How Can Malaysian Courts Consistently Perform Meaningful Constitutional Rights Review? Lessons from Past Cases and the Way Forward

Abstract

In the past, Malaysian courts performing constitutional rights review played a merely clerical role, applying a test that was trivially easy for legislation to pass. Then a more rigorous proportionality test took root. However, the Federal Court in the 2020 case of Letitia Bosman whittled the test down again, and the courts once more played a minimal role in checking state action. The reasons for this cannot be explained merely by diversity in judicial philosophy or political contextual factors. Rather, the near-demise of proportionality (and, with it, robust constitutional review) was made possible by a lack of a clear sense of the doctrinal foundations of proportionality (and, indeed, of constitutional rights review generally), and the relative roles of the courts and the legislature therein. As a result, there is a risk that the courts’ important role in safeguarding constitutional rights has been minimised to near vanishing point. This article aims, through an analysis of the case law and its foundations, to explain how this came to be, and hence highlight important issues which Malaysian constitutional law must grapple with if meaningful rights review is to take place.

Introduction

The Federal Constitution of Malaysia sets out various fundamental liberties¹ with which all state action must be compatible.² But the Federal Constitution does not clearly state a test for constitutionality. ³ It provides several examples of what the test is not,⁴ but does not tell us what the test is.

Following a ‘tortuous’ process of development,⁵ the Malaysian courts embraced proportionality as the dominant test for compliance with constitutional rights. Under this test, legislation that engages a constitutional right is only constitutionally valid if

(i) the legislative objective is sufficiently important to justify limiting a fundamental right; (ii) the measures designed to meet the legislative objective are rationally connected to it; and (iii) the means used to impair the right or freedom are no more than is necessary to accomplish the objective.⁶

According to this writer's chapter in a 2020 book on Proportionality in Asia, while the status of the proportionality doctrine was initially unclear, the Federal Court (Malaysia's highest court) eventually ‘ma[d]e abundantly clear … that proportionality is to be applied rigorously and not merely given lip service’.⁷ This took place in the 2019 case of Alma Nudo Atenza where the court held that ‘when any State action is challenged as violating a fundamental right … the action must meet the test of proportionality’.⁸ However, as for lower courts, this writer wrote that ‘only time will tell whether the proportionality doctrine in Malaysia is doomed to a hopeless cycle in which the Federal Court lays down authoritative guidance which the lower courts then misunderstand, misapply, or simply ignore.’⁹

By the time the book had been published, the state of the law had become the opposite of what this writer said it was. While lower courts came to embrace proportionality analysis, the Federal Court, in Letitia Bosman,¹⁰ turned its back on proportionality in favour of a doctrine of extreme permissiveness to state authorities. Subsequent Federal Court decisions warrant, at best, a sense of cautious optimism for the future of constitutional rights review in Malaysia.

This volte-face is so striking that it needs to be chronicled and explained. But it also prompts reflection on deeper questions. Is the change a bad thing? One might ask:¹¹ why does it matter whether the Malaysian courts embrace proportionality analysis or not? Indeed, one theme of the 2020 book was that, while proportionality is the ‘the most ubiquitous legal doctrine relied upon by judges in rights-adjudication’,¹² it is by no means the only possible doctrine. There are jurisdictions, such as Bangladesh and the Philippines, in which the courts apply proportionality analysis in substance even though not in name.¹³ There could also be other means of paying heed to constitutional rights, as in Singapore (a useful comparator given textual similarities between rights provisions in the Singaporean and Malaysian Constitutions) where the courts ask whether the state had ‘due regard’ for constitutional rights.¹⁴

This article is concerned not with which of these means Malaysian law should adopt, but rather with the perils of not adopting any substantial means of constitutional rights review – despite the pride of place that Malaysian law has (at least in some cases) given to proportionality. The problem is this: if proportionality has been swept away and nothing has emerged to replace it, then Malaysian law is left with no system of constitutional rights adjudication that is fit for purpose. Instead, the impression one often gets is that the courts can play a merely minimal role, with little meaningful checks on exercises of state power that burden constitutional rights.

I aim to explore how this has happened and, in turn, how the law may be improved. My aim will not be to produce a complete proposal as to how constitutional rights adjudication in Malaysia should take place. My aim is simply to demonstrate that it has not taken place in a consistently robust manner, and explain why this is so, with a view to aiding future conversations on how the law may be improved.

After remarking on my methodology in more detail, I aim to critically chronicle the development of constitutional rights review in Malaysian law up to the high-water mark of proportionality in Alma Nudo Atenza and subsequent decisions of the High Courts (Malaysia's first-instance superior courts) that followed suit. The focus will be on proportionality, not because that is necessarily the best means of rights adjudication, but because it is the one on which Malaysian courts have most recently concentrated. As we will see, while it may appear that proportionality had at least been firmly entrenched in Malaysian law, there were lingering problems; chief among these is the lack of a clear justification for proportionality analysis. This, in turn, reveals a lack of a clear theoretical sense of constitutional rights review generally, and the relative roles of the courts and the legislature therein.

These problems reared their head in the Federal Court's decision in Letitia Bosman, which, while not denying the existence of the proportionality doctrine, at best denuded it of all force and, at worst, side-stepped it altogether. The result was a form of rights review that reduced the courts’ role to a merely clerical one, asking whether an act that engages constitutional rights was enacted by Parliament in the proper form. After examining how and why this turn took place, I will examine subsequent Federal Court jurisprudence in the penultimate section, which demonstrates that rights review in Malaysia faces a quite uncertain future because of the problems mentioned earlier.

Some notes on methodology

Malaysian constitutional rights and rights adjudication

As we have seen above, proportionality is not the only method of rights adjudication. But proportionality is important in Malaysian constitutional law – and, hence, it is the lens through which we will conduct much of the analysis in this article – because it has been touted as the prime method. The Federal Court has called it ‘an essential requirement of any legitimate limitation of an entrenched right’.¹⁵ To illustrate why this is so, it will be useful to discuss the rights provisions in the Federal Constitution and then catalogue the various tests for compatibility with constitutional rights provisions other than proportionality.

With a few exceptions,¹⁶ rights provisions in the Federal Constitution have a two-part format: they set out a right, then permit state interference with the right, but only on certain grounds (for example, ‘Parliament may by law impose’ on the freedom of association ‘such restrictions as it deems necessary or expedient in the interest of the security of the Federation or any part thereof, public order or morality’)¹⁷ or in a certain manner (for example, ‘No person shall be deprived of his life or personal liberty save in accordance with law.’)¹⁸

Faced with such rights provisions, the Malaysian courts could conceivably apply one of four possible approaches. The first, and weakest, involves merely asking whether a law has been passed by the Legislature in accordance with certain criteria of formal validity (for example, whether royal assent has been given).¹⁹ The problem with this approach is clear: it reduces the courts to playing a merely clerical role of ensuring that the right number of votes were cast in Parliament and the right signature appears on the bill. Moreover, it would make a mockery of the fundamental liberties provisions since their very purpose is to impose legal limits on legislative action beyond merely formal ones – otherwise, the fundamental liberties provisions might as well not exist.

Second, the courts could ask whether Parliament, in its deliberations, had regard to the need to uphold the fundamental liberties. After all, Members of Parliament are oath-bound to ‘preserve, protect and defend’ the Federal Constitution.²⁰ The problem is that the Federal Constitution expressly forecloses many such arguments. For example, the ‘validity of any law shall not be questioned’ merely on the ground that, say, it engages the Article 10(1) rights to freedom of speech, assembly, or association, but the restrictions it imposes on that freedom ‘were not deemed necessary or expedient by Parliament for the purposes mentioned in that Article’.²¹ Clearly, the courts are forbidden from attempting to read Parliament's mind or peer into its thought process. Besides, given the aforementioned oath, it would be nearly impossible to prove that Parliament has not directed its mind to the fundamental liberties. Surely, if Parliament made such an omission, it would at least take steps to give the impression that it did.

Third, the courts could ask what the subject-matter of the legislation is. For example, they could say that a law that engages the freedom of expression is constitutionally valid if, objectively speaking, it has some link with the ‘security of the Federation’, ‘public order’, or ‘morality’. However, there are two problems with this approach. First, such an approach is not applicable to certain rights such as in Article 5(1) of the Federal Constitution, which define permissible state restrictions not in terms of their subject-matter, but instead more vaguely in terms of whether they are ‘in accordance with law’. Second, the stated categories of subject-matter are very broad indeed; it is not at all difficult to make out some relation with one of those categories. For example, according to the courts, the term ‘public order’ encompasses everything from ‘danger to human life and safety’ to the ‘disturbance of public tranquillity’ to ‘public health’,²² as long as it pertains to the ‘even tempo of the life of the community taking the country as a whole or even a specified locality’.²³ Even an act affecting ‘merely an individual leaving the tranquillity of the society undisturbed’ can fall within the ambit of ‘public order’, if it ultimately has some effect (even if indirect) on society generally.²⁴

Fourth, the court could ask about the effect of the legislation in question on fundamental liberties. Various flavours of this approach have emerged in Malaysian law. One old approach (which has since fallen into disuse) involves asking whether legislation makes the exercise of a constitutional right ‘ineffective or illusory’²⁵ Another is the proportionality test.

These are not necessarily the only possibilities. In particular, an additional possibility is that the courts could apply a relatively permissive approach when reviewing legislation, but apply a stricter standard when reviewing the constitutionality of executive action taken pursuant to that legislation. For example, the courts could hold that any law requiring a police permit for a public assembly to be held is constitutionally valid, but then subject individual decisions by the police to withhold (or impose conditions on) a permit to stricter analysis. However, it appears that the Malaysian courts have yet to explore the possibility of such an approach.

Why focus on proportionality?

So there is a diversity of possible approaches toward rights adjudication (even after discounting the patently incorrect ones). Nonetheless, proportionality deserves special attention for several reasons.

First, the Malaysian courts have drawn special attention to it. Even when they have arguably not performed proportionality analysis in substance, they have often at least seen the need to purport to do so.

Second, a positive case can be made in favour of proportionality analysis. Proportionality provides a structure – as Stone Sweet and Mathews put it, an ‘analytical procedure’²⁶ – through which ‘counsel [can] sequence their arguments’²⁷ and courts can not only justify individual decisions, but also ‘rationalize and defend rights review’ more generally.²⁸ Moreover, this structure is particularly suited to constitutional rights adjudication, which very often requires a court to balance, in a coherent manner, a right and the grounds on which the state is entitled to limit the right.²⁹

At this point, caution is needed. Some may also seek to justify proportionality analysis on normative grounds. Indeed, there are some normative benefits which are inherent in any flavour of proportionality analysis, such as the idea that proportionality creates a ‘culture of justification’³⁰ according to which ‘every exercise of power is expected to be justified’.³¹ As a matter of principle, it is not difficult to disagree,³² but others have proffered normative justifications for proportionality such as that it promotes the value of (a particular conception of) the ‘individuality and the dignity of the human being’.³³ These create a particular danger, because such values are not universal (or universally conceived of in the same way). If one were to seek to justify transplanting a proportionality doctrine into a given jurisdiction on the basis of such values, one could easily be rebuffed by pointing out such differences in values.

Nonetheless, while proportionality – like any exercise in constitutional rights adjudication – is not value-neutral, it does not necessarily import a particular set of values. As Yap points out, proportionality is a ‘doctrinal construction and an analytical procedure’, but it ‘does not – in itself – produce substantive outcomes’ because everything comes down to judges’ particular ‘choices’ about how to perform proportionality analysis.³⁴ In addition, to the extent that proportionality is necessarily value-laden, those values are simply those drawn from the Constitution itself.³⁵

Therefore, this writer, like Stone Sweet and Mathews, does not see proportionality as a ‘miracle cure-all that will make hard constitutional questions easier to answer’.³⁶ Neither is it this writer's view that proportionality analysis is desirable merely because the state is (to put it crudely) less likely to win. There is nothing intrinsically right about state action being found constitutionally valid. Indeed, we should be worried if it is not.³⁷ But what is important is that, if the state wins, it does so because the court has scrutinised its reasoning and/or conclusions and found them adequate, and not merely because some merely formal test has been passed.

That said, proportionality has become a locus of contemporary debates about constitutional adjudication in Malaysia. This has led to problems, because courts which do not favour proportionality analysis are left without a robust method of rights adjudication. In some cases, the courts incant the word ‘proportionality’, but their analysis collapses into the first, second, or third of the methods of rights adjudication mentioned above – which, for reasons already explored, is problematic because it risks reducing the judicial role to the vanishing point. In others, the courts have, in abandoning proportionality (whether in substance or in name), thrown the baby out with the bath-water and failed to perform any meaningful review of legislation vis-à-vis constitutional rights.

While there would in principle be nothing wrong with abolishing proportionality analysis, proponents of removing proportionality analysis risk being seen as undermining these worthy aims unless they provide a viable alternative means to achieve them.

On a more prosaic note, the value of consistency demands that any judicial departure from established doctrine be justified, or, at the minimum, recognised as a departure. Perhaps proportionality exists as a legal doctrine, or perhaps it does not. It may even be that proportionality analysis applies in certain cases but not others, or applies with a variable level of intensity depending on certain factors. But the rule of law would suffer if – as is presently the state of play in Malaysian law – it is not even possible to state with certainty the law surrounding constitutional adjudication. As the Federal Court put it in 2012: ‘It is of supreme importance that people may know with certainty what the law is … Little respect will be paid to our judgments if we overthrow that one day which we have resolved the day before. “We cannot say that the law was one thing yesterday but is to be something different tomorrow.”’³⁸

Possible attempts to explain inconsistency

One might respond³⁹ that what appears to be inconsistency in judicial decisions, or a failure to perform proportionality analysis in anything more than name, is merely the manifestation of a diversity of judicial philosophies toward rights adjudication. Of course, judges do not all think the same way, and indeed the law would suffer if they did. But the value of judicial disagreement lies in the fact that that disagreement is reasoned, and not merely asserted. Unfortunately, it appears that there are not even agreed ground rules or principles as to the very point of constitutional review, and when the courts have eroded rights adjudication, they have not even squarely confronted that this is the result of their doctrinal developments.

One might also speculate on possible reasons why the proportionality doctrine might or might not apply in certain categories of cases depending on the subject-matter. This writer would decline to do so. It would, of course, be possible to attempt to spot patterns, and perhaps even to isolate them from other variables such as judicial philosophy. Nonetheless, this article focuses on legal doctrine as articulated in the cases, and not so much on possible reasons underlying the articulation of the conclusion. After all, if one were to speculate that the courts are reluctant to apply proportionality in (say) cases involving politicians, one may as well speculate that the courts would have an incentive to disguise the bias (for such speculation is, in substance, an allegation of bias) by cloaking it in the language of proportionality. Therefore, this article takes the case law at its highest, presuming that each judgment represents the judge's best attempt to decide each case in accordance with the product of the best attempt to say what the law is, and, where appropriate, critiques the judgment on that basis.⁴⁰

Of course, this is not to deny the impact that contextual factors can have on the case law. The literature identifies the relevant context as the lingering spectre of the 1988 Malaysian judiciary crisis.⁴¹ The crisis began when several judges were impeached and removed from office after the then-Lord President of the Supreme Court of Malaysia, Tun Salleh Abas, wrote to the Yang di-Pertuan Agong (the King) expressing concern about comments made by Prime Minister Dr Mahathir Mohamad about the judiciary.⁴² The Prime Minister retaliated by initiating the process to have the Lord President removed from office.⁴³ Five judges granted the Lord President's application for an order restraining the removal process; they were suspended and two were eventually removed from office. The Lord President himself was eventually removed. These, says Sinnadurai, ‘were the most shocking incidents to affect the judiciary … the safeguard against executive interference on the judiciary were destroyed’.⁴⁴

One might suppose that courts would decide cases involving certain topics with this incident in mind. But the evidence does not support that theory. The recent case in which a majority of the Federal Court repudiated proportionality in all but name – Letitia Bosman – involved drug trafficking. Even if one calls this a politically sensitive area, one must note that the next most recent case in which the Federal Court embraced proportionality analysis – Alma Nudo Atenza – was also about drug trafficking. Even in cases involving political rights, such as the freedom of (political) expression, the courts do not uniformly refuse to engage with proportionality. For example, in Hilman – which concerned a ban on university students from ‘expressing support for or sympathy with or opposition to… any political party’ – the majority considered proportionality while the minority did not.⁴⁵ In Nik Nazmi, another case involving a political assembly, the Court of Appeal unanimously ruled in the applicants’ favour and against the state.⁴⁶

In short, contextual factors do not fully explain why the Malaysian courts have ‘t[aken] an unsteady approach to constitutional adjudication’, sometimes ‘assertive’ or ‘empowered’, and sometimes doing an ‘about-face’.⁴⁷ Therefore, I will undertake a primarily doctrinal approach instead, focusing on the courts’ reasoning and whether it stands up to scrutiny.

Choppy waters: rights review in Malaysia before Alma Nudo Atenza

Before proportionality: little meaningful check on state action

There was a time when the Malaysian courts read the fundamental rights provisions as creating mostly formal, and rarely substantive, constraints on legislative power.

Take, for example, the freedom of expression. Article 10(1)(a) of the Federal Constitution provides that ‘every citizen has the right to freedom of speech and expression’, but Article 10(2)(a) states that ‘Parliament may by law impose’ on this right ‘such restrictions as it deems necessary or expedient’ for various purposes (such as ‘public order’ and ‘morality’). In the 1994 case of Pung Cheng Choon, the Supreme Court (as the Federal Court was then known) held that ‘the scope of the court's inquiry is limited to the question whether the impugned law comes within the orbit of the permitted restrictions’.⁴⁸ The court inquired into whether the law (ostensibly) had some link with ‘public order’ or ‘morality’. But the court would not go further and ask whether the law was reasonable or proportionate. The courts in later cases followed suit, and therefore their role in constitutional rights review was often merely clerical.

Moreover, as we have seen, Parliament could even omit to direct its mind to whether that state interest is really engaged. For example, the Federal Constitution expressly states that a law may not be struck down as violating the rights to freedom of speech, assembly, or association merely because Parliament did not really consider it ‘necessary or expedient’ to pass that law.⁴⁹

The first and second waves of cases: ‘proportionality’ put in place, but built on foundations of sand

Then came the 1996 case of Tan Tek Seng,⁵⁰ where the Court of Appeal (Malaysia's intermediate superior court, between the High Courts and the Federal Court) mentioned proportionality for the first time. This began the first of what this writer has called five ‘waves’ of cases. In the first two ‘waves’, the courts developed what appears to be proportionality doctrine and cemented it in Malaysian constitutional law.⁵¹ But after the judge largely responsible for these developments, Justice Gopal Sri Ram, retired, it became clear that the foundations of proportionality analysis were not as solid as they first appeared.⁵² In the third ‘wave’ of cases, ‘[s]ome Court of Appeal cases simply ignored Federal Court jurisprudence on proportionality; some purported to apply proportionality but failed to do so properly; others applied proportionality reasoning in a more rigorous manner.’⁵³

This writer identified two main problems. First, the textual basis for proportionality analysis was unclear, and it was not clear precisely which constitutional rights were amenable to proportionality analysis. Second, even when the case law purported to apply a proportionality test, it did not always do so clearly.⁵⁴

There is a third problem. A close examination of the cases discloses little reasoning as to why any doctrine of proportionality should apply. There was, to be sure, the statement that Parliament cannot possibly be ‘free to impose any restriction [on constitutional rights] however unreasonable that restriction may be’.⁵⁵ But why should proportionality be the solution? The proportionality doctrine had no basis in either authority, principle, or the constitutional text; the cases merely asserted that it applied. This, in turn, gave rise to two more problems. First, in later waves of cases, the courts could sweep away proportionality (and, for reasons that will be explained later, other forms of substantive rights review) by exposing that proportionality rested on foundations of sand. Second, even when proportionality analysis does apply, because it is not clear why it does, the courts lack a clear sense of principle as to how to apply it.

Consider Tan Tek Seng, in which Gopal Sri Ram JCA attached importance to the ‘all pervading’⁵⁶ equality clause in Article 8(1) of the Federal Constitution. He asserted that the main principle is this:

When the constitutionality of State action; be it legislative … or administrative; is called into question on the ground that it infringes a fundamental right, the test to be applied is, whether that action directly affects the fundamental rights guaranteed by the Federal Constitution, or that its inevitable effect or consequence on the fundamental rights is such that it makes their exercise ineffective or illusory.⁵⁷

In support of this, Gopal Sri Ram JCA cited an older Supreme Court case, Nordin,⁵⁸ which in turn cited an Indian case called Mian Bashir,⁵⁹ which in turn cited another Indian case called Maneka Gandhi.⁶⁰

However, Gopal Sri Ram JCA then went on to cite Maneka Gandhi for another proposition, namely, that Article 14 of the Indian Constitution (equivalent to Malaysia's Article 5(1) of the Federal Constitution) contains a ‘great equalizing principle’ that should be interpreted not in a ‘narrow, pedantic or lexicographic’ manner, but rather having regard to the ‘all-embracing scope and meaning’ of equality. This, in turn, meant that ‘Art 14 strikes at arbitrariness in State action and ensures fairness and equality of treatment’, which in turn demanded ‘reasonableness’.⁶¹

So it appears that Gopal Sri Ram JCA – then a Court of Appeal judge, bound by the Supreme Court's decisions – treated the decision in Nordin as importing everything that Maneka Gandhi said into Malaysian law: ‘By reason of the decision of our Supreme Court in Nordin's case I do not think it is open to me to ignore the new approach to the construction of art 8(1).’⁶² The problem is that the Supreme Court in Nordin said no such thing. It made no reference to Article 8 of the Federal Constitution, nor to ‘equality’ or the prevention of ‘arbitrary’ state action.

In other words, in asserting that the Federal Constitution demanded that state action be proportionate, the Court of Appeal in Tan Tek Seng made a bare appeal to authority. However, the authority did not even support the principle that the Court of Appeal said it did. Maneka Gandhi did not speak of proportionality, save to say that extreme disproportionality could mean that a right was for all intends and purposes ‘abridge[d] or take[n] away’.⁶³ This is not Nyambirai/de Freitas-style proportionality either in form or in substance. Nordin, meanwhile, did not mention proportionality of any sort at all.

In short, Tan Tek Seng provided neither a normative basis for proportionality, nor even a basis for it grounded in stare decisis. Neither did the next case, Nasir.⁶⁴ There, Gopal Sri Ram JCA claimed that another Court of Appeal judge had applied a ‘principle of substantive proportionality’ in another case, Menara PanGlobal.⁶⁵ That case concerned a challenge to a decision of the Industrial Court, which held that an employer was justified in dismissing its employee. The Court of Appeal did not apply any proportionality test. Instead, it applied a test of irrationality. The Court of Appeal held that the decision to dismiss the employee was not ‘irrational’ because it was ‘proportionate’.⁶⁶ It does not follow that a decision may be struck down on the ground of disproportionality, because not all disproportionality necessarily amounts to irrationality.

Gopal Sri Ram JCA also cited an Indian case, Om Kumar. There, the Indian Supreme Court remarked obiter that Article 14 of the Indian Constitution calls for a proportionality test.⁶⁷ But it only said that proportionality is called for (a) when a constitutional right that expressly allows only ‘reasonable’ restrictions on the right is engaged; or (b) when the right engaged is the right to equality. By contrast, the right at stake in Nasir was not the right to equality, and no Malaysian constitutional rights provision provides that the state may only burden the right with ‘reasonable’ restrictions.

In short, all that Nasir boils down to is a bare assertion that restrictions on rights must not be rendered ‘illusory’; must be ‘reasonable’, and hence must be ‘proportionate’.⁶⁸ What is lacking is an explanation of how these concepts – only the first of which finds grounding in authority – relate to each other. And, as Ong pointed out earlier, it is not clear how the Court of Appeal's analysis in Nasir amounts to proportionality analysis.

In short, the early cases which claim to be about proportionality, on closer examination, only say that the state may not completely nullify a constitutional right. The ‘foundations’ which later cases were built on, then, were made of sand. The first-wave cases disclose no basis for the proportionality doctrine in the constitutional test, authority, or principle.

The same may be said of the ‘second wave’ of cases, which cite nothing more than the ‘first wave’ of cases (and cases citing those cases) as authority for the proportionality test. There is, at most, an assertion that proportionality applies by virtue of Article 8(1) of the Federal Constitution,⁶⁹ backed up by an equation – without support – of the concepts of ‘equality’, ‘reasonableness’, the prevention of ‘arbitrariness’, and proportionality. To the extent that these assertions are supported by foreign case law, that case law may be distinguished. For example, in Sivarasa, the Federal Court cited the de Freitas case,⁷⁰ which hung the proportionality test on the peg of a constitutional provision that restrictions on a right had to be ‘reasonably justifiable in a democratic society’⁷¹ – words that do not appear in the Malaysian Federal Constitution.

The third wave: again, little meaningful check on state action

These problems are not merely academic. The key issue was that the proportionality test rested on foundations of sand – on assertions, not on arguments of principle, nor even on sound appeals to well-established Malaysian authority. Meanwhile, there had emerged no sound alternative to anchor constitutional rights adjudication in the event that the sand gave way. As a result, when the ‘third wave’ of cases – decided after the architect of the first two waves, Justice Gopal Sri Ram, retired – washed the sand away, Malaysian law was left with no established robust approach to constitutional adjudication.

Recall that the proportionality doctrine was based on the assumption that ‘reasonable’ means ‘proportionate’. But as the Court of Appeal recognised in Yuneswaran,⁷² Malaysian constitutional rights provisions do not even call for ‘reasonableness’ analysis. Meanwhile, the old ‘ineffective or illusory’ test had withered away, having been obscured by the first- and second-wave cases on ‘proportionality’. The upshot was that – as this writer put it – ‘legislation is proportionate as long as it is passed by Parliament and has (or purports to have) some legitimate purpose’.⁷³ The court's role was, once again, reduced to a merely clerical one.

Yuneswaran demonstrates the dangers of a court laying down a test (say, proportionality) without foundation. When another court exposes the lack of foundation, it over-corrects against what it sees as a lack of fidelity to the constitutional text by instead applying an extreme textualist view, shorn of attention to matters of principle such as the need for the judiciary to play a meaningful role in constitutional rights review and not merely a clerical one. In Yuneswaran, the Court of Appeal did just that: it swept away not only the proportionality test, but also other possible tests. So, while the fourth wave of cases ‘rehabilitate[d]’ the proportionality doctrine,⁷⁴ the cases did little more than to cite the cases in the first two waves, flaws and all. In other words, all the Federal Court did was – again – to assert that proportionality analysis was to apply, without explaining why. The foundations laid were, once again, only foundations of sand.

The fourth wave: proportionality, but why?

At this point, one might think that a proportionality test was not necessary, and perhaps Malaysian law should focus on developing alternative approaches to constitutional rights adjudication⁷⁵ (such as, perhaps, the old ‘ineffective or illusory’ test). Such a test might not go as far as proportionality, but it would still allow the courts to play more than a merely clerical role. But if such a test is to apply, then the word ‘proportionality’ should not feature in the case law at all because it would obscure the real issue. Judges opposed to proportionality could, like the court in Yuneswaran, over-correct by applying a test that entails no meaningful constitutional review.

But the word ‘proportionality’ reared its head again in the ‘fourth wave’ of cases. The courts insisted on applying a proportionality doctrine, founded on the authority of the first and second waves of cases. For example, Azmi bin Sharom, the first of the fourth-wave cases, cited the first-wave case of Nasir and the second-wave case of Sivarasa as authority for the proportionality test.⁷⁶ The problem is – as we have seen – those cases themselves do not disclose any textual, normative, or precedential foundation for proportionality. Neither does Azmi itself.

In the next fourth-wave case, Gan Boon Aun,⁷⁷ the Federal Court considered whether a derogation from the presumption of innocence was proportionate. This time, the Federal Court was on firmer ground, because it tapped on long-established strands of case law, reasoned in a detailed manner, relating to the presumption of innocence. The problem, however, is that this reasoning was specific to the presumption of innocence, and was not a justification for proportionality analysis generally.

So this writer was not quite right to say that Azmi and Gan should have made it ‘abundantly clear to the lower courts that they must apply proportionality reasoning’ and to do so ‘in substance rather than merely in form’.⁷⁸ Except in a narrow category of cases (namely, those involving the right to presumption of innocence), the fourth-wave cases – like the first- and second-wave ones – did not disclose any clear reason why proportionality analysis should apply. Azmi is particularly problematic because it asserted that a proportionality test is to apply without explaining that that test is better than the alternatives. Nor was there any clearly articulated theory explaining why other styles of constitutional adjudication are deficient.

The Alma Nudo Atenza case: problems bubbling below the surface

We see these problems bubbling below the surface in the ‘fifth wave’ of cases, consisting of the Court of Appeal's and the Federal Court's decisions in Alma Nudo Atenza. Although the Federal Court eventually overruled the Court of Appeal, the Court of Appeal's decision is worth studying because it foreshadows the problems that became apparent in later cases.

By the time of the Court of Appeal's decision, the Federal Court had not articulated a textual basis for proportionality analysis other than that – as the second-wave cases put it – proportionality is ‘housed’ in Article 8(1) of the Federal Constitution. That is a strained reading of Article 8(1). At most, the Indian cases on the Indian equivalent of Article 8(1) only say that unequal treatment must be proportionate,⁷⁹ not that any incursion on a constitutional right must be proportionate.⁸⁰

But let us assume that the Federal Constitution does not foreclose proportionality analysis; after all, there are theories of constitutional interpretation that go beyond the text, and would justify applying proportionality analysis even though it is not mentioned explicitly in the Federal Constitution. Let us also assume that the Federal Court's statement in Alma Nudo Atenza that ‘when any State action is challenged as violating a fundamental right … the action must meet the test of proportionality’⁸¹ is a sufficiently clear precedential basis for proportionality. Indeed, even in Letitia Bosman – the contemporary nadir of constitutional review – the majority of the Federal Court purported to apply proportionality analysis.

The problem is that proportionality continued to lack a normative basis – particularly, a theory of the role of the various branches of the state in constitutional adjudication. The Federal Court's decision in Alma Nudo Atenza, having rightly disavowed a doctrine of parliamentary sovereignty in favour of constitutional supremacy,⁸² left the courts with plenty of guidance about their role in proportionality analysis. But the courts still had little to no guidance on the Legislature's role. So the status of dicta in previous cases, like the following extract from Loh Kooi Choon, remained unclear:

The question whether the impugned Act is ‘harsh and unjust’ is a question of policy to be debated and decided by Parliament, and therefore not meet for judicial determination. To sustain it would cut very deeply into the very being of Parliament. Our courts ought not to enter this political thicket, even in such a worthwhile cause as the fundamental rights guaranteed by the Constitution …⁸³

That is not merely a question of stare decisis. As the Court of Appeal's approach in Alma Nudo Atenza illustrates, if the language of proportionality had been well-established in Malaysian law, but a theory of constitutional adjudication was not, then proportionality analysis could easily become a mere afterthought. But this writer's point is not limited to proportionality. It is a more general point: A legal system cannot have legal rules – whether they be called ‘proportionality’, ‘reasonableness’, ‘non-illusoriness’, or something else – relating to constitutional rights analysis without a clear sense of the underlying foundation of theoretical principles; otherwise, there is a risk that the rules are applied anaemically.

Background

The case reported as Alma Nudo Atenza involved Alma Nudo Atenza and Orathai Prommatat, who had in separate instances been convicted of drug trafficking.⁸⁴ At the airport, each of them had had with her a bag containing drugs. This fact alone was sufficient to convict each of them. To understand how, it is necessary to examine various statutory provisions.

Section 39B(1)(a) of the Dangerous Drugs Act (DDA) makes it a crime to ‘traffic in a dangerous drug’. Section 2 defines ‘trafficking’ as including a variety of actions such as ‘importing’, ‘transporting’, ‘carrying’, ‘sending’, and ‘delivering’.

In seeking to prove that an accused person is guilty of drug trafficking, the prosecution has the benefit of several evidentiary presumptions. For present purposes, two are relevant:

1. a. Section 37(d) of the DDA: ‘any person who is found to have had in his custody or under his control anything whatsoever containing any dangerous drug shall, until the contrary is proved, be deemed to have been in possession of such drug and shall, until the contrary is proved, be deemed to have known the nature of such drug’.

2. b. Section 37(da) of the DDA: ‘any person who is found in possession of [a certain amount of drugs or more] …otherwise than in accordance with the authority of this Act or any other written law, shall be presumed, until the contrary is proved, to be trafficking in the said drug’.

In the 1998 case of Muhammed bin Hassan, the Federal Court had held that ‘it would be unduly harsh and oppressive’ for both presumptions to operate simultaneously.⁸⁵ In other words, it could not be said that a person found to be in ‘custody’ or ‘control’ of a bag containing drugs was presumed knowingly to possess the drugs, and in turn presumed to have been trafficking in the drugs. In so ruling, the Federal Court did not cite any constitutional provisions, but only ordinary principles of statutory interpretation.

The Legislature subsequently inserted section 37A of the DDA: ‘Notwithstanding anything under any written law or rule of law, a presumption may be applied under this Part in addition to or in conjunction with any other presumption provided under this Part or any other written law.’ On this basis, the High Court convicted each of the accused persons of drug trafficking once it was satisfied that each of them (a) had been in ‘custody or … control’ of a bag containing drugs, and (b) failed to rebut either presumption.

The Court of Appeal's decision: only a test of formal validity; proportionality a mere afterthought

Both accused persons appealed to the Court of Appeal. Alma Nudo Atenza argued, inter alia, that section 37A of the DDA was unconstitutional, in that it effectively reversed the burden of proof, ‘erode[d] the presumption of innocence’, and rendered the trial unfair.⁸⁶

The Court of Appeal, aware of the existence of the proportionality doctrine, cited the Federal Court case of Gan Boon Aun.⁸⁷ In that case, which also involved a statutory presumption, the Federal Court had shown that it had ‘come to internalize the nature and purpose of proportionality reasoning and apply it in adjudication … in an unprecedentedly rigorous manner’.⁸⁸

But instead of putting proportionality front and centre, the Court of Appeal essentially adopted the following as its first line of reasoning:

1. a. Article 5(1) of the Federal Constitution states: ‘No person shall be deprived of his life or personal liberty save in accordance with law’;

2. b. Section 37A of the DDA is a ‘law’; and

3. c. Therefore, the Constitution allows a person to be deprived of his ‘life or personal liberty’ in accordance with section 37A of the DDA.

This reasoning is clearly problematic, because it means that the only restriction on the state's ability to deprive someone of life or liberty is that the Legislature must enact a statute saying it can, and the court's only role is to check whether the statute was duly enacted. This reasoning assumes the answer to the very question that it seeks to resolve, namely, whether section 37A of the DDA is a constitutionally valid ‘law’ in the first place. An additional problem is that the Court of Appeal quoted Article 5(1) of the Federal Constitution, but said nothing about Article 8(1) – a glaring omission, considering past Federal Court cases which state that the ‘test of proportionality [is] housed in … the equal protection limb of Art 8(1)’.⁸⁹

As startling as these propositions are, they are the implication of the following statement by the Court of Appeal in Sivarasa Rasiah:

Learned counsel nevertheless did not dispute that the amending Act A 1457 which introduced the double presumption rule was a valid Act enacted by Parliament. Therefore we could safely hold that the new section 37A of the [DDA] is not illegal or ultra vires Article 5 or Article 8 of the Federal Constitution. This is so by virtue of the operative words used in the articles ‘save in accordance with law.⁹⁰

Moreover, everything that was said about proportionality was merely a set of additional reasons to justify this conclusion. This approach belies an underlying doctrine of de facto parliamentary sovereignty (rather than substantive constitutional supremacy) according to which all the constitution requires for an Act of Parliament to be valid is that it be ‘enacted by Parliament’.⁹¹ The Court of Appeal did issue several remarks about proportionality. But these were merely obiter, and essentially amount to a statement that any rebuttable presumption which is triggered upon the prosecution's proof of certain primary facts must pass the proportionality test:

The onus of proving the guilt of the appellant beyond reasonable doubt is still on the prosecution. The presumptions under sections 37(d) and 37(da) of the [DDA] is a rebuttable presumption [sic]. Before a presumption can safely be invoked, it is incumbent upon the prosecution to adduce positive evidence of facts from which the presumption can be relied on … The rights of the defence were maintained and the appellant was given an opportunity to rebut the presumptions on the balance of probabilities.⁹²

In short, the Court of Appeal appears to have thought that Gan Boon Aun stood for the principle that statutory presumptions are constitutionally valid, despite the constitutional right to presumption of innocence, as long as they (a) are triggered only upon the prosecution's proof of certain facts, and (b) can be rebutted by the defence's proof of certain other facts. This is hardly adequate, for three reasons.

First, what the Court of Appeal said would apply to any rebuttable presumption. By contrast, the Federal Court in Gan Boon Aun explained why the particular presumption before it passed the proportionality test: for instance, the presumption had to be ‘fair and necessary’, and not ‘difficult’ to rebut, and it also mattered whether a prosecutor would face ‘difficulty … in the absence of a presumption’ and ‘the importance of what is at stake’.⁹³ The Court of Appeal in Alma Nudo Atenza did not ask whether the presumptions in the DDA met these criteria.

Second, the Court of Appeal did not address the fact that while Gan concerned one statutory presumption, Alma Nudo Atenza involved two presumptions operating concurrently.

Third, in Gan Boon Aun, the Federal Court ‘left open the possibility that, even though [the presumption-creating statutory provision] was not in and of itself unconstitutional, a trial court's application of [that provision] could still be unconstitutional’.⁹⁴ By contrast, in Alma Nudo Atenza, the Court of Appeal stopped at considering the proportionality of the presumptions in the abstract, rather than considering whether the presumptions as applied in the particular case before it would operate in a proportionate manner.

To put the Court of Appeal's decision in Alma Nudo Atenza in perspective, it will be useful to outline the Federal Court's approach in Gan Boon Aun. That case concerned section 122 of the Securities Industry Act, which created a presumption that when a company was guilty of certain offences, then its directors and various officers were also guilty unless each ‘prove[d] that the offence was committed without his [or her] consent or connivance and that he [or she] exercised all such diligence to prevent the commission of the offence as he [or she] ought to have exercised’.⁹⁵ In holding that that presumption was constitutionally valid, the Federal Court explained:

1. a. why it was necessary to punish the directors and officers (namely, that they were the very individuals responsible for the company's offence, but fining the company alone would not suffice to deter them);⁹⁶

2. b. why the ‘deeming provision was fair and necessary’ (because, prima facie, ‘only [the directors/officers] could have caused or motivated the offence committed by [the company]’);⁹⁷ and

3. c. that ‘[i]t should not be difficult for [each director/officer] to prove the absence of his knowledge, consent or connivance’.⁹⁸

These are crucial points in explaining why the statute in question passed the proportionality test. The outcome may well have been different if, for instance, it had been ‘difficult’ to rebut the presumption, or if the presumption had not been ‘fair and necessary’. Unfortunately, in Alma Nudo Atenza, the Court of Appeal went into no such level of detail.

Conclusion on the Court of Appeal's reasoning

What is interesting is that the Court of Appeal appears to not have seen its reasoning – even the statement that any ‘valid Act enacted by Parliament’ is constitutionally valid even if it impinges on the rights to life or liberty – as an act of disobedience to the Federal Court's authority that commanded it to perform proportionality analysis. In fact, although the Court of Appeal in Alma Nudo Atenza did not use the term ‘proportionality’ at all, it still discussed Gan Boon Aun.

In other words, the underlying problem was that there was no general theory of constitutional rights adjudication, nor a general idea that state action that engages constitutional rights must be proportionate. Instead, the Court of Appeal saw Gan Boon Aun as an authority about statutory presumptions, not an authority about rights adjudication; and ultimately the Court of Appeal's primary line of reasoning is that any duly enacted statute allows the state to deprive anyone of life or liberty. The Court of Appeal saw itself as applying legal rules that used the new language of ‘proportionality’, but with only the old, deficient theory of de facto parliamentary supremacy.

The Federal Court's decision: the high-water mark of proportionality, and of rights review generally

Both accused persons appealed to the Federal Court, which unanimously held that section 37A of the DDA was unconstitutional. That meant that only one of the two statutory presumptions could apply. As a result, the accused persons could at most be convicted of drug possession, not drug trafficking, and so faced a maximum sentence of life imprisonment, not death.⁹⁹

The Federal Court could have reached this conclusion without discussing proportionality, or indeed any constitutional issue. Section 37(da) of the DDA provides that a person ‘found in possession’ of certain amounts of drugs is presumed to have been trafficking in those drugs. As the Federal Court pointed out, those words only denote a person who has been proven by evidence to have possessed drugs, not a person who was presumed to have possessed drugs. Section 37A of the DDA does not change that.¹⁰⁰ On this basis, alone, the convictions for trafficking could have been overturned.

So it is noteworthy that the Federal Court did not stop there, instead expressly rejecting the Court of Appeal's reasoning. It stated: ‘when any State action is challenged as violating a fundamental right, such as the right to life or personal liberty under art. 5(1) [of the Federal Constitution] … the action must meet the test of proportionality’.¹⁰¹ When the statute in question creates a presumption, relevant factors include:

1. (a) whether the presumption relates to an essential or important ingredient of the offence;

2. (b) opportunity for rebuttal and the standard required to disprove the presumption; and

3. (c) the difficulty for the prosecution to prove the presumed fact.¹⁰²

As we have seen, such factors are precisely what the Federal Court in Gan Boon Aun considered, but the Court of Appeal in Alma Nudo Atenza did not.

Having laid the groundwork, the Federal Court then held that section 37A of the DDA failed the proportionality test. Because it required the accused person to ‘prove some fact on the balance of probabilities to avoid conviction’,¹⁰³ it prima facie infringed on the right to presumption of innocence.¹⁰⁴ While it was rationally connected to a legitimate aim (fighting the ‘substantial and pressing’¹⁰⁵ problem of drug trafficking), it was disproportionate. This was because the presumptions, taken together, ‘relate to the three central and essential elements of the offence of drug trafficking’.¹⁰⁶ This placed a very heavy legal burden on the accused, which would

giv[e] rise to a real risk that an accused may be convicted of drug trafficking in circumstances where a significant reasonable doubt remains as to the main elements of the offence.¹⁰⁷

In short, in the face of the

essential ingredients of the offence, the imposition of a legal burden, the standard of proof required in rebuttal, and the cumulative effect of the two presumptions … [i]t is far from clear that the objective cannot be achieved through other means less damaging to the accused's fundamental right under art. 5 [of the Federal Constitution].¹⁰⁸

The upshot of the Federal Court's judgment was as follows. First, proportionality was now a firmly established doctrine of Malaysian constitutional law. Second – contrary to the Court of Appeal's decision – a court cannot avoid engaging in proportionality analysis by reading phrases in the Constitution such as ‘in accordance with law’ narrowly, as though the mere fact of having been passed by Parliament in a procedurally proper manner is enough to make legislation constitutionally valid. Third, the Federal Court had provided a clear example of how proportionality analysis was to apply. But the question remains: was there a clear sense of why, in principle, it should apply, other than that the Federal Court said so?

Riding the fifth wave

In two post-Alma High Court cases, the High Court did perform proportionality analysis in a manner that took constitutional rights seriously rather than treating them as trivial for the state to defeat.

High Court cases following Alma Nudo Atenza

Suresh Kumar Velayuthan

The first case, Suresh Kumar Velayuthan,¹⁰⁹ involved section 13 of the Security Offences (Special Measures) Act 2012 (SOSMA). The Act provides (with a few exceptions) that a man aged 18 or above who is not ‘sick or infirm’ shall not be granted bail if he is charged with a ‘security offence’.¹¹⁰

The applicant had been charged with a ‘security offence’. He applied for bail, arguing that section 13 of SOSMA was unconstitutional. The High Court, in rejecting this submission, performed proportionality analysis. It said that ‘acts of terrorism’ are so potentially dangerous that the state may respond through ‘preventive and not reactive action’.¹¹¹ The alternative would be to revoke bail only if, while on bail, the accused person re-offends – but that would be too late to prevent the ‘grievous harm’¹¹² that would ensue if that second offence turned out to involve terrorism. In other words, the legislation pursued a sufficiently important objective which could not be achieved through a less rights-intrusive means.

To be sure, the High Court's decision is not beyond criticism. Why do concerns about repeat terrorism offences justify a blanket rule against bail? Is it not enough that the court has the discretion not to grant bail? Moreover, at certain points, the High Court suggested that the question was whether there was a ‘rational nexus’¹¹³ or ‘reasonable co-relation’¹¹⁴ between section 13 and its objectives, rather than explicitly asking whether section 13 of SOSMA goes no further than is necessary to achieve its objectives.

These criticisms are valid, but the point remains: the High Court had required that the legislation be justified in proportionality terms. There is certainly a conversation to be had about the strictness of proportionality review, and the degree of deference owed to the Legislature.¹¹⁵ But, crucially, the state had to bear the burden of justifying a restriction on fundamental rights. This is in marked contrast to a statement that (as a different High Court judge held in another case) section 13 of SOSMA is immunised from constitutional challenge on any ground by Article 149 of the Federal Constitution.¹¹⁶

Najib Razak

The application of proportionality was more straightforward in Najib Razak, which did not involve national security.¹¹⁷ The accused had been charged with corruption-related offences. One issue involved section 50 of the Malaysian Anti-Corruption Commission Act, which provided that, in criminal proceedings in respect of certain corruption-related offences, a person who had received ‘gratification’ ‘shall be presumed’ to have done so ‘corruptly’. The question was whether ‘shall be presumed’ meant that the court had to apply the presumption, or merely that the court could apply it.¹¹⁸

The High Court concluded that it could choose whether or not to apply the presumption.¹¹⁹ In support of this view, the High Court drew from Alma Nudo Atenza the notion that the court must ‘consider the proportionality of applying a … presumption’ and only apply it if it appears ‘necessary’¹²⁰ to do so for the purpose of ‘address[ing] the unfairness prevailing from the near impossibility of demonstrating a relevant fact’.¹²¹ And even then, there must be a ‘clear notification made to the accused that the presumption has been invoked against him when the accused is called to enter his defence’.¹²²

This is noteworthy because the High Court did not make the same mistake as the Court of Appeal in Alma Nudo Atenza did, namely, to assume that if a presumption is constitutionally valid in the abstract, then it must be constitutionally valid as applied in any case.

For completeness, it is worth stating that the fact that the accused was a former Prime Minister is irrelevant. The High Court did not trigger the presumption¹²³ – which would have operated against the accused – yet ended up convicting the accused anyway.

Taking stock

By now, the Federal Court had clearly stated that ‘when any State action is challenged as violating a fundamental right … the action must meet the test of proportionality’.¹²⁴ It had laid down a three-step proportionality test¹²⁵ and demonstrated how a court may apply this test having regard to various relevant factors.¹²⁶ The High Court had followed suit and applied proportionality analysis in substance. The reasoning in these cases clearly does not, like the Court of Appeal's reasoning in Alma Nudo Atenza, reduce the judicial role in rights adjudication to a vanishing point.

The lingering problem: no clear sense of the Legislature's role

There was, however, a significant weakness in the case law. Even as the courts performed proportionality analysis, the role of the Legislature was not as clear. The issue of deference to the legislature, and how it interacts with proportionality analysis, had never been thoroughly addressed. In other words, the theoretical foundations of proportionality analysis had yet to be fully established.

In fact, the question of the legislative role had been raised in Tan Tek Seng, the case that introduced proportionality analysis into Malaysian law. That case involved a public service employee who had committed a criminal offence. The majority held that dismissing him from service was a disproportionate response to the offence. Unfortunately, the majority did not engage with the dissenting judge's point that the court ought not to ‘substitute [its] own views for that of the employers’¹²⁷ and ought instead to uphold any decision that a ‘reasonable employer’ could ‘reasonably have taken’.¹²⁸

The issue was also quietly raised by Suresh Kumar Velayuthan.¹²⁹ In holding that the law on bail in SOSMA cases was not unconstitutional, the High Court said this:

The particular objective to be achieved is therefore not out of proportion to the state action taken which is to prevent possible terrorist-related acts by the absolute prohibition of bail. After the carrying out of a balancing exercise between the individual right of the accused and the greater good of the public at large it was necessary in all the circumstances for Parliament to pass such a law. There is also a clear rational nexus between the relevant state action with the objective to be achieved.¹³⁰

This passage contains an ambiguity. The first and third sentences suggest that the court, having itself considered whether the law was proportionate, concluded that it was. The second sentence, by contrast, does not make clear who had ‘carr[ied] out … a balancing exercise between the individual right of the accused and the greater good of the public at large’: is it the court, or Parliament, which had done so? In other words, can the court defer to Parliament's judgment that legislation is proportionate?

No case has squarely addressed the issue of the nature and degree of deference which the courts should show toward legislative judgments. At most, there is Mohamad Ariff JCA's statement in Court of Appeal case of Nik Nazmi: ‘[t]he courts in testing the constitutionality of legislative action cannot substitute its own view on what ought to be the proper policy.’¹³¹ However, that was only obiter.

This situation is regrettable. On the one hand, constitutional rights review – whether under the rubric of proportionality or otherwise – cannot be an excuse for juristocracy, in which the courts, in the name of proportionality analysis, arrogate to themselves the power to strike down any state action without even considering what can be said in its favour. On the other hand, one instinctively recoils at the notion that the Legislature can validly pass rights-restricting legislation that is not ‘reasonable’¹³² or that is ‘harsh and unjust’.¹³³ The solution is that courts often apply a doctrine of due deference when testing state action for its proportionality. But the Malaysian courts had not explored this. Some judges had expressed fear of juristocracy, others of legislative dictatorship; but the courts had yet to explore how both extremes are in tension and how the tension could be resolved.

This gap in the case law, coupled with old cases hostile to proportionality which had never been explicitly overruled, paved the way for the Federal Court in Letitia Bosman to hold that the need for deference (and a related doctrine of the ‘presumption of constitutionality’) can override not only proportionality analysis, but also, it seems, any meaningful form of rights review.

The tides turn: the Letitia Bosman case

Introduction

The case reported as Letitia Bosman arose from four appeals to the Federal Court by persons who had received mandatory death sentences (three for drug trafficking¹³⁴ and one for murder) between 2015 and 2016.¹³⁵ The eight-judge majority of the Federal Court acknowledged that ‘Malaysian constitutional jurisprudence has developed so as to recognize that statutory provisions may be struck down on the grounds of proportionality’.¹³⁶ However, it did not apply any proportionality test at all. Instead, in the name of a ‘presumption of constitutionality’, it asserted that the ‘the matter is an inherently legislative issue for Parliament to decide’¹³⁷ due to its ‘social and moral features’,¹³⁸ which involved ‘controversial issues of legislative policy and social values’.¹³⁹ While the majority acknowledged the case law referring to proportionality,¹⁴⁰ it ended up stating: ‘If a judge were to decide that the mandatory death penalty is not proportionate, it would entail the judge enacting his or her personal views of what is just and desirable into legislation.’¹⁴¹ In other words, proportionality analysis existed in name only. Only Nallini Pathmanathan FCJ, the lone dissentient (who was also part of the Alma Nudo Atenza court), did.

One might argue that there is nothing intrinsically wrong with the majority's approach in Letitia Bosman. After all, as noted above, proportionality is not the only possible means of rights adjudication.¹⁴² But, as we will see, the implication of the court's reasoning appears to be that any legislation that engages constitutional rights is nonetheless ipso facto constitutionally valid. This is antithetical, not only to proportionality analysis, but also to any concept of constitutional rights review.

This writer will now justify this inference from the majority's judgment, and then explain why it cannot be justified in principle in the name of ‘deference’ or the ‘presumption of constitutionality’.

A narrow reading of the rights provision

It is useful to deal first with the section of the majority's judgment titled ‘The Fair Trial Point’,¹⁴³ which states that proportionality cannot apply in the specific context of the constitutional right to a fair trial.

The appellants’ argument was that the mandatory death penalty violated the right to a fair trial, because a ‘fair trial’ must involve allowing the accused to make ‘meaningful representations to the court on the appropriate sentence’.¹⁴⁴ The Federal Court did not dispute that ‘[t]he right to a fair trial is a constitutionally guaranteed right’.¹⁴⁵ However, it said that that right ‘can be taken away in accordance with law’,¹⁴⁶ and that the DDA is such a ‘law’. Therefore, the majority held that there was no unconstitutional violation of the right to a fair trial.

We have seen how similar problematic reasoning was raised by the Court of Appeal in Alma Nudo Atenza, but departed from by the Federal Court in that case. Now, in Letitia Bosman, the Federal Court undid this development.

The true meaning of ‘in accordance with law’

According to the Federal Court, the source of the constitutional right to a fair trial is Article 5(1) of the Federal Constitution. That provision does not speak explicitly of a fair trial; it merely says ‘No person shall be deprived of his life or personal liberty save in accordance with law.’ Nonetheless, according to the Privy Council in Ong Ah Chuan (a Singaporean case cited in Letitia Bosman),¹⁴⁷ this does not mean that life or liberty may be taken away by just any law passed by Parliament. Instead:

references to “law” in such contexts as “in accordance with law” … refer to a system of law which incorporates those fundamental rules of natural justice that had formed part and parcel of the common law of England that was in operation in Singapore at the commencement of the Constitution.¹⁴⁸

(Let us call this ‘the proviso’.)

The majority's narrow reading of ‘in accordance with law’: not supported by the cases

The Malaysian courts later followed suit. For example, in Lee Kwan Woh, the Federal Court stated, in effect, that punishment imposed other than following a fair trial does not count as deprivation of life or liberty ‘in accordance with law’.¹⁴⁹

However, in Letitia Bosman, the majority did not engage with this reading of ‘law’. Instead, it took ‘law’ in Article 5(1) of the Federal Constitution to refer to any statute: ‘a law that provides for the deprivation of a person's life or personal liberty is valid and binding as long as it is validly passed by Parliament’.¹⁵⁰

By this reasoning, Article 5(1) may as well not exist. This, alone, ought to have given the majority pause for thought. It should worry us, not because it is not proportionality analysis, but because it is antithetical to the very aim of rights review generally – namely, to set up rights as a meaningful constraint on state power.

For completeness, it is worth briefly exploring how the majority reached its striking conclusion. The majority cited three cases in support of this approach; on closer examination, none of them supports it.

The first case, Selangor Pilot Association, may readily be distinguished. That case involved Article 13 of the Federal Constitution:

1. 13. (1) No person shall be deprived of property save in accordance with law.

   (2) No law shall provide for the compulsory acquisition or use of property without adequate compensation.

In short, the Privy Council appears to have held that any statute which complies with Article 13(2) is a ‘law’ for the purposes of Article 13(1).¹⁵¹ In other words, the only substantial constraint on the content of a law that deprives one of one's property is that in Article 13(2): there must be ‘adequate compensation’. This might make sense in the context of Article 13(1): expressio unius est exclusio alterius. Besides, ‘adequate compensation’ is a test with clear substantive content. But, in the context of Article 5(1), there is nothing which expressly circumscribes the content of a law that deprives one of one's right to life.¹⁵²

The second case, Sugumar Balakrishnan,¹⁵³ made no mention of the proviso, but this approach must be treated as having been overruled by Lee Kwan Woh.

The third case, Bird Dominic Jude,¹⁵⁴ also made no mention of the proviso, and post-dates Lee Kwan Woh, but may readily be distinguished on the facts. The case involved a law allowing for a person acquitted by the High Court to be arrested and detained pending the prosecution's appeal to the Court of Appeal. While the Federal Court did not explicitly say so,¹⁵⁵ the legislation in question would arguably pass the proportionality test. It pursued the legitimate aim of preventing the accused person from absconding, and its impact on the right to personal liberty was attenuated by the fact that the Court of Appeal could grant bail (which, on the facts, it did).¹⁵⁶ This is nothing like the mandatory death penalty for drug trafficking, which allows for no judicial discretion.

In effect, the majority in Letitia Bosman treated the mandatory death penalty as ipso facto constitutionally valid merely because it was provided for by something that is on the statute books.¹⁵⁷ This precludes any meaningful review of a statute that deprives someone of life or liberty. The court's role is reduced to checking that the correct number of parliamentarians had voted for the Bill and that royal assent had been given. This goes against the very purpose of the proviso in Ong Ah Chuan, of proportionality analysis, and of rights review generally.

‘Nothing unusual and arbitrary’ about mandatory sentencing?

One may argue that none of this matters. According to the majority in Letitia Bosman, to the extent that there is a constitutional right to make submissions on sentencing, it can only be a right to make submissions on which of all the possible sentences prescribed by law one may receive. Therefore, when the legislation provides for only one possible sentence, ‘the mitigation plea plays no role in the sentencing process’.¹⁵⁸

It is notable that the fact that the right to ‘life or liberty’ are at stake plays no part in this reasoning. Nonetheless, let us take this reasoning on its own terms. It is premised on the idea that there is ‘nothing unusual and arbitrary’ in a mandatory death sentence – or, indeed, any mandatory sentence.¹⁵⁹ This followed from a statement in the Privy Council's decision in Ong Ah Chuan that

[t]here is nothing unusual in a capital sentence being mandatory … If [an argument to the contrary] were valid[, it] would apply to every law which imposed a mandatory fixed or minimum penalty even where it was not capital – an extreme position which counsel was anxious to disclaim.¹⁶⁰

But would that position be so ‘extreme’ nowadays? Ong Ah Chuan was decided in 1980, long before a proportionality doctrine had been established in Malaysia. This is also true of the other Federal Court decisions cited in Letitia Bosman that followed Ong Ah Chuan on this point.¹⁶¹ Surely the minimum requirements that a statute must meet in order to count as ‘in accordance with law’ must be higher than the level where Ong Ah Chuan set them.

One more point is worth mentioning. The majority did consider several cases from other jurisdictions holding that the mandatory death penalty violated a constitutional right against ‘inhuman or degrading’ punishment.¹⁶² The Federal Court held that these cases were of no relevance because Malaysia's Federal Constitution does not contain such a right.

This is unfortunate. True, there is no constitutional right against ‘inhuman or degrading’ punishment in Malaysian law, and cases involving different constitutional texts should not unthinkingly be followed. However, the point in these cases is that the mandatory death penalty is ‘inhuman or degrading’ because it is disproportionate. One wonders: if the mandatory death penalty is disproportionate, would that not mean that depriving someone of his life through the mandatory death penalty is, for the purposes of Malaysia's Federal Constitution, not ‘in accordance with law’? Surely that point deserved at least some consideration.

‘Deference’ to the Legislature's judgment?

A more formidable argument against the ‘inhuman or degrading’ test might be that it would give too much power to the courts. The classic response to this problem is a doctrine of due deference to the political branches. The majority alluded to this point:

In the present appeals, the questions at issue are fraught with the complex issue of the proper balance between judicial power and legislative power that goes into the heart of the doctrine of separation of powers. While the court has a substantive constitutional role in reviewing the legislative act where deemed necessary to ensure legality, the task is complex as it also involves the appropriate measure of judicial deference the court should give the Legislature on matters involving delicate and contentious areas of social policy.¹⁶³

As we will see, the majority's application of the doctrine of deference is highly problematic. Not only was it antithetical to proportionality analysis (which, recall, the majority acknowledged exists in principle);¹⁶⁴ it also collapsed into, as Allan puts it, ‘non-justiciability dressed up in pastel colours’.¹⁶⁵ The majority had in effect relied on the ‘supposedly superior qualifications of the decision-maker’, hence ‘divest[ing] the court of its role as independent scrutineer’.¹⁶⁶ The correct position, this writer submits, is that the constitutionality of legislation, being a question of law, is not a question as to which Parliament is entitled to have the last word,¹⁶⁷ even if ‘Parliament might have the means to consider the issue more fully or on a broader canvas’.¹⁶⁸ In other words, even if the court defers to Parliament's answers to certain questions, the court must make a considered decision to do so.¹⁶⁹ At the very least, the court must inquire into whether Parliament has asked and directed its mind to the right questions in the first place,¹⁷⁰ instead of merely assuming that the Legislature must know better.

The majority's problematic reading of foreign cases about deference

While the majority stated that ‘legislative decisions are entitled to an appropriate measure of deference and respect’,¹⁷¹ it did not consider what an ‘appropriate measure’ would be. Instead, the court gave conclusive weight to Parliament's judgment. This is evident from its treatment of several cases about ‘deference’. These bear closer examination, because these very cases could well form the start of a discussion on the role of judicial deference to political judgments in constitutional rights review. In other words, such cases may well contain the seeds of the sort of robust theory of constitutional review which Malaysian law currently lacks. Instead, the majority of the Federal Court overlooked the point of these cases.

(a) Reyes

One case was the Privy Council's decision in Reyes, from which the Federal Court quoted the following extract:

In a modern liberal democracy it is ordinarily the task of the democratically elected Legislature to decide what conduct should be treated as criminal, so as to attract penal consequences, and to decide what kind and measure of punishment such conduct should attract or be liable to attract … The ordinary task of the courts is to give full and fair effect to the penal laws, which the Legislature has enacted. This is sometimes described as deference shown by the courts to the will of the democratically elected Legislature.¹⁷²

But the Federal Court omitted to cite the following passage from Reyes, which followed the one just reproduced:

When (as here) an enacted law is said to be incompatible with a right protected by a Constitution, the cour[t] … must interpret the Constitution to decide whether the enacted law is incompatible or not … consider the substance of the fundamental right at issue and ensure contemporary protection of that right in the light of evolving standards of decency that mark the progress of a maturing society …¹⁷³

If nothing else, this latter extract tells us that the legislature is only to be deferred to in some matters, but not others; ‘deference’ does not automatically mean that the legislature gets its way. By contrast, the majority of the Federal Court appears not to have considered the ‘substance of the fundamental right at issue’ at all. Indeed, it is not even clear from the part of the majority's judgment labelled ‘The Proportionality Point’¹⁷⁴ what constitutional right was at stake.

(b) RJR-Macdonald

One would think that the views of the Supreme Court of Canada in RJR-Macdonald ¹⁷⁵ are instructive:

Care must be taken not to extend the notion of deference too far … To carry judicial deference to the point of accepting Parliament's view simply on the basis that the problem is so serious and the solution difficult, would be to diminish the role of the courts in the constitutional process and to weaken the structure of rights upon which our constitution and nation is founded.¹⁷⁶

But the majority in Letitia Bosman appears to not have engaged with the substance of this view at all, instead asserting that

controversial matters of policy involving differing views on the moral and social issues involved is one circumstance where Parliament is better placed to assess the needs of society and to make difficult choices between competing considerations. Courts should accept and recognise Parliament is better placed to perform those functions.¹⁷⁷

However, the majority did not explain precisely why the issue in the instant case was so ‘controversial’ in this sense.

(c) Lau Cheong

The majority also discussed two cases relating specifically to the death penalty. One of them, Lau Cheong, involved a challenge before the Hong Kong Court of Final Appeal to the mandatory sentence of life imprisonment for murder. The Federal Court in Letitia Bosman cited the Hong Kong court's statement that ‘the question of the appropriate punishment for what society regards, as the most serious crime is a controversial matter of policy involving differing views on the moral and social issues involved.’¹⁷⁸

However, this one quotation does not fully reflect what the Hong Kong court held. The court did not say that the legislature must have the final say on what the sentence for murder may be merely because the issue is a ‘controversial matter of policy’. Instead, the Hong Kong court deferred to the legislature's decision only because the court independently reached a considered conclusion that that decision was ‘tenable and rational’,¹⁷⁹ having regard to the ‘severity of the punishment’, the ‘nature and seriousness of the offence’, and ‘other legitimate sentencing objectives (such as protection of the public, preventing repetition of the offence, deterring others from committing like offences, and societal denunciation of the offence)’.¹⁸⁰ The Hong Kong court did not hold that it must accept the legislature's choice as a foregone conclusion.

(d) Yong Vui Kong

The next case is the Singaporean case of Yong Vui Kong, which involved a challenge to the mandatory death penalty. The Singapore Court of Appeal said that ‘what is an appropriate threshold of culpability for imposing the [mandatory death penalty] is, in our view, really a matter of policy, and it is for Parliament to decide’.¹⁸¹ But the case must be of limited application in Malaysia. First, the Singapore court's statement was made in response to an argument that the mandatory death penalty was ‘inhuman’;¹⁸² this, in turn, was obiter because the Singapore court eventually held that there is no constitutional requirement that punishment not be ‘inhuman’.¹⁸³ As for the right to life (which is a constitutional right in both Singapore and Malaysia), Singapore law, unlike Malaysian law, does not even purport to apply a proportionality test,¹⁸⁴ and there appears to be no Singaporean case in which counsels seeking to challenge the mandatory death penalty raised a proportionality-based argument.

(e) Nicklinson (Lord Sumption’s judgment)

The final case – which the majority of the Federal Court cited the most extensively – was the UK Supreme Court case of Nicklinson. That case was heard by nine judges. Lord Sumption, together with three other judges, held that the court not only should not declare that the UK's ban on assisted suicide violated Article 8 of the European Convention on Human Rights, but also that the court did not even have the authority to issue such a declaration. According to Lord Sumption, this was because:

1. a. ‘two fundamental but mutually inconsistent moral values, upon which there is at present no consensus in our society’ were at stake – these were personal autonomy (which would weigh in favour of allowing assisted suicide) and the sanctity of life (which would weigh against it);¹⁸⁵

2. b. Parliament had been actively debating the issue of assisted suicide;¹⁸⁶ and

3. c. Parliament, unlike the courts, had access to a ‘fuller range of expert judgment and experience’, as well as the ‘interests of groups not represented or not sufficiently represented before the court’.¹⁸⁷

The Nicklinson case's role in the Federal Court majority's decision bears closer scrutiny. The aim is not to hold up the views of any judge in Nicklinson as a model for rights adjudication. The point is simply this: Malaysia currently lacks a theory of the role of political branches in upholding constitutional rights; it is therefore worth considering at least the theory immanent in Nicklinson, given the weight that the Federal Court appeared to place on that case.

The majority's flawed appeal to the Legislature's views

The most obvious criticism is that the majority engaged only with Lord Sumption's judgment. It did not consider, for example, Lord Neuberger's statements that ‘the mere fact that there are moral issues involved plainly does not mean that the courts have to keep out’¹⁸⁸ and that ‘the mere fact that Parliament has recently enacted or approved a statutory provision does not prevent the courts from holding that it infringes a Convention right’.¹⁸⁹

But let us take the Federal Court at its highest and examine Lord Sumption's reasoning more closely. We will see why it was not sufficient for the Federal Court to assert that it was good enough that Parliament thought that the mandatory death penalty was proportionate.¹⁹⁰

Lord Sumption's second and third points are compelling, but they hinge on his first point, which is that there are ‘two fundamental but mutually inconsistent moral values’ at stake. If there were not, then there would be no controversy to speak of, and no ‘choice’ between them for Parliament to make in the first place.

According to Lord Sumption, the conflicting values meant that any decision about the permissibility of assisted suicide ‘cannot fail to be strongly influenced by the decision-makers’ personal opinions about the moral case for assisted suicide’, and that judges should not be the ones whose ‘personal opinions’ count.¹⁹¹ Lord Sumption did not say that the mere existence of some controversy justifies the courts’ taking a hands-off approach.

Anyway, the majority of the Federal Court had not even established what the two competing moral principles at stake were, let alone whether they are ‘mutually inconsistent’¹⁹² or, as Lord Neuberger put it, whether both sets of moral argument are ‘telling’.¹⁹³ Instead of Lord Sumption's three points, the Federal Court discussed only the second, namely, that Malaysia's Parliament had debated the mandatory death penalty for drug trafficking in 1983 and March 2017.¹⁹⁴

This is problematic because, at least before the Federal Court's decision, Malaysia's Parliament appears not to have directed its mind to whether the mandatory death penalty was compatible with the Federal Constitution. Parliament did eventually make the death penalty non-mandatory in 2018, and it did direct its mind to whether the Federal Constitution demanded that this apply to those already on death row.

In 1983, the DDA was amended to make death the only possible sentence for trafficking in cases involving more than a certain amount of drugs. On 23 March 2017, Dato’ Sri Azalina Dato’ Othman Said, Minister in the Prime Minister's Office, announced that Cabinet had once again decided that in principle the death penalty for drug trafficking should be made discretionary, not mandatory.¹⁹⁵ She pointed out that the mandatory death penalty was not necessarily watertight as a deterrent.¹⁹⁶ Various MPs were concerned about how long the proposed amendments would take, whether there could be a moratorium on the death penalty in the meantime, and whether the amendments would apply retroactively so that people on death row would have the chance to be re-sentenced to imprisonment and whipping instead. The Minister responded that, among other things, the question of a moratorium would be discussed in Cabinet.¹⁹⁷

Next, on 30 November 2017, Parliament passed the Dangerous Drugs (Amendment) Act 2017,¹⁹⁸ which made the death penalty non-mandatory. The main debate was whether the amendments should apply retroactively, or only where the accused person had not been convicted as of the date when the amendments came into force.¹⁹⁹ The amendments came into force on 15 March 2018.

Finally, on 10 October 2018, Liew Vui Keong, Minister in the Prime Minister's Department, announced a moratorium on executions pending possible further legislative changes to the death penalty regime.²⁰⁰

Throughout this entire process, Parliament never discussed the constitutionality of the mandatory death penalty itself. The debates on constitutional rights focused only on the question of whether the Article 8(1) right to equality demanded that the amendments apply to people convicted before the amendments came into force.²⁰¹

Now, let us return to the Letitia Bosman case. The Federal Court handed down its judgment on 13 August 2020, by which time all of the above developments had taken place. Yet, the majority of the Federal Court only referred to debates in Parliament in 1983 and in March 2017.²⁰²

The majority claimed, on the basis of these debates: ‘In the wide-ranging debate that had taken place, different views had been expressed, including the majority view that the mandatory death penalty should not only be retained, but should be carried into effect. Parliament has made the choice.’²⁰³ This is simply not true. As of March 2017, Parliament had not made a choice. It was in the process of making a choice. In November 2017, it made the choice not to have the mandatory death penalty for drug trafficking. Therefore, Lord Sumption's judgment in Nicklinson – which focused on the fact that ‘Parliament has made the relevant choice’²⁰⁴ – may immediately be distinguished. The process of making a choice continued through to 2018, while the case was before the Federal Court.

There can be no speaking of deference to a legislative judgment when the Legislature had not made a judgment. It is true that the Legislature made a judgment in November 2017, but the majority made no inquiry at all into what was said in November 2017 about the reasons for that judgment. Moreover, that legislative judgment did not squarely address the constitutional right to life.

It is true that the Legislature had decided that the mandatory death penalty would remain for offenders who had been convicted before 15 March 2018. But surely the Federal Court ought to have gone on and squarely considered the constitutionality of this legislative decision. Besides, this decision was, as Liew announced in October 2018, itself open to being revisited.

In short, even if one considers that judicial review on rights grounds must be tempered with a measure of deference to the Legislature, and indeed even if one considers that all the court has to do is to ask whether Parliament had considered the law to be constitutionally valid, the majority of the Federal Court had done neither. Because it focused on precisely the wrong legislative materials, the Federal Court cannot be said to have deferred to Parliament's view on the constitutionality of the mandatory death penalty (even assuming that Parliament had had any views on its constitutionality in the first place). It cannot even be said that the Federal Court had inquired into whether Parliament thought that retaining the mandatory death penalty for those convicted before 15 March 2018 was proportionate.

Finally, it bears repeating that one of the four conjoined appeals – that of Pubalan a/l Peremal – involved murder, not drug trafficking. The majority of the Federal Court said nothing about any legislative decision about the desirability of the death penalty for murder.

In short, the majority's decision cannot be said to have been the outcome of due deference to the Legislature. The majority referred to only a narrow subset of the materials evidencing the ongoing legislative debate, and treated them as conclusive rather than as indicative of an ongoing process of legislative deliberation. Recall that, in Nicklinson, because such deliberation was ongoing, the majority chose the course of action that best preserved the status quo – namely, to refuse to robustly assert the right to die, without prejudice to the possibility that legislation would more clearly enshrine that right in future. By contrast, in Letitia Bosman, the implication of the majority's reasoning, when seen in the light of the ongoing legislative debate, is that people can be deprived of their life despite the possibility that legislation to the contrary might be passed soon.

One might respond that the majority's decision had no such effect, because, at the time of the decision, a moratorium on executions was in place. Even then, the majority should have said so, and considered the impact of the moratorium on the extent of deference due to the Legislature as well as the proper order to be made by the court. It did not.

The majority's fear of allowing judges’ ‘personal views’ to override Parliament

One final point about deference should be made. The majority of the Federal Court feared that ‘[if] a judge were to decide that the mandatory death penalty is not proportionate, it would entail the judge enacting his or her personal views of what is just and desirable into legislation’.²⁰⁵ This fear is misplaced for three reasons.

First, constitutional rights review – whether under the rubric of proportionality, or otherwise – is not, and has never been, an exercise in a judge giving effect to ‘his or her personal views of what is just and desirable’. It is, and has always been, an exercise in applying legal criteria, as we see in Alma Nudo Atenza.²⁰⁶

Second, the majority had misunderstood what deference entails. It appears to have assumed that, because of deference, the court cannot reach a conclusion that the mandatory death penalty is disproportionate. But that is not how deference works. Instead, the nature and degree of deference (if any) that should be exercised is something that the court should consider at each stage of proportionality analysis.²⁰⁷ It might be, for instance, that the court should defer to the legislature as to whether some legislation pursues a legitimate aim, but not defer (or exercise a lower degree of deference) on the question of whether the legislation goes further than necessary to pursue that aim.

Third, the Federal Court's misgivings about ‘challenging the policies underlying Parliament [sic] decision to legislate the mandatory death sentence penalty’²⁰⁸ are also misplaced. Of course, the court cannot ‘challeng[e]’ those policies in the sense of overruling them for no reason other than bare disagreement. But surely the court can, at least, ensure that Parliament thought there was some rational basis for those policies. This is all the more so when – as demonstrated above – the policies are in flux. If anything, the court's treating those policies as having been set in stone in March 2017 amounts to ‘challenging’ the policy changes between then and the date of the judgment.

A complete analysis of the interplay between proportionality (or other forms of rights review) and deference in mandatory death penalty legislation would be outside the scope of this article. Suffice it to say that the doctrine of deference does not simply mean, as the majority effectively held, that proportionality analysis by the court must become totally impermissible. By the majority's reasoning leading to this point, any more-than-clerical rights review by the court must be impermissible too.

Finally, the majority feared that a judge striking down legislation in a ‘controversial’ area would, in effect, be deciding an issue which ‘by its nature is more suitable for determination by Parliament than by the courts’.²⁰⁹ But this is not true. If a law is unconstitutional, then the court would not be subverting Parliament's role by saying so; it would be aiding the work of Parliament by pointing out precisely what choices Parliament is and is not entitled to make, and leaving it to Parliament to make its choice from among the range of constitutionally permissible choices.

The ‘presumption of constitutionality’: yet again, little meaningful check on state action

Even if one disagrees with all of the analysis above, the majority's treatment of the ‘presumption of constitutionality’ is striking, because this ‘presumption’, as the court applied it, would foreclose any degree of deference other than total deference.

According to the majority, it is of ‘crucial importance’ that ‘[t]here is always a presumption in favour of the constitutionality of an enactment and the burden is upon him who attacks it to show that there has been a clear transgression of the constitutional principles.’²¹⁰ As we will see, this ‘presumption of constitutionality’ is virtually irrebuttable, which means that the court's test for compatibility with constitutional rights is virtually impossible for any legislation to fail.

The ‘presumption of constitutionality’ was first articulated in Malaysian law in Su Liang Yu,²¹¹ which cited a series of Indian cases from the 1950s, which in turn cited the 1919 case of Middleton v Texas Power & Light Co;²¹² the latter was itself one of several US cases that spoke of a ‘presumption of constitutionality’.²¹³ According to Middleton,²¹⁴ in a passage echoed in Su Liang Yu:²¹⁵

There is a strong presumption that a legislature understands and correctly appreciates the needs of its own people, that its laws are directed to problems made manifest by experience, and that its discriminations are based upon adequate grounds.

At first glance, it appears that the basis of the presumption is due deference to the Legislature. Indeed, the Malaysian Court of Appeal has stated that the presumption amounts to ‘deference to legislative determinations as to the desirability or adequacy of particular statutory objectives’;²¹⁶ and, in a later Federal Court case, one judge (albeit in a dissenting judgment) stated that the presumption ‘seeks to appreciate the importance of the role of the Legislature as the democratically elected body representing the will of the People and upon whose mandate it is formed to pass laws’.²¹⁷

Yet there are several serious issues with the presumption of constitutionality as applied in Malaysian law.

First, it is not clear how the ‘presumption of constitutionality’ adds anything to other established principles. If the point of the presumption is that ‘it is for the party who attacks the validity of a piece of legislation to place relevant materials and evidence before the court’,²¹⁸ then the presumption adds nothing to the ordinary rule of evidence that the plaintiff bears the burden of proof. If the point is that ‘it must be presumed that the Legislature understands and correctly appreciates the need of its own people’,²¹⁹ then the presumption would seek to pursue the same aims as the doctrine of due deference.

But the ‘presumption of constitutionality’ is not the same as the doctrine of due deference, because – and this is the second point – it has no regard as to what degree of deference is ‘due’. Even assuming that the ‘Legislature understands and correctly appreciates the needs of its own people’,²²⁰ as Jack Lee points out, (a) there is no guarantee that the Legislature has directed its mind to the question of whether the legislation is constitutionally valid;²²¹ and (b) in any event, it is the courts, not the Legislature, that have the last word on the constitutionality of legislation.²²² Instead, not only is it wholly unclear from the majority's decision what it would take to rebut the ‘presumption of constitutionality’;²²³ the majority also treated the presumption as virtually irrebuttable.

The majority stated that the mandatory death penalty is ‘one of the most sensitive and controversial social issues of our time on which there is no consensus’,²²⁴ ‘relates to a most delicate area of social policy’,²²⁵ ‘involve[s] a most problematic area of social policy’ that is ‘controversial’,²²⁶ and had certain ‘social and moral features’²²⁷ that made the controversy about it ‘an inherently legislative issue for Parliament to decide’.²²⁸ One might wonder whether the same cannot be said of most, if not all, issues involving the balancing of a right against some ‘area of social policy’.

Surely the mere existence of controversy does not make more-than-clerical constitutional rights review impossible in all but name. In Alma Nudo Atenza, the Federal Court said that there is a ‘presumption in favour of constitutionality’,²²⁹ yet that did not stop it from performing proportionality analysis. Even in Su Liang Yu, for all its problematic remarks about the presumption, the High Court did ask whether the legislation in question had any ‘prejudicial or hostile effect’ on accused persons.²³⁰ By contrast, the majority of the Federal Court in Letitia Bosman said:

whether or not the mandatory death penalty is a disproportionate response to the cases covered by these appeals involve controversial issues of legislative policy and social values, which by its nature is more suitable for determination by Parliament than by the courts. It would therefore be wholly inappropriate for the courts to declare the mandatory death sentence inconsistent with art. 8 on the basis that Parliament violated the proportionality principle.²³¹

So the mere existence of controversy immunises legislation from judicial review. If the court meant that only certain kinds of controversy do so, this is by no means clear. This problem was apparent to Nallini Pathmanathan FCJ, the lone dissentient, who pointed out that the presumption of constitutionality is not ‘unassailable’ and cannot be applied to render a law that is invalid, valid’.²³² This must be correct. The presumption of constitutionality, like all presumptions, can be at most a rule of evidence; the majority had no basis for morphing it into an inflexible rule of substantive law. This point is buttressed by the following recent observation by the Singapore Court of Appeal in Saravanan Chandaram:

a presumption of constitutionality in the context of the validity of legislation can be no more than a starting point that legislation will not presumptively be treated as suspect or unconstitutional; otherwise, relying on a presumption of constitutionality to meet an objection of unconstitutionality would entail presuming the very issue which is being challenged.²³³

In short, the presumption as the Federal Court applied it in Letitia Bosman is antithetical, not only to proportionality analysis, but to any meaningful constitutional rights analysis. The court's reasoning, taken to its logical conclusion, would mean that the courts can virtually never conclude that legislation violates constitutional rights. Again, the courts’ role is a merely clerical one.

Conclusion on Letitia Bosman

To sum up: the majority of the Federal Court concluded that any controversial matter must be beyond the reach of the courts, and that any ‘specific law authorising the deprivation of [a person's] personal liberties’²³⁴ that has been ‘duly enacted by Parliament’²³⁵ must be constitutionally valid.

The majority's decision is not merely a deferential application of proportionality analysis. If it were, the majority would at least have identified the right in question and considered how it stood in relation to the legitimate state aim, explaining which stage of proportionality analysis called for deference.²³⁶ The majority would also have identified the specific decision made by the Legislature and evidence of the reasoning behind that decision, before explaining why it deferred to that decision.

More importantly, nothing in the majority's approach to ‘deference’ or the ‘presumption of constitutionality’ is specific to proportionality analysis. If the majority is right on these points, then it is difficult to see what the judicial role in rights review is. It is even more difficult to see what the role of ongoing political debate is, given the majority's ignoring Parliamentary debates on the mandatory death penalty after March 2017.

Nor can we say that the majority had applied a doctrine of non-justiciability in a principled manner. If it had, it would have said more about when the court is to take a hands-off approach, rather than that the court must give way to the legislature in ‘controversial matters of policy involving differing views on the moral and social issues involved’.²³⁷ (Is there any case involving constitutional rights that does not involve such a controversy?) Nor would it have said that a law depriving one of the right to personal liberty (or, indeed, the right to one's life – which the Federal Court itself had previously accepted was ‘the most fundamental of human rights’²³⁸) is constitutionally valid, regardless of whether it derogates from the right to a fair trial, merely because it is ‘duly enacted by Parliament’.²³⁹

Finally, one would think that the Federal Court's sudden departure from its own decision in Alma Nudo Atenza – a case from just the previous year involving the right to life, which it said is the ‘most fundamental of human rights’²⁴⁰ – would be accompanied by an explanation for the departure and/or a description of an alternative method by which to perform proportionality analysis. Unfortunately, there was none.

In such circumstances, it is difficult to resist the conclusions drawn by Nallini Pathmanathan FCJ in her dissenting judgment that ‘despite the disparity in the widely varying circumstances in which those offence occurred, the perpetrators are dealt with identically’,²⁴¹ that there is no ‘reasonable basis for placing all persons who have been convicted for “trafficking in drugs” … within the same class, for the purposes of punishment’,²⁴² and that the mandatory death penalty is therefore not ‘[a] proportionate punishment for the crime’ because it does not allow for ‘consideration of whether the punishment is commensurate to the gravity of the offence in every case.’²⁴³ Whether or not one agrees with these conclusions, the majority should have engaged in such a discussion instead of merely asserting that the matter is ‘controversial’. If nothing else, Nallini Pathmanathan FCJ's judgment shows that the controversy has a legal dimension and cannot be waved away as being purely a matter for politicians.

A coda: A possible way forward might take inspiration from the Singapore Court of Appeal's recent decision in Tan Seng Kee,²⁴⁴ in which the Court of Appeal ordered that a statute that was the subject of ongoing controversy – coupled with political promises of non-enforcement – was not to be enforced. The Singapore court was concerned with using the legal means at its disposal to preserve space for political processes as ‘the more obvious choice than litigation for debating and resolving highly contentious societal issues’.²⁴⁵ This is the very aim of the doctrine of deference, whether applied in the specific context of proportionality analysis or otherwise. The Malaysian courts may well wish to consider such an approach, according to which it would be ‘contrived and unrealistic’ to ignore political developments relating to legislation when determining the constitutionality of that legislation.²⁴⁶

Later Federal Court cases: cause for cautious optimism, or false hope?

Following Letitia Bosman, there have been two Federal Court cases which, taken together, give us cause for cautious optimism – but no more than that – about the future of constitutional review in Malaysia.

Malaysiakini

The first case is Malaysiakini, which involved the freedom of expression. Malaysiakini is an ‘online news portal’.²⁴⁷ The case concerned comments which several readers had posted on one of Malaysiakini's articles. These comments made various allegations against the Malaysian judiciary.²⁴⁸ The Federal Court, by a majority of six to one, found Malaysiakini guilty of contempt of court and fined Malaysiakini 500,000 Malaysian ringgit (as of the date of judgment, £88,256 or US$123,749).²⁴⁹ According to the majority, although Malaysiakini had merely provided the platform on which those comments had been posted, it was ‘deemed to have published the impugned comments’.²⁵⁰ This was because section 114A of the Evidence Act created a presumption that a person who ‘in any manner facilitates to publish or re-publish [a] publication is presumed to have published or re-published the contents of the publication unless the contrary is proved’, and Malaysiakini had failed to rebut this presumption.

Malaysiakini argued that holding it liable would violate the constitutional right of freedom of expression. This appears to refer to the rights both of Malaysiakini to ‘disseminat[e] information’ and of commenters to engage in ‘public discussion on matters of public interest’.²⁵¹

The dissenting judgment: proportionality in substance, even if not in name

Nallini Pathmanathan FCJ was the only dissenting judge. She did not need to engage with the constitutional point in order to reach the conclusion she did, because she found that Malaysiakini had successfully rebutted the statutory presumption that it had published the statements. Nonetheless, she was anxious to prevent a ‘dilut[ion] [of] the protection to freedom of expression under art. 10 of the Federal Constitution’.²⁵²

Nallini Pathmanathan FCJ's judgment is worth studying because it illustrates how Malaysian courts can perform meaningful constitutional rights review other than through structured proportionality analysis. Pathmanathan FCJ did not deny that the case involved ‘third-party commentators utilising their anonymity to direct unwarranted abuse, amounting to contempt, at the Judiciary’.²⁵³ But she also said that if an Internet intermediary were to be held liable for all comments made by users that happened to be contemptuous, this would ‘plac[e] an undue burden on entities for the contemptuous publications of others’ and incentivise the intermediary to remove ‘non-contemptuous material’ out of an abundance of caution.²⁵⁴ This, in turn, would create a ‘chilling effect on freedom of expression in the media’, which would be ‘detrimental and anathema to art. 10 of the Federal Constitution’.²⁵⁵ One detects echoes of the old ‘ineffective or illusory’ test – could this not be a viable test for Malaysian law to adopt, even if proportionality analysis is ultimately to be rejected?

Pathmanathan FCJ concluded that it would be acceptable for the law to require Internet intermediaries to remove a contemptuous comment within a ‘reasonable time’ after gaining actual knowledge of the comment.²⁵⁶ In taking this approach, Pathmanathan FCJ was concerned to see to it that constitutional rights cannot be defeated by a mere assertion that the state has some countervailing interest. Whether or not we describe this approach as ‘proportionality’ or ‘unstructured proportionality’,²⁵⁷ the important point is that it is meaningful rights review. Pathmanathan FCJ did not simply apply a test that, taken to its logical conclusion, any duly enacted legislation could easily pass.

The majority's lack of principled rights review

By contrast, the majority's judgment is wanting. The majority began by noting – quite rightly – that an ‘Internet content provider’ can, in principle, commit wrongs such as defamation, just as a newspaper publisher can.²⁵⁸ The majority then said that a court would face ‘difficulties … in pinning down the role of publication on the Internet content provider when the comments were made and posted by third parties … It must be to resolve this difficulty that the Malaysian Parliament enacted s[ection] 114A of the Evidence Act.’²⁵⁹ In short, the legislation aims to pin liability on somebody identifiable.

The majority went on to hold that Malaysiakini had not rebutted the presumption. The reasons for this conclusion are best discussed elsewhere. What is important, for present purposes, is the majority's treatment of Malaysiakini's constitutional argument. The majority held that:

[Malaysiakini] cannot insist on exercising its fundamental right and at the same time violate the right of others. A proper balance must be struck between the freedom of speech and expression enunciated and guaranteed in art. 10 of the Federal Constitution and the need to protect the dignity and integrity of the courts and the Judiciary. Case laws are replete with this entrenched principle of law that the exercise of this right is never absolute given the phrase ‘subject to’ provision appearing at the forefront of art. 10.²⁶⁰

But what constitutional ‘right of others’ was engaged? As the majority itself acknowledged, the purpose of the law on contempt of court is ‘not to protect the dignity of individual judges, but to protect the administration of justice’. If there is a constitutional right to ‘the administration of justice’, this is by no means apparent from the face of the Federal Constitution.

And even if such a constitutional right exists, the majority's reasoning on the freedom of expression boils down to this: the freedom of expression is ‘never absolute’. This says everything and nothing: nobody can disagree that the right is not absolute, but the real question is when and to what extent it can be curtailed. The majority said nothing on the applicable legal test, whether it be proportionality or something else.

The majority's lack of engagement with previous cases taking a more robust approach

There are, to be sure, past cases containing similar dicta about the constitutionality of the law on contempt vis-à-vis the right to freedom of expression. Crucially, however, those cases do not stop at saying that the right is not unlimited (as though any limit to it would be constitutionally acceptable). Rather, they show a judicial effort to consider whether the right has been appropriately balanced against legitimate grounds for limitation.

For example, in the 2019 case of PCP Construction,²⁶¹ the Federal Court endorsed previous cases holding that a ‘proper balance … between the right of speech and expression as provided for in Article 10 of the Federal Constitution and the need to protect the dignity and integrity of the Superior Courts in the interest of maintaining public confidence in the Judiciary’ was struck by the rule that ‘criticism of a judgment, however vigorous … within the limits of reasonable courtesy and good faith’ is not contempt.²⁶² Whether or not this is proportionality analysis is beside the point. What is clear is that the right to freedom of expression cannot be defeated by just any assertion of a state interest, however legitimate that state interest may be and however severely the right is to be curtailed. It is puzzling that a similar style of reasoning is missing from the majority's judgment in Malaysiakini.

A more explicit example involving the protection of the administration of justice is Najib Razak. There, the High Court, following local authorities as well as cases from Canada, India, and England and Wales, held that an order prohibiting the discussion of an ongoing case could only be made if it would be ‘necessary to prevent an immediate threat of a real and substantial risk of serious prejudice to the administration of justice in the relevant proceedings, in the absence of alternative measures, and is proportionate in reference to the competing interests of free speech and risk of prejudice to a fair trial.’²⁶³ On the facts, there was no such ‘immediate threat’. The Court of Appeal²⁶⁴ and Federal Court²⁶⁵ agreed with this approach, which is not dissimilar to that of Pathmanathan FCJ in Malaysiakini.

One would think that, given such precedents, the majority in Malaysiakini would have said something more about the freedom of expression than that it is ‘never absolute’. But it did not.

Anwar Ibrahim

In Datuk Seri Anwar Ibrahim, the Federal Court did take a more robust approach. However, its judgment is haunted by the ghost of Letitia Bosman and its accompanying doctrine of what is effectively blanket non-justiciability.

The applicant²⁶⁶ applied to the High Court for a declaration that, among other things, the National Security Council Act 2016 (NSCA) – particularly section 22 – was unconstitutional. He then attempted to invoke a provision of the Courts and Justice Act that allowed the High Court to refer a constitutional question directly to the Federal Court.

Perhaps cause for optimism about constitutional review …

In what I will call the ‘first decision’, the majority of the Federal Court declined to answer the question because it was ‘abstract and purely academic’.²⁶⁷ The dissenting judges,²⁶⁸ having considered that the Federal Court should answer the question, went on to ask whether the NSCA passed the proportionality test. They concluded that it did not. Section 22 of the NSCA allows the Director of Operations to exclude any person from a ‘security area’ declared by the Prime Minister.²⁶⁹ But there are no limits on ‘what exactly comprises a ‘security area’, how large such an area may be declared, and for how long such a declaration may persist’.²⁷⁰ Moreover, the Director of Operations may ‘re-settle’ persons from the ‘security area’ to an indefinite place for an indefinite period.²⁷¹ According to the dissenting judges, because of the open-endedness of these powers, section 22 of the NSCA cannot be the ‘least intrusive’²⁷² measure to achieve the aim of ‘preserving national security’.²⁷³ Moreover, these judges said, because section 22 is ‘inextricably linked’ to the rest of the NSCA, the entire Act is unconstitutional.²⁷⁴

The following year, a differently-constituted panel of the Federal Court agreed (in what we will call the ‘second decision’) to re-hear the case.²⁷⁵

The Federal Court did so and issued its ‘third decision’. Here, the majority once again held that the question was ‘abstract and hypothetical’,²⁷⁶ but went on to answer the question anyway.²⁷⁷ The majority's analysis calls for cautious optimism. (We need not consider the minority judges’ view that the Act was unconstitutional, because this view was for reasons that did not pertain to constitutional rights.)

The majority, citing Alma Nudo Atenza, affirmed that it had to perform proportionality analysis.²⁷⁸ This time, the majority held that the Act passed the proportionality test. At first glance, it appears that the majority baldly asserted that section 22 of the NSCA ‘has a rational nexus and is proportionate to the objective to be addressed, namely, national security’.²⁷⁹ However, while the majority's judgment does not explain why section 22 is rationally connected to a legitimate aim, the majority ultimately took a nuanced approach. The following passage is worth repeating:

Even in the matter of declaring an area as a security area, the presence of such provisions does not ipso facto render the NSCA invalid and unconstitutional … [T]he circumstances in which an area is declared a security area are stringent, that it is only where the threat is grave and has potential to cause serious harm; where it would be imperative and necessary to exclude or evacuate persons from a security area. This is as provided in s. 18 itself … Hence, the gravity of the threat and the urgency of response are key or paramount elements to any valid exercise and recourse to s. 18. Section 18 implicitly recognises the doctrine of proportionality and has prescribed conditions before its aid may be resorted to. The availability of access to court and to the protection of the law is further undisturbed.²⁸⁰

This being so, the mere existence of the power to declare a security area was not disproportionate, but a particular decision to invoke that power can be.²⁸¹ One can assume that this reasoning also applies to the power under section 22 to exclude a person from a security area.

The majority's decision is to be welcomed because it makes clear that, even if the existence of a legal power is not disproportionate, each use of the power still stands to be tested for its proportionality (and, a fortiori, for its constitutionality). In this regard, it is worth noting that the Malaysian courts have previously demonstrated a readiness to hear similar challenges to individual exercises of executive power.²⁸² This is arguably a form of ‘due deference’, in that the court will not pre-empt the possibility that the executive acts compatibly with constitutional rights.

… but only heavily qualified optimism

But several dampeners must be put on this development.

The first problem is evident from the following passage from the majority's judgment in the third decision:

It must always be borne in mind that matters of security involve policy consideration which are within the domain of the Executive … courts do not possess knowledge of the policy consideration which underlay administrative decisions; neither can the courts claim it is ever in the position to make such decisions or equipped to do so … [R]egardless how challenge is mounted [sic], where matters of national security and public order are involved, the court should not intervene and should be hesitant in doing so as these are matters especially within the preserve of the Executive, involving as they invariably do, policy considerations and the like.²⁸³

Should the court ‘not intervene’, or merely ‘be hesitant in doing so’? In any event, the passage brings with it the spectre of the majority's judgment in Letitia Bosman, which amounted to a doctrine of strict non-justiciability and left the courts with little or no meaningful role. Now, one could see this passage as a positive development, in that this doctrine applies only to cases involving ‘national security and public order’. But the problem is that, as we have seen, ‘public order’ is a very broad term.²⁸⁴ Surely the court ought to consider which ‘public order’ issues call for greater or less deference, and not merely assume that any ‘public order’ issue must be exclusively for the executive to determine.

It is true that where ‘national security’ is involved, the degree of deference owed the political branches is higher: as the English courts have remarked, ‘[t]hose who are responsible for the national security must be the sole judges of what the national security requires’,²⁸⁵ and ‘the court is in no position to substitute its opinion for the opinion of those responsible for national security’.²⁸⁶ Even then, it is not true that ‘the court should not intervene’. At the very least, the court must satisfy itself as to ‘the essential facts to which the opinion or judgment of those responsible’²⁸⁷ and, as a Singapore court held, ‘determin[e] whether the decision was in fact based on grounds of national security’.²⁸⁸ At the risk of over-simplification: the court might do less, but the court cannot do nothing.

Finally, while the third decision in Anwar Ibrahim took its cue from Alma Nudo Atenza in applying proportionality analysis, it is notable that it also cited Letitia Bosman as authority for the ‘cardinal principle of the presumption of constitutionality’.²⁸⁹ That presumption, according to Letitia Bosman, would militate against proportionality analysis in a case involving a ‘controversial’ matter; a fortiori, one would think, in a case involving ‘national security’. So there are three Federal Court cases pointing in different directions: Alma Nudo Atenza calls for proportionality analysis; Letitia Bosman claims to perform it, but is in substance antithetical to it; and Anwar Ibrahim, confusingly, appears to endorse both. Then, there is the separate decision of Malaysiakini, in which the majority said nothing about the matter at all.

Conclusion

Alma Nudo Atenza may have appeared to be the high-water mark of proportionality review – and, more generally, constitutional rights review – in Malaysia. But, as this article has shown, there were serious underlying problems with the proportionality doctrine, which spilled over into how the courts have regarded the very enterprise of constitutional rights review. The Federal Court's majority judgment in Letitia Bosman casts a pall over even later cases where we see the courts playing anything more than a rubber-stamping role. Rights review in Malaysia is now lost at sea; it is very difficult for a lawyer to state the law on constitutional rights with certainty, or a legal subject to predict how their constitutional rights are to be protected by the courts. Even if one disagrees that proportionality should apply, other models of rights review, such as the ‘ineffective or illusory’ test,²⁹⁰ also seem to have disappeared.

Much of the discussion in this article has centred around proportionality analysis because, for better or worse, so has the discussion in the case law. But I have aimed to show that there are deeper problems relating to how the courts see themselves, and the Legislature, as figuring into the enterprise of giving effect to constitutional rights.

It is difficult to chart a course on where the case law should go next, nor how to get there. What is clear is that Malaysian law has lacked a nuanced account of judicial deference, and has clung to the flawed notion of the ‘presumption of constitutionality’. As a result, the law is haunted by the ghosts of several old cases which are completely antithetical to the very enterprise of constitutional review. So there is a risk that rights-related matters are placed entirely in the Legislature's hands, and reducing the courts’ important role in safeguarding constitutional rights to near vanishing point. It is hoped that this article has unpacked how the law got to where it is now, if nothing else.

What is now needed is deeper engagement with the issues which this article has aimed to highlight and explore. Judges opposed to certain forms of rights review must squarely confront the implications of this view for the separation of powers, and consider the question of what precisely the court's role is. Meanwhile, judges in favour of a more assertive judicial role must be prepared to defend their views against accusations – not entirely unfounded – that their views, taken to the logical conclusion, would slip into juristocracy.

In this spirit, it is hoped that this article may help to facilitate more meaningful discourse about constitutional review in Malaysia. It has not purported to furnish a complete solution, but, hopefully, has provided notes on how conversations to that end may take place, and urged that they take place soon.