Printer Friendly

The Protective Function of the Constitutional Amending Formula.

In the spring of 2014, the Supreme Court of Canada rendered two major advisory opinions: the Reference re Supreme Court Act (1) and the Reference re Senate Reform. (2) These were the Court's first decisions dealing directly with the interpretation of the amending formula for the Constitution of Canada adopted in 1982. In the first reference, the Court ruled that Parliament could not amend the provisions of the Supreme Court Act governing the eligibility requirements for appointment to the Court. In the second reference, it determined that Parliament could not enact a law providing for the holding of "consultative elections" intended to guide the Prime Minister's choice in appointing senators. The common element in these two initiatives was that Parliament claimed to be acting alone without amending the text of the Constitution. Furthermore, the legislative texts in question had been carefully drafted so as not to contradict any existing provisions of the Constitution.

The challenge to the validity of such initiatives made it urgent to address a blind spot in constitutional doctrine. Up to that point, most analyses of the amending formula had sought to determine what procedure would apply to efforts to amend a given provision of the current Constitution, or to add certain types of provision to the Constitution. (3) In other words, the amending formula has been viewed as a switching mechanism for selecting the appropriate procedure. A discussion of those questions is extremely useful, but it assumes a willingness on the part of politicians to amend the Constitution, a willingness that is rare indeed these days. (4) Few authors had analysed the constraints that the amending formula could impose on what the ordinary legislator can do, although some had examined the case of the Supreme Court (5) and studied the constraints on the power of the provinces to amend their own constitutions. (6 ) Thus, when debate began over the validity of the federal proposals for Senate reform, there was no consensus on a general theory of constitutional amendment that would have allowed the validity of this initiative to be determined without controversy. (7)

In this paper, we hope to make a contribution to the development of such a theory. We submit that the amending formula has a protective function, i.e., it protects certain rules, principles or institutions from the action of the ordinary legislator. (8) This protective function is in addition to the enabling function studied by the abovementioned authors, which permits certain legislative bodies, acting together, to amend or add provisions to the formal text of the Constitution.

We also submit that this perspective provides for a better explanation of the Supreme Court's two 2014 advisory opinions than the perspective adopted by most commentators. Due to the use of unwritten principles and the concept of "constitutional architecture," it has become commonplace to claim that the Court has "constitutionalized" certain rules or principles that were not previously included in the Constitution. Some have criticized this type of reasoning on the grounds that the Court has appropriated the power to infinitely expand the scope of political changes that would require a constitutional amendment. (9 ) This vision, which we will call the "open constitution," nevertheless raises significant conceptual problems, which can be avoided with the theory of the protective function. Moreover, a careful reading of the opinions rendered by the Court shows that these opinions are as compatible, if not more so, with the protective function theory as with the open constitution theory. Finally, we will demonstrate that certain aspects of the 2014 decisions can also be explained by an application of the ordinary principle of the supremacy of the Constitution.

The theory we intend to explain in the following pages is a doctrinal endeavour. It is important to clarify from the outset what that means. Legal doctrine is not intended merely to describe the status of the law, nor to explain how the law obeys a number of social, political or historical determinants as a sociologist or a political scientist would do. Unlike the social sciences, legal doctrine participates in the construction of its object, within the framework of the role attributed to it by the theory of the sources of law. (10) It adopts an "internal point of view" of the law. (11) It seeks to improve the law by presenting it as a coherent and morally justifiable whole. Of course, it must respect the facts as represented by statutes and court decisions, but it can criticize them or propose a general theory that transcends them. (12) This is the challenge that we intend to tackle.

Such an undertaking is all the more urgent given that the new federal government has proposed a number of changes to Canadian political institutions, and some opponents of these reforms argue that they can only be implemented by means of a constitutional amendment. Indeed, in the summer of 2016, the government was unclear as to whether it was willing to respect the tradition of regional representation in the Supreme Court. A group of Nova Scotia lawyers filed a lawsuit, arguing that any departure from that tradition constituted a constitutional amendment. The possibility of incorporating a requirement of bilingualism into the Supreme Court Act raises similar doubts. Other political initiatives, such as the reform of the electoral system, are also being challenged on the grounds that they would require an amendment to the Constitution. A logical framework is necessary to sort out all these assertions.

I. Three competing theories to explain the Supreme Court advisory opinions

For the sake of analytical clarity, it is important from the outset to describe the three theories that could explain the advisory opinions of the spring of 2014. We will call them: the theory of implicit finding of incompatibility; the open constitution theory; and the protective function theory.

In order to differentiate between them, it is useful to bear in mind the basic legal principle by which the holder of a delegated power is subject to two types of limits: it must act within the powers entrusted to it, and it must comply with the hierarchically higher norms. This principle is sometimes more clearly stated in the field of administrative law, (13) but it is equally valid for constitutional law. It is, in fact, a corollary of the principle of the rule of law. When it comes to federalism or the division of powers, the question is whether Parliament or the legislative assembly concerned has acted within the limits of its powers, while in matters of rights and freedoms the question is, rather, whether a statute is inconsistent with a hierarchically higher norm in the Canadian Charter of Rights and Freedoms. The three theories combine these two types of constraint differently.

A. The theory of implicit finding of incompatibility

The theory of implicit finding of incompatibility is simply a corollary of the principle of the supremacy of the Constitution (i.e., the necessity to respect higher-level norms). Section 52 of the Constitution Act, 1982 provides that ordinary law that is not compatible with the constitutional text may be declared inoperative. Based on this theory, the rulings on Senate reform and judicial appointments could be explained as simply the invalidation of federal statutes on the grounds that they would be inconsistent with a provision of the Constitution, even if it is not expressed this way by the Court (hence the "implicit finding").

In Canadian constitutional law, the concept of incompatibility (or conflict) has been developed primarily with respect to the paramountcy of federal statutes. Indeed, in order to determine when a provincial law must yield to a federal law, the courts have had to develop an analytical framework that clarifies this notion of incompatibility. Without going into all the intricacies of Supreme Court case law, (14) we can say that there are two scenarios where two laws will be declared incompatible: the first is where there is operational conflict, while the second involves conflict with the purpose of a federal law. In the first scenario, it is impossible to comply with both laws at the same time. (15 ) The second scenario arises when a provincial law frustrates the attainment of the purpose of a federal statute. (16) For example, such a conflict will occur when a provincial statute seeks to change the order of priority in which creditors will be paid in a bankruptcy case governed by federal law. It is also said, in such cases, that the purpose of the federal statute is to establish a "complete code," i.e., to thoroughly regulate a specific subject. (17)

A little-known ruling of the Supreme Court, Sutherland, (18) provides an example of an implicit finding of incompatibility outside the context of the paramountcy of federal laws over provincial laws. This case involved a conflict between a provincial law and a constitutional provision that protected the right of Aboriginal people to hunt on "unoccupied Crown lands." (19) The impugned Manitoba statute provided that certain wildlife areas were not "unoccupied Crown lands" for the purposes of the constitutional provision at issue. The Supreme Court ruled that the Manitoba statute was invalid because it was, in effect, intended to implicitly alter the constitutional provision at issue. According to Justice Dickson, the statute had the "effect" of amending the Constitution, even if the constitutional text remained officially unaltered. (20 ) It could also be said, using the concepts developed with respect to the doctrine of paramountcy, that Manitoba's law thwarted the attainment of the purpose of the constitutional provision, which was to guarantee certain rights to Aboriginal peoples. At the end of the day, the two statutes were found to be incompatible.

The concept of "constitutional architecture," which is often mentioned by the Supreme Court in its two advisory opinions, can be used to give effect to the principle of supremacy of the Constitution. When a statute is said to be incompatible with the constitutional architecture, this may well mean that it frustrates the achievement of the purposes of certain constitutional provisions, even if these objectives are not explicitly stated in the text. We will come back to that.

B. The open constitution theory

The open constitution theory is based on the wording of section 52(2) of the Constitution Act, 1982, which states that "the Constitution of Canada includes" a number of statutes listed in a schedule. (21) Since 1982, constitutional lawyers have debated whether this list is exhaustive or open-ended. If it is open-ended, laws that are not listed could acquire the distinctive characteristics of the formal Constitution, i.e., they would override incompatible ordinary laws (section 52(1)) and could only be amended through the special procedure for amending the Constitution (section 52(3)).

Authors who advocate recognition of the open nature of the Constitution often refer to the example of the Supreme Court Act to support their argument. The reason for this is quite simple. The amending formula refers twice to the Supreme Court (in sections 41(d) and 42(1)(d)). There is no mention of the Supreme Court elsewhere in the constitutional text, however, (22) which would make sections 4l(d) and 42(1)(d) irrelevant. This can only make sense if we conclude that certain provisions of the Supreme Court Act are implicitly part of the Constitution of Canada.

From this perspective, there is no limit to the powers of Parliament and provincial legislatures other than the requirement for compatibility with the Constitution (and, of course, the limits that flow from federalism). As such, the amending formula would not limit the powers of Parliament and the legislatures; it would merely indicate how to amend the other provisions of the constitutional text that, alone, can render an ordinary statute inoperative. It follows that if in the Supreme Court Act Reference the Court declared an Act of Parliament to be invalid, it must have been because of an inconsistency with a provision of the Constitution, which is hard to find elsewhere than in the Supreme Court Act itself. Moreover, in denying the existence of limits to the powers of legislatures other than those arising from the requirement of constitutional compatibility, this approach aligns with the oft-repeated idea that the Constitution Act, 1867 exhaustively distributed the power to enact laws between the two levels of government in Canada. (23)

This is how most commentators have read the two advisory opinions. (24) By following that line of thinking to its logical conclusion, one could go so far as to argue that any rule, principle, or institution that can be characterized as part of the "constitutional architecture" can no longer be modified in any way, without making a constitutional amendment in accordance with Part V. Some authors, no doubt aware of the extreme consequences of such a position, have instead proposed criteria for delimiting the new boundaries of the Constitution of Canada. (25)

C. The protective function theory

Unlike the open constitution theory, the protective function theory asserts that the amending formula found in Part V of the Constitution Act, 1982, not only clarifies the procedure for changing certain categories of provisions in the Constitution (the enabling function), but also excludes certain matters or areas from the jurisdiction of Parliament or provincial legislatures (the protective function). In other words, the constituent power has reserved certain questions to itself, even if the Constitution does not (or does not yet) include norms pertaining to these subjects.

A statute can therefore be declared invalid for two different reasons: it is incompatible with an existing provision of the Constitution; or it relates to a subject that the constituent power reserved for itself and has removed from the jurisdiction of the ordinary legislator. Part V of the Constitution Act, 1982 therefore has the effect of limiting the jurisdiction of the ordinary legislator, without it being necessary to prove incompatibility within the meaning of section 52 of that Act.

The simplest example of the protective function of the amending formula is found in section 42(1)(a), which provides that amendments to the "principle of proportionate representation of the provinces in the House of Commons" must be made in accordance with the procedure set out in section 38, i.e., the "7/50 formula." The representation of the various provinces in the House of Commons and its decennial adjustment is provided for in section 51 of the Constitution Act, 1867. According to section 44, this section may be amended by Parliament acting alone. Indeed, Parliament has amended it twice since 1982. (26) These amendments respected the principle of proportionate representation. They did not invade the area protected by section 42(l)(a)--that is, the choice between various principles of provincial representation (equal, proportionate to the population, proportionate to wealth, etc.). In reality, what is protected is a principle rather than the precise provision that implements that principle, as demonstrated by the two amendments made to section 51 of the Constitution Act, 1867, since 1982, and as clearly spelled out in section 52 of that same Act.

In most cases, the area covered by the enabling function of a provision of the amending formula is broader than the area covered by its protective function. Section 41(c) provides an example. It specifies that any amendment to the Constitution relating to the use of English and French (with certain exceptions) must be approved by all provinces. This section does not preclude Parliament from enacting legislation respecting the use of English and French, such as the Official Languages Act. Rather, it indicates how to go about adding new language rights to the Constitution or amending existing language rights provisions. If this section had a protective function, it would doubtless be limited to the choice of two, and only two, official languages. It is even possible that section 41(c) has no protective function and has only an enabling function. (27)

A comparison with civil law provides a better understanding of the distinction between enabling and protective functions. To a large extent, the Civil Code has an enabling function. Among other things, it sets out the rules enabling willing parties to enter into contracts. The Code also, however, includes a small number of rules of public order, i.e., rules that are binding on the parties, and that prohibit them from entering into certain types of contracts or including certain types of clauses in their contracts. (28) The protective function is, in a sense, a constitutional public order: it prohibits the ordinary legislator from adopting certain types of legislation.

In the area of the division of powers, the concept of interjurisdictional immunity could serve as an analogy to explain the protective function. As an exception to the double aspect doctrine, which allows for the coexistence of federal and provincial laws dealing with the same subject, the doctrine of interjurisdictional immunity protects the "core" competence (usually federal jurisdiction) arising out of a law of the other order of government. (29) For example, the Supreme Court ruled that Quebec legislation protecting agricultural land could not be applied to prevent the construction of an aerodrome, a matter at the heart of federal jurisdiction over aeronautics. (30) One important aspect of this doctrine is that it applies even in the absence of federal legislation dealing with the same subject matter as the provincial law, or in other words, even if the authority being protected has not been exercised. The same is true of the protective function of the amending formula, which applies even in the absence of a constitutional provision on the subject matter.

Before going any further, it is important to dispel any potential confusion. We use the term "protective function" in a specific sense, which is that the amending formula has the effect of protecting certain rules or principles that are not expressly provided for in the Constitution. In the broader sense, the amending formula obviously protects the entire Constitution against unilateral changes, but that is not the subject of this paper.

II. Assessment of the three theories

Having described the three theories that explain how the Supreme Court can rule that the adoption of an ordinary law actually requires an amendment to the Constitution, we can now assess which of these three theories constitutes the best explanation for the two 2014 advisory opinions. Drawing insight from the theories of Ronald Dworkin, (31) we will approach this question from two main perspectives: first, the compatibility of each of these theories with the facts (i.e., their capacity to explain what the Court said or did); and second, their justification (i.e., whether or not they are acceptable).

A. Compatibility with the Court's reasons

We will begin by examining whether each of the theories provides a plausible explanation for the Court's reasons in the two advisory opinions. In this assessment, we will accept the possibility that the reasons are not fully consistent themselves and that the judges who drafted them did not necessarily have these three theories in mind. We shall consider the reasoning actually employed by the Court as much as the principles it sets forth, because as the Court has already said, it is "wise to look at what the courts do as distinguished from what they say." (32)

1. The theory of implicit finding of incompatibility: a hidden but real foundation

The theory of implicit finding of incompatibility explains the result of the Senate Reform Reference, although the Court does not explicitly use this analytical framework. In fact, despite the oft-cited references to the Constitution's architecture, the Court was also careful to identify certain specific constitutional provisions that would be affected by the establishment of consultative elections: sections 24 and 32 of the Constitution Act, 1867. These provisions, the Court tells us, provide that senators shall be appointed, while other provisions provide that members of the House of Commons shall be elected. (33) After describing the basis for the selection process for members of both houses of Parliament, the Court concludes that "the proposed consultative elections would fundamentally modify the constitutional architecture we have just described." (34) In other words, even assuming that there is no operational conflict between section 24 and a federal law providing for consultative elections, there would at the very least be a conflict with the purpose of the higher norm, in this case section 24, which is to provide the Senate with a different type of political legitimacy than that of the House of Commons. (As mentioned above, the concept of the constitutional architecture is used here as a synonym for the intended purpose of certain provisions of the Constitution).

The concept of incompatibility may also have played a role in the Court's ruling on Senate abolition. There was general agreement that such action would require a formal amendment to the Constitution, but the Court had to determine whether the applicable formula was the general formula ("7/50") or that of unanimity. In deciding in favour of unanimity, the Court emphasized that the abolition of the Senate would render "inoperative" (35) the provisions of Part V that provide for Senate participation in the constitutional amendment procedure. This is, in my opinion, is another way of stating a finding of incompatibility.

However, there was no implicit finding of incompatibility in the Supreme Court Act Reference for the simple reason that there is no provision in the Constitution of Canada establishing the eligibility requirements for appointment to the Supreme Court. The Court does not reason in terms of incompatibility. The wording it uses when declaring invalid section 6.1 of the Supreme Court Act, added in 2013, focuses on the fact that it is an "amendment" to the composition of the Court. (36)

2. The open constitution theory: obvious answer or illusion?

Several passages in the Senate Reform Reference easily lend themselves to the interpretation that, for all intents and purposes, the Court has added to the constitutional text. In the introductory paragraphs of its reasoning, the Court begins by recalling its previous decisions in which it asserted that the list of constitutional texts set out in the Schedule to the Constitution Act, 1982 is not exhaustive. (37) It then introduces a concept that is central to its reasoning, that of the constitutional architecture:
As discussed, the Constitution should not be viewed as a mere
collection of discrete textual provisions. It has an architecture, a
basic structure. By extension, amendments to the Constitution are not
confined to textual changes. They include changes to the Constitution's
architecture. (38)

In many respects, this concept of architecture seems to be key to the Court's reasoning. As stated above, the Court puts consultative elections out of the reach of the ordinary legislator because they would transform the architecture of the Constitution. The Court also repeatedly refers to "the Senate's fundamental nature and role"--presumably a component of the constitutional architecture--which would be affected by the consultative elections. (39) It is tempting to conclude that the Court has added new rules or principles to the constitutional corpus that can only be amended in accordance with Part V. Nevertheless, before jumping to conclusions, it is important to note that at no time does the Court state that it has made such an addition. (40)

Even more than the Senate Reform Reference, the Supreme Court Act Reference has been viewed as a clear example of an addition to the Constitution. Is the assertion that "the Constitution Act, 1982 confirmed the constitutional protection of the essential features of the Supreme Court" (41) not an acknowledgement of such an addition? In fact, these features are nowhere to be found in the texts listed in the Schedule to the Constitution Act, 1982. The conclusion seems all the more inescapable in that the Court identifies three provisions of the Supreme Court Act, sections 4(1), 5 and 6, that "codify" (42) the composition of the Court referred to in section A1(d) of the Constitution Act, 1982. These statements led several authors to assert that these three provisions had been "constitutionalized" or had become part of the Constitution. (43)

3. The protective function theory's presence in the Court's reasons

Is it possible to argue instead that the Supreme Court attributed a protective function to Part V's amending formula? References to this protective function abound in the two advisory opinions. Indeed, the Court held that "the framers of the Constitution Act, 1982 extended the constitutional protection provided by the general amending procedure to the entire process by which Senators are 'selected,'" (44) and that these framers intended "to ensure that Quebec's representation was given special constitutional protection." (45)

The Court also associates the idea of a protective function with the intention of the framers of the Constitution Act, 1982 to "freeze the status quo" (46) in relation to certain institutions until a consensus emerged on the reforms that should be made. For example, with regard to the Senate, the Court notes that "the framers of the Constitution Act, 1982 intended to constitutionally entrench the status quo with respect to the Senate until the day when broad federal-provincial consensus could be obtained on the matter of Senate reform." (47) It also mentions certain specific types of reform that had been envisaged in the 1970s, (48) suggesting that reforms of such magnitude should be implemented through a constitutional amendment. In the Supreme Court Act Reference, this intention to protect is linked to the guarantee given to Quebec that three of the nine judges of the Supreme Court will be Quebec jurists. This is no more and no less than a right of veto for Quebec (and all the other provinces) on any amendment to this fundamental compromise.

It is worth mentioning that the idea of preserving the status quo had previously been developed by the Quebec Court of Appeal, which had also been called upon to give an advisory opinion on the question of Senate reform. The Court of Appeal stated:
The interpretation of section 42 must also take account, in particular,
that because of the inability of the federal government and the
provinces to agree in 1982 on a total reform of the Constitution,
including the Senate, amongst other institutions, the framers decided
to postpone further discussion of the matters it contains, while
specifying the applicable amending procedure to incorporate an eventual
consensus in the Constitution. (49)

The Court of Appeal drew the following conclusion, which succinctly describes the difference between the enabling and protective functions of the amending formula, although it undoubtedly exaggerates the scope of the protective function:
... section 42 prescribes not only the amendment procedure for such
matters, but recognizes that they are not within the sole jurisdiction
of Parliament. (50)

The protective function theory is also consistent with the Supreme Court's key assertion with regard to the general framework for analysis that the first step is to "determine whether the changes contemplated in the Reference amend the Constitution and, if so, which amendment procedures are applicable." (51)

Indeed, the protective function theory makes it possible to answer the first of these two questions not by broadening the concept of constitution, but by broadening that of amendment.

B. Justification for the different theories

Beyond the compatibility of each of the three theories with what the Court said and did in the two advisory opinions of spring 2014, it is also important to ask which of these three theories is more justified (or preferable). This is an essentially normative judgment. As Dworkin says, it is a question of presenting the law in its best light, i.e., as a coherent whole, compatible with the values and principles underlying the legal and political system, and possessing, as far as possible, the qualities usually associated with the law, such as intelligibility and predictability, even if absolute certainty is illusory. (52)

Presenting the law in its best light is a difficult exercise, particularly when there are disagreements about the role of law and the constitution in a democratic state. There are two possible strategies: either choose a particular vision of the role of the constitution and propose an interpretation that adheres to this vision, at the risk of it being rejected by those who refute this role; or, propose an interpretation that is compatible with a range of different visons of the role of the constitution. We have opted for this second approach. (53)

In the case at hand, it is clear that the relationship between the Constitution, the judiciary, and democracy is the subject of considerable debate. Some authors suggest that the courts have the power to review the validity of statutes passed by Parliament on the basis of general moral criteria (i.e., in the absence of a specific constitutional text against which to evaluate legislation). (54) The Court's assumption of a power of judicial review in the absence of a textual basis poses a democratic problem, however, given that unelected judges find themselves establishing the supreme norm. Justice La Forest summarized these concerns in his response to the majority opinion of the Chief Justice in Reference re Remuneration of Judges: (55)
Judicial review, therefore, is politically legitimate only insofar as
it involves the interpretation of an authoritative constitutional
instrument.... This legitimacy is imperiled, however, when courts
attempt to limit the power of legislatures without recourse to express
textual authority. (56)

The Court adopted this approach in subsequent decisions and refused to review the validity of statutes solely on the basis of the unwritten constitutional principles it had set out in the Quebec Secession Reference? (57) In particular, it ruled that the principle of the rule of law did not preclude Crown immunity, (58 ) did not prohibit the legislator from making exceptions to general rules of law" (59 ) and did not conflict with the imposition of a tax on legal services. (60) The Court's recent decision to invalidate a system of hearing fees was founded on an existing constitutional provision: section 96 of the Constitution Act, 1867. (61) This means that the Court itself, in the vast majority of cases, refuses to engage in judicial review based exclusively on unwritten principles. (62) In our analysis of the competing theories to explain the two advisory rulings of 2014, we will therefore favour a theory that avoids the exclusive use of unwritten rules.

1. The open constitution theory

Numerous authors, notably political scientists, formulated criticisms of the two advisory rulings based on the open constitution theory. (63) These criticisms can be summarized as follows.

The first target of criticism is the methodology employed by the Court; it essentially ignored the text of the Constitution and based its reasoning on abstract principles of uncertain origin. This is particularly evident in the Senate Reform Reference. The Court appears to have disregarded the meaning of the expression "method of selecting" found in section 42(1)(b), basing its reasoning instead on the concept of constitutional architecture. Critics claim, however, that this concept is unclear, making it difficult to predict what type of reform requires an amendment to the Constitution.

These methodological shortcomings would undermine the legitimacy of the Court's new jurisprudence. Indeed, recourse to the concept of constitutional architecture would be somewhat akin to judicial invention. Furthermore, the idea that non-listed texts may form part of the Constitution of Canada has always raised questions about the legitimacy of a judicial addition to the "constitutional block." (64) The Court would be arrogating to itself, without a democratic mandate deriving from the constitutional text, the power to prevent Parliament from undertaking certain types of reform. The lack of clarity in the concept of constitutional architecture would even allow the Court, according to some authors, to disguise a value judgment about the desirability (or lack thereof) of the reform that is the subject of a court challenge. (65) This would call into question the political neutrality of the Court and the Constitution.

On the practical front, these authors argue that the Court's interpretation of the amending formula would make it almost impossible to amend the Constitution even when the reform in question appears to be relatively limited in scope. Given the difficulty of implementing the amending procedure, the additional rigidity arising from the two advisory rulings would lead to constitutional paralysis in Canada.

2. The protective function theory

The protective function theory offers a better justification for the Court's advisory rulings given that, despite appearances, it allows them to be considered as the outcome of an exercise in interpreting the constitutional text that does not imply adding extrinsic elements or extending the constitutional block. This theory allows for the Court's reasoning to be seen as being based on an interpretation of the provisions of Part V of the Constitution Act, 1982, which interpretation is itself based on an analysis of the purpose of Part V derived from the discussions leading up to its adoption.

Part V of the Constitution Act, 1982 was not drafted overnight in November 1981. (66) The text of the constitutional amending formula is the result of a long process that began with the first serious discussions on the subject in the 1930s. (67) Studying these negotiations provides an understanding as to the function of the various components of Part V and an appreciation of the political consensus on certain principles, or in some cases, an understanding of how a position put forward by certain parties was accepted by the other parties to these negotiations. Without necessarily seeking to determine the actual intent of individual participants in these negotiations, this approach does provide for relatively precise indications as to the objective of certain provisions of Part V.

In particular, an historical review provides an understanding of the origin of the lists of subjects referred to in sections 41 and 42. In 1949, following the failed discussions of the 1930s, the Canadian Parliament obtained an amendment to the Constitution Act, 1867 from the British Parliament that authorized the Canadian Parliament to amend certain aspects of the Canadian Constitution without the intervention of British authorities. This provision, section 91(1), included a list of exceptions to this new jurisdiction of Parliament: matters within the jurisdiction of the provinces, denominational school fights, provisions on the use of English and French and the mandate of the House of Commons.

This addition to the Constitution was made without the consent of the provinces and despite opposition from some of them. According to these provinces, the formulation of exceptions to the federal power to amend the Constitution was too narrow and allowed for certain unilateral changes that could affect the balance of federalism or the interests of the provinces. Then Prime Minister Louis St. Laurent quickly acknowledged the validity of these criticisms and stated that a future amending formula should limit the unilateral power of Parliament more than did section 91(1). (68) All subsequent proposals for amending formulae (including the Fulton-Favreau formula and the Victoria Charter) included a list of subjects that were explicitly excluded from the scope of Parliament's unilateral power to amend certain aspects of the Constitution, and were therefore subject to a multilateral formula requiring a high degree of provincial consent. The precise content of these lists varied slightly from one proposal to another, but the principle of limiting the powers of Parliament has been the subject of political consensus since that time. In particular, this protection was a corollary of Quebec's claims regarding its right of veto over constitutional amendments. (69) It could certainly be argued that the Supreme Court took note of this consensus when it rendered its opinion in the Upper House Reference in 1979. (70) In that case, the federal government had asked the Court whether Parliament could use section 91(1) to unilaterally amend certain important features of the Senate. Although it did not answer all the questions, the Court did rule that Parliament could not change the method of selecting senators to provide for their election, as this would result in a profound change in the role of the Senate within the Canadian political system. According to the Court, section 91(1) did not apply to amendments affecting provincial interests. Given that the Senate has a regional representation role, however, the interests of the provinces are affected by any change in the Senate's essential characteristics. The Court's opinion in this case undoubtedly contributed to legitimizing the presence of lists of subjects removed from Parliament's unilateral amending power, which became sections 41 and 42.

Interpreted in light of this historical background, sections 41 and 42 of the Constitution Act, 1982 were not established primarily for an enabling purpose. (71) Rather, they were intended to restrict Parliament's unilateral authority to amend certain parts of the Constitution, initially provided for by section 91(1) of the Constitution Act, 1867, and subsequently by section 44 of the Constitution Act, 1982. It is therefore logical that sections 41 and 42 have a protective function, i.e., that they limit not only the powers deriving from section 44, but also any authority of Parliament to enact or amend ordinary laws dealing with those same subjects, pursuant to section 91 of the Constitution Act, 1867. In othet words, section 91 must now be interpreted in light of sections 41 and 42.

Such a conclusion is founded on the text and on usual methods of interpretation. (72) It does not propose an extension of the constitutional block that would be devoid of any textual hook. (73) It therefore presents a justification that is more likely to find consensus than the open constitution theory. From this perspective, the concept of "constitutional architecture" found in the Court's reasons appears in a different light. It is not a set of diffuse principles that would be added to the constitutional text and given independent normative force. The concept is more akin to what is usually called the purpose of legislation, and serves as a basis for a purposive interpretation of the constitutional text. This purpose emerges from the structure of the text, the history of its adoption and the function of the institutions it establishes. Thus, paradoxically, the constitutional architecture would be the invention of politicians and not judges.

It is true that there is little direct evidence that the "framers" of the Constitution specifically contemplated the protective function. (74) However, the Fulton-Favreau formula contained a provision explicitly providing for the full retention of the existing powers of Parliament and the provincial legislatures. (75 ) The abandonment of this provision in subsequent drafts may show that the drafters were aware that the adoption of a constitutional amending formula could result in a narrowing of the powers of Parliament and the provincial legislatures.

3. The theory of implicit finding of incompatibility

Explaining certain aspects of the 2014 references by an implicit finding of incompatibility with existing provisions of the Constitution does not raise any particular problem of justification. Indeed, to the extent that the invalidation of an ordinary law is simply based on the principle of the supremacy of the Constitution, this does not lead to any specific problems of legitimacy, beyond the issues usually associated with the interpretation of constitutional texts.

III. Delimiting the area protected by the amending formula

We therefore come to the conclusion that the protective function theory is superior to the open constitution theory given that it is more compatible with what the Court said and did in the Supreme Court Act Reference and the Senate Reform Reference and provides more convincing justification. The theory of implicit finding of incompatibility may be used to complement the protective function theory. We shall now propose a method for determining the scope of this protective function. We shall do so by drawing the logical conclusions from the protective function theory and tying them to some of the Supreme Court's statements in the two references.

The underlying principle is that the scope of the protective function should be determined by a purposive interpretation of sections 41 and 42 of the Constitution Act, 1982. Indeed, these sections reflect the will of the provinces, accepted by the federal government, to limit the power of Parliament (and that of provincial legislatures) to unilaterally change certain institutions. It is therefore logical that the wording of these sections would be the starting point for an analysis of the protective function. This was also the case in the two references: the Supreme Court based its reasoning on the wording of sections 41(d) and 42(1)(b), which deal with the composition of the Supreme Court and the method of selecting senators. It also follows that the protective function should not, in principle, be associated with matters that fall under the general procedure of section 38 but are not covered by sections 41 and 42. (76) In these areas, the concept of incompatibility should determine in most cases whether a legislative initiative is contrary to the Constitution. (77)

In some instances, the wording of sections 41 and 42 precisely describes the subject of constitutional protection. For example, the "extension of existing provinces into the territories" (section 42(1)(e)), the "principle of proportionate representation of the provinces in the House of Commons prescribed by the Constitution of Canada" (section 42(1)(b)), or "the right of a province to a number of members in the House of Commons not less than the number of Senators by which the province is entitled to be represented at the time this Part comes into force" (section 41(b)) are relatively well-defined concepts that leave only limited room for interpretation. In such cases, Parliament cannot legislate in a manner that carries out the action prohibited by sections 41 and 42, or that undermines the principle or right protected by these provisions.

In other cases, the wording of sections 41 and 42 is less specific and makes general references to certain institutions, such as the monarchy, the Senate and the Supreme Court. In such situations, the Supreme Court relied on the concept of "essential characteristics" to determine the area protected. Once again, reducing the scope of the protection afforded to the institutions concerned to what is "essential" is consistent with the intent that can be ascribed to the constituent power in the context of the negotiations leading to the adoption of the Constitution Act, 1982. The framers of the constitutional text were well aware that these institutions were already governed by a combination of constitutional provisions and ordinary statutes, or in the case of the Supreme Court, almost exclusively by ordinary statutes. (78) Knowing from experience that it is not easy to agree on a constitutional amendment, they would not have sought to prevent the normal evolution of protected institutions, but rather any fundamental changes affecting the essential characteristics of those institutions.

History can help determine what these essential characteristics are. First, we can examine the negotiations that led to the patriation of the Constitution. Although the negotiations were not successful with regard to the Senate and the Supreme Court, these institutions were nevertheless the subject of substantial discussions. The main topics of discussion provide a good idea of what politicians of the time considered to be essential characteristics. This is precisely the approach taken by the Supreme Court in the reasons for its decisions. In the Supreme Court Act Reference, the Court notes that the concept of the "composition" of the Court should be primarily understood as the reflection of the provinces' desire to guarantee Quebec's representation within the institution. In the Senate Reform Reference, the Court refers, albeit briefly, to the constitutional discussions of the 1970s and the various proposals regarding the Senate. It implicitly concludes that whether members of the Senate were appointed or elected was an essential characteristic of the Senate. (79) The Court does not limit itself to the travaux preparatories of the Constitution Act, 1982, however. It also examines the history of the establishment of the institutions in question--beginning in 1867 for the Senate, and in 1875 for the Supreme Court. The Court deduces from its inquiry that politicians had specifically intended, in the first case, that the Senate not be an elected chamber, and in the second, that Quebec's representation in the Supreme Court be ensured in order to guarantee the vitality of the civil law tradition. In this regard, the Court uses the concept of historical compromise to stress the importance of the choices made by the framers of the Constitution. What this means is that two historical wills combine to define what is protected by the amending formula: that of the politicians who shaped the institution in question and that of the politicians who agreed on the amending formula.

It is possible that certain categories of sections 41 and 42 have only a limited protective function. Section 41(c) deals with amendments to the Constitution relating to "the use of the English or French language." Most existing constitutional provisions on this subject confer rights. We cannot assume a desire on the part of the constituent power to prohibit the ordinary legislator from guaranteeing additional rights beyond those provided for in the Constitution. (80 ) Indeed, the Official Languages Act (81) and the language laws of several provinces confer rights that go beyond those entrenched in the Constitution. Section 41(c) therefore has no protective function with respect to language rights. (Obviously, a law that purported to reduce rights guaranteed by the Constitution would be inoperative due to incompatibility with the Constitution, but not because of the protective function.) The only protective function that could be envisaged in section 41(c) would involve the choice of official languages at the federal level: this provision could be interpreted as referring to the choice of two, and only two, official languages, such that Parliament could not add a third. (82 ) In fact, section 41(c) is primarily intended to subject certain provisions of the Constitution to the rule of unanimity rather than the general formula, so as to grant a right of veto to Quebec on any amendment to the principle of bilingualism. This analysis of section 41(c) reveals an interesting aspect of the protective function; it focuses on institutions rather than on rights.

What are the effects of the protective function? Concluding that an ordinary statute deals with a subject covered by the protective function does not prohibit amendments to that statute. The protective function does not apply to legislative texts, but rather to the essential characteristics of the institutions concerned. Let us take the example of section 42(l)(a), the principle of proportionate representation of the provinces in the House of Commons. This representation is provided for in section 51 of the Constitution Act, 1867, which the federal Parliament has amended twice since 1982. (83) This therefore means that the wording of ordinary legislation is not protected from amendment. The constraint on the legislator lies with the requirement to safeguard the protected principle when legislating. Thus, to the extent that amendments to section 51 maintain proportionate representation, they are valid. Another way of looking at the relationship between the protective function and ordinary legislation is to say that the latter does not suddenly become part of the "Constitution of Canada" as defined in section 52 of the Constitution Act, 1982, which is paramount over other statues and can only be amended under Part V. The Supreme Court's reasons demonstrate this. The Court held that section 5.1 of the Supreme Court Act, added by Parliament in 2013, was valid because it did not change the substance of the eligibility requirements that had existed up to that time. This would have been impossible had the provisions of the Supreme Court Act in question been part of the Constitution strictly speaking, just as Parliament cannot add sections to the Canadian Charter of Rights and Freedoms even if it asserts that the addition does not change its substance.

It is not always possible to precisely determine in advance what types of legislative changes are prohibited by the protective function of the amending formula. The concept of subterfuge or circumvention is likely to play an important role in this respect. Indeed, in the situations that led to the two advisory opinions of 2014, Parliament had obviously attempted to circumvent the constitutional amending procedure, or at the very least, to make maximum use of the flexibility it believed it had. Some authors even suggested that such processes were legitimate, provided that the constitutional text was not formally amended and the discretionary power granted to certain political actors was officially preserved. (84) Obviously, the question of what constitutes a subterfuge can hardly be the subject of abstract definitions.

At the end of this overview, a final question should be asked: Can the protective function be applied to institutions or principles that are not listed in sections 41 and 42? Let us recall here one of the reasons the protective function theory is more attractive than its rival, the open constitution theory: it is based on the constitutional text and does not purport to add to it. The reference to specific subjects highlights the special concern of the framers of the Constitution with regard to those subjects, and justifies the interpretation of these sections as conferring a protective function. In contrast, section 38 does not target any particular subject. It is much more difficult to deduce a protective function for this section and to determine its boundaries. Ultimately, associating a protective function with section 38 without a precise textual basis is tantamount to defining subjects or matters that can no longer be touched by ordinary legislators. In that case, the protective function theory would be confused with that of the open constitution. It is therefore preferable to limit the protective function to the subjects referred to in sections 41 and 42.

It may well be said that such a vision is too narrow, and that it is incompatible with the elements of the Senate Reform Reference dealing with amendments to senators' term of office. On that front, the Supreme Court held that the length of term was outside the jurisdiction of Parliament acting alone, even if this characteristic of the Senate is not mentioned in section 42, because it nevertheless constitutes an essential characteristic that "engages the interests of the provinces." (8) '' It is true that these passages in the Court's reasons suggest that the concept of the interests of the provinces is more decisive than a basis in the text of sections 41 and 42. (86) The fact remains that length of term may be intimately related to the explicitly mentioned characteristics, as pointed out by the Quebec Court of Appeal. According to that Court, the different characteristics of the Senate are interrelated, and "an amendment to the duration of that mandate could affect both the powers of the Senate and the method of selecting senators." (87)

IV. Application to specific cases

We shall now consider the application of the protective function theory to certain subjects that have recently attracted attention, namely the Supreme Court, the Senate, the judiciary and the electoral system. Other issues could also be considered, such as the monarchy and the constitutional amending formula itself, but they will be left for a subsequent study.

A. The Supreme Court

For at least 40 years now, the Supreme Court of Canada has been the subject of various reform proposals. (88) These proposals have dealt with subjects as diverse as the method of appointment, a requirement for judges to be bilingual, a gender balance requirement, the jurisdiction of the Court with respect to appeals from Quebec, and so on. Over the past dozen years, successive governments have also implemented administrative policies governing the process for appointing judges to the Court. Although provisions relating to the Court were included in several agreements to amend the Constitution (the Victoria Charter, the Meech Lake Accord, the Charlottetown Accord), politicians have always appeared to assume that Parliament possessed considerable flexibility for amending the Supreme Court Act. For example, when he was a member of Parliament, Yvon Godin introduced several bills to require judges of the Court to be bilingual, one of which was passed by the House of Commons. (89) During debates on this bill, there was never a serious suggestion that Parliament lacked the authority to adopt it. Has the Supreme Court Act Reference changed the situation? (90) Does the protective function associated with sections A1(d) and 42(1)(d) prevent Parliament from amending significant portions of the Supreme Court Act?

It is not easy to give an answer that is both compatible with every statement of the Court and satisfactorily justified by the protective function theory. The Court's reasoning, it must be said, includes some shortcuts that should be addressed if we are to arrive at conclusions compatible with the scheme of the constitutional text. We therefore propose an interpretation that reconciles as closely as possible the various constraints arising from the constitutional text and from the authority attaching to the Court's ruling.

As discussed above, the starting point for the analysis should be the recognition that, pursuant to section 101 of the Constitution Act, 1867, Parliament retains jurisdiction to legislate with regard to the Supreme Court. (91) In other words, and contrary to what has often been argued, the Supreme Court Act is not, in the formal sense, part of the "Constitution of Canada." However, Parliament's jurisdiction must now be exercised in a way that does not affect the areas reserved for the action of the constituent power. The Supreme Court explains this in the reference: "Parliament undoubtedly has the authority under s. 101 to enact routine amendments necessary for the continued maintenance of the Supreme Court, but only if those amendments do not change the constitutionally protected features of the Court." (92) Indeed, Parliament has made several amendments to the Supreme Court Act since 1982.

Thus, in order to determine whether a proposed amendment to the Supreme Court Act is within Parliament's authority, it is first necessary to determine whether the amendment relates to the "composition of the Court" and second, if it affects another essential characteristic of the Court. If not, the proposed amendment can validly be passed by Parliament without following the procedures set out in Part V of the Constitution Act, 1982. With a view to further clarifying these issues, we undertake a review of the Court's ruling in light of the constitutional text itself, the history of the discussions leading to its adoption and the essential characteristics of the Court.

1. The composition of the Court

The question of the composition of the Court was at the heart of the Supreme Court Act Reference. The Court's reasoning relied heavily on the history of the negotiations immediately preceding passage of the Constitution Act, 1982. The Court notes that Part V is a direct result of an agreement reached by eight provinces in April 1981. In that agreement, the explanatory note accompanying what became section 41(d) explicitly stated that this provision was intended to protect Quebec's representation on the Court. Without explaining the logic underlying the assertion, the Court equated the composition of the Court protected by section 41(d), the eligibility requirements for becoming a judge, and the precise provisions of its constituent Act:
The notion of "composition" refers to ss. 4(1), 5 and 6 of the Supreme
Court Act, which codify the composition of and eligibility requirements
for appointment to the Supreme Court of Canada as they existed in 1982.
By implication, s. 41(d) also protects the continued existence of the
Court, since abolition would altogether remove the Court's composition.

As Justice Moldaver also pointed out in his dissenting reasons, (94) the guarantee of three seats for judges from Quebec does not mean that the specific eligibility requirements under the current Act must be maintained as is. If we take a step back, we can see that this guarantee can be assured in two ways: by imposing special eligibility requirements to determine who can be considered a judge from Quebec; or by giving a representative political body in Quebec the power to appoint these three judges. In both cases, the methods employed have the same objective: to ensure authentic representation for Quebec. This means that the eligibility requirements affect an essential characteristic of the Court only insofar as they seek to ensure that authenticity. It is true that the Court completely avoids this question in rendering its opinion. It is easy to understand why: tackling it head-on would have required the Court to propose its own definition of "true Quebec jurist," whereas a very broad consensus already existed in Quebec's political circles and media that a Federal Court judge was not eligible to be appointed to one of the three Quebec seats on the Supreme Court. (95) By drawing parallels between composition and eligibility requirements, the Court avoided this trap.

In our view, this parallel between composition and eligibility requirements should be confined to the particular case before the Court; here, the definition of the eligibility requirements intended to give effect to the guarantee of Quebec representation, assuming that this guarantee is given effect by means of eligibility requirements rather than according the power of appointment to a Quebec political body. (96) The Court's remarks cited above should not be read as a statute that generally applies to any situation falling within the scope of the description contained therein, but as a jurisprudential statement that cannot be dissociated from the context of the case. Such a restriction would be consistent with the text of section 41(d), with history and with a structural analysis.

First, the expression "composition of the Court" must be contrasted with other expressions used in sections 41 and 42 of the Constitution Act, 1982, such as the "method of selecting Senators" and the "residence qualifications of senators." This choice of words reflects the intent of the framers of the Constitution to attribute a very specific meaning to the expression "composition," which does not include all the characteristics of the Court associated with the selection of judges, and only includes eligibility requirements to the extent that they have a significant impact on the "composition."

Second, there is no indication that the federal and provincial governments sought to protect anything other than Quebec's representation on the Supreme Court. It should be borne in mind that during the negotiations between 1978 and 1981, governments intended to incorporate a chapter on the Supreme Court into the Constitution. Due to the civil law/common law duality, the proposal that received majority support was to enshrine a Supreme Court composed of eleven judges, including five civil-law judges from Quebec. (97) Before that time, the Supreme Court had never been mentioned in drafts of the amending formulae that would become sections 41 and 42. When in April 1981, eight provinces (including Quebec) proposed to proceed with patriation and the adoption of an amending formula, while postponing changes to institutions such as the Senate and the Supreme Court, Quebec suddenly saw the opportunity to increase its quota on the Court disappear. This probably explains the appearance of references to the Supreme Court in the lists of matters subject to the general amending formula or unanimity. (98) As to the requirement for unanimous support for changes in the composition of the Court, the most plausible explanation is that it was intended to give a veto to Quebec on any change to its relative weight on the Court. We are unaware of any other justification that could explain the special treatment accorded to the composition of the Court in section 41. This is the conclusion reached by the Court in the reference:
The intention of the provision [section 41(d)] was demonstrably to make
it difficult to change the composition of the Court, and to ensure that
Quebec's representation was given special constitutional protection."

Third, from the standpoint of justifications associated with the political structure, or the "architecture" of the Constitution, it is difficult to discern justifications for a broad interpretation of the composition of the Court referred to in section 41. In its reasons, the Court suggests that the guarantee of its composition protects it against an attempt at abolition, which it states, "would altogether remove the Court's composition." (100) It is also conceivable that the guarantee relative to the composition of the Court could help to prevent serious attacks on its independence. For example, by enshrining the number of judges on the Court, section 41(d) prevents Parliament from implementing a "court-packing plan" similar to that envisaged by US President Roosevelt in the 1930s. None of this is related to the eligibility requirements, except perhaps the principle that the Court should consist exclusively of jurists. (101)

It follows that, notwithstanding what may be appear from a cursory reading of the Supreme Court Act Reference, not all aspects of the composition of the Court are covered by the protective function of the constitutional amending formula. The only aspects that are protected are those that are essential characteristics of the Court and that have implications for the guarantee given to Quebec concerning its representation within the institution. The eligibility requirements may be protected, but only to the extent that they contribute to guaranteeing the representation of Quebec. Subject to this reservation, Parliament may amend sections 4, 5, and 6 of the Supreme Court Act, in particular to create a requirement for bilingualism (102) or gender parity or a geographical distribution of seats other than those of Quebec.

2. Other protected characteristics of the Court

Beyond the composition of the Court, section 42(1)(d) of the Constitution Act, 1982, provides that amendments to the Constitution relating to the Supreme Court of Canada must be made using the general procedure in section 38. In its advisory opinion, the Court provided the following explanations:
Section 42(l)(d) applies the 7/50 amending procedure to the essential
features of the Court, rather than to all of the provisions of the
Supreme Court Act. The express mention of the Supreme Court of Canada
in s. 42(1)(d) is intended to ensure the proper functioning of the
Supreme Court. This requires the constitutional protection of the
essential features of the Court, understood in light of the role that
it had come to play in the Canadian constitutional structure by the
time of patriation. These essential features include, at the very
least, the Court's jurisdiction as the final general court of appeal
for Canada, including in matters of constitutional intetpretation, and
its independence. (103)

Do the text, the negotiating history, and the analysis of the political structure give indications as to the scope of this protection?

The text of section 42(1)(d) provides no guidance as to the scope of the protective function that flows from it. Indeed, as we pointed out earlier, the protective function is necessarily narrower than the enabling function of the amending formula. It follows that only a subset of the rules governing the Court is protected against unilateral amendment. Do the discussions that led to patriation of the Constitution make it possible to define this subset? Unfortunately, these discussions do not provide information as precise as that relating to section 4l(d). (104) In fact, the explanatory notes to the April Accord of 1981, upon which the Court relied heavily in its interpretation of section 4\{d), recall that the Court is established by an Act of Parliament and not by the Constitution itself. They go on to state that the provision that became section 42(1)(d) "anticipates constitutional amendments relating to the Court." (105) This tends to reinforce the arguments put forward by the Attorney General of Canada in the Supreme Court Act Reference that this provision has only an enabling function and not a protective function (the "empty vessel theory").

In deciding otherwise, the Court relied on the need to protect its essential characteristics against unilateral amendments. In order to better understand these characteristics, it is possible to draw some indications from the historical evolution of the role of the Court, to which it refers in its advisory opinion. In a federal system, one of the crucial roles of the Court is to arbitrate the division of powers. (106) Moreover, since 1982, the Court has also played the role of guardian of the rights and freedoms guaranteed by the Canadian Charter of Rights and Freedoms. (107) The role of the Court, however, is not limited to constitutional questions; its role in public and provincial law is equally essential "to the development of a unified and coherent Canadian legal system." (108) Indeed, in the absence of any clues arising from the text or constitutional negotiations, it is undoubtedly the structural analysis that will determine the scope of the protective function with respect to the characteristics of the Court other than its composition. (109)

It is not possible in the context of this article to consider all imaginable reforms involving the Court's jurisdiction. Nevertheless, Parliament remains competent to amend this jurisdiction, provided that it does change the essence of the Court's role, as described above. For example, the Supreme Court Act was amended in 1991 and 1996 to eliminate certain appeals as of right in criminal matters, and these changes are undoubtedly valid. However, one might wonder about an eventual total abolition of the Court's jurisdiction with respect to the application of provincial laws or other similarly significant amendments.

Any amendments to the Supreme Court Act should also safeguard the independence of the Court. In any case, this independence was certainly already protected as a result of the Reference re Remuneration of Judges. (110) The Court's case law on the independence of the judiciary may serve as a guide in this respect.

What about the appointment process? Unlike section 42(1)(b), which explicitly deals with the method of selecting senators, section 42(1)(d) does not mention the method of selecting judges. This is an important textual difference. Moreover, in its advisory ruling, the Court failed to include the appointment process among protected characteristics, although the reform of this process has been the subject of numerous proposals in recent years. Does this mean that there is no constraint on Parliament's action in this area? We believe that the response is somewhat more nuanced. Parliament could not fundamentally alter the way in which Supreme Court judges are selected--currently appointment by the executive--by replacing it, for example, with popular elections. That would certainly affect an essential characteristic of the Court. However, there is nothing to prevent Parliament from legislating a framework for the process of appointments by the executive, even if that circumscribes the discretion of political actors. (111) The same goes for the appointment of senators, which we will now discuss.

B. The appointment of senators

In the wake of the Reference re Senate Reform, it is now clear that Parliament cannot unilaterally implement a mechanism that transforms the Senate into an elective chamber. The new federal government's focus has shifted to less ambitious targets that do not require a constitutional amendment. Thus, the government has put in place a selection process that allows any citizen to apply. (112 ) An independent committee reviews candidacies and proposes a short list to the Prime Minister, who then makes the selection. This process is not legislated, however. To date, it is only an administrative policy that the government could modify or abandon at any time.

No doubt, certain passages in the Reference re Senate Reform discouraged the government from proposing legislation to implement its reform. The Court drew attention to the broad meaning of the term "method of selecting senators." This does not only cover the formal act of appointment. On the contrary, the Court tells us, "By employing this language, the framers of the Constitution Act, 1982 extended the constitutional protection provided by the general amending procedure to the entire process by which Senators are 'selected.'" (113) Does this mean that Parliament has no competence to legislate on this subject? That would lead to an absurd consequence whereby Parliament would be incapable of doing what the government can do by means of a simple administrative policy. (114)

The theory of the protective function that we have outlined above leads us to qualify the Court's assertions. It is true that the enabling function of section 42(1)(b) covers the entire selection process. However, the protective function is narrower in scope, and as the Court points out elsewhere, refers only to the essential characteristics of the Senate. What, then, are these essential characteristics, having regard to the selection process? The Court's ruling draws our attention to certain clues arising from the discussions surrounding Confederation: members of the Senate were to be appointed, not elected. Other indicators are provided by documents from the time of the adoption of the Constitution Act, 1982, which outline the main options for the selection of senators: appointment, election, or appointment as agents of provincial legislatures. (115) It seems to us that the protective function is aimed at the fundamental choices involving the method of selecting senators that have an impact on the source of their political legitimacy. The political importance of the selection method stems from the fact that it largely determines the type of legitimacy that the selected person can claim, and consequently, the political role of the institution of which he or she is a member. As such, an appointed person can possess legitimacy based on expertise and independence, an elected person enjoys democratic legitimacy, and a person designated by a provincial legislature acts as an agent of his or her province.

However, the protective function does not affect the detailed implementation of any of these three options. In other words, the choice between appointment, election or designation by the provinces is a matter for the constituent power, but Parliament still has jurisdiction over the specific means of giving effect to that choice. For example, if the Constitution were amended to provide for senators to be elected from now on, Parliament could govern the election process, the voting system, the role and funding of political parties, etc. The only constraint at this level of detail is that which arises from the requirement that ordinary legislation must be compatible with the Constitution. Thus, if the Constitution were amended to provide not only for senators to be elected, but to be elected by proportional representation, Parliament could not, by ordinary legislation, adopt another method of voting.

What are the implications of the above for the current system? In 1867, the constituent power indicated that senators would be appointed and not elected. The constituent power of 1982 indicated that this fundamental choice could not be set aside without changing the Constitution. However, there is nothing to prevent Parliament from legislating on the method of selecting senators, provided that it does not alter this fundamental choice and that it legislates in a manner consistent with the existing provisions of the Constitution. The only relevant provision is section 24 of the Constitution Act, 1867, which provides that senators will be appointed by the Governor General. This provision has a formal aspect (the appointment is signed by the Governor General) and a substantive aspect (the power of appointment rests with the federal executive). As long as these two aspects are preserved, an Act of Parliament can establish the process that the federal executive must follow in order to appoint a senator.

Some argue that an Act of Parliament that provides for such a process must nevertheless preserve the discretion of the ultimate decision-maker (the Governor General, or in practice, the Prime Minister) to choose as he or she sees fit. In other words, the process put in place by Parliament should remain purely advisory. Yet, once the limited scope of the protective function is conceded, there seems to be no reason why the selection process should not constrain the choice of the Prime Minister, for example, by requiring the Prime Minister to select senators from a short list prepared by an independent committee. Such a requirement would not be inconsistent with section 24. Formally, senators would still be appointed by the Governor General, and in substance they would still be appointed by the federal executive. In fact, the Canadian political system does not maintain a strict separation between the legislature and the executive. It is recognized that legislatures may, by statute, grant or withdraw powers of the executive or adjust the decision-making process followed by the executive. This explains why the Supreme Court has already recognized that Parliament can restrict the exercise of a power that the Constitution grants to the executive, or even exercise that power itself. (116)

In fact, it seems that the idea that Parliament can legislate, provided that it does not in any way restrict the exercise of a discretionary power conferred by the Constitution, was the corollary of the theories of "constitutional workaround." These theories were intended to enable Parliament to make major changes to political institutions, provided that it found a way to appear to maintain the discretionary power provided for in the constitutional text. (117 ) These theories focus on form rather than substance and are incompatible with the Court's 2014 advisory rulings. A rule derived from these theories should therefore not be applied.

C. The judiciary

It has occasionally been suggested that sections 96 to 100 of the Constitution Act, 1867, impose limits on Parliament's power to legislate on the eligibility requirements or the process for appointment to the federal judiciary. For example, one might question the validity of section 3 of the Judges Act, which provides that persons appointed to the federal judiciary must have been members of a bar for at least ten years, given that sections 97 and 98 of the Constitution Act, 1867, do not impose any time requirement. (118)

The approach to addressing this issue should begin with an examination of the scope of the protective function. Apart from the Supreme Court, the judiciary is not mentioned in sections 41 and 42 of the Constitution Act, 1982. The provisions of the Constitution Act, 1867, concerning the judiciary may be amended by following the ordinary procedure provided for in section 38 of the Constitution Act, 1982. As noted above, the protective function should be attached to the matters referred to in sections 41 and 42, but not to those falling under section 38 independently of those two provisions. It follows that the federal judiciary would not be subject to the protective function. Some might be alarmed by this conclusion, for fear that an institution essential to preserving democracy and the rule of law is being left to the mercy of Parliament. This view overlooks the fact that the protective function is complemented by the theory of implicit finding of incompatibility. In other words, while Parliament may legislate regarding the judiciary, it cannot do so in a manner incompatible with the provisions of the Constitution, interpreted in light of their purpose and the essential features of the institutions they establish.

This means that the validity of a provision such as section 3 of the Judges Act must be judged on the basis of its compatibility with the constitutional text. This incompatibility could stem from an operational conflict, i.e., the impossibility of respecting both texts at the same time. In the case of section 3, there would be no conflict since it is possible to comply with this statutory provision and sections 97 and 98 by appointing only persons with at least ten years of membership in the bar. Inconsistency may also arise from a conflict with the purpose of the hierarchically superior text. As pointed out above, this notion of conflict with the purpose can be synonymous with an amendment to the architecture of the Constitution. Here, the courts have identified several objectives underlying sections 96 to 100, including ensuring the independence of the judiciary, (119) the creation of a unitary judicial system modelled after the British courts, (120) and more recently, the guarantee of access to the courts. (121) It could also be said that these elements are part of the constitutional architecture with respect to the judiciary. Without going into details, it is nevertheless difficult to see how section 3 would prevent the achievement of any of these objectives.

More generally, it can be argued that the Constitution does not prevent Parliament from legislating with respect to the process of appointing federal judges, for the same reasons that it can legislate with respect to the appointment of senators. The constraints, if any, would arise from the requirement for compatibility in its two prongs. For example, if Parliament were to provide that judges be elected, this would probably be inconsistent with section 96, which, like section 24 regarding the Senate, provides that judges will be appointed and not elected.

D. The electoral system

Doubts have recently been raised about the constitutionality of a possible reform of the electoral system used to elect members of the House of Commons. Some have suggested that because of its political importance, such a reform would affect the architecture of the Constitution and could be implemented only through a constitutional amendment. Is the electoral system covered by the protective function?

Again, the analysis should begin with sections 41 and 42. Two features of the House of Commons are specifically mentioned: proportionate representation of the provinces (section 42(1)(a)) and the "Senate floor" (section 41(b)), i.e., the guarantee of minimum representation for small provinces. As stated above, it is clear that these provisions have a protective function and that Parliament is not competent to legislate without regard to these principles.

Does the protective function go further? We can compare the provisions regarding the House of Commons with those pertaining to the Senate. Sections 41(b) and 42(1)(a) can be read in conjunction with section 42(1)(c), which refers to the number of seats for each province in the Senate. However, the Senate is also mentioned in section 42(1)(b), which is much broader in scope. This is not surprising. In the discussions leading up to Confederation, the Senate occupied a large place because of the role of regional representation that it was intended to be given, in order to counterbalance a House of Commons governed by the principle of representation proportionate to the population. Similarly, the Senate featured prominently in constitutional discussions that have taken place since the 1960s, in the hopes of finding a way for the institution to better fulfill the role originally assigned to it. The House of Commons, on the other hand, does not play such a role of regional representation. For that reason, it does not appear that the House of Commons is at the heart of provincial interests. This has led the Supreme Court, on two occasions, to assert that Parliament cannot unilaterally alter the essential characteristics of the Senate.

It is therefore difficult to see why the constitutional amending formula would have a protective function with respect to the electoral system, beyond the principle of proportionate representation of the provinces and the guarantee of representation for the smaller provinces. It should be stressed, however, that these principles can impose significant constraints. Provincial representation implies that each MP represents a province, even if he or she does not represent a specific constituency. To ensure this representation, it is logical to require that MPs representing a province be chosen by the electors of that province. This requirement can be compared to the requirement that judges of the Supreme Court who represent Quebec must meet certain eligibility requirements to ensure that they can truly be considered Quebeckers. It follows that a proportional electoral system should be applied on the basis of lists established for each province, to ensure that MPs are from the province they represent and are chosen on the basis of the proportion of votes won by the various parties in the province in question, and not across the country.

It should also be noted that the reasoning that led the Supreme Court to rule that the term of office of senators could not be changed unilaterally by Parliament cannot be transposed to the question of the electoral system. Indeed, the term of office of senators was expressly provided for by a constitutional provision, section 29 of the Constitution Act, 1867. Consequently, any amendment had to follow the amending formula in Part V of the Constitution Act, 1982, and the Supreme Court ruled that section 44 did not allow for amending an essential characteristic of the Senate, whether or not it is mentioned in section 42. A change to the electoral system would be made under the general powers of Parliament flowing from sections 41 and 91 of the Constitution Act, 1867, and not Part V of the Constitution Act, 1982.

We have shown that the theory of the protective function of the amending formula of the Constitution of Canada is the best explanation for the reasoning used by the Supreme Court in the Senate Reform Reference and the Supreme Court Act Reference. In its reasons, the Court explicitly mentions this protective function. Moreover, since it avoids the constitutionalization of rules or principles outside the constitutional text, it preserves the legitimacy of constitutional judicial review against frequent objections. In reality, the protective function flows from the provinces' desire to protect the essential characteristics of certain institutions from any federal attempt at unilateral change. Sections 41 and 42 of the Constitution Act, 1982, which reflect the will of the provinces, provide a textual basis for the protective function. The protective function can be complemented by the principle of the supremacy of the Constitution, which allows for ordinary statutes that are not compatible with the constitutional provisions to be declared inoperative, either directly or because they thwart the attainment of the Constitution's purposes. That is what we have called the theory of implicit finding of incompatibility.

We have suggested a method for determining the scope of the protective function and have given examples of the results of its application to the Supreme Court, the Senate, the federal judiciary and the reform of the electoral system. Our findings show that the constraints imposed by the Constitution on several types of reform of these institutions are less onerous than several authors claimed following the two advisory rulings rendered by the Supreme Court in 2014. Our conclusions will certainly not achieve unanimity, particularly because they lead to qualifying certain statements of the Supreme Court, at least if these are taken literally.

It would be helpful for the Supreme Court to revisit these issues in order to avoid foreclosing reforms that might be desirable, and to refocus debate on their merits rather than their constitutional validity. The current government has announced its intention to move forward on several of these issues. In view of the anticipated legal challenges, it would no doubt be wise to refer certain questions to the Supreme Court, allowing it to rule on Parliament's authority to carry out specific reforms, and more generally, to continue the construction of a general theory of constitutional amendment.

Sebastien Grammond (*)

(*) D.Phil. (Oxon.), MSRC, Ad.E. Professor, Civil Law Section, University of Ottawa; Legal counsel, Dentons Canada. The author wishes to thank Yan Campagnolo, Warren Newman, Peter Oliver, and Mark Walters, who kindly commented on a previous version of this article, as well as anonymous reviewers. As a lawyer, the author represented certain interveners before the Quebec Court of Appeal and the Supreme Court of Canada in the Reference re Senate Reform and the Reference re Supreme Court Act. This is an English translation of a paper that is published in the Revue generate de droit: (2017)47:1 RGD 119.

(1) Reference re Supreme Court Act, ss 5 and 6, 2014 SCC 21, [2014] 1 SCR 433 [Reference re Supreme Court].

(2) Reference re Senate Reform, 2014 SCC 32, [2014] 1 SCR 704 [Reference re Senate].

(3) See in particular the monograph of Benoit Pelletier, La modification constitutionnelle au Canada (Toronto: Carswell, 1996), or the chapters on constitutional amendment in constitutional law textbooks, such as Henri Brun, Guy Tremblay & Eugenie Brouillet, Droit constitutionnel, 6th ed (Cowansville: Yvon Blais, 2014) at 214-256; Jacques-Yvan Morin & Jose Woehrling, Les constitutions du Canada et du Quebec: du regime francais a nos jours (Montreal: Themis, 1992), at 487-539; Peter W Hogg, Constitutional Law of Canada (Toronto: Thomson Carswell, 2007) (loose-leaf revision 2016 -Rel. 1), ch 4 [Hogg, Constitutional Law].

(4) See Richard Albert, "The Difficulty of Constitutional Amendment in Canada," (2015) 53:1 Alta L Rev 85.

(5) Some authors claimed that certain provisions of the Supreme Court Act had been constitutionalized: Patrick J Monahan & Byron Shaw, Constitutional Law, 4th ed (Toronto: Irwin Law, 2013) at 205. Others maintained that the reference to the Supreme Court in the amending formula was only intended for future additions to the Constitution regarding the Supreme Court: Pelletier, supra note 3 at 74, 214, 280; Morin & Woehrling, supra note 3 at 483; Hogg, supra note 3 at ch 4-14, 4-15. A detailed summary of this debate can be found in Warren J Newman, "The Constitutional Status of the Supreme Court of Canada" (2009) 47 SCLR (2d) 429.

(6) See e.g. Morin & Woehrling, supra note 3, at 488-501.

(7) Two opposing contributions illustrate the range of opinions on this subject: Robert E Hawkins, "Constitutional Workarounds: Senate Reform and Other Examples" (2010) 89:3 Can Bat Rev 513; Matk D Walters, "The Constitutional Form and Reform of the Senate: Thoughts on the Constitutionality of Bill C-7" (2013) 7 JPPL 37. See also Jennifer Smith, ed, The Democratic Dilemma: Reforming the Canadian Senate (Montreal and Kingston: McGill-Queen's University Press, 2009).

(8) For examples of authors who have considered this question, see Peter C Oliver, "Quebec and the Amending Formula: Protection, Promotion and Federalism" in Stephen Tierney, ed, Accommodating Cultural Diversity: Contemporary Issues in Theory and Practice (Aldershot: Ashgate, 2007) 167; Warren J Newman, "Living with the Amending Procedures: Prospects for Future Constitutional Reform in Canada" (2007) 37 SCLR (2d) 383 at 386.

(9) See e.g. Dennis Baker & Mark D Jarvis, "The End of Informal Constitutional Change in Canada?" in Emmett Macfarlane, ed, Constitutional Amendment in Canada (Toronto: University of Toronto Press, 2016) 185; Peter W Hogg, "Senate Reform and the Constitution" (2015) 68 SCLR (2d) 591.

(10) John EC Brierley & Roderick A Macdonald, eds, Quebec Civil Law: An Introduction to Quebec Private Law (Toronto: Emond Montgomery, 1993) at 127.

(11) Jacques Chevallier, "Doctrine juridique et science juridique" (2002) 50 Dr et soc 103.

(12) As Justice Beetz stressed with regard to the limits of the rule of unjustified enrichment, "it is really a matter for authors [doctrine] systematically to clarify these difficulties.": Cie immobiliere Viger v L Giguere Inc (1976), [1977] 2 SCR 67 at 77, 10 NR 277.

(13) See e.g. 114957 Canada Ltee (Spraytech, Societe d'arrosage) v Hudson (Town of), 2001 SCC 40, [2001] 2 SCR 241.

(14) See in particular the rulings in Multiple Access Ltd v McCutcheon, [1982] 2 SCR 161, 138 DLR (3d) 1; Saskatchewan (Attorney General) v Lemare Lake Logging Ltd, 2015 SCC 53, [2015] 3 SCR 419 [Lemare Lake].

(15) See the majority reasons in 407 ETR Concession Co v Canada (Superintendent of Bankruptcy), 2015 SCC 52, [2015] 3 SCR 397, Gascon J.

(16) See in particular Bank of Montreal v Hall, [1990] 1 SCR 121, 65 DLR (4th) 361. In that case, the Supreme Court concluded that for all intents and purposes, federal law must be considered to be a hierarchically higher norm. The Court also stated that the doctrine of paramountcy, particularly as it pertains to conflict with the purpose of the federal law, should be applied restrictively. See Lemare Lake, supra note 14 at paras 20-27.

(17) Husky Oil Operations Ltd v MNR, [1995] 3 SCR 453 at para 85, 128 DLR (4th) 1.

(18) R v Sutherland, [1980] 2 SCR 451, 113 DLR (3d) 374 [Sutherland].

(19) This involved paragraph 13 of the Memorandum of Agreement approved under the Manitoba Natural Resources Transfer Act, RSM 1970, c N30, constitutionalized by the Constitution Act, 1930, RSC 1985, Appendix II, No 26.

(20) Sutherland, supra note 18 at 456.

(21) Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11, s 52(2). The origin of this non-exhaustive definition of the Constitution of Canada can be found in the resolution tabled by the federal government in October 1980 with a view to patriating the Constitution without the consent of provinces: Anne F Bayefsky, Canada's Constitution Act, 1982, and Amendments: A Documentary History (Toronto: McGraw-Hill Ryerson, 1989) vol 2 at 757. Previous plans to patriate the Constitution included a clause that exhaustively defined the Constitution of Canada. A document that was apparently prepared for the Minister of Justice at the time provides the following explanation: "The definition is not exhaustive; it includes the documents specifically listed. The Constitution of Canada is found in other documents as well as those listed, such as the letters patent appointing the Governor General, the instructions to Lieutenant Governors, provincial statutes relating to the constitution of the province, federal statutes such as the Succession to the Throne Act. To try to enumerate such documents would be too time-consuming. There would be a danger of leaving some out.": Briefing book for clause-by-clause consideration of the proposed resolution (Book 111) (January 1980), Ottawa, Library and Archives Canada (R11344, vol 406, files 7-9).

(22) For the sake of precision, it should be mentioned that section 101 of the Constitution Act, 1867 empowers Parliament to create a "General Court of Appeal," wirhout further details as to its jurisdiction ot organization: Constitution Act, 1867 (UK), 30 & 31 Vict, c 3, reprinted in RSC 1985, Appendix II, No 5.

(23) Ontario (AG) v Canada (AG), [1912] AC 571 (PC) at 581, 3 DLR 509; See also Murphy v Canadian Pacific Railway, [1958] SCR 626 at 643, 15 DLR (2d) 145.

(24) See e.g. Hogg, Constitutional Law, supra note 3 at ch 4-23, which states that according to the Court, some provisions of the Supreme Court Act "have mysteriously migrated into the Constitution of Canada"; Kate Glover, "Structure, Substance and Spirit: Lessons in Constitutional Architecture from the Senate Reform Reference" (2014) 67 SCLR (2d) 221 at 248.

(25) See e.g. Adam Dodek, "Uncovering the Wall Surrounding the Castle of the Constitution: Judicial Interpretation of Part V of the Constitution Act, 1982" in Macfarlane, supra note 9, 42 at 51-53; Philippe Lagasse & Patrick Baud, "The Crown and Constitutional Amendment after the Senate and Supreme Court References" in Macfarlane, supra note 9, 248; Noura Karazivan, "De la structure constitutionnelle dans le Renvoi relatif an Senat: vers une gestalt constitutionnelle?" (2015) 60:4 McGill LJ 793.

(26) Representation Act 1985, SC 1986, c 8, Part I; Fair Representation Act, SC 2011, c 26, s 2.

(27) See infra note 82 and accompanying text.

(28) See in particular art 1732 CCQ, on sales warranties.

(29) See the landmark ruling in Canadian Western Bank V Alberta, 2007 SCC 22, [2007] 2 SCR 3 [Canadian Western Bank].

(30) Quebec (AG) v Canadian Owners and Pilots Association, 2010 SCC 39, [2010] 2 SCR 536.

(31) Ronald Dworkin, Law's Empire (Oxford: Harvard University Press, 1986); see also Stephen A Smith, Contract Theory (Oxford: Oxford University Press, 2004) at 7-32.

(32) Canadian Western Bank, supra note 29 at para 52.

(33) Reference re Senate, supra note 2 at para 55.

(34) Ibid at para 60.

(35) Ibid at para 107.

(36) Reference re Supreme Court, supra note 1 at paras 104-106.

(37) Reference re Senate, supra note 2 at para 24.

(38) Ibid at para 27.

(39) Ibid paras 52, 69, 79.

(40) Karazivan, supra note 25 at 816.

(41) Reference re Supreme Court, supra note 1 at para 90.

(42) Ibid at paras 91, 104.

(43) See e.g. Hogg, Constitutional Law, supra note 3 at ch 4-23.

(44) Reference re Senate, supra note 2 at para 65.

(45) Reference re Supreme Court, supra note 1 at para 92. See also paras 93, 99.

(46) Ibid at para 100; See also Reference re Senate, supra note 2 at paras 31, 100.

(47) Reference re Senate, supra note 2 at para 11.

(48) Ibid at pata 18.

(49) Renvoi relatif au projet de loi federal relatif au Senat, 2013 QCCA 1807 at para 40, 370 DLR (4th) 711 [Renvoi relatif au Senat].

(50) Ibid at para 48.

(51) Reference re Senate, supra note 2 at pata 21.

(52) Dworkin, supra note 31 at 254-258. See also Consolidated Fastfrate Inc v Western Canada Council of Teamsters, 2009 SCC 53 at para 45, [2009] 3 SCR 407.

(53) This is the strategy recommended by Daniel Weinstock to find a compromise between the proponents of two sets of conflicting fundamental principles: Daniel Weinstock, "So, Are You Still a Philosopher?" (Lecture delivered at the Big Thinking Lecture Series, University of Victoria, 5 June 2013), 5 Trudeau Foundation Papers 127, online: <>.

(54) See e.g. TRS Allan, Constitutional Justice: A Liberal Theory of the Rule of Law (Oxford: Oxford University Press, 2001); TRS Allan, The Sovereignty of Law. Freedom, Constitution and Common Law (Oxford: Oxford University Press, 2013).

(55) Reference re Remuneration of Judges of the Provincial Court of Prince Edward Island, [1997] 3 SCR 3, 150 DLR (4th) 577 [Reference re Remuneration].

(56) Ibid at paras 315-316. See also Jean-Francois Gaudreault-DesBiens, "Le juge comme agent de migration de canevas de raisonnement entre le droit civil et la common law? Quelques observations a partir devolutions recentes du droit constitutionnel canadien" in Ghislain Otis, ed, Le juge et le dialogue des cultures juridiques (Paris: Karthala, 2013) 41 at 56.

(57) Reference re Secession of Quebec, [1998] 2 SCR 217, 161 DLR (4th) 385.

(58) Babcock v Canada (AG), 2002 SCC 57, [2002] 3 SCR 3.

(59) British Columbia v Imperial Tobacco Canada Ltd, 2005 SCC 49, [2005] 2 SCR 473.

(60) British Columbia (AG) v Christie, 2007 SCC 21, [2007] 1 SCR 873.

(61) Trial Lawyers Association of British Columbia v British Columbia (AG), 2014 SCC 59, [2014] 3 SCR 31 [ Trial Lawyers Association].

(62) By way of analogy, in France, the extension of the "Constitutional block" has been attached, in the vast majority of cases, to certain elements of the text of the 1958 Constitution: Louis Favoreu et al, Droit constitutionnel, 13th ed (Paris: Dalloz, 2013) at 121-128.

(63) See supra note 9.

(64) Unless a text attained supralegislative status prior to 1982 based on the principles of British Imperial law. This could be the case, for example, with The Quebec Boundaries Extension Act, SC 1912, c 45; see R v Sparrow, [1990] 1 SCR 1075 at 1104, 70 DLR (4th) 385.

(65) Baker & Jarvis, supra note 9 at 195.

(66) Even if some elements of the final version of this text were drafted in a hurry: Mary Dawson, "From the Backroom to the Front Line: Making Constitutional History" (2012) 57:4 McGill LJ 955 at 965.

(67) For the history of these negotiations, see in particular Paul Gerin-Lajoie, Constitutional Amendment in Canada (Toronto: University of Toronto Press, 1950); J Peter Meekison, "The Amending Formula" (1982) 8 Queen's LJ 99; Peter H Russell, Constitutional Odyssey: Can Canadians Become a Sovereign People?, 2nd ed (Toronto: University of Toronto Press, 1992); Nadia Verrelli, "Searching for an Amending Formula: The 115-Year Journey" in Macfarlane, supra note 9, 19 [Verrelli, "Amending Formula"]; James Ross Hurley, Amending Canada's Constitution: History, Processes, Problems and Prospects (Ottawa: Canada Communications Group, 1996).

(68) Verrelli, "Amending Formula", supra note 67 at 26.

(69) See Oliver, supra note 8.

(70) Reference re Authority of Parliament in relation to the Upper House, [1980] 1 SCR 54, 102 DLR (3d) 1.

(71) As we will see further on (infra note 82 and accompanying text), this philosophy of protection may translate into the protective function analyzed in this paper, but also in the fact that the constitutional provisions that establish bilingualism at the federal level cannot be amended without the support of all the provinces, which means that Quebec has a right of veto on this subject.

(72) And in particular, it could be said, on the interpretation methods specific to civil law.

(73) Glover, supra note 24 at 237.

(74) See e.g. Meekison, supra note 67 at 115-116, who appears to envisage an enabling function for sections 41(d), 42(1)(A) and 42(1)(d). Meekison was Deputy Minister of Alberta Intergovernmental Affairs at the time.

(75) Reproduced in Hurley, supra note 67 at 186.

(76) See infra.

(77) See infra, the section dealing with the judiciary. Furthermore, as mentioned earlier, the amending formula also serves to "protect" the provisions of the Constitution from unilateral amendment, but the concept of the protective function that we are developing here deals with amendments to rules or principles that are not explicitly mentioned in the Constitution.

(78) The authority to create the Supreme Court is set out in section 101 of the Constitution Act, 1867, supra note 22.

(79) Reference re Senate, supra note 2 at paras 54-63.

(80) See e.g. Jones v New Brunswick (AG), [1975] 2 SCR 182 at 192-195,45 DLR (3d) 583.

(81) RSC 1985, c 31 (4th Supp).

(82) Andre Braen, "Le statut des langues autochtones au Canada: le cas de l'lnuit au Nunavut" (2008) 87 Can Bar Rev 741 at 751 argues in fact that section 16 does not prohibit creating official status tor Aboriginal languages.

(83) Supra note26. These amendments were adopted pursuant to section 44 of the Constitution Act, 1982, supra note 21.

(84) Hawkins, supra note 7.

(85) Reference re Senate, supra note 2 at para 78.

(86) See Catherine Mathieu & Patrick Taillon, "Le federalisme comme principe matriciel dans 1'interpretation de la procedure de modification constitutionnelle" (2015) 60:4 McGill LJ 763.

(87) Renvoi relatif au Senat, supra note 49 at para 82.

(88) For an overview, see Nadia Verrelli, ed, The Democratic Dilemma: Reforming Canada's Supreme Court (Montreal and Kingston, McGill-Queen's University Press, 2013).

(89) Bill C-232, An Act to amend the Supreme Court Act (understanding the official languages), 3rd Sess, 40th Parl, 2010 (as passed by the House of Commons 31 March 2010).

(90) See e.g. Josh Hunter & Padraic Ryan, "The Entrenchment of Discretion: Prospects for Judicial Appointment Reform after a Trio of References" in Lisa M Kelly & Ivo Entchev, eds, Judicious Restraint: The Life and Law of Justice Marshall E. Rothstein (Toronto: LexisNexis, 2016) 117 at 153-156; Emmett Macfarlane, "The Uncertain Future of Senate Reform," in Macfarlane, supra note 9, 228 at 242.

(91) Warren Newman, supra note 5, reaches a similar conclusion in suggesting that Parliament's powers flowing from section 101 of the Constitution Act, 1867 are now subject to an implicit prohibition on amending the essential characteristics of the Court.

(92) Reference re Supreme Court, supra note 1 at para 101.

(93) Ibid at para 91.

(94) Ibid at para 115.

(95) With regard to the political context in this matter, see Hugo Cyr, "The Bungling of Justice Nation's Appointment to the Supreme Court of Canada" (2014) 67 SCLR (2d) 73.

(96) For example, we might ask what the Court would have done with a federal Act providing that the three Quebec judges be appointed by the Government of Quebec, and leaving it up to the Quebec National Assembly to determine the eligibility requirements...

(97) See in particular the "Report of the Continuing Committee of Ministers on the Constitution to First Ministers: Supreme Court" (September 1980), in Bayefsky, supra note 21 at 718; Roy Romanow, John Whyte & Howard Leeson, Canada... Notwithstanding: The Making of the Constitution 1976-1982, 2nd ed (Toronto: Thomson Carswell, 2007) at 80-82.

(98) This is an assumption that cannot be confirmed through published sources. Archival research would no doubt be required to shed more light on the reasons for this inclusion.

(99) Reference re Supreme Court, supra note 1 at para 92.

(100) Ibid at para 91.

(101) Contrary, for example, to the French Constitutional Court: Favoreu, supra note 62 at 310-311.

(102) See contra Leonid Sirota, "The Comprehension of 'Composition'" (16 May 2016) Double Aspect (blog), online: <>; Kate Glover, "A Third View on Legislating Two Languages at the SCC" (18 May 2016) Double Aspect (blog), online: <https://doubleaspectblog.wordpress. com/2016/05/18/a-third-view-on-legislating-two-languages-at-the-scc/>.

(103) Reference re Supreme Court, supra note 1 at para 94.

(104) Bear in mind that only published sources have been used; exhaustive archival research may reveal little-known information.

(105) Reproduced in Bayefsky, supra note 21 at 811. It should be mentioned that the French version of these notes, reproduced in Hurley, supra note 67 at 251, does not mention the idea of "anticipation." See also Meekison, supra note 67 at 116.

(106) Reference re Supreme Court, supra note 1 at para 83.

(107) Ibid at para 89.

(108) Ibid at para 85.

(109) See also Newman, supra note 5. at 439.

(110) Reference re Remuneration, supra note 55.

(111) Remember that the Constitution does not give "discretion" to the Governor General (in practice, the Prime Minister) to appoint judges to the Supreme Court, contrary to the situation with senators and superior court judges. The appointment power is found in the Supreme Court Act. See also Hunter & Ryan, supra note 90 at 149-153.

(112) See Annex: Qualifications and Merit-Based Assessment Criteria (Consulted December 13, 2016), online: <>.

(113) Reference re Senate, supra note 2 at para 65.

(114) Macfarlane, supra note 90 at 234-235.

(115) Memorandum from Michael Kirby to Prime Minister Pierre Elliott Trudeau (21 August 1980), reproduced in Linda Cardinal & Sebastien Grammond, Une tradition et un droit: le Senat et la representation de la francophonic canadienne (Ottawa: University of Ottawa Press, 2017) in Appendix; Special Joint Committee of the Senate and of the House of Commons on Senate Reform, Reform of the Senate: A Discussion Paper (16 June 1983) (Chairs: Honourable Gildas Molgat, Honourable Paul Cosgrove).

(116) Munro v National Capital Commission, [1966] SCR 663 at 669-670, 57 DLR (2d) 753.

(117) Hawkins, supra note 7.

(118) Gilles Pepin, Les tribunaux administratifs et la Constitution: etude des articles 96 a 101 de l'A.A.N.B. (Montreal: Presses de l'Universite de Montreal, 1969) at 116-129. Followed to its logical conclusion, this reasoning would lead to calling into question the validity ot large segments of the Judges Act, notably those dealing with judicial discipline and the periodic adjustment of remuneration: judges Act, RSC 1985 c J-1.

(119) Martineau & Sons v Montreal (City of), [1932] AC 113 (PC) at 120; Toronto (City of) v York (Township of, [1938] AC 415 (PC) at 426; R v Beauregard, [1986] 2 SCR 56 at 73, 30 DLR (4th) 481; MacMillan Bloedel Ltd v Simpson, [1995] 4 SCR 725 at 741, 130 DLR (4th) 385, Lamer CJC [MacMillan].

(120) MacMillan, ibid at 741, Lamer J, and 759-760, McLachlin J dissenting; Reference re Residential Tenancies Act, [1981] 1 SCR 714 at 728, 235 DLR (4th) 176.

(121) Trial Lawyers Association, supra note 61.
COPYRIGHT 2017 Centre for Constitutional Studies, University of Alberta
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Canada
Author:Grammond, Sebastien
Publication:Review of Constitutional Studies
Date:Sep 1, 2017
Previous Article:All I Really Needed to Know About Federalism, I Learned from Insurance Law.
Next Article:Surfing the Surveillance Wave: Online Privacy, Freedom of Expression and the Threat of National Security.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |