our courts have readily accepted (see for instance EG v AG, holding that the High Court cannot borrow the experiences from other jurisdictions such as India that have decriminalized consensual same sex relations, because those other jurisdictions do not have the equivalent of our Article 45(2) on heterosexual marriages—as if same sex relations and same sex marriages are the same concepts).
This logic—that “our constitutional experience is different from other jurisdictions and so we should not borrow blindly”— is actually used mainly by conservatives, in the most inappropriate contexts, as seen in EG v AG, and not by progressives like Joshua. There are indeed peculiar constitutional experiences in every jurisdiction that must never be ignored. Still, I have consistently refused to be persuaded that certain high constitutional ideals are unique to any specific jurisdiction. I believe it is the Swahili that told us: “Chema cha jiuza, kibaya cha jitembeza”, loosely translated to mean “a good thing sells itself; a bad thing must be advertised.” This proverb highlights the idea that true quality or goodness will naturally gain recognition and be sought after. In contrast, something of poor quality needs constant promotion and effort to be noticed. I subscribe to the school of thought that believes sound and transformative constitutional order and principles sell themselves everywhere and cannot be constrained to any specific jurisdiction; they naturally gain recognition and are sought after. In this case, the deployment of suspended declarations in exceptional circumstances stated by the Supreme Court at paragraph 236 of the Finance Act Judgment is a high constitutional order worthy of emulating.
Joshua’s Articles 22 & 23 Claim
Joshua’s second argument is that Article 23 of the Constitution, on which the Finance Act Judgment situates the remedy of suspended declarations, does not in actual fact support the grant of such an anachronistic remedy in the post-2010 constitutional dispensation. More pointedly, Joshua argues that first, Article 23 reliefs are only granted in proceedings brought under Article 22. Article 22 provides locus standi for people to approach the court where there is a violation of the Bill of Rights. Accordingly, although Article 23 does not define appropriate relief, an effective relief is limited only to a relief that is capable of protecting and enforcing the Constitution. To Joshua, suspended declarations cannot be read into the open-ended reliefs under Article 23 because they neither protect nor vindicate the Constitution, nor do they deter future violations.
Rebuttal to Articles 22 and 23 Claim
Joshua’s arguments on Article 23 are reminiscent of Vargas’ “For my friends everything, for my enemies the law”. The people represent Joshua’s friends, while the government is seen as the adversary. To be fair, the Kenyan government is mainly seen as rogue and deserving of punishment. I’ve personally argued that the progressive aspects of the Finance Act Judgment are hindered by the government’s questionable conduct, which often makes it unworthy of indulgence. Nevertheless, since the Constitution, especially Article 23, does not expressly or implicitly prohibit suspended declarations, courts should not be blamed when Parliament passes unconstitutional laws that courts are condemned to temporarily uphold because invalidating those laws immediately would be far more costly than beneficial. After all, there is only so much a court can do when confronted with widespread lawlessness.
Joshua’s argument that suspended declarations neither protect nor vindicate the Constitution, nor do they deter future violations, is also not honest for a number of reasons. First, the mere fact that an appropriate relief can only be granted under proceedings brought under Article 22 by people having locus standi claiming that the Bill of Rights is violated or is threatened with violations, Article 23 does not limit relief to parties bringing the suit only. It is trite law that for a party to be part of a suit, they must have standing to sue and/or to be sued. The open-ended appropriate reliefs intended under Article 23 of the Constitution are designed to be granted to every party in the proceedings under Article 22, including the State. To limit the appropriate reliefs to people suing is a classic, to my friends, everything, to my enemies, the law. Secondly, Joshua ignores a very critical aspect of the Social Contract that informs the constitutional democracy we live in. Although we have a government that is hellbent on doing the wrong things, it needs to be allowed to function. This would explain the Supreme Court’s test in the Finance Act Ruling on Public interest tilting in favour of seamless government functioning. It may be a controversial test that we can discuss later, but for now, it is good to appreciate where the Supreme Court was coming from when developing the test.
Joshua also forgets that there are instances in which a suspended declaration will ensure that the government proceeds to discharge a constitutional and pro-people mandate while Parliament adopts corrective measures. For instance, if Parliament enacted a law for the compensation of a specific class of people, for instance, persons who lost family members and/or property during demonstrations, but such a law is only found faulty on account of being underinclusive wouldn’t the interests of justice demand that a court suspends the declaration until Parliament corrects by adding the excluded instead of striking the whole law down and rendering the class of beneficiaries destitute?
Thirdly, Joshua erroneously crafts his arguments as if the Supreme Court issued a carte blanche for the grant of suspended declarations in every constitutional challenge. This was not the case. The Supreme Court said that this would only be appropriate in specific instances and even issued guidelines on when and how they should be issued at paras 236, 240 and 241 in the Finance Act Judgment. This was a properly thought-out guideline for courts to use only in appropriate times.
Joshua’s Article 10 Claim
Joshua’s final central argument is that suspended declarations violate Article 10 of the Constitution to the extent that they countenance a situation where citizens are compelled to comply with an unconstitutional law through judicial fiat contrary to the normative state based on values such as good governance, the rule of law and accountability that Article 10 establishes. Additionally, since constitutional values represent public interest, public interest militates against suspended declarations to the extent that they give more room to more violations or undermining of constitutional values.
Rebuttal to the Article 10 Claim
The Finance Act Judgment interpreted Article 10, too. This is one of those findings I see as problematic in the judgment. The Supreme Court effectively stated that the national values and principles of governance outlined in Article 10(2) of the Constitution are broad guiding principles. Courts should be careful not to treat them as strict normative rules. These values and principles act as frameworks that guide decision-making, specifying considerations for duty bearers rather than exact duties or actions. They are inherently open-ended, providing direction without prescribing specific steps. These are optimising commands that allow duty bearers to develop appropriate measures to fulfil their obligations, without dictating exact actions. Overall, the Supreme Court held that government bodies have discretion in executing their constitutional duties under the values and principles in Article 10(2), and courts should only assess whether their measures are reasonable and compliant with the Constitution and law. When Article 10(2) is alleged to be breached, such as in the 2023 Finance Act, where the National Assembly was alleged to have violated transparency and accountability principles, the Court’s role was limited to establishing if the Assembly took reasonable steps to promote transparency and accountability during public consultations on the 2023 Finance Bill. While the Supreme Court’s interpretation of Article 10 and the subsequent effective test of “reasonable efforts” is questionable, it is still difficult to establish the validity of Joshua’s argument that suspended declarations violate the Constitution. The claim would be more convincing if Joshua argued that the deployment of suspended declarations violates Article 10 when they are deployed in inappropriate cases outside those specified in paragraph 236 of the Finance Act Judgment.
Conclusion
While concluding his post on this issue, Joshua seems to concede that, if at all, suspended declarations should be issued to avoid disruption caused by immediate declarations of unconstitutionality; they should be combined with the remedy of reading in to ensure that the Constitution remains intact at all times. This is a reasonable concession (that could have applied, for instance, in Kahinga), though it may not always be feasible to combine the two measures. Joshua’s argument appears to be that if the two cannot be combined or if the Court fails to do so, then the Court is in the wrong. I am not sure which among Articles 2(4), 23, and 10 the Court would be wrong for violating, because Joshua’s overarching claim is that a court should not issue a suspended declaration at all. His hypothesis falls apart if he admits there are exceptional circumstances where suspended declarations are appropriate. This flaw is present in all three of his arguments, which seem to be omnibus in nature.
Joshua’s arguments would be more convincing if his overarching claim were his concluding remarks. Specifically, I would be more persuaded if Joshua’s claim against suspended declarations were: “Generally, courts should not issue suspended declarations. However, if they must, in exceptional circumstances, they should do so only if they can combine suspended declarations with the remedy of reading in. This is because a law that cannot be read in or severed during the suspension period is inherently unconstitutional and thus not eligible for suspended declarations.” Nonetheless, this approach raises further questions. For example, what should happen to laws that are constitutional in their specific provisions but are invalid due to other constitutional requirements, such as public participation or the Speakers’ concurrence? The Finance Act, the Privatisation Act, and the 23 Laws cases come to mind in this respect. This is a complex issue I may explore further in the future.