Stephen Gardbaum

As a recent and ongoing experiment in constitutional design, the new Commonwealth model of constitutionalism may be something new under the sun. It represents a third approach to structuring and institutionalising basic constitutional arrangements that occupies the intermediate ground in between the two traditional and previously mutually exclusive options of legislative and judicial supremacy. It also provides novel, and arguably more optimal, techniques for protecting rights within a democracy through a reallocation of powers between courts and legislatures that brings them into greater balance than under either of these two lopsided existing models. In this way, the new Commonwealth model promises to be to forms of constitutionalism what the mixed economy is to forms of economic organization: a distinct and appealing third way in between two purer but flawed extremes. Or, it may be, as some have claimed, more like a comet that shone brightly and beguilingly in the constitutional firmament for a brief moment but quickly burned up, a victim of the inexorable law of the excluded middle. In exploring the theory and practice of the new Commonwealth model, this book assesses whether ink or eraser is the better response to its current pencilled-in status on the short list of alternatives from which constitutional drafters everywhere make their momentous decisions.

« The new Commonwealth model of constitutionalism » — « the new model » for short — refers to a common general structure or approach underlying the bills of rights introduced in recent years in Canada (1982), New Zealand (1990), the United Kingdom (1998), the Australian Capital Territory (2004) and state of Victoria (2006). This approach self-consciously departs from the old or traditional Commonwealth model of legislative supremacy, in which there is no general, codified bill of rights; rather, particular rights are created and changed by the legislature through ordinary statutes on an ad hoc basis. Under this model, courts have no power to review legislation for infringing rights, as rights are not limits on legislation but its product, and are changeable by it. In this way, legislatures are supreme because they ultimately determine what legal rights there are and how rights issues are resolved. The judicial function is limited to faithfully interpreting and applying whatever laws the legislature enacts.

At the same time, however, the new model also contrasts with the alternative standard option for institutionalising basic constitutional arrangements: namely, judicial or constitutional supremacy. Here, there is a general, codified bill of rights, which imposes constitutional limits on legislative power. These limits are enforced by authorising courts to review legislation for consistency with the bill of rights and to invalidate or disapply statutes that, in their final view, infringe its provisions. As a result, courts are supreme because they have the last word on the validity of legislation and the resolution of rights issues, at least within the existing bill of rights.

As we shall see in detail in the next section, the new model’s novel third approach calls for the enactment of a bill of rights — although not necessarily one that imposes constitutional limits on the legislature — and its enforcement through the twin mechanisms of judicial and political rights review of legislation, but with the legal power of the final word going to the politically-accountable branch of government rather than the courts. In this way, the new model treats legislatures and courts as joint or supplementary rather than alternative exclusive protectors and promoters of rights, as under the two traditional models, and decouples the power of judicial review of legislation from judicial supremacy or finality.

I. What is the new Commonwealth model and what is new about it?

In essence, the new Commonwealth model of constitutionalism consists in the combination of two novel techniques for protecting rights. These are mandatory pre-enactment political rights review and weak-form judicial review.

The first technique requires both of the elective branches of government to engage in rights review of a proposed statute before and during the bill’s legislative process. The formalized, mandatory and deliberate nature of political rights review under the new model distinguishes it from characteristic practices under both other forms of constitutionalism, where if any such review occurs it tends to be ad hoc, voluntary and unsystematic^[1]. Political rights review is a direct and alternative response to the standard concerns about legislative/majoritarian rights sensibilities that underlie the traditional argument for judicial review of legislation. Political rights review is designed to take this concern seriously and to address it directly, at the horse’s mouth as it were, by ensuring that the general rights consciousness of the executive that proposes bills and the legislature that considers and enacts them is raised and that specific rights concerns are identified and aired during the legislative process^[2]. In other words, political rights review provides an internal solution to this potential problem that transfers some of the responsibility for rights protection from the external and more indirect mechanism of judicial review to the legislature itself. As such, it also supplements a purely ex post technique of rights protection with an ex ante one, with many of the associated general advantages of this type of regulation. In this context, ex ante regulation provides the only protection against those outputs of the legislative process that are never litigated for one reason or another,^[3] and a second layer in addition to ex post review for those that are.

The second technique of rights protection that is constitutive of the new model is weak-form judicial review. It is this technique that decouples judicial review from judicial supremacy, meaning that although courts have powers of constitutional review they do not necessarily or automatically have final authority on what the law of the land is. Unlike the case under judicial supremacy, their decisions are not unreviewable by ordinary legislative majority. This is because one of the defining features of the technique (and so of the new model) is that it grants the legal power — but not the duty — of the final word to the legislature. That is, in giving political discretion to the legislature whether or not to use it in any particular case, the new model creates a gap between this legal power and its exercise that distinguishes it from both legal and political constitutionalism. Whereas under strong-form judicial review and legislative supremacy, the institution with the power of the final word is essentially bound to exercise it and does so routinely, almost automatically — courts in the context of deciding a case or abstract review and legislatures because the act of passing a law is the final word — this is not so under the new model. In deciding whether (rather than how) to use their power, legislatures may be heavily influenced by the prior exercise of weak-form judicial review.

Here it is necessary to clarify both the relevant sense of judicial supremacy that the new model rejects and what is novel about the technique. The term judicial supremacy has become a little clouded as a result of the rise of dialogue theory. In Canada, where the theory originated and has its strongest hold, its proponents argue that, quite apart from the formal section 33 override power, the frequency of « legislative sequels » following the judicial invalidation of a statute means there is judicial-legislative dialogue and often de facto legislative supremacy, especially where such sequels are upheld by the courts^[4]. Even in the United States, it has been noted that a similar practice of legislative sequels and inter-institutional dialogue sometimes occurs, as exemplified by Congress’s continuing to create hundreds of legislative vetoes of executive action after the practice was declared unconstitutional by the Supreme Court in INS v. Chadha^[5]. This, it has been argued, means that in reality the meaning of the Constitution depends on interpretations put forward by legislators in opposition to those proposed by the judiciary and that no single institution, judiciary included, has the final word on constitutional questions^[6]. Putting aside the fact that this Chadha episode is unrepresentative of U.S. constitutional law as a whole because on separation of powers (as distinct from rights) issues it is well-known that legal resolutions generally play a lesser role than political ones^[7], this train of thought misses the specific and relevant finality issue. This is who has the final legal word on the validity and continuing operation of the particular existing law at issue in the litigation, not whether the judicial decision binds future legislative or executive acts — an issue about which there has long been divided opinion in the United States^[8]. But on this relevant issue for our purposes, there is no doubt or controversy: the judiciary has the final word on whether the specific law (or part of it) challenged in Chadha is the law of the land — and indeed, on the validity of any of the subsequently enacted legislative vetoes that may come before them. This is what, in context, strong-form judicial review refers to.^[9] By contrast, weak-form judicial review under the new model means that the legislature and not the judiciary has de jure finality, the legal power of the final word with respect to the specific law at issue — unlike in the United States or other regimes of judicial supremacy.

On the novelty of the technique, the concept of weak-form judicial review per se may not be original to the new model. This is because there are arguably other pre-existing constitutional theories that have a similar basic structure of judicial review without judicial finality and so can perhaps properly be called such. These include certain versions of departmentalism (each branch of government is the final interpreter of its own powers) and popular constitutionalism (the people are the final interpreters of constitutional meaning)^[10]. Nonetheless, weak-form judicial review as institutionalized within the new model is innovative in at least three ways. First, it is the general mode of judicial review under the new model, whereas it is only a partial or supplementary mode under these other theories, employed in certain areas but not others (e.g., separation of powers type issues under departmentalism) or triggered exceptionally or only periodically (e.g., popular constitutionalism). Secondly, the new model’s general mechanism of « penultimate judicial review »^[11] followed by possible exercise of the legislative override power is not one that seems to be present in the other theories, because either courts defer to the relevant other branch in the first place or it is the people themselves who have the final say. Indeed, the new model’s distinctive allocation of powers provides a far more tangible and concrete institutional mechanism of judicial non-finality than is present in most versions of popular constitutionalism and departmentalism^[12]. Thirdly, two of the new model’s specific mechanisms of weak-form review were entirely novel when introduced: namely, the « notwithstanding mechanism » contained in section 33 of the Canadian Charter of Rights and Freedoms (the Charter) and also section 2 of its predecessor, the Canadian Bill of Rights 1960 (CBOR),^[13] and the power of the higher UK courts to issue declarations of incompatibility under section 4 of the Human Rights Act 1998 (HRA)^[14].

These two techniques of political rights review and weak-form judicial review, which in combination define and distinguish the new model, can be further broken down into the following four essential institutional features, or jointly necessary and sufficient conditions. The first is a legalized and codified charter or bill of rights — as distinct from purely moral and political rights, residual common law liberties or a piecemeal collection of specific, stand-alone statutory rights. This bill of rights forms the subject-matter or focus of both political and weak-form judicial review and may have either constitutional or statutory status. In principle, it could even be judicially created, like pre-Charter EU human rights law^[15], which would satisfy the criterion of legal enforceability, although the codification requirement is likely to be inconsistent with the case-by-case, accretive methodology of the common law.

The second feature is mandatory rights review of legislation by the political branches before enactment. As we shall see, this is typically institutionalized by a requirement that a government minister provide a formal statement where he or she is of the opinion that a bill is incompatible with protected rights on its introduction in the legislature, which triggers both prior executive vetting and subsequent legislative scrutiny.

The third is some form of constitutional review of legislation by the courts. That is, a form of judicial power to protect and enforce these rights going beyond an interpretive presumption that the legislature does not intend to violate them or ordinary modes of statutory interpretation. From the perspective of traditional parliamentary sovereignty, these are enhanced or greater judicial powers to protect rights than previously existed. As we shall see momentarily, the required form of constitutional review may range from a duty to interpret legislation consistently with protected rights where reasonably possible to a judicial power of invalidation.

The fourth feature, notwithstanding this judicial role, is a formal legislative power to have the final word on what the law of the land is by ordinary majority vote. The specific form of this legislative power will vary according to the version of the constitutional review power granted to the courts — ranging from the power to amend legislation as interpreted by the courts under their rights-respecting duty to the power to override the judicial invalidation of legislation, with others in between.

In combination, the first and third features distinguish the new model from traditional parliamentary sovereignty; the fourth from judicial or constitutional supremacy. These essential features of the new model are quite general and permit a range of different specific instantiations, particularly with respect to the second and third features, some of which have in fact been adopted in various countries. So, on a spectrum in which traditional judicial and legislative supremacy mark the two poles, the new model has at least five different possible variations, thereby occupying five slightly different intermediate positions.

Starting from the judicial supremacy pole, the first of these is exemplified by the Charter: (1) a constitutional bill of rights (2) granting the judiciary power to invalidate conflicting statutes but (3) with a formal legislative final word in the form of the section 33 power exercisable by ordinary majority vote^[16]. The second is a statutory bill of rights granting the judiciary the same power to invalidate conflicting statutes, with a similar legislative override power. This position is most closely, although not exactly, illustrated by the still operative CBOR.^[17] The third version is exemplified by the HRA, the ACT Human Rights Act 2004 (ACTHRA) and the Victorian Charter of Human Rights and Responsibilities Act 2006 (VCHRR): a statutory bill of rights without the power of judicial invalidation of legislation but instead one new judicial power to declare statutes incompatible with protected rights that does not affect their continuing validity, and a second new judicial power (and obligation) to give statutes a rights-consistent interpretation wherever possible. Both types of judicial decision – declaratory and interpretive — are subject to the ordinary legal power of the legislature to have the final word: a default power in the case of the former and requiring affirmative action in the case of the latter. The fourth variation is a similar statutory bill of rights containing the second judicial power, the interpretive power/duty, but lacking the first or declaratory power. This was exemplified by the New Zealand Bill of Rights Act 1990 (NZBORA), at least until 2000 when the latter power was seemingly implied by the courts^[18]. A fifth variation would be granting the courts the declaratory power but only ordinary and traditional powers of statutory interpretation^[19].

A statutory bill of rights alone without either the interpretive duty or the declaratory power would not satisfy the third necessary feature of the new model and thus, whatever its independent merits, does not depart from traditional parliamentary sovereignty. Similarly, pre-enactment political rights review alone, with or without a bill of rights^[20]. Weak-form judicial review by itself is also insufficient, which is why certain stand-alone legislative override mechanisms in non-Commonwealth jurisdictions noted above amount to no more than a « partial » adoption of the new model. Analytically, once again, a common law version of the new model is perhaps possible but would (among other things) require a judicially-created bill of rights, along the lines of the one developed by the European Court of Justice, to satisfy the first criterion. Most problematic of all would seem to be the second, mandatory political rights review. If this could somehow be overcome, perhaps by development of a constitutional convention, there could in theory be common law versions of all three models of constitutionalism.

We have already seen that what is new about the new model is the following: (1) it transcends the standard dichotomy in institutional forms of constitutionalism, providing a third choice; (2) it does so by combining two novel techniques of rights protection; and (3) as part of this second, it provides a clear institutional mechanism for decoupling judicial review from judicial supremacy. Also as part and parcel of these characteristics, the new model establishes a distinctive and more balanced allocation of powers between courts and legislatures than under the two lopsided existing models. Thus, with their authority to engage in constitutional review, courts have greater powers than under political constitutionalism but their lack of de jure finality means less power than under any form of legal constitutionalism. And conversely, legislatures are faced with greater legal and judicial constraints on their actions than under political constitutionalism, but fewer than under legal constitutionalism.

This allocation of powers demonstrates that the new third option is specifically an intermediate one in between the two standard and traditional choices. Its intermediate nature can be further elaborated and explained in the following ways. First, it takes certain key ideas from each of the other two models and combines them into a distinct third option. By borrowing from both, the new model creates something in between. From the « big-C » version of legal constitutionalism, the new model first takes the importance of a comprehensive set of affirmative legal rights^[21], as distinct from the (a) mostly moral and political, (b) ad hoc statutory and/or (c) default, or negative, conception of rights and liberties as whatever is left unregulated by government that characterizes the traditional model of parliamentary sovereignty. From both forms of legal constitutionalism — « big-C » and common law — it also takes the importance of judicial protection and enforcement of rights, as compared with exclusively political. And from legislative supremacy, the new model takes the importance of the notion that there is no form of law set above and wholly immunized from legislative action.

Secondly, the new model can be said to create a distinct blending of legal and political constitutionalism across the board. Although the discourse of political versus legal constitutionalism tends to suggest that the choice is an either-or one, in reality most legal systems have elements of both even where one or the other is predominant^[22]. Thus, a paradigmatically legal constitutionalist regime such as the United States still has swathes of putatively constitutional law that are typically politically rather than judicially enforced, such as separation of powers between Congress and the President^[23]. Australia is perhaps the best example of a formally « mixed regime » at the national level, with a legal constitutionalist treatment of structural issues — federalism and separation of powers — and a mostly political constitutionalist treatment of rights^[24].

By contrast with such formally or informally mixed regimes that apply one or other model to different substantive areas, the new model blends political and legal constitutionalism across the board. It provides a sequenced role for both legal and political modes of accountability as its general mode of operation. As we have previewed above and shall see in more detail in chapter four, in its various instantiations the new model begins with political rights review at the legislative stage, whereby the government is required to consider whether proposed legislation is compatible with protected rights and make its conclusion known to parliament^[25]. The second stage involves judicial rights review, whereby in the context of a litigated case courts may exercise one or more of their enhanced powers to protect and enforce the rights. The third and final stage involves post-legislative political rights review, whereby the legislature may exercise its power of the final word and enforce any disagreement with the courts. Indeed, the new model not only combines legal and political modes of accountability, but also (1) legal and moral/political conceptions of rights and (2) judicial and legislative rights reasoning^[26], rather than a general systemic choice of one rather than the other.

Thirdly, and most formally, the new model offers a set of intermediate legal positions to the essential and conflicting postulates of constitutional and legislative supremacy. Despite interesting differences in the institutionalization of the first form of legal constitutionalism — or « big-C » constitutional law — since the end of World War II, most notably between centralized and decentralized judicial review, contemporary systems of constitutional supremacy around the world uniformly adhere to the basic principles first established by the United States in its legal revolution against Great Britain that closely followed the political one. These, of course, are that the written — or, rather codified — constitution, including its rights provisions, is (1) the supreme law of the land, (2) entrenched against ordinary majoritarian amendment or repeal and (3) enforced by the judicial power to invalidate or disapply conflicting statutes and other government actions, against whose decisions the legislature is powerless to act by ordinary majority vote. The contrary principles of traditional parliamentary sovereignty, which the U.S. Constitution was deliberately designed to reject, are that statutes are (1) the supreme law of the land, (2) not entrenched against ordinary majoritarian amendment or repeal and (3) not subject to a judicial power of review and invalidation on substantive grounds^[27].

The new model provides intermediate positions on each of these three basic issues. In a legally significant sense, the protected rights have some form of higher law status compared to ordinary statutes but not one that wholly immunizes them from legislative action. This may, for example, be conventional constitutional status but subject to a legislative override, as in Canada, or « constitutional statute » status as has been argued for under the HRA^[28] and occasionally applied in practice in New Zealand, whereby the earlier statutory right prevails over a conflicting later ordinary statute unless expressly amended or repealed^[29]. Such non-application of the normal doctrine of implied repeal also provides a mode of partial entrenchment that straddles the full and no entrenchment of the other two models^[30]. And, as discussed, the new model grants courts greater powers to protect rights than under traditional parliamentary sovereignty, powers that amount to forms of constitutional review, but not powers against which legislatures are wholly powerless to act by ordinary majority, as under constitutional supremacy. These include the power of Canadian courts to disapply conflicting statutes subject to the legislative power in section 33, the power of higher UK courts to issue declarations of incompatibility under section 4 of the HRA, and the power/duty of UK and New Zealand courts to interpret statutes consistently with rights provisions whenever possible^[31]. These new, « weak-form » powers occupy the space in between strong-form judicial review against which there is no legislative recourse by ordinary majority vote vis-à -vis the particular statute at issue and no constitutional review at all.

The Commonwealth model does not only, however, provide a new form of judicial review; it also provides a new justification of judicial review. For once shorn of judicial supremacy, the task of defending a judicial role in rights protection is a different — and easier — one. A model of constitutionalism that provides for judicial rights review of legislation but gives the legal power of the final word to ordinary majority vote in the legislature is normatively, and not only practically, different from one that does not. Indeed, even if it turns out (as certain critiques maintain) that there is little or no practical difference between the power of courts under certain instantiations of the new model and judicial supremacy, there is still a normative difference between them. Despite the current fairly strong political presumption against use of the legislative override in Canada, there is still a straightforward sense in which exercises of judicial review are more democratically legitimate than in the United States because of the existence of the override power

The fuller spectrum

From a systemic perspective, the new Commonwealth model suggests the novel possibility that the universe of constitutionalism, rather than a bifurcated one clustered around one or other of two mutually incompatible poles, is more of a continuum based on the scope and role of legal/judicial versus political/legislative decision-making in resolving rights issues and enforcing other limits on political power. The continuum stretches from pure political constitutionalism or strong legislative supremacy at one end to pure legal constitutionalism, or what has been termed « the total constitution »^[32], at the other. On this continuum, unlike on the bipolar model, many constitutionalist systems will occupy positions somewhere between the two ends.

For pure political constitutionalism, the answer to the general question of what type or number of rights-relevant issues and conflicts in a society should be resolved by judicially enforceable higher law is zero. All such issues/conflicts should be resolved politically, through ordinary, non-constitutional laws made and executed by political actors who remain fully accountable for them to the electorate. The judicial role is limited to fairly interpreting and applying this law. The opposite answer is given by pure legal constitutionalism. Its instrument is the « total constitution », a constitution that decides or strongly influences virtually all rights-relevant issues and conflicts in a society. It does this by broadly defining the rights it contains, imposing affirmative duties on government and/or by creating greater horizontal effect on private law and private individuals^[33]. In this way, the total constitution effectively constitutionalizes all law by requiring it to be not merely consistent with, but effectively superseded by, the comprehensive higher law of the constitution. Here there is relatively little room for discretionary, autonomous political decision-making or lawmaking as the total constitution provides mandatory answers to almost all issues, leaving ordinary law in effect as a form of administrative law. What defines this polar position, then, is the scope or reach of legal constitutionalism.

Moving along the continuum from total constitutionalism, we come to more standard or limited versions of legal constitutionalism, in which the written or unwritten higher law as construed and applied by the constitutional judiciary resolves some but not all of the rights-relevant issues and conflicts in a society. Again, as compared with the polar version, this will typically be because of its fewer and more narrowly defined rights, lesser reach into the private sphere and/or fewer affirmative duties on government. Here, legal constitutionalism (in either its « big-c » or common law form) still leaves significant space for discretionary and autonomous political decision-making in that it removes some but not all topics from the political sphere and, within those remaining, some but not all approaches to those topics. In other words, within conventional legal constitutionalism, higher law (as interpreted and applied by the courts) provides answers to certain issues and narrows the range of permissible political options on others, but its lesser scope compared to the pure or polar version maintains greater space for politically accountable decision-making. Just as important as its better-known function of taking some issues off the political agenda^[34] is that ordinary legal constitutionalism leaves others on it — and this has been central to its appeal in an era that has seen the rise of world constitutionalism alongside, and as part and parcel of, the rise of world democracy^[35].

The new model occupies that part of the continuum in between this more limited and common form of legal constitutionalism on the one side and pure political constitutionalism on the other. With its blending and sequencing of legal and political accountability and modes of reasoning, its form of judicially enforced higher law influences but does not automatically or necessarily resolve any rights-related issues, distinguishing it from the neighbouring positions on either side. Within the space occupied by the new model and on the basis of the introductory discussion of the range of different specific instantiations above, it might be suggested that Canada is slightly closer to the limited legal constitutionalism part of the continuum than the other new model jurisdictions, with the original version of the NZBORA slightly closer to the political constitutionalism pole than the HRA, ACTHRA and VCHRR.

To give a concrete example of how these various positions on the continuum affect how and by whom rights issues are decided, let us consider the case of abortion. On this issue at least, Germany approximates pure legal or total constitutionalism^[36]. As interpreted by the Federal Constitutional Court, the Basic Law largely determines how this most controversial issue is resolved, leaving relatively little space for discretionary political decision-making. As is well-known, because the foetus » right to life is protected by Article 2(2)^[37] and the state has a constitutional duty to protect this life even against its mother, the state must treat all abortions as unlawful with the exception of the few judicially defined « unexactable » situations, such as rape, incest or sever birth defects^[38]. Discretionary political decision-making is limited to the narrow window of selecting constitutionally permissible means, apart from the criminal law, for effectively fulfilling the state »s duty whilst still maintaining the required general unlawfulness of abortion. Even here, however the Federal Constitutional Court has prescribed much of the content of mandatory counselling as a permissible alternative^[39].

The United States exemplifies the second position on the continuum, the more conventional or limited version of legal constitutionalism, in its written or enacted form. Here, judicially enforced higher law determines what legislatures cannot do — namely, as currently interpreted by the Supreme Court, prohibit or place « undue burdens » on pre-viability abortions or post-viability ones necessary to protect the life or health of the mother — but leaves a greater amount of space for discretionary political decision-making within the parameters of the constitutionally permissible^[40]. Thus, the scope of legislative choice runs from no regulation of abortion at all to twenty-four hour waiting periods, prohibiting so-called partial birth abortions, and perhaps mandatory viewing of foetal ultrasounds^[41].

In the UK, the HRA as interpreted and applied by the judiciary may influence the abortion issue but does not definitively decide any aspect of it — either what legislatures must or cannot do. So, even if a higher court were to interpret Convention rights as bestowing a right to life on the foetus and declare the current UK abortion statute inconsistent with it — or, conversely, declare a future statute criminalizing abortion inconsistent with a woman’s right to privacy — Parliament would be free to exercise its power to disregard the declaration^[42]. Indeed, this first was the specific scenario cited by the Home Secretary during legislative debate on the HRA as the type of situation where Parliament might reject a declaration^[43]. Similarly, if a court were to interpret the current abortion statute narrowly to render it consistent with its finding of a right to life, Parliament would be free to amend the statute to make its intention and disagreement with the judicial decision clear.

At the federal level in Australia, one of the last surviving bastions of a fairly pure form of political constitutionalism in the rights context, the abortion issue is fully and exclusively decided by politically accountable lawmaking, with no substantive role for the judiciary — apart, of course, from interpreting it according to traditional principles of statutory interpretation and applying it in litigated cases.

To be sure, other factors than the four defining the new model and differentiating it from both conventional legal and pure political constitutionalism may also help to locate the relative position of any particular system on this continuum. These are factors that might be said to affect the depth or strength of legal/judicial decision-making, as distinct from its breadth or scope, such as the ease or difficulty of constitutional amendment^[44], the independence and tenure of the judiciary, and access to (individual standing) and systemic consequences of judicial review. Thus, on these issues, the U.S. system, with its very high bar for constitutional amendment, life tenure for federal judges without a mandatory retirement age, relatively easy access to judicial review due to individual standing and decentralization, and system-wide effects of judicial decisions is closer to the polar position than most other systems of conventional legal constitutionalism or constitutional supremacy. At the margin, this may even result in some blurring of the boundary between pure and ordinary legal constitutionalism, especially if or where a total constitution bestows lesser depth to legal/judicial decision-making through its position on these issues. Ultimately, however, depth issues of this sort are subordinated to the prime criterion of the scope of such decision-making within the political system.

II. The case for the new Commonwealth model

If the new Commonwealth model is a distinct institutional form of constitutionalism, then the next issue is how attractive or compelling an alternative is it. In what ways might it be thought preferable or advantageous?

The essential case for the new Commonwealth model is that it is to forms of constitutionalism what the mixed economy is to forms of economic organization: a distinct and appealing third way in between two purer but flawed extremes. Just as the mixed economy is a hybrid economic form combining the core benefits of capitalism and socialism whilst minimizing their well-known costs, so too the new model offers an alternative to the old choice of judicial supremacy or traditional parliamentary sovereignty by combining the strengths of each whilst avoiding their major weaknesses. Like the mixed economy’s countering of the lop-sided allocation of power under capitalism to markets and under socialism to planning, the new model counters legal and political constitutionalism’s lopsided allocations of power to courts and legislatures respectively. It recalibrates these two existing choices by effectively protecting rights through a reallocation^[45] of power between the judiciary and the political branches (adding to judicial power if starting from parliamentary sovereignty and reducing it if starting from judicial supremacy) that brings them into greater balance and denies too much power to either. As such, it is largely an argument about greater subtlety in constitutional engineering. The result is a more optimal institutional form of constitutionalism within a democratic polity than provided by either traditional model alone, one that provides a better working co-existence of democratic self-governance and the constraints of constitutionalism, the twin concepts underlying constitutional democracy.

After the latest round of the debate about judicial review conducted within the conventional bi-polar framework, it seems clearer than ever that there are powerful arguments both for and against legal constitutionalism and that no unanswerable, knock-down case — for one side or the other — that persuades all reasonable people is likely anytime soon. Although political constitutionalists have generally been more comfortable in critical mode, focusing rather more on presenting arguments against legal constitutionalism than on the positive case for their own position^[46], these are simply two sides of the same coin within a bi-polar debate so that which one to pick mostly reflects choice of rhetorical strategy. Indeed, one of the benefits of the new three-way debate ushered in by the new model is that it becomes necessary to specify what position is being argued for and not only against, as there is no single, dichotomous default option but rather two separate alternatives. So, for example, a successful critique of judicial supremacy is no longer sufficient to justify traditional parliamentary sovereignty, for whilst the new model is also opposed to judicial supremacy, it supports a judicial role in rights issues. The net effect is that this high quality bi-polar debate has helpfully isolated the two key issues as (1) which model better protects rights and (2) whether judicial review is politically legitimate within a democracy^[47], and also provided an enhanced assessment of the strengths and weaknesses of both traditional models with respect to them.

This enhanced assessment is particularly helpful because in order to explain how the new model combines the core strengths of both traditional ones whilst avoiding their major weaknesses, it is of course first necessary to specify what these are. As an institutional form of constitutionalism in a democratic political system, political constitutionalism (or legislative supremacy) has two major strengths or benefits. Firstly, on the issue of legitimacy, by institutionalizing limits on governmental power as political in nature and enforcing them through the twin mechanisms of electoral accountability and structural checks and balances — such as parliamentary oversight of the executive — political constitutionalism coheres easily and unproblematically with democracy as the basic principle for the organization of the governmental power that it limits. Whether these limits that protect individual rights and liberties remain exclusively in the political sphere as moral or political rights, or are given legal effect as common law or statutory rights, they are ultimately within the scope of the democratic principles of equal participation and electorally-accountable decision-making as determined or changeable by ordinary legislative act. Secondly, on the issue of outcomes, given the nature of many, if not most, rights issues that arise in contemporary mature democracies — including the existence of reasonable disagreement about how they should be resolved — legislative reasoning about rights may often be superior to legal/judicial reasoning. As powerfully argued by Adam Tomkins and Jeremy Waldron, high quality rights reasoning often calls for direct focus on the moral and policy issues involved free of the legalistic and distorting concerns with text, precedent, fact-particularity and the legitimacy of the enterprise that constrain judicial reasoning about rights^[48]. Moreover, electorally-accountable representatives are able to bring a greater diversity of views and perspectives to bear on rights deliberations compared to the numerically smaller, cloistered and elite world of the higher judiciary.

At the same time, proponents of judicial review have identified two major weaknesses of political constitutionalism on the key issues. The first is the risk of either understating or under-enforcing constitutionalism’s limits on governmental power, especially individual rights, as the result of various « pathologies » or « blind spots » to which electorally-accountable legislatures (and executives) may be prone. These include sensitivity to the rights and rights claims of various electoral minorities — whether criminal defendants, asylum seekers, or minority racial, ethnic or religious groups — given the exigencies and logic of re-election, legislative inertia deriving from tradition or the blocking power of parties or interest groups, and government hyperbole or ideology^[49]. Under-enforcement of rights may also result from the circumstance that however high the quality of legislative rights reasoning, it inevitably competes in this forum with other deliberative and decisional frameworks. Undoubtedly, these standard, well-known concerns were primarily responsible for the massive switch away from political constitutionalism towards judicial supremacy around the world during the post-war « rights revolution », as the resources of representative democracy alone were perceived to provide insufficient protection.

Secondly, just as political constitutionalists have attempted to turn the tables on the conventional argument that rights are better protected with judicial review in the way we have just seen, legal constitutionalists have tried to do the same with the standard argument that judicial review is democratically illegitimate. Thus, Richard Fallon has argued that important though democratic legitimacy undoubtedly is, it is not the exclusive source or type of legitimacy in constitutional democracies and that the substantive justice of a society also contributes to is its overall political legitimacy. Accordingly, to the extent that political constitutionalism may undermine substantive justice by under-enforcing rights for the above-stated reasons, it also detracts from the overall political legitimacy of a democratic regime^[50]. More generally, Mattias Kumm has argued that in addition to electorally-accountable decision-making, a second precondition for the legitimacy of law in constitutional democracies is the requirement of substantively reasonable public justification for all governmental acts, including legislation, burdening individuals’ rights. As part of our commitment to constitutionalism, legislation unsupported by a reasonable public justification for the burdens it imposes on individuals is illegitimate regardless of majority support. Political constitutionalism, however, provides no adequate forum for critically scrutinizing the justification for a piece of legislation to determine if it meets the minimum standard of plausibility in terms of public reasons. Given the various potential pathologies noted above, legislative deliberation and political accountability are insufficient to ensure that burdened individuals are provided with the reasonable justification to which they are entitled, as evidenced by many decisions of domestic and international constitutional courts^[51].

If these are the most important strengths and weaknesses of political constitutionalism that emerge from the recent academic debate, what are the equivalents for legal constitutionalism? One of its strengths is fostering public recognition and consciousness of rights. A reasonably comprehensive statement of rights and liberties, as found in the typical constitutional bill of rights, renders rights less scattered and more visible or transparent, more part of general public consciousness than either an « unwritten » set of moral and political rights or a regime of residual common law liberties supplemented by certain specific statutory rights.

A second strength of legal constitutionalism — in either its « big-C » or common law variations — is that it may help to protect against the above-mentioned tendency towards the under-enforcement of rights resulting from the potential pathologies and blind spots affecting politically accountable legislatures and executives. Where they are politically independent in the sense of not needing to seek re-election or renewal in office after initial appointment, judges exercising the power of judicial review are in a better institutional position to counter or resist such electorally-induced risk of under-enforcement^[52]. This is not so much an argument about expertise as about incentives and institutional structure. Courts also decide cases upon concrete facts, some of which may have been unforeseen by legislators^[53], and indeed bring a more context specific or « applied » dimension to rights deliberation that complements the necessarily greater generality of that undertaken by legislatures.

Thirdly, in the positive version of the argument noted above, legal constitutionalists have made the case that judicial review is essential to the overall legitimacy of a constitutional democracy. Thus, Richard Fallon argues that to the extent judicial review promotes substantive justice by helping to protect against under-enforcement of rights, it might « actually enhance the overall political legitimacy of an otherwise reasonably democratic constitutional regime »^[54]. In this sense, judicial review may result in a trade-off among different sources of legitimacy but not between rights protection and overall political legitimacy. Mattias Kumm has argued that judicial review provides the forum, required for the legitimacy of legislation, in which individual rights claimants can put the government to its burden of providing a reasonable public justification for its acts. As he puts it:

Human and constitutional rights adjudication, as it has developed in much of Europe, … is a form of legally institutionalized Socratic contestation. When individuals bring claims grounded in human or constitutional rights, they enlist courts to critically engage public authorities in order to assess whether their acts and the burdens they impose on the rights-claimants are susceptible to plausible justification…. Legally institutionalized Socratic contestation is desirable, both because it tends to improve outcomes and because it expresses a central liberal commitment about the conditions that must be met, in order for law to be legitimate^[55].

Thus, for example, judicial review aims to ensure that an individual burdened by a statutory ban on gays in the military is able to put the government to the task of providing a reasonable public justification for the enacted law, one not relying on prejudice, tradition, disproportionate means, etc., failing which it is illegitimate^[56].

And what are the weaknesses or costs of legal constitutionalism as an institutional form in a democracy? Starting with the issue of rights protection, one is that just as there may be under-enforcement of rights due to electorally-induced or other legislative pathologies, there may also be under-enforcement resulting from certain judicial pathologies^[57]. These include (1) the risk of rights-relevant timidity that comes with responsibility for the final decision and its real world consequences; (2) concerns about lack of policy expertise or legitimacy in the context of assessing justifications for limiting rights — the universal second stage of modern rights analysis; (3) the artificially and legalistically constrained nature of judicial reasoning about rights; and (4) the relative lack of diversity of perspectives among the elite members of the higher judiciary. Now, it might be thought that, even if it exists, the risk of judicial under-enforcement of rights is not much of a concern because it is premised on, simply mirrors, a prior under-enforcement by the legislature. Where it occurs, it is true that the countering force of judicial review does not take place, but we are no worse off in terms of rights-enforcement than before the judicial decision.

This response strikes me as at least partially misguided for two reasons. First, assuming a court has under-enforced the right, it is not true that we are no worse off. The judicial decision formally legitimates the statute and the legislative under-enforcement in a way that would not be the case without; there would simply be a controversial statute on the books which many people reasonably believe violates rights and should be repealed. Moreover, there is now a judicial precedent in place, which may affect the political and/or legal treatment of other or future statutes. It is for these reasons that Justice Jackson famously chided the U.S. Supreme Court for taking the case of Korematsu v. United States^[58]. It is one thing for the elective branches to under-enforce rights during a perceived national emergency; it is another for the highest court to give its seal of legitimacy to that under-enforcement. Secondly, the response assumes that the existence of judicial review has no effect on the rights deliberations otherwise undertaken by the legislature itself in the course of enacting the statute, that judicial review provides an additional and supplementary layer of rights scrutiny — a safety net — over and above the legislative one. There are plausible reasons to believe, however, that judicial review within a legal constitutionalist framework results in the processes of political rights review being reduced or even bypassed altogether in favour of relying on the courts, which after all have the final word^[59]. Why spend precious time on matters you do not decide? That is, judicial and political review may well be more substitutes for each other than supplements within legal constitutionalism, so that before opting for the latter one would need to be persuaded that on balance the rights under-enforcement stemming from judicial pathologies is likely to be less than from legislative ones.

A second weakness of legal constitutionalism is that may also lead to the overstatement or over-enforcement of constitutionalist limits on governmental power. There is a term for this weakness and it is « Lochner »^[60]. So even if, very generally speaking, potential under-enforcement of rights is worse than potential over-enforcement,^[61] over-enforcement of the Lochner variety is far from harmless error. That is, where courts use their supreme interpretative power to read into a constitutional text certain controversial rights that are the subject of reasonable disagreement, they may be artificially limiting the scope of governmental power in the service of substantive injustice. This type of over-enforcement undermines the overall political legitimacy of an otherwise democratic constitutional regime.

A third weakness of legal constitutionalism is the general weakness and relative ineffectiveness of relying on ex post regulatory mechanisms to the exclusion of ex ante ones^[62]. If constitutionalism imposes limits on governmental power, some of which take the form of individual rights, then relying primarily or exclusively on courts to enforce them will often be tantamount to closing the barn door after the horse has bolted. Some laws that raise serious rights issues may never be challenged in court, others may be challenged but under-enforced, and in most cases laws will not be challenged until at least some of the damage they are judicially assessed to impose has already been caused. Abstract judicial review acknowledges, and is designed to deal with, this problem but several systems do not permit this type of review and those that do usually limit standing to elected representatives of a certain number or office, whose political interest in challenging a law may or may not coincide with those likely to be adversely affected by it.^[63]

Fourthly, there is a strong tendency within legal constitutionalism for courts to become the primary expositors of rights in society and yet there are serious weaknesses in judicial modes of rights deliberation from the perspective of this important function. Judicial review may be conceptualized and defended (in common law jurisdictions at least) as incidental to the ordinary judicial function of deciding a case,^[64] but deciding a specific case is far from all that a highest court typically does when exercising this power in the context of a controversial rights issue. Rather, depending on the scope of its judgment, it resolves not only the case but the rights issue raised in it as far as lower courts in future cases are concerned, and, depending on its accepted or perceived interpretive supremacy within the entire political system, its resolution becomes the authoritative one for all purposes. In this way, the highest court tends to speak for, and in the name of, society as a whole. Here again, the « limitations inherent within judicial forms of decision-making »^[65] discussed by Tomkins and Waldron come to the fore, as does the concern with over-legalization or judicialization of principled public discourse generally, whereby the legal component or conception of rights is over-emphasized at the expense of the moral and political^[66].

These first four weaknesses mostly address the issue of whether or not rights are better protected with judicial review. Last, but by no means least, is the familiar and standard concern with legal constitutionalism from the perspective of legitimacy in a democratically-organized polity, the concern that Fallon and Kumm have attempted to outflank. As this concern is so familiar, I shall be brief. It may perhaps be expressed or captured this way: in the name of attempting to ensure against under-protection of rights, legal constitutionalism gives to an electorally-unaccountable committee of experts unreviewable power to decide many of the most important and weighty normative issues that virtually all contemporary democratic political systems face, even though it turns out that these issues are not ones for which the committee’s expertise is especially or uniquely relevant.

The easy, conventional and mostly rhetorical response to this concern is premised on a legal fiction; namely, that a supermajority of citizens has self-consciously, deliberately and clearly pre-committed to a set of higher law solutions to rights issues, and the function of the courts is simply to apply these — in essentially the same way as any other type of law^[67]. The legal reality is that many of the most important rights issues as and where they present themselves are inevitably the subject of reasonable disagreement among and between judges, legislators and citizens — as routinely evidenced by closely divided courts, legislatures and referenda on some of the most controversial and difficult topics. Such disagreement — about which rights exist, their meaning, scope and application, as well as permissible limits on them — persists whether or not rights and rights claims are left in the realm of moral and political discourse only, are deemed part of the common law or have been incorporated into the particular textual formulas of a statutory or constitutional bill of rights. As Jeremy Waldron puts it, « the Bill of Rights does not settle the disagreements that exist in the society about individual and minority rights. It bears on them but it does not settle them. At most, the abstract terms of the Bill of Rights are popularly selected sites for disputes about these issues »^[68].

In this context, the case for some of the most fundamental, important and divisive moral and political issues confronting a self-governing society of equal citizens being subject to the rule that the decision of a judicial majority is final and effectively unreviewable, on the legal fiction that they are wholly questions of law akin to the interpretation of a statute or a contract, appears weak — if not duplicitous. So too on the frequently proffered alternative basis that they concern matters of principle (as distinct from policy) best left to, and answered by, courts alone^[69]. Even were the distinction between principle and policy to be successfully explained and justified, if « constitutional democracy » is taken to require excluding the participation and reasonable judgments of equal citizens and their electorally-accountable legislative representatives on all rights-relevant issues of principle in favour of the reasonable judgments of judicial majorities, then the qualifying adjective has largely swallowed what it qualifies.

The new model as a normatively appealing third way

The persistence of these weaknesses with both traditional models alongside each of their strengths is a major problem because of the structure of the choice between them. In the either-or universe of the bipolar model, we are stuck with one or the other in a « winner-take-all » institutional system that requires the weaknesses of the chosen model to be endured alongside its strengths, whilst the complementary merits of the other model are lost entirely. It is legal constitutionalism versus political constitutionalism, judicial supremacy or no judicial review at all. But this « warts-and-all » structure of institutional design choice is unnecessarily crude and disproportionate with respect to the normative costs and benefits of the two models. By contrast, a major advantage of the new model as an intermediate hybrid is that it makes possible a form of « proportional representation » among the strengths of both legal and political constitutionalism, whilst also severing or minimizing the major weaknesses of each.

The core of the case for the new model is the argument for both weaker-form judicial review and weaker-form legislative supremacy versus either strong-form judicial review or strong-form legislative supremacy. The central problem with strong-form judicial review is not that rights-based judicial review has no value or cannot be justified at all, but that it is too strong. In the familiar language of proportionality, it is not the least restrictive way of achieving this value with respect to others that are also central and essential within a constitutional democracy. Moreover, as already previewed in the previous section, there are good reasons for believing that at least part of this value — protecting against under-enforcement of rights — may not be optimally or best promoted by strong judicial review, even if it were the case that on balance it affords better protection than political constitutionalism.

Similarly, the central problem with traditional strong-form legislative supremacy is also that it is unnecessarily strong. Just as judicial supremacy effectively gives exclusive voice to the highest court, traditional strong-form legislative supremacy needlessly creates a monopoly for elected representatives in terms of whose voice counts or has legal authority on rights issues. If the core concept of parliamentary sovereignty is perfectly consistent with the existence of moral, political and procedural constraints on legislative decision-making, as Jeffery Goldsworthy reminds us^[70], the new model adds two concrete and specific types of such constraint: the procedural requirement of pre-enactment rights review and the very visible political constraint of a formal, but not necessarily legally final, judicial opinion on rights issues raised by enacted laws. By challenging the legislature’s institutional monopoly of authoritative voice on rights issues, this second constraint in particular can be said to weaken legislative supremacy compared to the traditional version that remains part and parcel of political constitutionalism.

I have claimed that the general case for the new model, like the arguments for the mixed economy, it that it is combines the strengths of the two purer but flawed extremes whilst avoiding their weaknesses. It is now time to make good on this claim by explaining how this is achieved. As we have seen, to the extent that proponents of legal and political constitutionalism have engaged each others’ arguments, it has mostly been in a debate about judicial review in which the common ground is that the two main issues are whether there is reason to suppose that rights are better protected with or without judicial review and whether judicial review is democratically legitimate. Although at times, political constitutionalists almost seem to rue the focus on rights — which they acknowledge has been the trigger for the growth of legal constitutionalism^[71] — as misplaced, it is too late in the rights revolution (at least in the context of mature liberal democracies) to cede this territory to the opposition.

How exactly does the new model accommodate and combine the strengths of both polar positions whilst severing their weaknesses as inessential and dispensable? And what is the argument that the resulting intermediate position better protects rights whilst also maintaining political legitimacy in a democracy? First, on the issue of rights protection, the case for the new model accepts almost everything that critics of legal constitutionalism say as to why legislative reasoning about the sorts of rights issues confronting all modern societies is or may be better/more appropriate than judicial reasoning, with its inherently artificial and constrained nature and relative inability to focus directly on the moral issues involved. This acceptance is institutionalized in pre- and post-enactment political rights review. At the same time, it also accepts and accommodates the legal constitutionalist argument that judicial review may sometimes help to reduce the risk of certain types of under-enforcement of rights, hence the role of courts in between the two stages of political review. Given what has just been said, this is obviously not because courts are better or more expert than legislatures at rights deliberation but because each institution comes to the task from a different perspective, has different strengths and weaknesses that may usefully be brought to bear on rights issues to help improve outcomes and protect against under-enforcement. Again, the relative strengths of legislatures are those expressed by Tomkins and Waldron, as well as the greater diversity of views mentioned above. The relative weaknesses of legislatures are the potential rights-relevant pathologies to which they may be subject. The relative advantage of courts here is independence from these potential electorally-induced pathologies and the dimension of fact-specific, applied rights deliberation versus the more general and abstract approach of legislatures, but the weaknesses are the parallel tendencies towards pathologies of their own and the general problem of relying exclusively on ex post regulation discussed above.

What the argument for the new model rejects as uncompelling, disproportionate and dispensable in the two polar models on this issue is the following. First, in the case for political constitutionalism, it does not accept the consequence of concluding that, on balance, legislative reasoning about rights is superior to (or no worse than) judicial; namely, that rights issues should be left exclusively to the former. This consequence is a function of the either-or universe of the bi-polar framework, in which it is necessary to choose between legislative and judicial modes of reasoning about rights. The appeal of the new model here is that it revises the standard implication of this argument by recognizing the respective strengths and weakness of courts and legislatures and providing a significant and appropriate role for both. Accepting the net superiority of legislative over judicial reasoning about rights may determine which has the formal power of the final word but it does not entail that no role is served by, or afforded to, the latter.

Secondly, with respect to the legal constitutionalist case for judicial review, the argument for the new model rejects the implication that under-enforcement concerns justify not only a judicial role in the protection of rights but also a judicial veto over legislation — what Fallon refers to as one of the « multiple veto points » in the system^[72] — or at least one that is not defeasible by ordinary majority vote of the legislature. Rather, for the new model, under-enforcement concerns mean that courts should be a « checking point » in the system, having an interpretive, alerting and informing function with respect to rights issues, somewhat akin to the delaying power of the House of Lords as the second legislative chamber versus the veto power of the U.S. Senate^[73]. This revision, of course, reflects and expresses the difference between weak-form and strong-form judicial review. To the significant extent that the case for legal constitutionalism turns on the incentives and potential rights-relevant pathologies of elected officials, the case for the new model here is that the combined impact of mandatory political rights review and non-final judicial review will sufficiently alter those incentives and counter the pathologies to render the solution of judicial finality unnecessary and disproportionate. This distinct mode of judicial input into rights discourse can be helpful as the legally penultimate word in both informing/spurring rights review by the political branches and raising the costs of legislative disagreement through an alerted citizenry. As with the criminal jury trial to which Fallon analogizes the argument for judicial review as protection against under-enforcement of rights, we may give citizen-members of the jury a veto power in order to minimize erroneous conviction of the innocent, but (and this is the limit of the analogy) we do not give such a power to second-guess their decisions to judges. Accordingly, unlike the two traditional models, the new model recognizes and reaps the respective benefits of both legislative and judicial reasoning in terms of their contributions to rights deliberation and protection against under-enforcement, but within an institutional structure that affords the power of the final word to the former.

Let us now turn to the issue of legitimacy. Once again, the case for the new model is that it is able to combine and accommodate the core insights of both opponents and proponents of judicial review into a package that is more compelling and proportionate than either alone. The democratic legitimacy of collective decision-making procedures (and especially higher lawmaking procedures) is obviously a centrally important value within constitutional democracies. By granting the legal power of the final word to the legislature, the new model preserves and promotes this value. At the same time, the new model acknowledges and accommodates the broader legitimacy concerns raised by Fallon and Kumm in their defences of judicial review. To the extent that weak-form judicial review helps to protect against under-enforcement of rights by giving courts checking, alerting, informing and decision-making functions that supplement legislative rights deliberations and counter characteristic potential pathologies, it promotes justice and so enhances overall political legitimacy. But it does so, too, when also countering judicial under- and over-enforcement of rights, against which legal constitutionalism is generally powerless.

With respect to Kumm’s argument, it is first necessary to distinguish reasonable public justification for general legislative acts that burden individuals from administrative and judicial decisions, which are typically subject to forms of judicial review for reasonableness even in systems that do not provide for constitutional review of legislation^[74]. These are not at issue and clearly perform the legitimating, rule of law function that Kumm prescribes. As for legislative acts, the new model obviously provides the judicial forum for the required critical assessment of reasons. The question, therefore, is whether strong-form judicial review rather than weak is necessary or essential to fulfil this condition of legitimacy and so is justified as a proportionate departure from the norm of democratically-accountable decision-making^[75]. I believe the answer is no. To explain why, let me begin by making explicit what has been left implicit in the argument so far and will be discussed at greater length in the next chapter: the case for the new model’s override power is premised on reasonable disagreement with the courts on a rights issue. The basic principle at work here is that democracy requires a reasonable legislative judgment to trump a reasonable judicial one^[76]. In one sense, therefore, if courts and legislatures both adhere to their normatively assigned roles and (as in Kumm’s theory) courts only invalidate legislation for which there is no reasonable public justification, then legislatures would never exercise their override power — which perhaps becomes redundant. But by the same token, under this scenario it cannot be said that strong-form judicial review is necessary as weak-form review would achieve exactly the same result.

More realistically, however, the risk that both will depart from their normatively circumscribed powers must be taken into account: that courts will invalidate reasonable legislative decisions in favour of the court’s view of the correct one and legislatures will exercise their override power in support of unreasonable legislative decisions. In these circumstances, is strong-form judicial review rather than weak justified? In current practice, Kumm’s normative standard is not in fact the one that is generally understood to govern judicial review and courts regularly overturn legislative decisions which cannot be said to be unreasonable^[77]. But what if it were? Under strong-form review, there is little to counter the risk of judicial overreaching on this issue — as by reason of their very independence, courts face no direct political constraint — and the legislative override power would be a useful institutional check in the absence of others as a form of separation of powers. Moreover, we are by hypothesis here — a court has invalidated a reasonable legislative act — in the situation where the principle of a reasonable legislative judgment trumping a reasonable judicial one applies, so that use of the override would be justified. By contrast, unlike the strong-form judicial power, this legislative power would be subject to a significant institutional or political constraint against the risk of misuse; namely, the fact that a court has issued a formal judgment finding there to be no reasonable public justification for the legislation violating individual rights. Finally, so far we have been discussing the situation in which there have been clear departures from the standard of reasonableness, but as Kumm notes, the limits of reasonable disagreement may also be subject to reasonable disagreement^[78].That is, courts and legislatures may reasonably disagree about whether a legislative act is within the bounds of the reasonable. For the same two reasons just noted — the checking function of the override and the default or tie-breaking nature of legislative power that democracy requires — weak-form review also seems the more justified solution here.

In sum, the conventional democratic legitimacy concerns with judicial review are genuine and powerful in the context of pervasive rights indeterminacy. Again, given this context, the argument that democratic legitimacy requires the reasonable view of a legislative majority to trump the reasonable view of a judicial majority seems compelling. Fallon and Kumm are correct that democratic legitimacy is not the only source or type of political legitimacy in constitutional democracies, but it is a critically important and presumptive one. Departures from it carry a strong burden of justification. If protecting against under-enforcement of rights and/or the requirement of reasonable public justification for legislative burdens on individuals are the potential bases for such a justified departure, the means of furthering these components of political legitimacy must be proportionate; in particular, they must promote their objectives in ways that least restrictively depart from the democratic legitimacy of electorally-accountable decision-making. Weak-form judicial review is that least restrictive means; strong-form judicial review is not.

Institutionally, then, the strengths of legal and political constitutionalism that the new model combines in its hybrid status are as follows. From the latter, it employs the benefits of the more unconstrained and all-things-considered legislative style of moral reasoning about rights both before and after the exercise of weak-form judicial review. As part of the « after », of course, the new model also retains the possibility of ultimate reliance on the principles of electorally-accountable decision-making and political equality. From legal constitutionalism, the new model first takes the enhancement of general rights-consciousness that generally comes with a specific and fairly comprehensive statement of legal rights. It then attempts to counter potential legislative under-enforcement of rights in part by empowering politically independent and unaccountable judges to give their considered opinions on the merits of rights claims filed by individuals, thereby providing a forum to critically assess the public justification of laws and bolstering the broader legitimacy of the political system.

At the same time, the new model also avoids or seeks to minimize the major weaknesses of both traditional models. From political constitutionalism, it counters the rights-relevant pathologies or blind spots to which electorally-accountable institutions may be prone by, first, mandating rights consciousness and review in the legislative process itself and, secondly, establishing judicial review. Of the weaknesses of legal constitutionalism, the new model counters certain judicial pathologies that may result in both the under- and over-enforcement of rights by not relying solely on courts for protection of rights but also on rights review and deliberation by the political institutions. This enables the benefits of legislative reasoning about rights to supplement the limitations of judicial rights reasoning. At the pre-enactment stage, this political rights review also introduces the advantages of ex ante regulation in addition to the ex post regulation of judicial review, which may help to prevent rights violations from occurring in the first place. And at the post-enactment stage, it permits the new model to neutralize legal constitutionalism’s democratic legitimacy problem.

As part of its hybrid nature, and like the analogous mixed economy, the new model not only selectively incorporates and combines certain existing features (i. e., the strengths) from each of the two polar ones whilst discarding others (the weaknesses), but also revises them and in the process creates at least two wholly novel features that are not part of either traditional model. The normative appeal of these two exclusive features contributes substantially to the overall case for the new model. The first of these is the checking and alerting rights-protective roles of the courts compared to the full veto power of judicial supremacy just discussed in the context of Richard Fallon’s arguments. More akin to the delaying power of the UK’s second legislative chamber, the House of Lords, than the outright veto of the U.S. Senate — and for similar reasons of democratic legitimacy — one version of these more limited powers is institutionalized in the judicial declaration of incompatibility, a novel judicial power when enacted as part of the HRA. The second exclusive feature is the new model’s dispersal of responsibility for rights among all three branches of government rather than its centralization in either the courts (judicial supremacy) or the legislature (legislative supremacy). It is achieved in the three sequenced stages of mandatory pre-enactment political rights review by the executive and legislature, post-enactment judicial rights review, and post-litigation political rights review by the legislature. In this way, the new model not only produces a better, more proportionate general balance of power between courts and legislatures than the two more lopsided models of legislative and judicial supremacy, but also specifically with respect to the recognition and protection of rights.

This dispersal of rights responsibilities has the goal of fostering a stronger and deeper rights consciousness in all institutions exercising public power and is an essential part of the aggregate rights protective features of the new model. Overall, in the three following ways, it creates a different, and arguably more attractive, rights culture than the one produced under judicial supremacy. First, in the context of reasonable disagreement about rights, the dispersal rather than the concentration of responsibility is likely to affect the content of the recognized rights. This is due to both types of « judicial pathologies » about rights discussed above: (1) the artificially and legalistically constrained nature of judicial reasoning about rights that largely excludes direct engagement with the moral issues involved; and (2) the greater diversity of views and perspectives that electorally-accountable representatives can openly bring to rights deliberations compared to the numerically smaller, cloistered and elite world of the higher judiciary. Secondly, in terms of procedure, rights discussions will be far more inclusive and participatory leading to greater rights consciousness among both elected representatives and electorate. In affirming rather than denying Waldron’s « right of rights »^[79], the new model here institutionalizes a democratically legitimate rights regime. Thirdly, for standard checks and balances reasons the dispersal rather than the concentration of rights responsibilities reduces the risk of under-enforcement that comes with relying exclusively on any one institution — whether courts or legislatures. As noted above, although better known, under-enforcement concerns are hardly limited to the legislature. The key innovation here is the distinctive new model feature of supplementing ex post judicial rights review with ex ante political rights review by the executive and legislature. For its goal is to internalize rights consciousness within the processes of policy-making and thereby reduce or minimize rights violations in legislative outputs at the outset.

III. An Internal Theory of the New Model

The new model is defined in large part by its novel allocation of powers (and duties) among legislatures, executives and courts resulting in a blending and sequencing of political and legal deliberations about rights. It is this overall package of powers, and not any one viewed in isolation, that renders the model distinct. The previous chapter presented the « external » normative case for the new model as against the other two standard forms of constitutionalism. By contrast, this chapter will develop the « internal » normative case for it, in the sense of articulating a theory of how — in light of its distinctive institutional features, objectives and comparative justification — the new model ought to operate. In other words, I shall be developing a sort of ideal-type, a general normative account of how a well-functioning version of the new model operates to ensure that its distinctness is maintained and its objectives and benefits realized. This, I believe, is currently the most pressing and least developed theoretical issue surrounding the new model. In developing a general account, or ideal-type, this chapter helps to establish the basic criteria and lays the groundwork for the assessments of specific instantiations of the new model in the second part of this book. It also provides a platform from which to consider reforms of, or alternatives to, these versions.

As an intermediate institutional form of constitutionalism occupying what was previously understood to be inherently bipolar conceptual space, it is unsurprising that there should be concerns about the new model’s stability and distinctness. Some of these concerns are more formal or conceptual; others are more practically orientated, focusing on how the new model is likely to, does or must actually work. For current purposes, however, it is timely and relevant to state my view that the new model does not in practice collapse into traditional parliamentary sovereignty in all but name simply because (or where) final authority with respect to legislation is given to the legislature across the board. This, after all, is one of the general and defining features of the model. Rather, it collapses in practice either where courts do not use their (also model-defining) powers of constitutional review but routinely defer to the political branches, or where legislatures too routinely use theirs. Similarly, the new model does not in practice collapse into judicial supremacy simply because (or where) courts have new powers of constitutional review that are not so different in effect from the power to invalidate legislation^[80]. For one thing, as we shall see, how these powers operate varies considerably from jurisdiction to jurisdiction; they have no fixed or inherent strength. For another, Canada gives its courts a formal invalidation power but is still a new model jurisdiction. Rather, it collapses in practice either where legislatures do not use their power of the final word to the extent that it becomes irrelevant and so act like legislatures under judicial supremacy, or where courts misuse theirs — similarly acting as if operating in a system of judicial supremacy^[81]. In short, it is non-use or misuse of powers that threatens the practical stability and/or distinctness of the new model and not where these powers are placed at the outset.

Accordingly, a theory of when and how respective legislative and judicial powers should be used under the new model is essential but is mostly lacking. Not only would this help to clarify and maintain the distinctness of the new model but might also bolster legislative use as there is currently no general normative account of the proper, legitimate exercise of the power of the final word^[82]. In this context, arguably legislatures are likely to be risk averse in political cultures long governed more generally by the rule of law and its constitutive norm that governments obey their own courts. It is this theory that I hope to develop in this chapter.

One final preliminary word concerning the general nature of the internal normative account is in order. In practice, even (de jure) judicial supremacy can effectively collapse into legislative supremacy, where there is a strong norm that the formal power of judicial invalidation is rarely, if ever, used or triggered by only the clearest, most blatant violation of a higher law provision, as in several Scandinavian countries and Japan^[83]. No one believes, however, that such examples undermine the general stability or distinctness of the model of judicial supremacy. So, too, with the new model qua model. Although it is obviously true that the number of sample jurisdictions is far smaller — more akin to that for traditional legislative supremacy — the general critique of the new model cannot be that in practice a particular instance of it may collapse, or has collapsed, into one or other of the two traditional forms of constitutionalism, but rather that this is a general, and perhaps even inherent and unavoidable, practical consequence of its combination of defining features. Accordingly, part of the point of developing an internal normative theory is to help build a bulwark against such collapse.

Pre-enactment political rights review

The new model contains a general blending and sequencing of political and legal deliberation about rights that constitutes both its distinct character and part of the normative case for it as compared with the other two traditional models of constitutionalism. Consequently, in order to develop an account of the proper employment of respective powers, it is important that the role and contribution of each sequenced stage be understood.

As discussed in chapter two, the deliberate institutional engineering of the new model combines mandatory political rights review and weak-form judicial review. This combination of novel mechanisms breaks down into three distinct stages at which the protected rights may come into play, three stages of rights review rather than the single one associated with judicial supremacy. The first stage is pre-enactment political rights review; the second is judicial rights review; and the third and final is post-enactment, post-litigation political rights review.

As its name suggests, pre-enactment political rights review is defined by three features which it may be helpful to unpack. Firstly, by who engages in it; namely, political actors including executive officials, legislators and administrators, subject to the normal oversight of the media and citizenry. Secondly, by when it is undertaken; namely, at the outset of and during the legislative process and, in particular, prior to any form of judicial review. Thirdly, by what it consists in; namely, political and moral deliberation and not only or primarily legal reasoning. That is, political rights review refers to the content of the review as well as to the identity of the reviewers. Accordingly, it should not be excessively legal in nature; that is, focused on reasonable interpretive pluralism within the law or (worse still) predicting what the courts will ultimately do. Rather it should bring a broader, freer perspective of principle to the issue than is typical of judicial reasoning. This is because part of the point of political rights review is to distinguish rights deliberation under the new model from that under judicial supremacy, to provide space for legislative-style rights reasoning — for both the outcome-related and democratic legitimacy reasons discussed in the previous chapter.

Having explored the empirical dimensions of pre-enactment political rights review, Janet Hiebert concluded (in 2004) that « for now, the political interpretation of rights before rather than after judicial review is what most differentiates parliamentary institutional practice with respect to rights [under the new model] from that in the United States »^[84]. Whether or not one agrees with this specific judgment (« most differentiates »), there is no doubt that this first stage is an important and distinctive feature of the new model. It is distinctive both normatively, as part of the bottom-up or dispersal of responsibility for rights approach, and practically in that a form of it is legally mandated in each of the new model jurisdictions but not generally elsewhere.

In Canada, the federal Minister of Justice is required by statute to certify that bills have been assessed in light of the Charter and, when inconsistent with its provisions, to report such inconsistencies to Parliament. Section 7 of the NZBORA requires the Attorney-General to « bring to the attention of the House of Representatives any provision in the Bill that appears to be inconsistent with any of the [protected] rights ». Section 19 of the UK’s HRA requires the minister in charge of a bill in either House of Parliament « to make a statement in writing to the effect [either] that in his view the provisions of the Bill are compatible with the Convention rights…or…if unable to make a statement of compatibility the government nevertheless wishes the House to proceed with the Bill ». In Australia, the ACTHRA and the VCHRR similarly require the Attorney-General and a Member of Parliament introducing a bill respectively to make a « compatibility statement » as to whether or not the bill is consistent with human rights^[85]. By contrast, in both judicial and legislative supremacy jurisdictions, any such pre-enactment review that occurs tends to be a voluntary and ad hoc, rather than a mandatory and formal, part of the legislative process.

Pre-enactment political rights review performs at least six functions within the overall matrix of the new model. First, it aims to ensure that the executive thinks and acts in a rights-conscious way when considering and preparing legislative proposals (and so perhaps also more generally), thereby helping to disperse responsibilities for rights review among all three branches of government. It is not something that should just be left to the courts. Mandating formal ministerial consideration of the rights implications of bills is likely to have ripple-effects at all departmental levels and also, of course, implications for the legislative text. Secondly, full and free legislative deliberation about potential rights issues raised by proposed statutes is a central component of the new model and may helpfully be triggered, informed and alerted by the required ministerial statements. Thirdly, as part of a broader, more ground-up rights culture, executive and legislative rights deliberations at this pre-enactment stage may, in turn, spur popular engagement with rights-issues in a more direct, open and relevant way than is typically possible under judicial supremacy where such issues are raised and considered primarily in the context of an ongoing lawsuit. Fourthly, pre-enactment review is intended to provide a forum for the type of freer, unrestricted political and moral deliberations about relevant rights issues that (as Hiebert implies) may only be practically possible before a specific and potentially constraining or framing legal decision is handed down by the judiciary^[86]. Fifthly (and subject to rules of admissibility), this first stage also makes it more likely that there is some record of political deliberation and judgment on the relevant rights-issues for the courts to take into account if and when legislation is the subject of litigation, rather than only the subsequently prepared legal arguments of the government’s lawyers. Finally, and perhaps most obviously, it helps to protect rights by potentially identifying and resolving rights concerns that might never be litigated (i.e., for which this will be the only stage of rights review), that the judiciary might not otherwise find or address (here, countering possible judicial under-enforcement), or simply at an earlier stage than under judicial supremacy. As compared with this latter, by transferring some of the responsibility for rights protection from the courts to the electorally-accountable institutions, the new model seeks to reap the benefits of ex ante review compared to the judicial ex post.

These general functions or objectives of pre-enactment review within the new model directly suggest the ideal way in which such review would operate in practice. From the very beginning, as soon as the substance of a proposed bill starts to take form, its rights impact and implications would be taken seriously at all political and administrative levels within the sponsoring government department — and not only by its lawyers — and ultimately by the cabinet in deciding whether and where to place it on the legislative agenda. This would culminate in a detailed written report presented by the relevant minister (rather than the Attorney General or other senior specifically legal official) to the legislature setting out whether and why the minister is of the opinion that the bill is or is not compatible with protected rights. This report would be submitted to a specialized and non-partisan legislative committee with the exclusive task of bringing to the attention of the full legislature any rights concerns in pending bills whether or not deemed compatible by the minister and aiding it in its future debates of them. After holding hearings in which independent witnesses would provide opinions on the rights issues, the committee would present a clear, non-technical statement of any rights concerns it has to the legislature, which in turn would have sufficient time to debate them in full. Like the minister’s report, this statement should be fully available to the public, triggering discussion in the media and elsewhere ahead of the legislative debate. This debate would be understood to be informed by, but not limited to, the types of legal reasoning found in relevant previous judicial decisions, with the aim of achieving a broader, freer discussion of the moral and political aspects of the rights issues involved. Although framed within the context of a parliamentary system, these norms are readily convertible for use within a presidential one^[87].

Judicial rights review

If and when it arises, the second stage of rights review is undertaken by the judiciary under its new model powers of constitutional review. As we have seen, these range from giving statutes rights-consistent interpretations where possible to invalidation. Adhering to the « who, when, what » formula above, in the case of legislation this second stage takes place post-enactment in the course of concrete litigation^[88] and consists in the attempt by the relevant members of the judiciary to reach the best legal view on the merits of the relevant rights issue. That is, a good faith, independent and professionally skilful employment of the tools, techniques and reasoning styles of the courts.

As with pre-enactment political rights review, this second stage has several distinct functions within the overall matrix of the new model. First, although the new model departs from judicial supremacy in directly and deliberately seeking to foster greater rights consciousness, and to distribute rights responsibilities, among all three branches of government (as well as the citizenry) and not only the courts, it also of course departs from the traditional model of legislative supremacy that sees no rights reviewing role for courts in the context of enacted statutes. So this second stage is the designated one in which judicial rights-consciousness and responsibility is developed and exhibited. It provides the form and the space that rights review of state legislative action takes in the courts.

Secondly, this stage institutionalizes and executes the specific, affirmative reasons for granting the courts powers of constitutional review presented in the previous chapter. Most notably, it provides for a relatively fact-specific, applied dimension to rights deliberation that contrasts with the legislative approach of generality, and addresses the risk of under-enforcement of rights resulting from certain potential tendencies or pathologies of electoral majoritarianism that are not filtered out during the first stage. Moreover, to the extent that the risk of under-enforcement is also or additionally a general function of institutional responsibility for the final practical consequences of decision-making, the withdrawal of this responsibility from the courts under the new model, as compared with judicial supremacy (i.e., the existence of the third stage of review), provides another basis for their role in countering this phenomenon.

Thirdly, and connectedly, the courts exercise their rights responsibilities during the second stage under the shadow of the third. That is, one important function that the courts perform — in addition, of course, to their adjudicatory one — is to inform the legislature and alert the citizenry of their rights concerns from a legal perspective posed by a piece of legislation. Here, the virtues of skilled professionalism and judicial independence from electoral accountability within a majoritarian political system, especially a parliamentary one, play their role — not by conclusively or automatically rendering the ultimate decision but by bringing a perspective to bear on it that may otherwise not be brought. If the legislature has the power (but not the duty) to have the final word, the decision whether to exercise that power should be as fully informed as possible, with the addition of an authoritative legal perspective and its dissemination among the voters marking an important new piece of information as compared to when it reached its previous decision at the end of the first stage.

In light of these functions, how should the constitutional review powers of the courts be exercised? What norms ought to govern their use? What standard of review should the courts employ? Because of the new model’s conception of rights responsibility and deliberation as a joint institutional enterprise^[89], courts should take seriously the political rights review at stage one and not ignore or treat contemptuously as a usurpation of the judicial function any rights deliberations undertaken by the political branches, as sometimes occurs in US-style systems^[90]. This judicial posture also properly acknowledges the reality and inevitability of reasonable, good faith disagreement about the resolution of many rights issues, which is one of the major justifications for the new model in the first place. Obviously, to the extent that the political rights review at the first stage approaches the ideal version sketched above, the basis for this judicial posture toward it is further strengthened. At the same time, however, their informing and alerting function — which, under the model, exists alongside their adjudicatory one — requires that the courts provide an independent judgment that seeks to present the best legal view on the merits. That is, they should take into account but not be foreclosed by, or formally deferential to, the views of the political branches expressed at the previous stage. Judicial rights review should be respectful but unapologetic. Not only is it unconstrained by full practical responsibility for the final decision and its consequences that can lead to under-enforcement of rights within judicial supremacy, but cultivating the « passive virtues »^[91] would be structurally misplaced and counterproductive in a system of penultimate judicial review. For here it is the non-use or under-use of deliberately granted powers that threatens to undermine the distinctness and stability of this system.

Accordingly, as an exercise in reasoned persuasion rather than « raw judicial power »^[92], judicial rights review should be characterized by transparency and candour, permitting a full airing of the issues from a legal perspective. It should attempt to be as clear and accessible as possible for its most important intended audience is not fellow lawyers but members of the legislature and the public. Because the judicial decision is not necessarily the final one, there is no need to assume an air of « infallibility »^[93] or false objectivity; its authority will be as much a function of its quality as its pedigree. Here, colloquies and differences among judges will be particularly important and their absence, except in very clear cases, will likely detract from rather than add to its influence.

To be specific about the standard of judicial review, the better practice is for the courts to engage in merits rather than reasonableness review. That is, courts should ask whether the legislation in question is consistent with protected rights as they see it, and not whether the prior political judgment is a reasonable one^[94]. Not only does this standard of review enable the courts to best — most usefully and clearly — perform their informing and alerting function, but the seemingly « weaker » reasonableness standard would threaten to undermine the legislative power of the final word. This is because, given the norm of substantive reasonableness that (I shall argue) constrains the legitimate use of this power, for courts to effectively declare in finding an incompatibility at the second stage that the legislature has not acted within the bounds of legal reasonableness would be to create an excessively narrow window for the legislature at the third. Its power could legitimately be used only where it (1) reasonably disagrees with the court that its position is unreasonable or (2) finds that the legal perspective on the rights issue is too narrow and supplements it to arrive at a broader but substantively reasonable overall resolution of the rights issue. This first possibility is rather a subtle and complex one for the third stage political debate and its condition would only fairly rarely be satisfied. Moreover, the political costs of rejecting a judicial decision of legislative unreasonableness are likely to be very high, so that overall legitimate use of the power would be infrequent. There would also be the additional risk of judicial over-reaching whereby the courts in effect disguise merits review within the approved language of reasonableness, further stymieing the operation of the third stage.

It is at most only an apparent paradox that, as a result, the standard of judicial review is stronger under weak-form review than under strong-form. For the most part, merits review is either the acknowledged norm under contemporary strong-form judicial review or reflects the actual practice even where it is not, especially given the vagaries of self-policing and the absence of any checks on judicial over-reaching noted above^[95]. But even where reasonableness review is both norm and practice under strong-form judicial review, it reflects the fact that democratic concerns are built into the standard of review — in part because they have nowhere else to go^[96]. By contrast, under weak-form review, these concerns are institutionalized in the external check provided by the independent legislative power of the final word, which is why it is structurally important that the norms of its legitimate use are clarified. Moreover, given the broad scope of reasonable disagreement on rights issues, it will be relatively rare that the legislative resolution is objectively unreasonable so that there will be relatively few exercises of the new judicial powers and even fewer exercises of the new legislative one. As a result, the new model would likely operate overall in a less distinct manner under a reasonableness standard of judicial review.

Although this « better » judicial practice is thus based on the likelihood that a reasonableness standard of review would result in too few exercises of both judicial and post-enactment legislative powers under the new model, it is possible that the proposed merits review would lead to too many judicial findings of inconsistency that in turn overwhelms the political ability of the legislature to reject them. If this is the general or widespread consequence, then the norm should be adjusted to incorporate the second-best.

Legislative reconsideration

The third and final stage of rights review is the possible exercise of the final legislative word in light of the judicial review at the second stage. Again, it is the gap between this power and its exercise that is a distinguishing feature of the new model. Obviously, this review is conducted by the legislature — although it may be instigating by the executive — and takes place following an exercise of judicial rights review that is, in some significant sense, in tension with the legislative one undertaken at the first stage. The most obvious example is where a court finds that legislation is incompatible with a protected right, in exercise of its invalidation or declaratory power, but it can also be triggered by exercise of the judicial power/duty to give statutes a rights-consistent interpretation where possible, even where such a reading would not be the result of more traditional modes of interpretation^[97]. Like the first stage, this is a form of political rights review not only in the sense of who engages in it but also what it consists in. It would make little structural sense to insist that this third stage be exclusively restricted to duplicating and reviewing the reasoning style of the courts at the second stage.

Conceptually, this stage is the most distinctive one of the three, in that the pre-enactment political rights review of the first stage is not necessarily inconsistent with either judicial or legislative supremacy^[98]. That is, it appears to be a contingent and not a necessary feature of both systems that this stage is not legally mandated. By contrast, the function performed by the third stage is the key one of decoupling constitutional review from judicial supremacy, of permitting the legislative view about rights to prevail over the expressed judicial one.

The critical normative question that has not yet been fully aired and addressed is when this ought to happen^[99]. The routine use or non-use of this power of the final word, although obviously legally permitted, would undoubtedly reduce — although perhaps not eliminate altogether — the distinctness of the new model in practice. Accordingly, it is easy and obvious to say that its exercise should fall somewhere in between these two extremes, but what is more difficult and urgently required is a general normative theory of its legitimate use. This is because there is a widespread perception in practice, at least in Canada and the United Kingdom, that there is no such use, which may in turn help to explain its relative dormancy. As a result, development of this norm of legitimate use is a critical and urgent task for the new model.

To begin to address it, it may be helpful to return to the general normative goals of the new model and the specific reasons for granting the legal power of the final word to the legislature, as discussed in chapter three. The fundamental objective of the new model is to provide an institutional form of constitutionalism that effectively protects rights whilst maintaining greater balance or equality of power between legislatures and courts than under either of the two lopsided traditional ones. As an integral part of this greater balance, the new model disperses responsibility for rights among the three branches of government rather than allocating it more or less exclusively to either the courts or the legislature, thereby seeking to create a « ground up » or more democratic rights culture. In a nutshell, the specific reasons it is both unnecessary and unjustified to always give courts the legal final word on rights issues are threefold. First, judicial reasoning is inherently too narrow and artificially constrained for courts to definitively resolve most rights issues, issues that are not conclusively legal or interpretive in nature. Secondly, certain potential, rights-relevant legislative pathologies can be countered by less restrictive procedural and/or substantive constraints on outcomes than a full judicial veto. Thirdly, whilst enhancing the specifically democratic legitimacy of the rights regime, these same lesser constraints can also satisfy more general criteria of political legitimacy, such as reasonable public justification of government action or preventing the under-enforcement of rights.

The norms for the legitimate exercise of the legislative power of the final word are strongly suggested by these reasons for rejecting judicial supremacy. The first is procedural. To counter concerns about legislative pathologies and rights-relevant blind spots – indeed legislative deliberative capacities more generally — the legislature must engage in serious and principled reconsideration of the judicial decision on the rights issue^[100]. The criteria for such reconsideration cannot be exhaustively fixed in advance, but at a minimum require sufficient time allotted for debate and genuine, good faith deliberation on the issues of principle involved versus mere lip service or purely partisan discussion. In a similar but inverse way from the second stage, the legislature should engage seriously and respectfully with the judicial view but not automatically defer to it. Indeed, overall, the process here at the third stage is the most important thing and not the outcome, so that principled and serious legislative reconsideration resulting in decisions to comply with the courts’ view manifest what the new model seeks to achieve as much as do decisions not to comply, at least so long as the latter is generally taken to be a realistic political possibility. For it enhances both the democratic legitimacy of the rights regime and the content/quality of rights deliberation. In other words, compliance per se is not a problem, although a « culture of compliance » is^[101].

The second norm is substantive. Where, as a result of the above process, the legislature concludes that the judicial reasoning on the relevant rights issue, though perhaps compelling in its own terms, is nonetheless too narrow, technical or path-dependent, and that broader, fresher, or more direct engagement with applicable moral and political principles leads it to a different resolution, this would be a legitimate basis for exercise of the final word as long as this resolution remains within the bounds of overall substantive reasonableness. So, too, where the legislature concludes that the judicial reasoning on any of the relevant sub-issues —interpretation, scope, application of the right or the justification of limits on it — is not compelling in its own terms and reasonably disagrees with it, typically because after due consideration it is unpersuaded by the merits of the majority opinion and agrees with a dissenting one. For within the normative framework of the new model, there is no good reason for not permitting a procedurally sound and substantively reasonable legislative view to prevail. By the same token, however, it is not a legitimate exercise of the legislative power to sustain either a non-deliberative or an unreasonable decision on rights.

In practice, the legitimacy of a potential use of the legislative power as just enunciated would be a necessary but not a sufficient condition for its exercise; for the legislature will, in addition, most likely take an assessment of the political costs into account. In a sense, as discussed, an important practical goal of developing norms of legitimate use is to reduce the general political costs widely thought to be currently associated with exercise of the power — to explain that, and how, its use may reflect « rights disagreements » and not only « rights misgivings »^[102]. But even once this general case is successfully made, it does not of course mean that on a specific rights issue there will not be (prohibitively) high political costs of a legitimate exercise, because, for example, a particular judicial outcome is widely supported by the public. On the other hand, as legitimate use is a necessary condition, political benefits (rather than costs) alone would not justify an exercise of the legislative power.

Two objections to my position might be anticipated at this point. Firstly, mere norms — as distinct from legal limits — will likely inadequately constrain exercise of the legislative power so that the attempt to maintain a gap between the power of the final word and its exercise collapses. Of course, the greater practical problem at the moment, at least according to many, is the non-use of the power rather than its misuse. Putting this to one side, however, as a general matter the entire « norms literature » of recent years testifies to the powerful and independent conduct-shaping role of norms in social life^[103]. Closer to home and more specifically, the consensus view is that a norm of non-use with respect to the section 33 power is governing the conduct of federal and provincial legislatures in Canada. Relying on a legal limit, certainly a judicially enforceable one, may not only be unnecessary but would risk undermining the new model’s careful balance between judicial and legislative power by giving at least a final word to the courts. Conceivably, in the case of a section 33-type power, the two requirements of legitimate use could be expressed in the relevant text as legally binding but not judicially enforceable, although this would not work for declaration of incompatibility mechanisms because they have no legal effect; legislatures are not required to do anything in order to « override » the judicial decision.

Secondly, the particular norms I have suggested are too vague to be useful and/or too easy to abuse. Whilst certainly not wedded to these precise formulations, given the goal of placing political/moral conditions on the exercise of the power of the final word rather than legal ones for the reasons just stated, I believe that anything significantly more specific or determinate would (1) likely make use of the power too rare or difficult, (2) be too formalistic, as, for example, with a procedural norm of at least X hours of legislative debate, and (3) not be possible with respect to outcomes as reasonable legislative judgment is the only normative standard that coheres with the reasons for granting the power in the first place.

Stephen Gardbaum is the MacArthur Foundation Professor of International Justice and Human Rights at the University of California, Los Angeles (UCLA) School of Law. A recent recipient of Guggenheim and Straus Fellowships, his research focuses on constitutional design, comparative rights jurisprudence, constitutional theory and federalism.