Three Models of Rights Protection

Over the last two centuries, constitutional experience from around the world has shown that there are two divergent models of rights pro­tection. They represent differing constitutional arrangements in the way in which they balance decision-making power between the legis­lature and the judiciary.

Political parties that represent contrasting interests and offer dif­fering perspectives safeguard rights through the democratic process. Governments that impinge upon the rights of citizens through primary legislation are not directly sanctioned by courts, but are confronted with paying the political price to the electorate. Parliamentary sovereignty, a concept most famously expounded by A.V. Dicey,^[116] the Vinerian Professor of English Law at the University of Oxford, emerged in Britain and trav­elled to former British colonies, including Australia^[117] and New Zealand.

On the other hand, the model of judicial supremacy (or 'strong-form judicial review'^[118]) prevails in jurisdictions that have a codified, canoni­cal constitution. In systems of judicial supremacy, legislative power is subjected to constitutional limitations that are enforced by courts by the exercise of a power to strike down, invalidate, or disapply primary legislation. Courts enjoy the last word on the resolution of issues relating to rights, at least within the existing constitution or bill of rights.^[119] In other words, judges, who have more modest representative credentials than legislators, are entrusted with policing the boundaries of rights.

These models reflected a binary choice that lay before newly independent nations or those undergoing constitutional transition. Based on the logic of the US Supreme Court's judgment in Marbury v. Madison,^[122] they would need to decide between a higher-law written Constitution or bill of rights policed by independent courts, and a constitution that is 'on par with regular law', which left questions of rights resolution to elected representatives.^[123] It should be noted, how­ever, that these traditional models did not represent monoliths, and there remained scope for variation within them. So, for example, within parliamentary sovereignty, a sovereign legislature may or may not have been empowered to enact legislation prescribing the 'manner and form' of future legislation, thereby effectively imposing limits upon itself.^[124] Within models of judicial supremacy, the authority to determine the rights-compatibility of legislation could be centralized in a constitu­tional court or superior courts, or dispersed among courts at all levels.^[125] Further, courts could be empowered simply to disapply legislation or to strike it down, removing it from the statute book altogether as though it had been repealed.^[126] These variations represented options within each of the traditional models, but left little room for any significant conver­gence between them.

Each of these traditional models is, in some sense, considered imperfect. Parliamentary sovereignty, although democratically legiti­mate, bears the risk of leaving behind minorities that suffer inadequate representation in the democratic process. Judicial supremacy, which leaves the process of adjudication of rights to independent courts, gives rise to apprehensions of democratic legitimacy that are well captured by the notion of the 'countermajoritarian difficulty⁷.^[127] The principal concern is whether it is legitimate, in systems of judicial supremacy, to allow unaccountable judges to thwart decisions that are reflective of the public will.

Over the last three decades, a third alternative to these traditional models has emerged. This model seeks to capture the best aspects of the two traditional models, while leaving aside their weaknesses. The high­light of the new model is that it enables both courts and the legislature to assert their competing understanding of rights. Courts have a limited power of review over primary legislation, which can be assailed by the legislature, acting in its ordinary capacity, in different ways. In Canada, section 33 of the Charter of Rights and Freedoms, 1982 empowers the legislature to declare, by ordinary majority, that a statute (or one of its provisions) will operate 'notwithstanding' certain provisions of the Charter. In New Zealand (under the New Zealand Bill of Rights Act, 1990), the UK (under the HRA), the Australian Capital Territory (under the Human Rights Act, 2004), and the Australian state of Victoria (under the Victorian Charter of Human Rights and Responsibilities Act, 2006), courts can make a declaration of inconsistency or incom­patibility when they consider primary legislation to be inconsistent with rights.^[128] However, it remains the task of the legislature to decide whether to take heed of such a declaration by amending the legisla­tion in any way. In this manner, the model seeks to tactfully reconcile democratic legitimacy with rights-based judicial review.

Constitutional scholars ascribe many labels to this third alternative. These include 'the new commonwealth model of constitutionalism',^[129] 'weak-form judicial review',^[130] the 'parliamentary bill of rights' model,^[131] the 'hybrid model' of judicial review,^[132] the 'third way' or 'third wave' bill of rights,^[133] the 'democratic dialogue' model,^[134] and 'weak-form bills of rights'^[135] (for the sake of convenience, these will collectively be referred to as the 'new model labels'). While recognizing subtle differences between them, Stephen Gardbaum, amongst the most well-known proponents of this new model, acknowledges that all these descriptions are alternative terms for the same phenomenon.²³ This is probably an overstatement—the more accurate way of understanding these labels is that all their proponents agree about the centrality of one aspect of the model: the fact that courts have the power to review primary legislation without being ascribed a definitive role in the process.²⁴

However, apart from this central aspect of the third model, the dif­ferent labels used by scholars reflect underlying differences in focus. So, for example, Tushnet's use of the label 'weak-form judicial review' is suitable to his heavy focus on decisions of courts in his vast scholarship on the subject.²⁵ Although he defends the model as a general one,²⁶ Gardbaum's use of the 'commonwealth' tag suggests that the new model will be better suited to commonwealth jurisdictions.²⁷ Klug's

²³ Gardbuam (n. 6), p. 14.

²⁴ Aileen Kavanagh, 'What's So Weak about "Weak-form" Review? The Case of the UK Human Rights Act 1998' (2015) 13 International Journal of Constitutional Law 1008.

²⁵ Tushnet (n. 17); Mark Tushnet, 'Forms of Judicial Review as Expressions of Constitutional Patriotism' (2003) 22 Law and Philosophy 353; Mark Tushnet, 'Weak-form Judicial Review: Its Implications for Legislatures' (2004) 2 NZJPIL 7; Mark Tushnet, 'Social Welfare Rights and the Forms of Judicial Review' (2004) 82 Texas Law Review 1895; Mark Tushnet, 'Weak-form Judicial Review and "Core" Civil Liberties' (2006) 41 Harvard Civil Rights-Civil Liberties Law Review 1; Mark Tushnet, Weak Courts, Strong Rights: Judicial Review and Social Welfare Rights in Comparative Constitutional Law (Princeton University Press 2008); Mark Tushnet, The Rise of Weak-form Judicial Review' in Tom Ginsburg and Rosalind Dixon (eds), Comparative Constitutional Law (Edward Elgar 2011), p.

²⁶ Stephen Gardbaum, The Single and General New Commonwealth Model: A Response to Ran Hirschl and Sujit Choudhry' (2013) 11 International Journal of Constitutional Law 1100.

²⁷ See Ran Hirschl, 'How Consequential Is the Commonwealth Constitutional Model?' (2013) 11 International Journal of Constitutional Law 1086, p. 1088, suggesting, in response to Gardbaum's work, that there could be something 'in the combination of characteristics shared by the countries under study—all relatively prosperous, stable democracies, and all current or former British colonies sharing a roughly similar legal tradition—that makes them con­ducive to this particular model of judicial review'.

use of the 'third wave' bill of rights idea clearly reflects her attention to the changing understanding of human rights across different periods of history. Alison Young's chosen label—the 'dialogue model'—reflects her emphasis on the importance of dialogic interactions between the courts and the legislature under this new model of judicial review, an issue which is addressed later in the chapter.

Scholars also hold different views about what the new model includes—more specifically, whether 'pre-enactment political rights review'^[136] or executive/legislative review of statutory provisions before they are enacted as law, forms an essential element of the new model of constitutionalism. In his early work on the subject, Gardbaum did not include political rights review in expounding upon the new model of constitutionalism.^[137] However, in his book that was published more than a decade later, he described the new model as a 'combination' of mandatory pre-enactment political rights review and weak-form judi­cial review.^[138]

What becomes clear from these discussions is that the ability of courts to review primary legislation for compliance with rights without enjoying primacy over the legislative will is the sine qua non of the new model of rights protection, however one chooses to package it overall.

At this stage, it is useful to consider whether this framework ascribes undue importance to section 4 of the HRA, ignoring the equally impor­tant directive under section 3 for British courts, 'so far as it is possible to do so', to read legislation in compliance with rights under the European Convention for the Protection of Human Rights and Fundamental Freedoms, 1950 (Convention). Two things should be made quite clear. First, section 3 (rather than section 4) is the principal remedy under the HRA. This was evident from the Lord Chancellor's observation in Parliament during the debates on the HRA that in 99 per cent of cases, there would be no need for declarations of incompatibility.^[140] Over the last few years, the case law has reflected this bias in favour of sec­tion 3.^[141] As Appendix B shows, between the years 2000 and 2015, there have been only 20 declarations of incompatibility under the HRA that have survived appeal. Second, section 3 enables courts to depart from the intention of the legislature enacting the statute that needs to be read in compliance with Convention rights.

Both these matters are unexceptionable. In most constitutional sys­tems, courts do their best to interpret legislation consistently with rights before exploring alternatives. Indeed, such interpretation sometimes departs from legislative intent, and there are well-known examples of this in Indian case law,^[142] as well as UK case law predating the HRA.^[143] But rights-compliant interpretation tends to convey the message that the government and Parliament have a 'clean bill of health' in human rights terms.^[144] Declarations of incompatibility are different—they involve open recognition of the fact that Parliament has legislated contrary to Convention rights, and that such legislation can neither be quashed nor be saved from the shadow of inconsistency through a rights-compliant interpretation. Therefore, even though section 3 is the HRA's primary remedy, section 4 is, in common with its equivalents in other jurisdictions embodying the new model, its distinctive feature.

Continuing the discussion on the meaning and scope of the new model, some scholars have used one or more of the new model labels to refer to more than just the ability of courts to review, without presump­tive finality, statutes enacted by democratically elected representatives. Aileen Kavanagh points out, for instance, that Tushnet has, across his different writings, used the 'weak-form' terminology to refer to 'insti­tutions, courts, issues, rights, statutes, cultures, or remedies'.^[145] On the other hand, Dixon has used the 'weak-form' label to refer to approaches to rights adjudication, rights themselves, as well as remedies,^[146] while Khosla uses it in the context of the intensity of judicial review.^[147] This book will avoid these expansive uses of the new model labels, and focus on the new model as a choice of formal constitutional design. It will consider the new model as a way of seeking to design institutional arrangements on the basis of 'speculative predictions' (made at the time of enactment of the constitution, bill of rights, or constitutional document), of how institutions will 'actually function'.^[148]

Treating the new model as a function of constitutional design enables us to effectively judge the promise of the new model against its performance in particular jurisdictions (in this case, the UK). For exam­ple, the formal design of the HRA makes it clear that it set up the new model, a compromise between parliamentary sovereignty and judicial supremacy. This is done principally (though not exclusively) through section 4, which empowers courts to make a declaration of incompat­ibility when primary legislation is inconsistent with Convention rights.

The expectation was that it would offer a more balanced allocation of the power of rights resolution between the Westminster Parliament and the courts. After examining the functioning of the HRA, it may turn out that this balanced allocation has not been achieved, and that either Parliament or the courts enjoy the upper hand in asserting their understanding of rights. It is precisely this exercise of setting out the promise of the HRA against its performance in practice (compared with the performance of the model of judicial supremacy in India), that this book engages in.

Similarly, judicial supremacy (in India) will also be defined from the standpoint of constitutional design. As explained in the Introduction, from the institutional arrangements established by the Indian Constitution, it was clear that courts were entrusted with the power to strike down primary legislation that violates rights. The expectation in jurisdictions that establish judicial supremacy is that courts will have the definitive word on the meaning of rights. Practice may meet this expectation—for example, the legislature may indeed have little option but to go along with courts' understanding of rights with which it disagrees. On the other hand, constitutional practice may defy expecta- tions—by giving the legislature a greater role in rights resolution than expected.

It is worth pointing out that constitutional design, which assists us in deciding which of the three options (parliamentary sovereignty, judicial supremacy, or the new model) has been put into place in a new constitutional settlement, does not simply boil down to the constitu­tional text.^[149] This would be an under-inclusive understanding of consti­tutional design. A better approach would be to consider constitutional design as relating to the institutional arrangements put in place by a constitutional document or bill of rights—the text being an important part of, but not the only, evidence of such arrangements. The most obvi­ous example of this understanding comes from the US. Although the power of judicial review did not flow directly from the constitutional text, scholars have argued that Marbury was not simply an act of 'self­articulation' by judges, but reflected institutional arrangements that were put in place at the time of the Constitution's founding.^[150]

The Indian Constitution offers us another good example. There is no single provision of the constitutional text that categorically states that courts have the power to review legislation on substantive, rights-based grounds. Some scholars have pointed to Art. 13(2), which provides The State shall not make any law which takes away or abridges the rights conferred by this Part and any law made in contravention of this clause shall, to the extent of the contravention, be void.'^[151] But this provision seems under-inclusive, because it only applies to one part of the Constitution (the chapter on fundamental rights), whereas Indian courts have the power to strike down legislation that impinges upon any mandatory provision of the Constitution. Others have pointed to Arts 3 72(1)^[152] and 245(1)^[153] of the Indian Constitution.^[154] Once again, the former is under-inclusive, because it applies only to laws that were enacted before the Constitution entered into force, whereas the latter only seems to mark the division of legislative competence between the Union and the states. Nevertheless, it has always been clear that the institutional arrangements established by the Constitution empowered courts with reviewing (and striking down) primary legislation that vio­lates constitutional rights.^[155]

Understanding these models of constitutionalism as choices in constitutional design also leads us towards a useful starting point for the comparative analysis of two systems which have, in design terms, adopted quite different models of rights protection. Failing to do so gives rise to classificatory challenges that have prompted scholars to arrive at widely differing categorizations of the two constitutional systems being studied. The UK and India have each, on different occa­sions, been labelled as 'weak-form' (or new model) and 'strong-form' (or judicial supremacy) systems.^[156] This book avoids such classificatory knots by safely categorizing the HRA as embodying the new model of judicial review (since it empowers courts to review primary legislation without presumptive finality) and the Indian Constitution as represent­ing the traditional model of judicial supremacy (since it enables courts to strike down legislation without the possibility of being overridden by ordinary legislative procedures). As mentioned earlier, whether these models actually achieve the outcomes that they were expected to achieve is an issue that this book explores.

Subject to what has been said about conceiving of models of rights protection as functions of formal constitutional design, this book will not insist upon the use of any one of the new model labels to describe this third model of judicial review. Throughout the book, the 'new commonwealth model', 'new model', or 'weak-form review' labels will be used interchangeably. The only label that will be consciously avoided is the phrase 'dialogic review'. This is because the metaphor of dialogue is over-inclusive. It is difficult to conceive of any system of judicial review which does not facilitate inter-branch dialogue, or some kind of interaction between courts and the legislature.^[157]

An example from India sheds light on the inter-branch dialogue that takes place in systems of judicial supremacy, just as it does in systems of weak-form judicial review. In 2005, the legislature of Maharashtra enacted primary legislation^[158] imposing a ban on dance performances in bars, except in certain establishments (such as hotels rated three stars and above). The Indian Supreme Court struck down this ban as being unconstitutional on two grounds: that it was discriminatory and that it deprived the dancers of their right to livelihood.^[159] The state govern­ment sidestepped compliance with the judgment and soon thereafter, the legislature passed a bill banning dance performances in all bars in the state.^[160] Even though it is unlikely that the newer legislation will survive judicial review, this clearly instantiates the dialogic interactions that can, and often do, take place between courts and the legislature in systems of judicial supremacy.