2. Canada in Context

Section 33 of the Canadian Charter of Rights and Freedoms provides that the Parliament of Canada or a provincial legislature “may expressly declare in an Act of Parliament or of the legislature … that the Act or a provision thereof shall operate notwithstanding [certain specified Charter rights]”;Footnote ³⁷ and that a legislative provision or statute subject to such a declaration “shall have such operation as it would have but for the provision of this Charter referred to in the declaration.”Footnote ³⁸ Though a majority of Canadian constitutional commentators believe that the Canadian override should not be used pre-emptively— i.e., before a court has spoken on the issue—section 33 is not textually restricted in this manner.Footnote ³⁹ Indeed, it is notable that section 33 does not mention courts or judicial rulings at all. Therefore, its wording permits legislatures to derogate from the specified Charter rights both pre-emptively (i.e., regardless of whether there has been a judicial decision on the matter—or as a way of shielding legislation from an adverse judicial decision ex ante);Footnote ⁴⁰ and responsively (i.e., to avoid or overcome the effects of a judicial decision ex post).Footnote ⁴¹ In other words, the notwithstanding clause can be used as a defensive shield (protecting legislation from a judicial strike-down) and a pro-active sword (overriding a judicial decision).

Whilst section 33 empowers legislatures to override a range of Charter rights—and to override, ignore, or counteract court rulings on what those rights require—it does not create a constitutional carte blanche. On the contrary, there are a number of significant limits on the invocation of the override in Canada. First, there must be an express declaration to override a particular Charter right. Rights cannot be overridden covertly, implicitly, or obliquely.Footnote ⁴² If Canadian legislatures wish to override rights, they must make their intentions clear and public. Second, the declaration must be contained within an Act of Parliament or the legislature. Therefore, any proposed override must go through all the stages of the legislative process and, ultimately, garner majority support in order to be approved. Third, the override is only allowed with respect to some specified Charter rights, not all. Thus, democratic rights, mobility rights, and language rights are explicitly excluded from the ambit of the override.Footnote ⁴³ Therefore, the override has a restricted scope. Finally, any override is subject to a sunset and re-enactment clause. If enacted, the override ceases to have effect five years after it comes into force, whereupon Parliament may re-enact the override if it so wishes for a further five years.Footnote ⁴⁴ By requiring review and renewal at five-year intervals, section 33 ensures that any derogation from Charter rights requires repeated mobilisation of a political majority across different electoral cycles.Footnote ⁴⁵

The upshot is that instead of creating “an untrammeled opt-out”Footnote ⁴⁶ from rights or a constitutional free-for-all, section 33 permits a limited derogation from a restricted set of rights, subject to all the publicity, transparency, and justificatory requirements of the legislative process and broader public debate.Footnote ⁴⁷ Of course, as constitutional lawyers, we are prone to thinking of publicity requirements and political limits as weak and ineffectual constraints on political behaviour.Footnote ⁴⁸ However, seasoned politicians and political scientists are well-versed in the constraining effects of the “power of publicity” and the ways in which transparency requirements can structure political incentives, making certain policy choices harder to pursue.Footnote ⁴⁹ By requiring the override to be publicly visible and subject to the deliberative demands of the legislative process, section 33 places significant political and accountability hurdles in its path.

It is worth noting that section 33 does not speak in strident terms of legislatures ‘overriding’ rights or ‘disregarding’ court rulings. Indeed, it does not mention courts at all. Instead, section 33 employs much more careful and conditional language, acknowledging that legislatures ‘may’ proceed with legislation ‘notwithstanding’ (i.e., in spite of) the fundamental rights otherwise entrenched in the Charter, subject to specified conditions and constraints. As the general heading to section 33 makes clear, the power to enact legislation ‘notwithstanding’ specified rights is an ‘exception’ to the general norm of Charter compliance, conditional on political actors making an ‘express declaration’ to that effect in an Act of Parliament.Footnote ⁵⁰ Viewed as a whole, therefore, section 33 sends a message of seriousness about any proposal to legislate notwithstanding rights. It signals that a derogation from rights should not be taken lightly. Instead, section 33 mandates that such a derogation requires political care, parliamentary debate, and probing public scrutiny.

Section 33 began life as part of a political compromise to get agreement between the federal government and the provincial leaders in the negotiations leading up to the Canadian Charter.Footnote ⁵¹ Some provincial leaders feared that judicial decisions under the Charter would unduly constrict their ability to protect deeply-held social values and institutions in their province. Therefore, they sought some reassurance that democratic legislatures would have some meaningful input into the constitutional debate about “which rights are fundamental in Canadian society and which should prevail when rights are in conflict.”Footnote ⁵² Whilst they accepted that rights should be entrenched in the Charter—and that judges should be charged with the task of interpreting those rights—they nonetheless wanted to ensure that the legislature would retain the political capacity to depart from court rulings in exceptional circumstances, and have “the last word” on particularly contentious issues.Footnote ⁵³

Drawing on the historical precedents of notwithstanding clauses in the Canadian Bill of Rights 1960 and in various provincial Bills of Rights,Footnote ⁵⁴ some provincial leaders argued that the Charter should include a legislative power “to override a court decision which might affect the basic social institutions of a province or region.”Footnote ⁵⁵ In order to reconcile entrenched rights with Canadian traditions of parliamentary democracy, they supported “a constitutionalization of rights, subject to a final political judgment in certain instances, rather than a final judicial determination as to the extent of all rights.”Footnote ⁵⁶ Common justifications for including the notwithstanding clause were that it would 1) give the democratically elected legislature “the final say” over rights in circumstances of intractable disagreement;Footnote ⁵⁷ 2) preserve parliamentary sovereignty; 3) allow democratically elected legislatures to prioritise rights or values not contained in the Charter; 4) correct misguided judicial decisions on rights;Footnote ⁵⁸ and 5) allow for diverse provincial identities in a flourishing federalism.Footnote ⁵⁹ Ultimately, the inclusion of the override in the Canadian Charter broke the impasse between rival political positions, enabling some otherwise reluctant provincial leaders to come onboard with the Charter project.Footnote ⁶⁰ In short, without the override, the Canadian Charter might never have come into existence.

Whilst the notwithstanding clause eventually basked in the glow of international admiration as a normatively desirable “model of constitutionalism,”Footnote ⁶¹ it received decidedly mixed reviews in its early days on the Canadian constitutional scene. Some defended it as a way of keeping the flame of parliamentary sovereignty alive and saving Canada from the spectre of US-style judicial supremacy.Footnote ⁶² The then Premier of Saskatchewan, Allen Blakeney, argued that it was a valuable “constitutional escape valve”Footnote ⁶³ allowing legislatures “to make the ultimate social decisions”Footnote ⁶⁴ in a democracy. For others, however, the notwithstanding clause was an alien graft onto an otherwise rights-respecting document, which left minorities at the mercy of majorities.Footnote ⁶⁵ By ceding ‘the last word’ on rights to a simple legislative majority, some queried whether the override undermined the very reason for entrenching rights in the first place.Footnote ⁶⁶ After all, if the purpose of the Charter was to put rights beyond the vagaries of majoritarian politics, then the override seemed to subvert that aim. What the Charter gave with one hand, it took away with the override. For these commentators, the override was as ominous as it was odious. It seemed anathema to the Charter project of entrenching rights.

Amongst the political elites who brokered the deal on the Charter, some harboured deep reservations about the override. The then Prime Minister—Pierre Trudeau—believed that it was wrong to allow governments to suspend any part of the Charter, only reluctantly agreeing to section 33 on condition that it contained a sunset clause as a form of damage limitation.Footnote ⁶⁷ For those like Trudeau who opposed the clause, the concern was that section 33 gave politicians an untrammelled opt-out clause to trample over rights whenever it was politically expedient to do so.Footnote ⁶⁸ Therefore, part of the Charter negotiation centred on putting adequate constraints in place so that it could not be used lightly, hastily, or without adequate parliamentary and popular deliberation.Footnote ⁶⁹ The fact that the override was viewed as the product of a “grubby late-night deal”Footnote ⁷⁰ struck in the “the raw politics”Footnote ⁷¹ of constitutional negotiation behind closed doors, tainted the reputation of the override from the outset, casting it as an illegitimate child born into a family of noble rights-provisions, all of which could claim much higher breeding.Footnote ⁷²

However, the fears of those who worried that the override would undermine the Charter and obliterate rights were not borne out in practice. Mirroring the pattern of non-use of its historical precursor in the 1960 Canadian Bill of Rights, the clause was never invoked by the Parliament of Canada, and only rarely used in the Provinces.Footnote ⁷³ Indeed, the clause was so rarely used in the first three decades of the Charter that it was widely assumed to have fallen into desuetude.Footnote ⁷⁴ Even factoring in the recent flurry of provincial threats to use the clause, the notwithstanding clause has only been successfully invoked 19 times at provincial level over four decades, with Quebec accounting for 16 of those uses.Footnote ⁷⁵ Outside of Quebec, the invocation of the override is in single digits, and even within Quebec, 16 uses in over four decades is hardly a tsunami.Footnote ⁷⁶ The overall picture, therefore, is still one of relatively rare occurrence at the provincial level. So, what explains the underuse of the override in the Canadian context? The following four reasons are commonly advanced.

The first is the path dependence argument. Footnote ⁷⁷ Just nine weeks after the Charter was proclaimed, the Quebec legislature used the notwithstanding clause in an omnibus fashion to encompass all statutes enacted in Quebec and even to immunise future statutes from constitutional review.Footnote ⁷⁸ Since Quebec was the sole province not to have signed up to the Constitution Act, 1982, its omnibus use of the override was widely regarded as a form of political protest against the Charter project as a whole.Footnote ⁷⁹ Many Canadian commentators argued that Quebec’s “sweeping and indiscriminate use of the clause” during its delicate incubation period, discredited and delegitimised the override as an acceptable constitutional tool which legislatures could use to disagree with the courts.Footnote ⁸⁰ As Canadian political scientist, Christopher Manfredi, put it: “Canadians experienced a use of the notwithstanding clause that they found outrageous before they experienced a Supreme Court decision of equivalent political unpopularity.”Footnote ⁸¹ Therefore, many argue that this historical contingency set Canada on a path of underuse from which there was no going back.Footnote ⁸²

The second reason is the familiar political costs argument.Footnote ⁸³ This is the claim that although the Parliament of Canada and the provincial legislatures wanted to use the override, they were effectively disabled from doing so by the exorbitant political costs accompanying such a move.Footnote ⁸⁴ Since Charter rights are highly prized in Canadian political culture, a political attempt to override them would inevitably cast politicians in the role of ‘human rights transgressors’ and Charter cheaters, determined to ride roughshod over rights and the rule of law.Footnote ⁸⁵ Canadian politicians are acutely aware that they would suffer significant popular, parliamentary, and political pushback if they attempted to override rights or disregard a decision of the Supreme Court of Canada, especially as judges are highly respected actors in Canadian constitutional culture.Footnote ⁸⁶ Indeed, some have argued that the courts have contributed to a culture of judicial supremacy by presenting themselves as the ultimate guardian and custodian of rights, thereby delegitimising any attempt to overturn judicial decisions.Footnote ⁸⁷ Whatever the source of the political costs, many observers believe that they were sufficient to render the override “politically toxic”Footnote ⁸⁸ and even taboo, at least at the Federal level.Footnote ⁸⁹ Whilst the political mood-music surrounding the notwithstanding clause has now changed such that invocation of the override is no longer seen as taboo, the political costs argument still has purchase as a plausible reason why many politicians across Canada are reluctant to use the override, right up to contemporary times.Footnote ⁹⁰

The third reason is rooted in textual constraints. The argument here is that the various textual constraints built into section 33 increased the political costs of using it, thus rendering it effectively impossible to use.Footnote ⁹¹ The sunset and re-enactment clauses operate as a particularly strong disincentive in this regard, because even if a government manages to mobilise sufficient political support to use it on one occasion, this is only a “temporary alliance” which requires the expenditure of further political capital five years down the line.Footnote ⁹² Given the political costs of invoking the clause and the “short duration of the reprieve it offers,” the political actors are dis-incentivised from incurring those costs in the first place.Footnote ⁹³ Some comparative constitutional scholars argue further that the wording of section 33 “requires the legislature to misrepresent its position on rights.”Footnote ⁹⁴ Instead of signalling that the legislature disagrees (reasonably) with the judicial interpretation of rights, these scholars suggest that section 33 creates the false impression that legislatures wish to violate or deny rights altogether.Footnote ⁹⁵ As Jeremy Waldron put it, the notwithstanding clause misrepresents the legislature as having disreputable “rights-misgivings” instead of being engaged in reasonable “rights-disagreements” with courts.Footnote ⁹⁶ This reputational cost deters political actors from invoking the override.

The fourth reason commonly advanced for the underuse of the override is the less drastic measures argument. In Canadian constitutional discourse, the override is often characterised in dramatic terms as a “nuclear bomb,”Footnote ⁹⁷ a “political sledgehammer,”Footnote ⁹⁸ a “majoritarian cudgel,”Footnote ⁹⁹ a “loaded rifle,”Footnote ¹⁰⁰ a “sword of Damocles,”Footnote ¹⁰¹ or even a “dagger pointed at the heart of our fundamental freedoms.”Footnote ¹⁰² Given this context, using the clause can seem like “radical overkill.”Footnote ¹⁰³ It is hardly surprising, therefore, that politicians will seek out less confrontational and dramatic ways of achieving the same or similar effect.Footnote ¹⁰⁴ Why use a sledgehammer when you have more subtle and tailored tools to do the work? The most obvious option is to attempt to justify a rights-limitation as justifiable in a free and democratic society under section 1 of the Charter.Footnote ¹⁰⁵ Another is simply to enact a law retaining much of the original policy objective, whilst implementing the judicial decision in a minimal way. These more subtle and accommodating approaches enable the government and legislature to achieve their policy objectives to some degree, whilst avoiding the popular outcry and political costs involved in seeking to override a court ruling outright. The availability—and utility—of these less drastic measures leads to an underuse of the override.

3. Reconciling Rights and Democracy, UK-Style

When the Human Rights Act 1998 was enacted in the UK, the central concern across the political spectrum was to find a way of allowing judges to enforce human rights whilst preserving parliamentary sovereignty.Footnote ¹⁰⁶ Previous attempts to incorporate Convention rights into UK law had foundered on a deep-seated aversion to the idea of allowing judges to strike down legislation enacted by Parliament.Footnote ¹⁰⁷ Such a power was perceived to be “anathema to the political and legal culture of the United Kingdom under which ultimate sovereignty rests with Parliament.”Footnote ¹⁰⁸ Therefore, all the key political actors knew that a Bill of Rights which included a strike-down power would never get enacted. The Canadian solution of including a judicial strike-down but offsetting it with a legislative override, was considered but ultimately rejected, because the combination of rights-entrenchment and judicial invalidation was thought to violate the fundamental doctrine of parliamentary sovereignty which was “so uncompromisingly embedded in [the British] political and legal culture.”Footnote ¹⁰⁹ The central challenge facing the political architects of the HRA was to devise a scheme which would give the courts “as much space as possible to protect human rights, short of a power to set aside or ignore Acts of Parliament.”Footnote ¹¹⁰

The solution to this conundrum was a carefully crafted legislative scheme containing two crucial components. The first was the legally non-binding judicial “Declaration of incompatibility” under section 4 of the HRA, which:

The second was the possibility of a negative Statement of Compatibility under section 19(1)(b) of the HRA. Section 19 (1) provides that:

The heavily qualified and conditional wording of section 19(1)(b) “signals the seriousness of proceeding without a positive Statement of Compatibility,”Footnote ¹¹³ suggesting that a “nevertheless statement”Footnote ¹¹⁴ of incompatibility will be the exception, not the norm. In practice, the government operates a “strong presumption”Footnote ¹¹⁵ in favour of issuing a positive Statement of Compatibility under section 19(1)(a)—a presumption which has only been rebutted on two occasions. Of course, on a surface reading, both the judicial declaration of incompatibility and the political statement of incompatibility seem to leave the UK government and Parliament completely free to legislate contrary to Convention rights and contrary to court rulings on what those rights require. Indeed, they seem specifically designed to do so. However, the UK Parliament has complied with almost every single declaration of incompatibility issued by the UK courts.Footnote ¹¹⁶ Similarly, UK governments have hardly ever invoked the negative ‘Statement of Compatibility’ under section 19, and Parliament has hardly ever enacted such legislation.Footnote ¹¹⁷ So, what explains the consistent practice of complying with rights and court rulings on rights under the Human Rights Act 1998? Two key reasons are typically advanced.

The first reason is rooted in the crucially important international law dimension of the HRA. If the UK government and Parliament decide to ignore or override a declaration of incompatibility by the UK Supreme Court, then the aggrieved litigant can take their case to the European Court of Human Rights at Strasbourg (ECHR), armed with a considered declaration by the UK Supreme Court that domestic legislation clearly violates their rights.Footnote ¹¹⁸ In this situation, the Strasbourg court is highly likely to find in the litigant’s favour, thus triggering an international law obligation on the UK to remedy the rights-violation declared by the ECHR.Footnote ¹¹⁹ Therefore, UK Governments are well aware that little will be achieved, and a good deal may be lost, “by not amending the incompatible legislation” once the declaration is issued.Footnote ¹²⁰ Given the rhetorical force of a domestic judicial ruling that legislation violates human rights, combined with the political and legal repercussions at the international level, Parliament is placed under enormous political pressure to amend statutes to accommodate the judicial ruling.Footnote ¹²¹ This helps to explain why compliance with—rather than defiance of—declarations of incompatibility is the norm within the UK system.

The second reason is the familiar argument from political costs. As in Canada, many UK-based scholars argued that the legislative freedom to enact laws notwithstanding rights, or to disregard declarations, was effectively negated by the political costs of using that power.Footnote ¹²² They suggest that reverence for rights and esteem for the courts made it “not only politically difficult but also constitutionally questionable for parliaments to reject a court’s particular interpretations or even question a court’s interpretive methods.”Footnote ¹²³ Moreover, whilst a public ‘Statement of Incompatibility’ is legally possible under section 19, the worry is that it is politically unpalatable because politicians supporting such a Statement would be pilloried as ‘human rights transgressors’.Footnote ¹²⁴ The upshot is that although the democratic legislature is entitled to override rights and disagree with courts, it was forced to adopt a “compliance oriented mentality,”Footnote ¹²⁵ capitulating to the judiciary rather than questioning or challenging their rulings. Instead of unleashing a laudable “culture of controversy”Footnote ¹²⁶ where legislatures and courts could disagree openly and fervently about the meaning of rights, the political dynamics created a lamentable “culture of compliance”Footnote ¹²⁷ where legislatures were expected to kowtow to courts.

4. Underuse by Design, not Demonisation

Though there are varied reasons for the underuse of the override in Canada and the UK, the dominant narratives in both countries nonetheless converge on a common theme—namely, that the override was thwarted by various political costs, contingencies, and cultural constraints which prevented legislatures from seizing ‘the last word’ on rights. On this view, the override was subtly subverted by a set of political dynamics which rendered it politically impotent, thereby neutering the notwithstanding clause and thwarting it from realising its full potential.Footnote ¹²⁸ Since the override was embedded in a culture of judicial supremacy and reverence for rights, so the story goes, the legislative override became delegitimised and, therefore, fell into desuetude.Footnote ¹²⁹

If this diagnosis of demise and desuetude is accepted, then a natural cure presents itself. In order to rescue the override from a descent into desuetude and realise its “unfulfilled potential,” we should try to lower the political costs of using it, and recalibrate the constitutional culture in order to resurrect the override.Footnote ¹³⁰ At least in the pre-populist era, the scholarly landscape on both sides of the Atlantic was strewn with normative claims of this kind, where we were urged to ‘reframe’ popular conceptions of the override in order to realise its untapped potential as a legitimate and democratically desirable form of legislative disagreement with courts.Footnote ¹³¹ Many comparative constitutional lawyers in Canada and beyond argued that we should rethink,Footnote ¹³² reconstruct,Footnote ¹³³ and reconfigureFootnote ¹³⁴ the role of the legislature and the courts, in order to “revive,”Footnote ¹³⁵ reincarnate,Footnote ¹³⁶ “rehabilitate,”Footnote ¹³⁷ and reinvigorateFootnote ¹³⁸ a flourishing legislative practice of overriding court rulings on rights.Footnote ¹³⁹ In the UK, too, scholars issued a clarion call to the Westminster Parliament to “assert its own supremacy” on rights by disagreeing more robustly and regularly with courts, and rejecting declarations of incompatibility when the legislature disagrees with them.Footnote ¹⁴⁰ For some scholars, inaugurating a regular practice of overriding judicial decisions would “[reaffirm] the ‘genius’ of the HRA” and realise the lost promise of dialogue.Footnote ¹⁴¹

But this narrative of thwarted potential and unrealised promise is deeply problematic. For one thing, it is not clear that the political costs of using the override are as unremittingly severe as the argument assumes.Footnote ¹⁴² In the UK, popular buy-in to the HRA is weak, in part because political elites have aligned with the tabloid press to portray the HRA as a “villain’s charter.”Footnote ¹⁴³ In a political climate which is inhospitable or even hostile to rights, political actors who defy court rulings may be hailed as heroic defenders of British values, rather than the villainous violators of fundamental rights.Footnote ¹⁴⁴ Given this political atmosphere, the claim that the political actors are politically disabled from defying court rulings on the rights of sex offenders, terrorist suspects, prisoners, and immigrants, seems tenuous at best.Footnote ¹⁴⁵ Indeed, the most remarkable feature of the UK experience is that despite the political costs of abiding by controversial court rulings, the UK Government and Parliament have consistently complied with them.

The political context in Canada is different, given the Charter’s role as an important symbol of Canadian self-identity.Footnote ¹⁴⁶ Popular support for the courts and the Charter seems relatively strong across the Canadian polity, and there is a relatively strong culture of compliance with judicial rulings amongst political elites.Footnote ¹⁴⁷ It seems to follow that the political costs of using the override will typically be high.Footnote ¹⁴⁸ Indeed, one of the reasons why some of the recent provincial proposals to invoke the notwithstanding clause have failed is precisely because the political and popular pushback has been formidable.Footnote ¹⁴⁹ However, even in Canada it cannot be assumed that the political costs will invariably impede the override, especially in today’s more polarized and populist political environment. The recent uptick in the override across the Provinces suggests that the political aversion to the override may be more fragile than previously assumed. Indeed, there is a sense that we are entering a new era of the override where the political costs have plummeted, thus enabling (primarily Conversative) politicians to use the notwithstanding clause more freely than heretofore.Footnote ¹⁵⁰ And the political costs have never been as high in Quebec as in the other provinces.Footnote ¹⁵¹ In fact, leading Canadian constitutional commentators have long suggested that political commitment to the Charter may vary from issue to issue, time to time, and province to province, such that public opinion could well go against a court ruling on a particularly contentious and sensitive issue.Footnote ¹⁵²

But there is a deeper problem with the narrative of thwarted potential and unrealised promise, and it is this: The narrative rests on the assumption that the override was intended to realise the promise scholars impute to it—namely, that of facilitating ongoing dialogue and disagreement between legislatures and courts in circumstances where they disagree.Footnote ¹⁵³ However, when we look at the legislative history of the override mechanisms in both countries, it seems clear that these assumptions were not shared by the political architects of these mechanisms on either side of the Atlantic.Footnote ¹⁵⁴

In Canada, the political actors who brokered the compromise on the Charter typically presented the override as “a safety valve to correct absurd situations,”Footnote ¹⁵⁵ which should only be used in the “unlikely event of a decision of the courts that is clearly contrary to the public interest.”Footnote ¹⁵⁶ Rather than viewing it as a vehicle for ongoing dialogue or disagreement between courts and legislatures on the meaning of rights, the key political architects conceived it more narrowly as an “exceptional”Footnote ¹⁵⁷ way of resolving intractable disagreements about what rights require in a democracy, or as a mechanism for judicial error-correction in extremis to counteract ‘absurd decisions’. Like a safety valve in a boiler, the override was designed to defuse tension in order to avert a potentially explosive situation, thus keeping the boiler functioning in the longer term.

Once it was apparent that the notwithstanding clause would be included in the Charter, the crux of the political negotiations turned to placing adequate safeguards and constraints on its use, so that it could not be used lightly, hastily, covertly, or without due legislative and popular deliberation.Footnote ¹⁵⁸ By requiring an express declaration in an Act of Parliament, section 33 was crafted to ensure that any proposed override would be subject to the full glare of political, parliamentary, media, and public attention.Footnote ¹⁵⁹ The sunset and re-enactment clause was inserted precisely to dis-incentivise its use ex ante, as well as to limit its effects ex post.Footnote ¹⁶⁰ For political actors who were philosophically opposed to the override—including the then Prime Minister, Pierre Trudeau—they accepted the override in the hope and belief that, just like its forerunner in the 1960 Canadian Bill of Rights, it was “unlikely ever to be used.”Footnote ¹⁶¹ Even those who advocated for the override never argued that it should be used on a routine or regular basis, whenever the legislature disagreed with court rulings. Instead, they defended the notwithstanding clause as a safety valve measure to be used with due deliberative focus in those instances where no other solution could be found.Footnote ¹⁶²

Thus, whilst the Charter permits legislatures to enact legislation ‘notwithstanding’ some of the Charter rights, section 33 ratcheted up the political costs and reduced the political gains. By heightening the “enactment costs” of legislating notwithstanding Charter rights, section 33 was designed to ensure that an override would not be an everyday occurrence.Footnote ¹⁶³ Political safeguards were hardwired into the provision in order to forestall frequent, hasty, or ill-considered invocation of the override. The aim of section 33, therefore, was not to enable legislatures to override rights whenever they disagreed with judicial rulings, but rather to facilitate a limited derogation from the Charter in circumstances of deep and intractable disagreement where no other solution could be found.Footnote ¹⁶⁴ At the very least, section 33 was designed to ensure that an invocation of the override would be exposed to political deliberation and significant accountability effects.

Rather than being “a bottle labelled ‘Drink me’ that cheapens the Charter,”Footnote ¹⁶⁵ the override was more like a bottle bearing a large health-warning, saying: ‘This drink may endanger the long-term health of the constitutional system. Only use this strong medicine if you have no other option—and be mindful of negative side-effects.’ The Charter may have given Parliament and the provincial legislatures the power (in the sense of the capacity or the option) to override court rulings in specified circumstances, but it was an option which would, they hoped, rarely arise. “If the purpose of the override was to give the legislature ‘the last word’, then it was very much the last word as last resort.”Footnote ¹⁶⁶ As Caitlin Salvino argues, the power to override was accompanied by a “political norm of rare use.”Footnote ¹⁶⁷ Section 33 erected some political and structural safeguards in order to create some ‘constitutional slow-down’,Footnote ¹⁶⁸ thereby avoiding “abusive invocation.”Footnote ¹⁶⁹

Turning to the UK, a strikingly similar picture emerges from the legislative history of the HRA. In the White Paper preceding the HRA, it was emphasised that although the judicial declaration of incompatibility would “not of itself have the effect of changing the law … it will almost certainly prompt the Government and Parliament to change the law.”Footnote ¹⁷⁰ This expectation of regular political compliance with judicial declarations was affirmed in parliamentary debates on the Human Rights Bill, where the Government Ministers proposing the Bill—Jack Straw MP in the House of Commons and Lord Irvine of Lairg in the House of Lords—stated their belief that “the Government and Parliament will in all cases almost certainly be prompted to change the law following a declaration.”Footnote ¹⁷¹ True, they acknowledged that there might be a rare case “of great controversy” where Parliament might not wish to accept a declaration of incompatibility.Footnote ¹⁷² However, they emphasised that this would be a rare and exceptional occurrence. As Jack Straw MP observed:

The use of obligatory language here is telling. It reflects the potent combination of constitutional norms, legal principles, political pressures, and international law requirements which, together, give the UK government numerous incentives to comply with a declaration of incompatibility, rather than allow further unnecessary and costly litigation in Strasbourg.Footnote ¹⁷⁴

The expectation of the drafters of the HRA that the government would comply with judicial declarations of incompatibility is further emphasised by section 10 of the HRA. Following a declaration of incompatibility, section 10 of the HRA provides for a ‘fast-track amendment procedure’ which enables a Minister to “make such amendments to the legislation as he considers necessary to remove the incompatibility” declared by the courts.Footnote ¹⁷⁵ Though this ‘fast-track amendment procedure’ is an option, not an obligation, it shows that the HRA was designed to incentivise, ease, and expedite legislative compliance with declarations of incompatibility.Footnote ¹⁷⁶ Indeed, the ‘declaration of incompatibility’ was partly modelled on the declaratory rulings of the European Court of Human Rights, which were intended to generate a strong sense of political responsibility to comply with those rulings.Footnote ¹⁷⁷ In short, the aim was not to facilitate an ongoing inter-institutional dialogue about rights where legislatures could routinely override judicial decisions, but rather to inculcate a practice of complying with judicial rulings on rights.Footnote ¹⁷⁸

A similar expectation of compliance with rights underpinned section 19 of the HRA. Although section 19 allows the UK government to propose a Bill to Parliament which the government believes to violate Convention rights, the firm expectation was that a negative Statement of Incompatibility would be a rare rather than regular occurrence. When discussing section 19, Jack Straw MP stressed that “Ministers will obviously want to make a positive statement whenever possible”Footnote ¹⁷⁹ and that it was “obviously … incumbent on Ministers … to do their best to ensure that bills are compatible with the Convention.”Footnote ¹⁸⁰ The rationale of requiring governments to make a Statement of Compatibility “openly, in the full glare of parliamentary and public opinion,” was not to encourage such statements or make them a regular or routine occurrence.Footnote ¹⁸¹ On the contrary, it was to incentivise political actors to think long and hard before legislating incompatibly with rights, and ultimately avoid doing so if at all possible.Footnote ¹⁸² As Janet Hiebert observed:

Rather than giving politicians a carte blanche to disregard rights, section 19 harnessed the ‘power of publicity’ and allied mechanisms of political accountability to force Ministers “to stand up and be counted for human rights.”Footnote ¹⁸⁴ At the very least, section 19 sought to ensure that any deviations from the requirements of the Convention would be “conscious and reasoned departures, and not the product of rashness, muddle or ignorance.”Footnote ¹⁸⁵ Just as the judicial declaration of incompatibility was expected to incentivise prompt political compliance, a governmental Statement of Incompatibility in the Westminster Parliament was intended to heighten political scrutiny and raise the ‘enactment costs’ of a Bill which threatened to violate rights. Instead of finding ways of reducing those costs and making a legislative override of Convention right easy to achieve, the HRA deliberately increased them in order to ensure that the override could not be used lightly, covertly, obliquely, or rashly. Compliance was eased; defiance was exacerbated. The overall message was that disregarding rights or defying court rulings on rights was not intended to be an easy or an everyday occurrence. On the contrary, it was envisaged as a rare and exceptional event, which should be justified in the full glare of probing political and public scrutiny.

When we look at the origins of the override in both jurisdictions, it seems clear that the key political architects of these clauses viewed the override as a safety valve for exceptional circumstances, rather than a regular mode of dialogic engagement between the courts and legislature. Whilst the Canadian Charter and the UK Human Rights Act undoubtedly gave the last word to the legislature, it was a ‘last word as last resort’. This opens up a novel solution to the mystery of the underuse of the override, namely, that ‘underuse’ or ‘rare use’ was hardwired into the design of the Bills of Rights from the outset. It was a feature, not a bug, in the system. Rather than treating the political costs as unforeseen or unwelcome impediments to the ideal functioning of the override—or as tragic failures of its dialogic promise—the historical narrative suggests that the political costs were deliberately harnessed to incentivise compliance with court rulings on rights. Underuse of the override was the product of design rather than demonization.

Of course, it could be argued that the political motivations in Canada and the UK were a matter of historical happenstance or, alternatively, that the political architects of the Canadian Charter and UK HRA were normatively misguided about how the override should operate. Maybe the political drafters should have internalised the lessons of dialogue theory and made it easier rather than harder to invoke the override as a regular form of inter-institutional exchange? In the next section, I will argue—contra that claim—that the legislative override is correctly conceived as a safety valve to be used in rare and exceptional circumstances. Grounding the argument in a deeper constitutional story about the need for the branches of government to forge constructive working relationships in a collaborative scheme, I will argue that legislatures should adopt—and in Canada and the UK largely do adopt—a presumption in favour of complying with rights and court rulings on rights, unless there is a sufficiently strong reason to do otherwise. This argument suggests that the political motivations of the drafters of the Canadian Charter and the UK HRA were more than mere historical happenstance. Instead, they were rooted in a proper understanding of the roles and relationships between the branches of government in a well-functioning constitutional democracy.

5. From Dialogue and Disagreement to Comity and Collaboration

When the Canadian notwithstanding clause appeared on the global constitutional scene, it was feted by comparative scholars and constitutional theorists who viewed it as the epitome of democratic constitutionalism and a seductive solution to the counter-majoritarian difficulty.Footnote ¹⁸⁶ Instead of giving the courts the ‘last word’ on which rights require, the override allowed legislatures to speak back and make their views prevail. Based on an abstracted conception of constitutional roles, scholars argued that legislatures should embrace the override and disagree with courts as part of a constitutional dialogue on the meaning of rights.Footnote ¹⁸⁷ However, as experience accumulated under both the Canadian Charter and the UK HRA, these normative claims about the desirability of dialogue and disagreement on rights were not internalised by the legislative actors on either side of the Atlantic. Indeed, it is interesting to note that when provincial premiers started to use the override more often in recent times—or at least threaten to do so—this was not celebrated by dialogue theorists as the apotheosis of democratic and dialogic constitutionalism. On the contrary, the uptick in the override—particularly its controversial invocation by Doug Ford in Ontario, or in the Laicity Act in QuebecFootnote ¹⁸⁸ —has been widely criticised as a crude manifestation of populist politics designed to undermine rights, violate the rule of law, and delegitimise courts.Footnote ¹⁸⁹ Though inter-institutional dialogue and ‘reasonable’ disagreement sounded nice in theory, they began to look a lot more problematic in practice.

So, what explains this deep disjuncture between the leading scholarly narratives about the override, on the one hand, and common constitutional practice, on the other? How can we reconcile the commitment to bind legislatures to constitutional norms in Bills of Rights, on the one hand, and apparently free them from those obligations, on the other? My argument is that instead of perceiving the override as a vehicle for ongoing dialogue and open disagreement between the branches on the meaning of rights, we should view it instead as a constrained mode of respectful engagement within a collaborative constitutional enterprise.Footnote ¹⁹⁰ Once we situate the override within an understanding of the roles and relationships between the branches of government in a workable constitutional scheme, we can see that the legislature should treat the override with caution and care, out of respect for the value of rights, the rule of law, and the independence of the courts.

In the collaborative constitution, the relationship between the branches of government rests on a constitutional division of labour between the three branches of government, where each play distinct but complementary roles whilst working together in a spirit of comity and mutual respect.Footnote ¹⁹¹ Constitutional comity has two dimensions: mutual self-restraint and mutual support.Footnote ¹⁹² The requirement of mutual self-restraint means that each branch should refrain from interfering with the capacity of the other branches to carry out their role in the constitutional scheme, or from undermining each other’s authority. The duty of mutual support captures the ‘affirmative obligations’ which require the branches to positively assist and support one another in carrying out their respective roles.Footnote ¹⁹³ On the collaborative vision of constitutionalism, the relationship between the branches is conceived as a “long-term partnership over time,” where each branch has a distinct role to play in making constitutional government work as a joint enterprise.Footnote ¹⁹⁴

As in any well-functioning partnership, mutual respect is the foundational normative requirement. Therefore, the key value underpinning the collaborative constitution is comity, i.e., the duty of each branch to respect and support the proper functions of the others.Footnote ¹⁹⁵ Comity does not require the branches to agree with each other. Comity and contestation are not mutually exclusive options.Footnote ¹⁹⁶ However, it does require the branches to respect the distinct but valuable contributions of each branch to the constitutional scheme. The fact that the branches of government are embedded in a long-term working relationship over time shapes the interaction between them. In order to sustain a successful long-term relationship, the interaction between the branches is framed by the requirements of “reciprocity, repeat-play and reputation.”Footnote ¹⁹⁷ An antagonistic move against another branch might trigger open retaliation and a break-down in reciprocal respect. Opportunistic power-play might secure a short-term advantage, but it can undermine a branch’s reputation as a reliable partner in the constitutional endeavor in the longer-term.Footnote ¹⁹⁸ An unduly confrontational move against another branch may trigger open retaliation, unleashing repeated “cycles of escalating constitutional brinkmanship” and a break-down in reciprocal respect.Footnote ¹⁹⁹ In order to sustain good working relations over the long-term—and accrue the mutual benefits of a stable constitutional order over time—the branches of government should adopt a norm of conflict-avoidance or at least conflict-minimisation.Footnote ²⁰⁰ These three norms—comity, collaboration, and conflict-management—frame the relationship between the branches of government and constrain the interaction between them.

So, how do these norms bear on the operation of the override? They suggest that a legislative override of court decisions should be used with caution and care, attentive to the need to treat the other branches with comity and respect as part of a collaborative enterprise. Constitutional comity does not preclude political actors from overriding court decisions in some circumstances—but it precludes from regularly or lightly overriding court decisions merely because they disagree with the outcomes of judicial decisions. This gives rise to a constitutional presumption in favour of compliance with court rulings on rights, which should only be rebutted in exceptional circumstances.Footnote ²⁰¹ Though the elected branches of government may have first-order reasons to disagree with court decisions, they have second-order institutional reasons to respect and implement those decisions, unless exceptional circumstances arise.Footnote ²⁰²

Consider the following problems which would arise if legislatures regularly overrode judicial decisions. First, if judicial decisions were cast aside whenever political actors disagreed with them, this would potentially undermine the institutional credibility and integrity of the courts in the constitutional scheme.Footnote ²⁰³ After all, if judges know that their decisions will be ignored if they do not find favour with government, this will render their decisions toothless and undercut their ability to hold the government to account. In systems which recognise the value of judicial independence, the aim is to create an institutional environment where judges can make decisions “without fear or favour,” confident that they will not be sanctioned or ignored if their rulings do not find favour with the powers that be.Footnote ²⁰⁴ Therefore, the rule of law creates a rule of thumb that the government and legislature should generally comply with court rulings, even if they disagree with them.Footnote ²⁰⁵ Judicial independence only survives in a system where there is a culture of respect for courts and the rule of law across the political spectrum.Footnote ²⁰⁶ If the legislature regularly overrode court rulings, this would undermine the position of the judiciary and strain the relations of comity between the branches.Footnote ²⁰⁷ In a system based on the separation of powers, it is “an important principle of comity … that political bodies do not attack the integrity of the accountability institutions established to monitor them.”Footnote ²⁰⁸ If political actors routinely second-guessed court decisions, this would subvert the constitutional division of labour on which the system of constitutional democracy rests.Footnote ²⁰⁹

Second, to treat court rulings merely as “thoughtful opinions”Footnote ²¹⁰ to be cast aside whenever another branch disagrees with them is to misconceive—and potentially undermine—the authoritative nature of judicial decisions in the constitutional scheme.Footnote ²¹¹ After all, it is the mark of authority that it binds even when we disagree.Footnote ²¹² Of course, if we view the interaction between the courts and the legislature as a dialogue between co-equal branches, we might be tempted to think that constitutional government is “a dynamic process involving the interchange of proposals for constitutional meaning” where “each institutional actor brings forth its understanding for consideration and examination by the other.”Footnote ²¹³ However, the role of the courts is not merely to offer some thoughtful suggestions for the legislature to ponder, but rather to provide an authoritative resolution of legal disputes by declaring what the law is, in a context where that resolution is treated with due regard and respect.Footnote ²¹⁴ Regular override of judicial decisions would undermine the authority of the courts in the constitutional scheme.

Finally, a regular practice of overriding the courts would create enormous uncertainty, unpredictability, and unfairness for litigants.Footnote ²¹⁵ If judicial decisions were merely provisional pronouncements subject to regular reversal by the legislature, this would mean that litigants could not rely on the courts to give them a final, authoritative ruling about what their rights require.Footnote ²¹⁶ Whilst a democratic dialogue between the courts and the legislature might be “normatively appealing” from the point of view of scholars attuned to the counter-majoritarian difficulty, it is not so appealing to litigants, who look to the courts for an authoritative ruling on what their rights require, which should then be honoured and respected by the political powers that be.Footnote ²¹⁷ Allowing parliamentarians to consider each case anew and reject any decisions they dislike would be a cruel waste of time and energy for litigants, who need an authoritative resolution of their constitutional claim.Footnote ²¹⁸ Indeed, one of the aims of the UK HRA was to give litigants real rather than illusory remedies for violation of Convention rights.Footnote ²¹⁹ A political practice of regularly ignoring or overriding judicial decisions on rights would undercut this ambition. Judges have a constitutional responsibility to litigants—and to the legal system as a whole—to resolve these cases in an authoritative, fair, and impartial way. A regular legislative override of their decisions would undercut this responsibility and damage the role of the independent courts.

Once we situate the override in the context of the constructive working relationships between the branches of government based on comity and reciprocal respect, we can see that the legislature does not—and should not—have the kind of “equal, revisory or final role” attributed to them by dialogue scholars.Footnote ²²⁰ Instead, the legislature is under a presumptive obligation to comply with judicial rulings, unless there are strong and weighty reasons to do otherwise. The upshot is that the override is correctly conceived as a safety valve for exceptional circumstances, to be treated with caution and care, rather than as a regular outlet for inter-institutional dialogue and disagreement about rights. Just as courts have content-independent duties to respect the decisions made by the democratic legislature, the elected branches of government have content-independent institutional reasons to comply with judicial decisions on rights, even if they disagree with them.Footnote ²²¹ This does not deny the possibility or, indeed, the legitimacy of a legislative override in some circumstances. Courts can get it wrong. Therefore, it is valuable, in an appropriate case, for legislatures to be able to correct or even pre-empt that error. However, constitutional principles—such as, e.g., the separation of powers, inter-institutional comity, the independence of courts, and the rule of law—counsel that an override of rights or court rulings on rights should be reserved for rare and exceptional circumstances, when the presumption in favour of compliance is rebutted by sufficiently strong reasons to the contrary.Footnote ²²² Mere disagreement is not sufficient to displace this constitutionally grounded presumption.

When political actors comply with judicial decisions on what rights require, they are not just acting on the basis of first-order reasons which they might have for agreeing with a particular ruling on rights. Nor are they motivated by the threat of political costs alone. Instead, they are acting on second-order institutional reasons for respecting the work-product of a coordinate branch as part of a collaborative enterprise which respects judicial independence and the rule of law.Footnote ²²³ By focusing on the first-order disagreements between legislatures and courts about the meaning of rights, dialogue scholars underplayed the second-order institutional norms which shape and constrain the interaction between the branches of government in a collaborative constitutional scheme.Footnote ²²⁴ The collaborative constitution brings those norms back in, placing them at the heart of the working relationships between the branches of government in a well-functioning constitutional democracy.Footnote ²²⁵

So, what would be an example of reasons sufficiently weighty to rebut the general presumption in favour of complying with rights and judicial decisions about what those rights require? A good example of a justified derogation was when the UK Parliament enacted a blanket ban on paid political advertising in the Communications Act 2003, thereby departing from a decision of the European Court of Human Rights, which held that such a ban might contravene freedom of expression.Footnote ²²⁶ When attaching a negative Statement of Incompatibility to the Bill under section 19(1) of the HRA, the Minister clarified that this was an “obviously exceptional”Footnote ²²⁷ course of action, made “after careful deliberation and a full examination of both the legal arguments and the policy alternatives.”Footnote ²²⁸ Whilst the government engaged in extensive pre-legislative scrutiny to try to devise a more circumscribed ban in line with the Strasbourg decision, it concluded that a blanket ban was the only workable solution to protecting political life from the corruption of big money.Footnote ²²⁹ In the end, both the UK Supreme Court and the ECtHR upheld the rights-compatibility of the UK Communications Act 2003, with the UK Supreme Court noting that it was “reasonable to expect that our democratically-elected politicians will be peculiarly sensitive to the measures necessary to safeguard the integrity of our democracy.”Footnote ²³⁰

Rather than adopting a disdainful or dismissive attitude towards the courts, the government and legislature departed from the initial Strasbourg ruling, whereupon the courts gave weight to Parliament’s considered, balanced, and proportionate approach when reviewing the legislation for compliance with rights. In clarifying that a negative Statement of Incompatibility should be restricted to “exceptional cases, triggering strict legislative scrutiny and a heightened burden of justification on government,” the political actors signalled the seriousness of departing from judicial decisions and emphasised their general commitment to protecting rights.Footnote ²³¹ This is a good example of collaborative constitutionalism in action, where the power to override judicial decisions was exercised with due caution and constitutional care.

6. Upshots of the Override

The collaborative explanation for the underuse of the override has broader implications for theoretical and comparative debates about the nature and dynamics of constitutional government. In this section, I will draw out some of those implications. First, if the override is understood as a safety valve to be used in exceptional circumstances, then its rare use is not necessarily evidence of constitutional failure. On the contrary, the underuse of the override may be a constitutional success story, manifesting the relative robustness of unwritten norms of constitutional behaviour which constrain and shape political behaviour in Canada and the UK. Since underuse is a feature—not a bug—in the system, then the Canadian and UK overrides may be operating as they should, lying in wait for the rare or exceptional circumstance which would warrant their use.Footnote ²³² If my boiler has worked well over the last twenty years without needing to release the safety valve, this does not mean that the safety valve is defective or defunct. It may simply mean that the boiler is working well and that the safety valve is still operating as an important fail-safe, which will kick into action if excess pressure builds up at some future point. On the collaborative understanding, rare rather than regular use of the override can be lauded as a legitimate outworking of the collaborative constitutional order, where the branches of government treat each other with comity and respect.

Second, the collaborative explanation of the underuse of the override can help us resolve the apparent paradox that a Bill of Rights which includes an override seems to simultaneously “bind and unbind ordinary legislative majorities to constitutional law.”Footnote ²³³ This has led many Canadian scholars to argue that the override is anathema to the project of entrenching rights, because it seems to cede supremacy to majoritarian whim rather than constrain majorities from encroaching on rights.Footnote ²³⁴ On this view, the Canadian Charter made a choice in favour of legal constitutionalism, but the override allows political constitutionalism to subvert that choice.

But this choice between legal and political constitutionalism is a false dichotomy which belies the more complex and collaborative mix of institutional roles under the Canadian Charter.Footnote ²³⁵ The paradox is resolved once we appreciate that rights are presumptively binding on the political actors but open to override in exceptional circumstances, in much the same way as precedent is presumptively binding on a court but open to overruling for strong reasons in suitable cases.Footnote ²³⁶ A legislative override does not set rights at nought; nor does it eviscerate the constitutional obligation on all three branches of government to take rights seriously as part of their institutional role-morality. Instead, it allows the political actors to derogate from those rights in exceptional and limited circumstances, when they have strong and weighty reasons to do so. The notwithstanding clause allows for some deviation from rights and court rulings on rights, but makes that deviation difficult to achieve.

Third, and relatedly, my analysis casts some critical light on Adrian Vermeule’s suggestion that the underuse of the override in Canada and the UK represents the “atrophy of constitutional powers,” borne out of a slippage in the modal status of the override power “from optional to prohibited” over time.Footnote ²³⁷ The assumption underpinning this argument is that the override was originally conceived as a regular mode of inter-institutional dialogue and disagreement, which then lamentably—and somewhat mysteriously—fell into desuetude or “withered on the vine.”Footnote ²³⁸

However, a flourishing and vigorous practice of legislative override was never intended in either country. Nor did it ever take place. The true modal status of the power is one of presumptive non-use, unless rebutted by strong and exceptional reasons to the contrary. Though Vermeule presents the modal status of the override in dichotomous and disjunctive terms—i.e., as either ‘optional’ or ‘prohibited’—there are a number of more complex, intermediary alternatives, such as ‘permissible but not desirable on a regular basis’ or ‘presumptively to be avoided unless absolutely necessary’.Footnote ²³⁹ The override occupies this more complex middle ground. Though legally permissible, the override should be treated with political caution and constitutional care.

Fourth, my argument takes issue with the widespread scholarly plea to ‘reinvigorate’ or ‘reincarnate’ the override as a regular mechanism for ongoing dialogue and disagreement between legislatures and courts.Footnote ²⁴⁰ I oppose such reincarnation because it would violate the norms of comity, conflict-avoidance, and collaboration on which a well-functioning constitutional democracy depends. Indeed, a political practice of regular override would not be a matter of ‘re-instating’ or ‘reviving’ anything, but rather of inculcating a novel practice of ongoing legislative disagreement with court rulings which violates core elements of the constitutional separation of powers.Footnote ²⁴¹ A liberal use of the override would be less like a ‘reincarnation’, and more like the dawning of a new constitutional day—and not a good day at that.Footnote ²⁴² Rather than usher in a new era of desirable democratic constitutionalism, the inauguration of such a practice would risk corroding the foundational constitutional norms of mutual respect, institutional “role recognition,” and political restraint which undergird the separation of powers in a well-functioning democracy.Footnote ²⁴³

Fifth, my analysis also casts doubt on the recommendation made by Jeremy Waldron, Jeffrey Goldsworthy, Stephen Gardbaum, Richard Albert, and others that we should amend section 33 of the Canadian Charter in order to resurrect and re-invent the override as a legitimate mode of ongoing political disagreement between legislatures and courts on what rights require.Footnote ²⁴⁴ For these scholars, section 33 should be reworded to reflect the reality that legislators are advocates rather than antagonists of rights.Footnote ²⁴⁵ Rather than harbouring dishonourable “rights-misgivings,” these scholars suggest that legislators are simply engaged in reasonable disagreements with courts about what those rights entail.Footnote ²⁴⁶ Therefore, section 33 should be amended to reflect that reality.

There are serious problems with this recommendation. For one thing, unpacking the political motivations for invoking the override is a complex empirical question which requires rigorous testing in different contexts at different times. A good-faith political belief in the value of rights cannot be assumed as a theoretical prior, still less as a political inevitability. Most likely, there will be mixed political motives in relation to any controversial rights-issues, and those motives will likely shift over time.Footnote ²⁴⁷ Consider Prime Minister David Cameron’s statement in the Westminster Parliament that it would make him “physically ill” to give prisoners the right to vote;Footnote ²⁴⁸ or the various political statements made about criminals, sex offenders, and immigrants in a turn towards “penal populism.”Footnote ²⁴⁹ In this context, the language of “rights-misgivings” is putting it mildly.Footnote ²⁵⁰ The fact is that section 33 does not prevent elected politicians from presenting their decision to invoke the override as rights-affirming rather than rights-encroaching. Nor does it prevent them from arguing that whilst their decision contravenes a judicial interpretation of rights, it nonetheless upholds broader political or moral values, including protecting pluralist provincial identities in a federal Canada.Footnote ²⁵¹

Finally, my analysis has implications for the argument that the Supreme Court of Canada should adjudicate whether an invocation of the override under section 33 infringes or limits rights, and then make a judicial declaration to that effect.Footnote ²⁵² Grounded in the values of educating the public about the relevant trade-offs and enhancing democratic accountability for invocation of the override, some Canadian scholars argue that judicial adjudication and eventual declaration could have “informational value where the effects of activating the notwithstanding clause are opaque to the electorate.”Footnote ²⁵³ In this framing, such declarations would be legally non-binding but democratically informative.Footnote ²⁵⁴

It is true that section 33 does not explicitly preclude judicial evaluation of the override. But it does not provide for it either. If Canadian courts developed this novel adjudicatory and declaratory jurisdiction, they would leave themselves open to the charge that they were seeking to judicialize an issue which was explicitly left to the democratically elected actors by the terms of section 33. In such a debate, concerns about American-style judicial supremacy would loom large. Comparative analysis of declarations of incompatibility under the UK HRA gives us further reason for caution about the suggestion that Canadian judges should issue advisory, educative, or ‘informational’ declarations on the override.

As noted above, although UK declarations of incompatibility are legally non-binding, they have nonetheless been treated as authoritative pronouncements on what human rights law requires, not advisory ruminations in an inter-institutional or society-wide dialogue. That is why they have garnered a near-perfect record of compliance by the UK government and Parliament. True, there are jurisdiction-specific and HRA-specific reasons for this remarkably high rate of compliance, including the likelihood of repeat litigation before the European Court of Human Rights if domestic compliance is not forthcoming.Footnote ²⁵⁵ However, there are also deeper and more generic reasons for political compliance with judicial declarations—ones which, I suggest, have significant resonance in the Canadian constitutional culture. Such reasons include the deeply-held belief that a considered and reasoned decision by the highest court in the land—culminating in a judicial declaration that legislation unjustifiably infringes rights—warrants serious respect and presumptive compliance by the legislature and the government of the day.Footnote ²⁵⁶ To dismiss such a declaration as merely advisory or informational would risk discrediting the courts and undermining the authority of judicial decisions.

Empirical analysis of the political responses to declarations of incompatibility under the HRA bears out this concern, revealing that UK parliamentarians take these declaratory rulings seriously as judicial decisions about what human rights law requires.Footnote ²⁵⁷ The political actors do not generally draw a distinction between declaratory rulings and other (authoritative) decisions of the higher courts.Footnote ²⁵⁸ Despite their legally non-binding status as a matter of strict law, the political actors have treated declarations as authoritative rulings, not advisory ruminations.Footnote ²⁵⁹ When the UK Supreme Court engages in a considered legal analysis of whether legislation violates fundamental rights—resulting in a public declaration that the UK Parliament has unjustifiably violated those rights—it is hard to present that decision as mere information or advice which politicians and the general public can take or leave as they see fit. The likelihood that such declarations will be viewed as more than mere informational input into a society-wide dialogue in Canada is further borne out by the Canadian experience with advisory opinions and constitutional references, which have engendered a consistent pattern of presumptive political compliance unless exceptional circumstances arise.Footnote ²⁶⁰

The lesson of this comparative experience is that we should beware of soft-pedalling judicial declarations as optional opinions or provisional pronouncements devoid of significant legal effect. Abstracted from the norms of inter-institutional comity in a system based on the rule of law, it is easy to claim that such declarations would simply play an educational or informational role. But in political reality, such declarations may confuse the electorate about the status of the Charter, discredit the courts for meddling in political matters, increase political polarisation on an already contentious issue, and put judges and politicians on a collision course where the courts will be in the direct firing line. Section 33 allows for a limited legislative derogation from certain rights under specified circumstances, subject to various political, parliamentary, and public constraints. A choice was made in section 33 to limit and curtail the exercise of the override through political rather than judicial means. For the most part, the political limits built into section 33 have sufficed to secure the underuse of the override in Canadian politics.

Of course, once the norm of rare use begins to erode, it is natural for lawyers to presume that the courts should provide the necessary corrective. But that route is fraught with danger, not least for the courts themselves, who will find themselves fighting a political battle they will likely lose. The solution to the increased use of the override is not, I suggest, to wheel in the judges. Instead, it is to call out abusive invocation of the override in public and political discourse, to raise its profile as an important electoral issue, and to highlight the fundamental constitutional point that rights and the rule of law are precious constitutional resources we squander at our peril.Footnote ²⁶¹ If politicians play fast and loose with the override—or use the notwithstanding clause as an ‘untrammelled opt out’ from their obligations to protect rights—they threaten to undermine the independence of the judiciary, the rule of law, and the rights guaranteed in the Canadian Charter. The urge to curb abusive invocation of the override is laudable. But expecting the courts to fix the problem may be counter-productive. Ironically, it may also play into the hands of populist leaders who wish to free themselves from the constraints of an independent judiciary and the rule of law.