In July of last year, the Kenyan Court of Appeal had struck down mandatory life imprisonment as unconstitutional. As the analysis on this blog noted, the judgment was delivered in a context in which the death penalty in Kenya has not definitively been held unconstitutional (just last week, in fact, a Kenyan Court controversially handed down a death sentence). However, even as the debate around the death penalty continues, the High Court – in a judgment delivered on 19th March 2024 – held life imprisonment itself to be unconstitutional.
The judgment – Justus Ndung’u Ndung’u vs Republic, authored by Justice Nixon Sifuna, is remarkably short, coming in at six pages. It was an appeal from both a conviction and a sentence (for incest), and indeed, much of the judgment is concerned with re-appraising evidence. The Court finds that the conviction was justified, and upholds it. This then brings it to the question of sentence: the magistrate had imposed a life sentence. The Court finds it unjustified not specifically on the facts of this case, but rather, on the basis that the sentence of life imprisonment itself is unconstitutional. The heart of the analysis is in paragraph 10, where the Court notes:
A life sentence is a sentence sui generis. In that, whereas it is philosophically supposedly imprisonment for a duration of time only, it is in actual sense imprisonment that is indeterminable, indefinite, uncompletable, mathematically incalculable, and therefore quantifiable only for the convict’s entire remainder of his lifetime.
Variants of this analysis are repeated in the succeeding paragraphs, before Sifuna J concludes that the sentence is, therefore, archaic, unreasonable and absurd, and violates the right to human dignity under Article 28 of the Kenyan Constitution (paragraph 17). Reconstructing the reasoning, at its heart, the issue appears to be that unlike all other sentences, life imprisonment is not definite, but pegged to a contingent event (the end of the convict’s life), which could happen at any given time. Sifuna J. therefore compares it to the death penalty (paragraph 16), and also highlights the potential absurdity of a person who dies in prison soon after being sentenced for a heinous crime, as opposed to another person who spends years behind bars for a less serious crime (paragraph 12).
Neither of these two arguments are, however, entirely convincing. The analogy with the death penalty is striking and powerful, but it is unclear if it supports the argument for unconstitutionality, given that the death penalty itself has not been struck down yet (in fact, a situation where life imprisonment is unconstitutional but the death penalty is constitutional feels somewhat anomalous!) And the potential absurdity of someone dying an early death behind bars is not quite an absurdity if we consider that the primary penological goal of life imprisonment is prevention – i.e., to prevent a convict from committing a crime again. From that perspective, there is nothing particularly absurd about a convict dying soon after being sentenced, as there is no question of recidivism after death. It is, of course, another matter whether punishment based solely on prevention, and completely ignoring reformation or rehabilitation, can pass constitutional muster; that, however, is not considered in the judgment.
We therefore come back to the question of indefiniteness, and the violation of the right to dignity. I think that the argument – although it is not spelt out in the judgment itself – is essentially one of dehumanisation, or considering the convict purely in instrumental terms. In assuming that an individual can never be re-integrated into society, the life sentence entirely strips them of agency, or the ability to make different choices in the future. The locus of the violation of human dignity, I would suggest, lies in this assumption.
Two points then arise with respect to the judgment itself. The first – as noted above – is that the judgment does not, in its consideration of the dignity question, engage with penology, or the goals of criminal punishment. In my view, striking down a sentence provision as unconstitutional is difficult without at least considering what the stated goal of the punishment is, and how the punishment itself relates to that goal. Indeed, that is a vital element of the proportionality test, which is the overarching basis of constitutional challenges, especially of this nature. Indeed, the paragraph above – that attempts to excavate the normative basis for the indefiniteness argument – finds itself going back to the penological goals of the life sentence (as it must).
The second point is a point of procedure: notably, it does not appear that the constitutionality of the life sentence was challenged in this case (if it was, then the entirety of this paragraph can be ignored). I do not have access to the pleadings, but let us go by the Court’s own framing of the question of sentence: “Whether the imprisonment sentence imposed by the trial court was unreasonable, excessive, or too harsh.” This is not the language of a constitutional challenge, but a plea for sentence mitigation on the facts of the case. The question then arises: can the High Court strike down the life sentence without it being under challenge? Would not, for example, the State have to be put to specific notice, so that it can defend the constitutionality of the sentence in those specific terms?
While, therefore, I agree with the High Court’s decision to strike down the life sentence, and I find locating the analysis in how indefiniteness violates the right to dignity, the Court’s reluctance to engage in a full-blown analysis of the dignity question (including applying the proportionality test), as well as the possibility that there was no constitutional challenge made, might leave the judgment vulnerable upon appeal. It will be interesting to see what happens at the Court of Appeal!
This essay considers the doctrine of “double proportionality” in the Electoral Bonds Case. In brief: the majority opinion, written by Chandrachud CJI, holds the various amendments brought through the Finance Act 2017 unconstitutional for being in contravention of right to information protected under article 19. Once it is established that the foundational design of the scheme was to make political donations through electoral bonds confidential, the enquiry proceeds to analyse whether the right to information is engaged (i), and whether the restrictions placed on right to information are reasonable and justified (ii). On aspect (i), the court, following Association for Democratic Reforms (2002) and PUCL (2003), holds that information about political funding is essential for the voter to exercise choice and their freedom to vote, and as a corollary, for the effective exercise of freedom of speech and expression. Aspect (ii) is where the application of proportionality and double proportionality is used by the majority to determine whether the restriction of the right to information is reasonable vis-à-vis the purpose(s) of the scheme. Hence, In part I, I highlight the necessity of application of the double proportionality. In part II, I analyse whether the test of double proportionality is correctly laid down.
Balancing fundamental rights
Chandrachud CJI recognizes two purposes of the scheme in restricting access to information. These are then assessed on the touchstone of proportionality i.e., whether the abridgement of the right is disproportionate to the legitimate aim sought by the purpose(s)/objective(s) of the scheme.
Chandrachud CJI identifies the two purposes of the scheme as a) curbing black money and b) donor privacy. The majority opinion accepts the State’s’ submission that donor privacy or confidentiality is not only a means to achieve the aim of incentivizing political donations through legitimate banking channels, but also a substantive end in itself. It therefore holds that “the Constitution guarantees the right to informational privacy of political affiliation” under article 21, and that extends to political donations being confidential (para 141). Having held so, the question now is whether the Electoral Bond Scheme adequately balances the right to information against right to informational privacy of political affiliation.
Donor privacy as a legitimate aim and double proportionality
According to the majority, the conflict is between the right under restriction i.e., the right to information (Right A) and the countervailing right i.e. right to privacy of the donor (Right B). The traditional balancing exercise to resolve or avoid the conflict can be conducted by various means, for instance:
It can be held that the right A trumps right B because the former right subserves the ‘larger public interest.’
Alternatively, it can be held that held that there is ‘no real conflict’ between the rights either because one of the rights is not engaged, or the boundaries of the rights are circumscribed in such a way as to avoid conflict.
However, the application of proportionality in the ‘contextual balancing’exercise between two competing rights is considered to bring in a ‘structured balancing exercise’ where both rights are given equal importance and weightage. It will be fruitful to highlight the reasons for double proportionality in balancing two fundamental rights, in order to analyse the test laid down by Chandrachud J (Para 157), and whether it is in consonance with the principled reasons behind the application and structure of the double proportionality test.
Reasons for Double-Proportionality
Reason 1: Equal Importance of both competing rights
In Re W, reason 1 was expressed in these terms: “… each Article propounds a fundamental right which there is a pressing social need to protect. Equally, each Article qualifies the right it propounds so far as it may be lawful, necessary and proportionate to do so in order to accommodate the other”. Application of the double-proportionality test to balance two competing rights of equal importance is said to secure ‘procedural justice’ by recognizing that both rights require full protection, and that this is accomplished by application of the proportionality test to the restriction on each right. It structures the enquiry in ways that minimizesthe interferences with both right A and right B, thereby giving fair and equal value in the enquiry to both the interests exemplified by competing rights.
Reason 2 : Single-Proportionality favours the right under restriction (Right A) over the competing interest of Right B
Chandrachud CJI recognises that the ‘priority-to-rights’ effect of proportionality enquiry, as it tends to give prominence to the fundamental right under restriction over the competing interests (para 153), and that single proportionality might not be appropriate when two fundamental rights are involved. When the competing interest is a fundamental right in itself, the prominence given to the ‘invoked right’ i.e. restriction of right A under challenge in the proportionality enquiry cannot be a ‘balanced exercise.’ This is because single proportionality enquiry, especially in the third stage, requires the minimization of interference to the right under restriction (right A) at the expense of the competing interest.
For example, the conflict in Campbell v MGNwas between freedom of speech (Right A) and the right to privacy (Right B), both of which are equally protected rights under the ECHR. If the court were to adopt a single proportionality test to review the restriction of the invoked right (A), the furthering of competing right B would form the legitimate aim and the restriction of the right to freedom of speech (A) would have to have a rational nexus to the protection of privacy (right B). However, in the third stage, the court would have to consider whether the restriction on right A is the “least restrictive measure” to further the competing interest of right B. This entails prioritizing and maximizing the protection of right A, but not maximizing the interests of competing right B. This is incompatible with the proposition that both rights require equal and full protection.
Reason 3 : Single Proportionality is insufficient to take into account the interest of the competing right
It should be noted that it is not conceptually impossible to account for the fundamental importance of the competing interest (Right B) within the single proportionality test while testing the restriction of right (Right A). However, it would require modification, or what some call ‘distortion’, of the single proportionality test in order to recognise the importance of the competing fundamental right. For example, the third stage of proportionality would require not the maximum protection of right under restriction to which proportionality is applied, but maximum protection for both the rights. Instead of asking the question whether the measure is least restrictive of the right under restriction (A) and achieves the end in ‘real and substantial degree,’ the question would be whether the alternative measure would be least restrictive of both rights and achieves the purpose in ‘real and substantial degree.’
Then, the third stage of proportionality test is no longer about prioritizing and providing maximum protection to the right under restriction, but an equilibrium position of right under restriction and competing interest (right B).This modification is considered unsatisfactory for giving maximum protection and importance to both right A and right B. Therefore, double proportionality is envisaged as the same question being examined from two perspectives, rather than two different perspectives being considered under a single proportionality review of restriction on one right (invoked right). This flows directly from the reason 1 that both the rights are fundamental and deserve equal importance and consideration.
Chandrachud CJI’s three-step test
The major case after Campbell v. MGN where the ‘balancing of rights’ was required was In Re S. The facts were that there was a gag order on news outlets, to prevent the publication of details and photographs of a 5-year-old child whose brother had been killed by their mother. The case concerned a conflict between freedom of speech of the newspapers (invoked right) and privacy of the child (competing right). Lord Steyn laid down the four propositions for the ‘ultimate balancing test’, last three of which include those laid down by Baroness Hale in Campbell, which Chandrachud CJI relies on.
First, neither Article has precedence over the other.
Secondly, where the values under the two Articles are in conflict, an intense focus on the comparative importance of the specific rights being claimed in the individual case is necessary.
Thirdly, the justifications for interfering with or restricting each right must be taken into account.
Finally, the proportionality test must be applied to each.
Chandrachud J modifies this test by holding that that second and third propositions are subsumed within the balancing (fourth) prong of the proportionality analysis. Further, he holds that:
a. if under the constitution, no hierarchy has been presented for the rights under consideration (firstly above), then :
the following standard must be employed from the perspective of both the rights where rights A and B are in conflict:
b. Whether the measure is a suitable means for furthering right A and right B.
c. Whether the measure is least restrictive and equally effective to realise right A and right B; and
d. Whether the measure has a disproportionate impact on right A and right B.
However, it is argued below that this distorts the double-proportionality enquiry and prioritizes the right under restriction (right A) over competing right B, and approximates towards a modification of single proportionality test, rather than a double-proportionality test. This is essentially evident from the subsequent analysis which Chandrachud CJI undertakes, where the contradictions of the test he lays down become apparent.
Subsequent Analysis and Contradictions
The first prong of proportionality is satisfied as each right ( A and B) provides a legitimate aim for the restriction of other.
On proposition (b):
(b): Whether the measure is a suitable means for furthering right A and right B.
The question is framed in such a way that it is not possible to fulfil the requirement of suitability from both sides but is only possible from the side OF the invoked right (Right A). As Chandrachud CJI holds, the measure that places restriction on right to information(right A) is suitable for purpose of realizing the informational privacy of the donor (right B). However, the measure can never be suitable for purpose of realizing the right to information (right A).
“… the purpose of securing information about political funding can never be fulfilled by absolute non-disclosure. The measure adopted does not satisfy the suitability prong vis-à-vis the purpose of information of political funding. However, let us proceed to apply the subsequent prongs of the double proportionality analysis assuming that the means adopted has a rational nexus with the purpose of securing information about political funding to voters.”(para 162)
This is obvious. It is because the anonymity of the contributor (privacy of the donor, right B) is intrinsic to the Electoral Bond scheme (para 158). The measure can never be suitable to realise right to information (right A) because the measure places no restrictions on the right to privacy (B). It is the invoked right A i.e. right to information which has the potential to place restrictions on competing right B if the challenge is successful. The court’s responsibility is to balance the involved right against the competing right by application the proportionality analysis separately to right each, by considering protection of each as a restriction on another.
Is the question posed by Chandrachud CJI analytically sound for application of Double-Proportionality analysis? The answer is no. As Prof. Huge Collins commenting on Campbell v HGN and In re S writes : “Given that there are [equal]competing interests, rights…the correct approach appears to be a double proportionality test. In other word, the case for interference with the separate rights of each party needs to be assessed separately according to a test of proportionality. The legitimate aim that may justify such an interference with a fundamental right ….include the protection of the fundamental right of the other party”
As Chandrachud CJI himself opines, the “standard must be employed from the perspective of both the rights where rights A and B are in conflict”. So, the question must be whether the restriction of the invoked right A (right to information) by the measure is suitable/rational to achieve right B (right to privacy). The answer might be yes/no. On inverse application, the question would be whether the potential restriction on competing right B (right to privacy) by protection of invoked right A is suitable/rational to achieve right A (right to information). The answer might be yes/no. There is no logical impossibly, as Chandrachud CJI finds to be engaged by the question framed. The assumption which Chandrachud CJI makes could be avoided if there was true application of the double proportionality test as shown in the table below.
It might be said that separate application of proportionality to each right considering the other right as legitimate aim is merely unnecessary repetition that has no substantive bearing on the final conclusion reached. However, that is not the case, as a single proportionality review in case of conflict of two equal rights has the potential to skew the balance in towards the invoked right to the disfavour of the competing right that is furthered by the measure. The ‘preferential framing’ negates the reasons mentioned above for balancing two equal fundamental rights by applying double-proportionality test. The test ensures that both the rights deserve maximum judicial protection by application of proportionality test to the restriction on each right.
Restriction of invoked right A (first stage)
Potential Restriction of competing right B by the invoked right ( second ‘inverse’ stage)
The test Chandrachud CJI lays down
Legitimate aim: Conflicting operation of right B of equal value.
Suitability/rational nexus: Whether the restriction of right A is suitable to realize right B.
Whether the restriction on right A is least restrictive measure and equally effective and whether the alternative such measure realizes the purpose of right B in real and substantial degree.
Balancing stage: Comparative importance of fundamental rights and justifications for restrictions on the same;
Whether the restriction on right A is disproportionate to the purpose of right B
Legitimate aim: conflicting operation/protection of right A of equal value.
Suitability/rational nexus: whether the restriction of right B suitable to realize right A
Whether the restriction on right B is least restrictive measure and equally effective to realise and whether the alternative such measure realizes the purpose of right A in real and substantial degree
Balancing stage: Comparative importance of fundamental rights Justifications for restrictions on the same. Whether the possible restriction on right B by invoking of right A disproportionate to the purpose of right A
b. Whether the measure is a suitable means for furthering right A and right B.
c. Whether the measure is least restrictive and equally effective to realise right A and right B; and
d. Whether the measure has a disproportionate impact on right A and right B
On proposition (c)
Consider question (c): Whether the measure is least restrictive and equally effective to realise right A and right B.
If we bifurcate the question into two parts as Chandrachud CJI does for the above question (b), then we see that the question is either illogical (as it is similar to the above), or prioritizes the invoked right against the competing right B. Non-bifurcation essentially becomes a modified version of the necessity stage of single-proportionality test (reason 3).
i. whether the measure is least restrictive [on right A] and equally effective to realised right B [right to privacy].
The framing of this question itself prioritizes right A over right B. This is because right B i.e., the right to privacy is intrinsic to the measure. The question effectively being asked is whether the restriction on right to information (right A) by the measure (in furtherance of right B) is the least intrusive. This is prioritizing and maximization protection offered to the invoked right against the competing right B that is effectuated by the measure.
However, if the question is asked differently, i.e. whether the alternative measure would be least restrictive of both rights and achieve the purpose in a real and substantial manner( conjoined question), it is then merely a modification of the least restrictive stage in the single-proportionality test to take into account of the competing right B, with limitations of that approach to effectively maximize the protection of both the competing rights (as explained in Reason 3).
ii. whether the measure is least restrictive (on right B) and equally effective to realise right A (right to information).
The measure is not placing any restrictions on right to privacy (right B), but is placing restrictions on right to information (right A). In fact, right B is intrinsic to the measure. If that is the case, the question is redundant. However, it is not redundant if the question is whether the alternative measure would be least restrictive and equally effective to realise rights A and B. But the question is not framed in that way.
Chandrachud CJI previously holds that the measure can never be suitable to realise right to information (right A). If that is the case, then this question is also one of logical impossibility, as a measure which can never be suitable to realise A can never be the least restrictive means of doing so. However, Chandrachud CJI avoids the second logical impossibility by framing the question conjoinedly as mentioned above.
Example: How would the third stage work in proportionality, applied separately from both perspectives
It would involve asking the same question from perspective of both right A and right B. From the perspective of right A, the question would be whether the restriction of right A is least restrictive means of realizing interest of right B. From the perspective of right B, the question would be whether the restriction on right B would be the least restrictive measure to realize right A. On the latter question, it might be said that there is no restriction on right B. However, the double-proportionality enquiry is merely tasked with balancing competing interests of right A and right B in the judicial setting. The competing rights mark a conflict for the space to be occupied by the judicial outcome, where giving protection to one involves restriction on another.
Consider section 29C of the Representation of Peoples Act 1951, which Chandrachud CJI holds to be minimally restrictive of both rights and secures the purpose of both rights in a real and substantial manner (para 165). The said alternative measure has to be considered from both sides as placing some restrictions on each right. Then, the question would be whether such restriction is minimized interference with each right vis-à-vis the purpose (competing right) considered from both perspectives, rather than a single question of whether the measure under challenge is least restrictive of both rights.
On Proposition (d) whether the measure has a disproportionate impact on right A and right B.
This question suggests that the measure can have disproportionate impact on the right B. However, at the risk of repetition, the right B is intrinsic to the measure. The question makes sense only if the question is modified into whether the identified alternative has disproportionate impact on either right. As table A shows, this involves separately asking two questions from two perspective by considering protection of each right (example: in the identified alternative measure) as a restriction on each other.
However, Question (d) is not applied at all as the Chandrachud CJI holds that the necessity stage (c) has not been satisfied, and therefore there is no need of applying the balancing stage (para 167). Having held that double proportionality standard formulated by Baroness Hale in Campbell v MGN is adopted, and having held that the first two components of the Campbell standard of “comparative importance of the actual rights being claimed in the individual case” and “justifications for interfering with or restricting each of those rights” is to be submerged within the balancing stage, Chandrachud J does not complete the said double-proportionality standard adopted. As discussed above, even the last component of Campbell standard of “applying the proportionality test to each” is not satisfactorily applied, as the questions are framed in a way that prioritizes the invoked right A against the competing right B: thus negating the very reasons for application of the double-proportionality test.
Conclusion
The Court fails to apply proportionality separately to each of the rights. Having held that there were two competing rights, the balancing exercise by application of proportionality in relation to each other is required to give due recognition to the equal importance and maximum protection to each right. Although the outcome of the case might not differ, the propositions laid down suffer from lack of clarity for future application. The question is framed in a way that would either lead to no answers or one which would prioritize the right under restriction i.e., the invoked right A. It is argued in this essay that what the majority opinion effectively applies is a modified single-proportionality test with all the analysis happening in the necessity (third) stage, making most of the questions in the test laid down redundant without a modification in the framing. This, it is submitted, makes the test incomplete.
In a previous post (“The Mysterious Disappearance of the Stay Hearing“), I had pointed to the Supreme Court’s refusal to hold stay hearings in cases involving time-sensitive constitutional challenges. Perhaps the only thing worse than a refusal to substantively decide a stay application is a judicial order that purports to decide it, but on scrutiny, turns out to be a caricature, or a parody, of legal reasoning. This is the Supreme Court’s order of 22nd March, that dismisses the applications for staying the Election Commissioners Act (that we have discussed extensively on this blog).
The striking thing about the Supreme Court’s ten-page order – authored by Khanna and Datta JJ – is its refusal to engage with the core of the constitutional challenge before it – indeed, its refusal to even advert to it. Recall that the challenge was, in effect, that by replacing the Chief Justice of India in the selection committee for the Election Commissioners (the “interim arrangement” set out in the Anoop Baranwal judgment) with a cabinet minister nominated by the Prime Minister, the impugned legislation impermissibly modified, or overturned, the Court’s judgment. In paragraph 9 of its order, the Court purports to deal with this argument by stating that the committee constituted in Anoop Baranwal (the CJI, the PM, and the LoP) was a “pro-tem measure.” This is correct on its own terms, but also, it is a highly selective reading of Anoop Baranwal. As noted in this blog’s analysis of Anoop Baranwal, the interim committee constituted by the Court did not pop out of thin air. The basis of Anoop Baranwal was that Article 324 of the Constitution encoded a constitutional expectation that the manner of appointment of the Election Commission would ensure institutional independence from the executive. This is repeated multiple times in Anoop Baranwal, most specifically in paragraphs 32-33, and then even more categorically in paragraph 215:
We have set down the legislative history of Article 324, which includes reference to what transpired, which, in turn, includes the views formed by the members of Sub-Committees, and Members of the Constituent Assembly. They unerringly point to one conclusion. The power of appointment of the Members of the Election Commission, which was charged with the highest duties and with nearly infinite powers, and what is more, to hold elections, not only to the Central Legislature but to all the State Legislatures, was not to be lodged exclusively with the Executive. It is, accordingly that the words ‘subject to any law to be made by Parliament’ were, undoubtedly, incorporated.
The “pro-tem measure” that the Supreme Court’s order refers to – as is therefore obvious from a very basic reading of Anoop Baranwal – was pro-tem because Parliament had failed to enact a law consistent with its obligations under Article 324; these obligations were – to repeat – to ensure that, under the legislation, the Election Commission would be institutionally independent, and not subject to executive dominance. The “pro-tem measure”, thus, was not “pro-tem until parliament passes any law that it pleases”, but “pro-tem until parliament passes a law consistent with maintaining EC independence from the executive.”
The real question before the Court in the stay application, therefore, was whether the impugned law – on a prima facie reading – violated the institutional independence of the EC. This is a question which the Court entirely evades answering, for the very good reason that it has no answer: under the impugned law, the members of the EC are, in effect, appointed by the executive. Quite apart from the fact that this would be impossible to reconcile with institutional independence, it would also by hypocritical, given that the Court, in considering the question of its own independence, has held that executive dominance in appointments was inconsistent with institutional independence (the NJAC judgment). No wonder, then, that the stay order passes delicately over the real question that it had to answer, and instead answers a question that nobody has asked.
To mask the absence of reasoning, in paragraph 10, the Court then recites well-worn principles of judicial deference when considering stay of legislation, but – unsurprisingly – does not apply those principles to the case before it, and does not explain how they apply. Remarkably, however, the order then gets even worse. In paragraph 11, the Court states that “any interjection or stay by this Court will be highly inappropriate and improper as it would disturb the 18th General Election, which has been scheduled and is now fixed to take place from 19.04.2024 till 1.06.2024.” On a bare reading of this paragraph, one would think that the Petitioners had only shown up to Court the day before; however, as the order sheet demonstrates, the case was first called for hearing on 12th January, 2024, and posted to April 2024 for hearing. One would be hard-pressed to think of a more self-serving piece of judicial reasoning: the Court itself schedules a stay hearing more than two months after the case is filed, and then uses its own delay as a ground to deny stay, because it is now too close to the general elections. One is reminded of the Chief Justice Ranjan Gogoi’s infamous order declining stay of the electoral bonds scheme ahead of the 2019 General Elections, on exactly the same grounds – and after the Supreme Court had, itself, refused to consider the issue for two years. Readers of this blog will, of course, be well aware of what finally happened to the electoral bonds scheme, and consequently, how much damage the Court’s six-year delay in considering the issue caused.
Quite apart from the self-serving nature of this reasoning, the Court never explains why the administrative difficulties caused by a stay outweigh the potential “irreparable injury” caused by an election conducted by an Election Commission where two out of three Commissioners have been unconstitutionally appointed: indeed, one would imagine that when placed in the balance, electoral integrity would outweigh administrative inconvenience (for which, again, no empirical evidence is actually provided). However, there is no reasoning as to why the administrative concern outweighs the constitutional concern, other than a statement of simple judicial fiat.
To sprinkle salt on the (constitutional) wound, the Court then records numerous procedural issues with the selection, including lack of information provided to one of the Committee members (the LoP) (paragraph 14). Having said this, it then goes on to note that “in spite of the said shortcoming, we do not deem it appropriate … to pass any interim order or directions.” One wonders why, if the Court is not minded to pass any orders, it felt the need to make comments on the procedure: it behooves a Constitutional Court to decide the legal questions before it, and not to engage in moral sermons which are not attended by any legal consequences. Instead, the Court finishes with a classic exposition of its unique brand of “hope and trust jurisprudence”, noting that “the assumption is that they [the election commissioners] shall adhere to constitutional role and propriety in their functioning.” No justification is provided for this assumption, but there is – of course – the obligatory quote from Dr. Ambedkar that follows (paragraph 15).
Socrates was said to have observed at his trial that the unexamined life is not worth living. For the Supreme Court, however, it seems that the unexamined law is certainly worth upholding. Its order on the Election Commissioners stands as a stark example of that aphorism.
With the recent controversies over the process for the appointment of the Election Commissioner, it might be time to consider various models for independence of 4th branch, or guarantor institutions. In this regard, a possible model emerges from the arguments in the PJ Thomas case, that concerned the validity of appointments to the CVC, another 4th branch institution. In the history of the Court’s engagement with guarantor institutions (which the EC undoubtedly is), this judgement is unique because it saw the Court quashing the appointment of PJ Thomas, relying on the principle of institutional integrity. However, in this case, another argument brought forward by the petitioners (ADR, one of the petitioners in the EC petition as well) was regarding the interpretation of Section 4(1) of the then CVC Act, and to read the recommendation required by the High-Powered Committee to be unanimous (instead of by majority).
The argument of the petitioners was based on a purposive reading of the text, as well as the circumstances of the case, where the then Leader of Opposition Sushma Swaraj had already dissented over the appointment of Thomas, highlighting flaws in his candidature. However, the Court did not agree with the same. At this juncture, it is important to point out that the circumstances in the PJ Thomas judgement are very similar to our own, with respect to the dissent of the Leader of Oppositon with respect to the appointment. Further, the controversies over the appointment procedure for the Election Commissioner also reflect an impasse over different models regarding the composition of committees tasked with the appointment of persons to guarantor institutions in India. Roughly speaking, the two prominent models are:
the procedure set down in the Anoop Baranwal judgement, which created a 3-member committee consisting of the PM, CJI and the Leader of Opposition.
the procedure set down in legislation governing various 4th branch institutions, such as the CEC and the CVC.
At this juncture, it is pertinent to point out that neither model is perfect. However, in the current scenario, the main model that seems to have been articulated is for a return to the model set down by the Court in the Anoop Baranwal judgement (that is, a selection committee consisting of the Prime Minister, the Leader of Opposition, and the Chief Justice of India). This is understandable as such a suggestion has already been made by various committees, as well as the Law Commission in 2015. Not only this, a comparative overview will show that in other countries such as South Africa and South Korea, a Supreme Court justice is either on the appointment committee (South Africa) or is one of the 3 people that nominates individuals to the Election Commission. As such, the desire for the CJI to be on the committee is not surprising at all.
The Problems with the Anoop Baranwal Model
However, despite such arguments, I believe we must reflect more on the viability of such a model in India. In this regard, we must remember that the practice of including the CJI on the committee already exists for the appointment of individuals to certain positions, and it is not clear yet that such a practice has been better than alternatives. More importantly, it is also important to remember the controversies over post-judicial appointments in India, and the allegations surrounding the executive bias of the judiciary across different political eras. I do not bring these points up to endorse these claims, but rather to highlight the sorts of issues that are already affecting public perception of the judiciary, and the further partisan conflicts that the judiciary could get embroiled in if it is included in the committee for the appointment of the Election Commissioners. Further, it can be argued that the presence of the CJI on the Committee also puts the Court in an awkward position in the case of controversies regarding appointments, as past practise will attest to the Court’s ‘awkward’ handling of cases where it is required to review behavior by the CJI or other judges.
Decision-Making by Unanimity
This brings us to the second model. As is clear from the criticisms of the Bill and its provisions, the model is inadequate because it clearly gives a majority to the Executive. However, a close reading of the text shows that it leaves room for an interpretation in which the Committee, despite its composition, has to take decisions unanimously. In the PJ Thomas case, it was such a suggestion that the petitioners suggested reading into the case, based on a purposive reading of the statute. This was the suggestion that was rebutted by the Court in the PJ Thomas case. However, the text in the CEC Bill remains very similar to the text of the CVC Bill when the PJ Thomas case was adjudicated.
The only difference in the text between the CVC Bill and the CEC Bill is that the text of the CEC Bill explicitly allows the committee empowered to appoint the Commissioner to set its own procedure. This, along with the Section 7(2) of the Bill, which allows the Committee to decide regardless of a vacancy in the Committee, are the provisions that reflect legislative intent to entrench executive dominance in the Committee. An approach that the Court could take here is to interpret these provisions to require unanimity from the committee in decision-making. The Court will also have to strike down Section 7(2) of the Bill. To ensure that such a provision is not misused, the Court can also put in place a deadline for appointments by the committee after it sits down to choose an Election Commissioner (a problem that has already reared its head in such appointment committees that include the CJI). Such an interpretation is also plausible in the aftermath of the Anoop Baranwal judgement, where the logic of the Court’s judgement required setting up an appointments committee that was not dominated by the Executive.
In this regard, the Court can take inspiration from UK jurisprudence in this regard. In the UK, Section 3(1) of the Human Rights Act requires the Court to interpret legislation in a manner that is consistent with the Convention. This has led to the Court to ‘read down’ or ‘read in’ provisions that are Convention incompatible. As Kavanagh discussed in her article, this approach is now a part and parcel of the Court’s interpretation in such matters. A similar interpretive strategy could be followed in this case, and other cases of this nature. Further, in Para 55 of the PJ Thomas judgement, the Court stressed that all individuals in the committee must publicize their decisions and the rationale for the same, especially in the case of dissent. This should be required regardless of dissent, to ensure trans,parency and clarity regarding the reasons for the appointment of the Commissioners.
Final Reflections
The PJ Thomas case remains a landmark case in the history of guarantor institution litigation in India because of the Court’s articulation of the concept of institutional integrity, as well as its application of the concept to the specific facts of the case. Since the case, the jurisprudence on guarantor institutions has been further strengthened by the ruling of the Court in Anoop Baranwal. As the guarantor institution litigation in India has been piecemeal, a coherent theory might emerge upon a combined reading of the various cases, particularly Vineet Narain, where the Court highlighted the importance of the concept in the Indian context, PJ Thomas, where it articulated the importance of institutional integrity for these institutions, in their appointments and working, and Anoop Baranwal, where it combined various strands to ensure that these institutions have structural autonomy. Such a reading will also help us see how the holdings of Courts in older cases might get updated based on jurisprudence that has emerged in later cases, and that now represents the final view of the Court in such matters.
In this regard, it could be said that the holding of the PJ Thomas judgment, that did not require the 3-member committee in charge of appointing the CVC to make its decision with unanimous consent, is in need of re-evaluation, especially in light of the jurisprudence on the importance of the independence of such institutions articulated by the Court in the Anoop Baranwal case. Thus, the jurisprudence of the Court from the Anoop Baranwal case, specifically the Court’s emphasis on the importance of an election commission independent from executive dominance, should now anchor any future decisions the Court takes regarding how a 3-member committee might operate to prevent executive dominance. While this does not seem to have been the consensus from the different committees set up on this matter, I do believe this might be an acceptable middle-ground approach with multiple advantages.
First of all, as has been pointed out earlier, while putting the CJI in the committee seems like a great idea, it doesn’t seem to have been any better than existing ones in practise. By inserting the CJI into the committee, it places the Court in the centre of partisan conflict, but this time in a role that it is new to it. Further, it also places the Court in an awkward position with regards to the adjudication of any future controversies into appointments, as the CJI’s inclusion in the committee could create potential conflicts of interest that might affect the Court in its core duty as an adjudicator of such disputes, not to mention the awkward manner in which the Court has carried out enquiries in controversies that implicate its own members. Another point to consider here is the extent to which the Court’s decisions seem to have created a hierarchy in which the Court is superior relative to 4th branch institutions. The inclusion of the CJI on the committee might further entrench this hierarchy, which might not be the best approach. As such, while the idea of the inclusion of the CJI in the committee for the appointment of the Election Commissioner is not a bad idea, I believe the Court would be better served by not directly involving itself in the process, while requiring the selection to be based on unanimity, transparency, and timely consultation, values that are common threads that run through its judgements on guarantor institutions in India.
This Monday, the High Court of Kenya at Nakuru delivered a landmark judgment on freedom of speech (Katiba Institute and Ors vs The Director of Public Prosecutions and Ors). At issue was Section 77 of the Penal Code, which criminalised acts done or words spoken with a “subversive intention.” “Subversion”, in turn, was defined as prejudicing public order or security, incitement to violence, promotion of enmity or hatred between different races, spreading disaffection, and other very similar phrases.
One Joshua Otieno Ayika was prosecuted under this section for a 2022 tweet that speculated that the army might be taking over the government. In this context, the Katiba Institute – and certain other parties – filed a challenge before the High Court, contending that Section 77 was unconstitutional. The High Court agreed, and struck it down.
The judgment of the High Court is a textbook application of some of the core doctrines of free speech. Justice Mohochi begins by noting that at the time of Independence, Kenya inherited a repressive colonial system; sedition prosecutions in the Independence era marked a continuation of this system (paras 105-106). In this context, the Mohochi J notes that the 2010 Constitution is transformative, and consciously designed to mark a break with the repressive past – including the post-colonial past, which retained colonial legal structures with “very minor and cosmetic variations that were intended for self-preservation and colonial repression.” (para 110) This requires pre-existing legislation and legislative structures to be levelled up in order to comply with the new Constitution’s standards (para 110).
The High Court’s analysis then proceeds within this historical and conceptual framework. In the present dispute, it was nobody’s case that Section 77 of the Penal Code infringed the right of free speech; the only question was whether it fell within the specific excluded categories of the free speech clause under Article 33 (propaganda for war, incitement to violence, hate speech, or advocacy of hatred based on discrimination), or whether it was protected by the general justification clause of the Bill of Rights (Article 24). The Court finds that although the State had invoked Section 77, it was unable to explain which exclusionary clause in the Constitution covered the offending tweet; this, the Court notes, is a chilling reminder of how a liberal and broad interpretation on making a decision to prosecute (something, we may add, specifically enabled by the section) makes possible the abuse of such a provision. (para 127)
Indeed, Mohochi J uses this to segue into the substantive challenge itself, noting that “the meaning of “Subversive” takes in quite a variety of activities, and that its contents are therefore broad and wide that it is vague or indefinite.” (para 128) This is, of course, a variant of the familiar free speech doctrine of “void for vagueness,” and the Court echoes the primary concern of this doctrine by noting the vast latitude that it gives to prosecutors and the police. The Court then folds the doctrine into the first limb of the proportionality standard by noting that Section 77 of the Penal Code – because of its vagueness – fails the test of being “prescribed by law.” (para 137) Strikingly, the Court finds that even the clarificatory sub-sections of Section 77 suffer from the vice of vagueness:
… there exists confusing definition of “subversion” especially about the meaning of “prejudicial to public order, security of Kenya and administration of justice”, “in defiance of or disobedience to the law and lawful authority; unlawful society” or “hatred or contempt or excite disaffection against any public officer or any class of public officer”. None of the terms used in the offence are defined or capable of precise or objective legal definition or understanding. (para 137)
One important thing to note here is that a lot of terms that the High Court singles our are staple features of repressive colonial legislation across the British commonwealth, which were then adopted into post-colonial penal codes. Too often, courts in these post-colonial countries have interpreted their Constitutions in light of these terms, rather than the other way round: that is, the scope of the free speech guarantee has been defined and limited by phrases such as “disaffection” or “prejudicial to public order,” rather than these phrases being considered on the touchstone of the Constitution. It is here that the High Court’s invocation of history (both colonial and post-colonial) becomes important, because it is that history that is the basis of the Court’s doctrinal approach towards interpreting the Constitution as responding to that history.
While Section 77 therefore failed on the first prong of the proportionality test, the Mohochi J also examines it on the third prong: necessity. In an interesting analysis, he notes that the stated goals of the provision – that is, combating hate speech, or preserving State security – are fulfilled through other existing laws, with lower penalties. This allows the Court to hold that Section 77 is not the “least restrictive measure” to serve the State’s goals, given that lesser restrictive measures exist. This is particularly interesting, as we don’t often see the “least restrictive measure” prong of the proportionality test being deployed to compare existing legislative measures.
On this basis, the Court therefore finds the relevant part of Section 77 to be unconstitutional. By way of conclusion, a final point: I have written above about the Court’s analysis, but what of the facts? It must be said, putting out a tweet stating that a military coup is imminent skirts a very fine line on the boundaries of free speech: given the potential panic that such communication might cause, one can easily see a court comparing it to the proverbial “fire in a crowded theatre” test. One must, therefore, commend the Mohochi J, in particular, for not allowing (potentially) bad facts to make bad law, and by completely separating the constitutional analysis from the individual behaviour that gave rise to the constitutional claim. Experience teaches us that while Courts always claim they do this, it is not as easy as it seems!
The artificial cleavage separating civil and political rights and socio-economic rights was brought to an end in 2010, when Kenyans promulgated the Constitution, which expressly makes socio-economic rights justiciable. By doing so, Kenyans ensured that the most vulnerable, marginalised and disadvantaged people were protected, and that the state occupied the tiller to ensure the continual improvement of their conditions. However, despite the progressive measure of entrenching these rights in the Constitution, their enforcement and implementation have not been without challenges. While accepting that the debate on whether socioeconomic rights are justiciable is old and well-worn, the state (like other governments) has consistently deployed the argument that socioeconomic rights are subject to progressive realisation, and to the availability of resources.
This was the case in a recent decision of the Court of Appeal in the County Government of Bungoma and others v Josephine Oundo Ongwen and others (2024) eKLR, where the state argued that since the right to maternal health is a socio-economic right, it is to be realised progressively and subject to the availability of funds. Had the Court accepted this argument, it was expected to, like Pilate, wash its hands and leave the rights to appear like ‘defected footballers who have lost a match’. However, the Court of Appeal did the precise opposite: in an innovative judgment, Justice Joel Ngugi employed the “minimum core obligation” concept to hold that notwithstanding socio-economic rights being subject to progressive realisation requirements, rights contain minimum obligations which are realisable immediately.
Setting the scene: Dereliction of duty and undignifying treatment by the state
On 8 August 2013, Josephine, a woman from a marginalised socioeconomic setting, went to one of the district hospitals for childbirth. Upon arrival at the hospital, she was admitted, and she informed the doctors that she was overdue. The doctor informed her that she would have to undergo induced labour. Due to the space constraints at the hospital, she was forced to share a bed with another patient. As if this was not bad enough, the nurses informed her that once the labour pains started, she would have to walk to the delivery room. She was left unattended. When the labour pains started, she attempted to walk to the delivery room, but unfortunately, she had to deliver on the floor along the corridors of the hospital.
This, however, was not the end of the ordeal. When she was giving birth, she lost consciousness and was woken up by shouts and physical and verbal assaults from two nurses who were displeased with the fact that she delivered on the floor. Josephine was then forced to carry her placenta and walk to the delivery room to have it expelled. She approached the High Court and awarded her Kshs 2.5M for violation of her right to maternal health, among other rights. Dissatisfied with this judgment, the state moved to the Court of Appeal.
The context for this case is, however, critical. In Kenya, three in four women report that they have been subjected to at least one form of obstetric violence during labour and delivery. Pregnant women have been mistreated at the hands of the healthcare workers. As a result of this background, one of the members of parliament has introduced a bill. While introducing the bill, the member of parliament stated that:
Across delivery rooms and maternity wards in Kenya, many women remain similarly unassisted and neglected as they struggle through childbirth; it is even worse if it involves women with disability. This is gender-based violence and needs to stop immediately.
The importance of this case is better expressed if it is understood with this context in mind.
A case for minimum core obligations in Kenya
The Committee on Economic, Social and Cultural Rights (CESCR) articulated the concept of minimum core obligation in General Comment 3 in para 10. The committee formulated this requirement as follows;
The committee is of the view that a minimum core obligation to ensure the satisfaction of, at the very least, minimum essential levels of each of the rights is incumbent upon every State party……………….. If the Covenant were to be read in such a way as not to establish such a minimum core obligation, it would be largely deprived of its raison d’être.
In simple terms, the concept seeks to provide the ’bare level ground’ for socioeconomic rights. In doing so, the concept sets ‘a quantitative and qualitative floor of socio-economic and cultural rights that must be immediately realised by the state as a matter of top priority’. The concept can further be broken down into three principles. The first is immediacy: the principle requires that the content identified must be realised immediately and not subject to progressive realisation; secondly, unique content: the identified content must be so peculiar and of high priority, without which the right loses its substantive significance as a human right; third, non-derogability: the principle requires that the identified content is non-derogable.
Courts and judges are called upon to deploy human rights language in adjudicating human rights. While considering socio-economic rights, courts must look for a proper doctrine. My assessment of the jurisprudence emanating from the Kenyan courts is that judges have failed to correctly point us to the standard of review that Kenya has chosen. However, in most decisions, the courts seem to have adopted the reasonableness test, with the Supreme Court choosing to remain silent in Mitu-Bell.
In this case, however, Justice Joel Ngugi was not shy of finding that the case involved the right to maternal health. The next obstacle was to deal with the state’s case that the hospitals are overstretched and the right should be subject to progressive realisation. In dismissing this obstacle, the Ngugi J observed that the right to maternal health has a “minimum core” that is realisable immediately and not subject to progressive realisation. He held:
It is not, at all, contested that under our Constitution, every woman is entitled to respectful maternal care during childbirth as part of their social and economic rights enshrined in Article 43 of the Constitution. That aspect of the right to health is not subject to progressive realization. It is part of the minimum core of the right that must be realizable immediately and not progressively.
Ngugi J further identified the minimum core of the right to maternal health to include:
a. The right to be free from physical violence and verbal abuse during labour and childbirth;
b. The right to be free from discrimination during labour and childbirth;
c. The right to a dignified and respectful care – including being granted acceptable levels of privacy and confidentiality during labour and childbirth.
He further underscored the point that whereas the state can argue the non-availability of resources with regards to the availability of drugs, hospital beds and even shortage of medical personnel, the ‘human rights-based maternity care commanded by a purposive reading of Article 43 of the Constitution’ includes the right of women to ‘dignified, respectful health care throughout pregnancy and childbirth as well as freedom from violence and discrimination’: and that this is realizable immediately. (see para 32).
Although the Judge did not spend much time on whether this minimum core obligation content test is applicable in Kenya, it is easy to answer. First, article 24 of the Constitution provides a general limitation clause asking judges to consider the proportionality standard. Secondly, the provision further requires that ‘[A] provision in legislation limiting a right or fundamental freedom shall not limit the right or fundamental freedom so far as to derogate from its core or essential content’. Article 20(5) also requires the state to prioritise delivering socio-economic rights. Therefore, even if it were to be argued that the position adopted by the committee is not binding (which is wrong), the concept of minimum core obligation has already been endorsed by the Constitution of Kenya (see Orago on this).
Articles 259 and 20(3) of the Constitution lay down the canons of interpretation of rights. Apart from requiring that the Constitution be interpreted liberally and purposively while giving effect to values and principles, the Constitution demands that the idea is that of maximisation of rights, and not minimalism. Notably, the Constitution imposes an injunction on judges to adopt the interpretation that most favors the enforcement of a right or fundamental freedom. Additionally, the Constitution requires judges to continuously develop the law to bring it in line with the Constitution. This rejects any dream of judicial timidity or rigidity but calls for an interpretation that enforces the letter of the Constitution and the aura of the Constitution. In accepting to use the test adopted by the committee on economic and social rights, Justice Joel Ngugi, it is submitted, correctly appreciated the injunction imposed on him as a judge by the Constitution. .
Conclusion
This decision is welcome for various reasons. First, the Constitution’s quest to alleviate the dire poverty, economic inequalities and marginalisation in Kenya can only be achieved if judges are willing to appreciate the language of the Constitution and properly mine an appropriate approach to adjudicating socio-economic rights. Second, the decision revitalises the right to maternal health in Kenya by reiterating the minimum obligations of the state. Third, the test adopted by the court of appeal leads to a more stringent scrutiny of the state’s arguments on the non-implementation of socioeconomic rights. Put differently, it leads to jurisprudence of exasperation, which the Constitution has been loudly crying out for.
Fourth and importantly, the decision by the Court of Appeal can be argued to be one of the very first progressive decisions of that specific court in Kenya’s history of adjudication of socio-economic rights. The Court of Appeal has had a rough start with judgments such as Mitu-Bell and Musembi, hence earning the term the “graveyard of jurisprudence, where jurisprudence dies.” For the first time, Justice Joel Ngugi promises Kenyans that the Court of Appeal might be the new Shiloh of jurisprudence, the sanctuary of jurisprudence, instead.
Constitutional Law and Philosophy 