In the Senate Reference, the Supreme Court held that the government’s proposed reforms to the Senate can only be achieved by following the constitutional amendment procedures in Part V of the Constitution Act, 1982. There were six reference questions. The first concerned the legislative authority of Parliament to unilaterally impose senatorial term limits of various lengths. The second and third questions concerned the legislative authority of Parliament to pass legislation that would allow for consultative elections in the provinces for nominees for Senate appointment. The fourth question asked whether Parliament could unilaterally repeal the wealth and property qualifications for Senators. The fifth and sixth questions considered the issue of whether the abolition of the Senate could be achieved under the general “7/50” amending formula or if Senate abolition could only be accomplished under the unanimous consent procedure.
This article argues that the Senate Reference is ultimately a decision about how decision making ought to be conducted when the role and function of fundamental democratic institutions are themselves at stake. The Supreme Court found, in essence, that unilateral decision making by Parliament is not permitted even if from a substantive standpoint the proposals are “more democratic” than the status quo. The Court held that consultative elections and senatorial term limits must be implemented under the “7/50 rule”, which requires the consent of the Senate, the House of Commons, and the legislative assemblies of at least seven provinces representing half the population of all the provinces. The wealth and property qualifications of Senators could be repealed in part by Parliament, although a full repeal would require the consent of the Québec legislature. The abolition of the Senate would require the unanimous consent of Parliament and all the provinces.
This article claims that the Court’s interpretation of the requirements for constitutional amendment is based upon a deeper democratic commitment to ensuring dialogue and deliberation between and among the relevant stakeholders. The process by which democratic institutions are to be reformed must itself be democratic in this deeper sense regardless of the substance of the proposed reforms. Although the Court’s approach has the unfortunate effect of making large-scale institutional reform difficult (if not impossible), the alternative is arguably worse. If it were possible for the government to unilaterally reform democratic institutions, then it could unilaterally reform them in an anti-democratic direction as well.
Despite its commitment to a deliberative democratic process for constitutional amendment, the Court’s general approach to constitutional amendment questions is susceptible to criticism. By setting itself up as the exclusive arbiter of the Constitution’s “internal architecture” and the primary decision-maker as to what constitutes an institution’s “fundamental role and nature”, the Court has enhanced its own authority over the evolution of the constitutional order while significantly narrowing the possibilities for constitutional change. For instance, the Court’s determination that the Senate does not have democratic legitimacy unnecessarily undermines the Senate’s representative function in the Canadian political order. The Court’s position runs the risk of locking into place a very narrow view of the function of the Senate. More generally, the Court’s approach runs the risk of freezing into place the status quo operation of democratic institutions and processes.
This article is organized into four sections. Part I addresses the critiques of the Senate and suggested reforms. Part II sets out the Supreme Court’s approach to constitutional amendment. It focuses, in particular, on two noteworthy features of the Court’s approach. First, the Court’s use of the Constitution’s “internal architecture” was crucial to its determination that the Constitution could be amended even in the absence of actual revisions to the constitutional text. Second, the Court held that the general amending procedure in section 38 is the default procedure for constitutional amendment. In addition, Part II argues that the Court’s approach is based upon a commitment to democratic decision making. The article both identifies the democratic commitments that underlie the Court’s decision and offers an evaluation of the benefits and disadvantages of the Court’s approach. Part III examines the Court’s treatment of consultative elections and the role of the Senate. It argues, in particular, that the Court’s decision turned on its assessment of the “fundamental nature and role” of the Senate. This Part offers a critical analysis of the Court’s assessment of the Senate’s role in the constitutional order. Part IV addresses the property and wealth qualifications of Senators, senatorial term limits, and the abolition of the Senate.
I. Critiques and Reforms
The Senate has been criticized almost since its inception. Although there exists a wide array of critiques, there is a notable consistency in the types of problems that have been identified over the decades. As the “whipping post of democratic institutions”, the Senate is routinely charged with being ineffective, unaccountable, unimpressive, and lacking in legitimacy. As noted by one observer, “[i]t would be idle to deny that the Senate has not fulfilled the hopes of its founders; and it is well also to remember that the hopes of its founders were not excessively high.”
The Senate is also criticized for a being an institution of patronage and partisanship. The partisan complexion of the Senate arises because the Prime Minister enjoys a monopoly in choosing Senators. Critics argue that the appointed nature of the Senate undermines representation. Not only are Senators unelected, but as a result of the selection process, they represent the interests of the political parties rather than the regions from which they are appointed. Senate “appointments are used first and foremost as a political reward for party faithful, both elected and those who toil behind the scenes.” Senators are also chosen for their ability to raise funds for the parties’ “war chests”. Although the Senate has been used to boost the number of parliamentarians from under-represented groups, it is still widely viewed as a “legislative hall of shame.” There have also been long-standing complaints that Senators act as lobbyists for business interests, although “Senators are not allowed to lobby on behalf of companies on whose boards they sit.” In short, many Canadians view the Senate as “a useless, expensive, undemocratic appendage of government” and as “a refuge and dumping ground for bagmen, party apologists, and failed politicians.”
The efforts to reform the Senate are as numerous as the critiques. In the government’s factum, the Attorney General of Canada provided a detailed discussion of the history of proposed reforms. In addition to the failed Meech Lake and Charlottetown Accords, there have also been countless legislative reports, policy proposals, and studies on the issue of Senate reform. Proposals have ranged from suggesting that the Senate should be elected by the people, or appointed by the provinces, or that Senators should be subject to term limits. Advocates of a Triple-E Senate, for example, argue that institutional reform is required to secure greater regional (and western) influence over national policy-making. There are some who argue that “reform of the Senate might prove the best strategy for revitalizing Canada’s political system and for strengthening the ties that hold the country together.” But others argue that any discussion of Senate reform “is purely an academic exercise with little chance of becoming a reality.”
The reference questions were based on proposed (but since lapsed) Senate reform legislation, in particular Bill C-7 (the Senate Reform Act), Bill S-4 (the proposed Constitution Act, 2006) and Bill C-20 (the Senate Appointments Consultations Act). In the words of the Attorney General, the proposed reforms “seek to address longstanding concerns that undermine the Senate’s legitimacy as a democratic institution.” There is little doubt that Bill C-7 was intended to democratize the Senate. For instance, the preamble states that “it is important that Canada’s representative institutions, including the Senate, continue to evolve in accordance with the principles of modern democracy and the expectations of Canadians” and that the government of Canada “has undertaken to explore means to enable the Senate better to reflect the democratic values of Canadians and respond to the needs of Canada’s regions.” The government’s objective was to redress some of the glaring deficiencies of the Senate.
II. Democracy and Constitutional Amendment
The Senate Reference established a set of principles and guidelines for constitutional amendment. At issue in the Senate Reference was whether Parliament could unilaterally undertake various reforms to the Senate by enacting ordinary legislation. The Attorney General of Canada argued for a textualist reading of Part V of the Constitution Act, 1982. On a strict textualist reading of section 42, there are only four items with respect to the Senate that require provincial consent: the powers of the Senate, the method of selecting Senators, the number of Senators apportioned to each province, and residence qualifications. Except for these four items, argued the Attorney General, Parliament under section 44 has “the exclusive authority to make laws amending the Constitution in relation to the Senate.” For this reason, Parliament may single-handedly impose term limits, provide for public consultations with respect to Senate appointments, and eliminate the real property and wealth requirements.
The Attorney General also argued that there is no need to rely on any unwritten constitutional principles to determine the correct approach for reform. Constitutional interpretation should value the “primacy of the text”. Part V provides an “exhaustive” account of the relevant procedures. Unwritten constitutional principles, claimed the Attorney General, “should not be used to create rights or impose requirements not found in the written text.” The government referred to the Secession Reference for the idea that such unwritten principles are to be used only when situations arise that were not expressly addressed in the constitutional document. The unwritten principles are not “an invitation to dispense with the written text of the Constitution.”
In addition, the government argued for a progressive approach to constitutional interpretation. The Attorney General noted that “[s]lavish adherence to original intent has been rejected” by the Court. For this reason, there is no need to follow what the 1867 framers thought was the role of the Senate, particularly in view of the fact that the British have reformed their upper house to better comport with democratic principles.
In the Senate Reference, however, the Supreme Court rejected the Attorney General’s approach to constitutional interpretation. Drawing heavily from its earlier opinion in the Secession Reference, the Court set forth the basic principles of constitutional interpretation. It defined the Constitution as “‘a comprehensive set of rules and principles’ that provides ‘an exhaustive legal framework for our system of government.’” Crucially, the Constitution has an “internal architecture” or “basic constitutional structure.” The Constitution is not simply comprised of the constitutional text itself; it also must be understood by reference to previous constitutional cases and the historical context. Constitutional provisions must be “interpreted in a broad and purposive manner and placed in their proper linguistic, philosophic, and historical contexts.” Constitutional interpretation must also be informed by the foundational principles of democracy, federalism, the rule of law, and the protection of minorities.
The Court’s analysis of constitutional interpretation was critical to its conclusions in this case. The concept of the Constitution’s “internal architecture” proved to be of crucial importance since it served as a bar to changes in constitutional practices that do not result from changes to the constitutional text. The notion that the Constitution has an internal architecture means that the various elements of the Constitution must be interpreted within the larger structure of the entire document. Accordingly the Court stated that that same interpretive approach applies to constitutional amendments. The Constitution is not simply a “collection of discrete textual provisions”. For this reason, the Constitution can be amended through changes to its architecture even though the text of the Constitution is left untouched.
Another significant aspect of the opinion was the Court’s determination that the general amending procedure was the default procedure for constitutional amendment, and that the other procedures in Part V were all exceptions to the general procedure. The general amending procedure is found in section 38 and is complemented by section 42. Under the general amending procedure, also known as the 7/50 procedure, constitutional amendments must be authorized by the resolutions of the House of Commons, the Senate, and the legislative assemblies of at least seven provinces whose population represents at least half of the population of the provinces. Section 38 also provides an opt-out mechanism to the provinces with respect to those amendments that derogate from the powers and rights of the provincial legislature. According to the Court, the general amending formula is based on the idea that “substantial provincial consent must be obtained for constitutional change that engages provincial interests.”
Section 42, which works in tandem with section 38, has two purposes. First, it expressly identifies certain kinds of amendments that are subject to the general amending formula in section 38. Particularly relevant to the decision is section 42(1)(b), which provides that the general amending formula applies to changes to “the powers of the Senate and the method of selecting Senators.” Second, section 42 does not provide the provinces with an opt-out mechanism, thus ensuring uniformity across the country for certain categories of amendment.
The Court held that the three other procedures are exceptions to the general amending procedure. The unanimous consent procedure, found in section 41, requires the unanimous consent of the Senate, the House of Commons, and all ten provincial legislatures for certain categories of amendments. The unanimity procedure, which is an exception to the general amending formula, applies to “fundamental changes” to the Constitution. In essence, section 41 amounts to a veto power. The special arrangements procedure, set forth in section 43, applies to those amendments that apply to some, but not all, of the provinces. The unilateral federal procedure, found in section 44, allows the federal legislature to amend the Constitution without the consent of any other body. Section 44 provides that “[s]ubject to sections 41 and 42, Parliament may exclusively make laws amending the Constitution of Canada in relation to the executive government of Canada or the Senate and House of Commons.” Finally, section 45 provides for unilateral provincial procedures of amendment.
The Court referred to its 1980 opinion in the Upper House Reference, which concluded that section 91(1) (now repealed) did not provide Parliament with the unilateral power to reform the Senate. In the Senate Reference, the Court interpreted sections 44 and 45 in the same way: these provisions enable the federal and provincial governments to unilaterally amend only those aspects of the Constitution that do not engage the interests of the other level of government. The limited scope of unilateral amendments is based on the “principle that Parliament and the provinces are equal stakeholders in the Canadian constitutional design.” For this reason, neither the federal government nor the provinces can single-handedly alter the “fundamental nature and role of the institutions provided for in the Constitution.”
The Court’s approach to constitutional amendment is consistent with basic democratic principles of deliberation, dialogue, and equality, albeit with important qualifications. In general, the Court emphasized the idea that Part V “reflects the principle that constitutional change that engages provincial interests requires both the consent of Parliament and a significant degree of provincial consent.” The purpose of the amending formula is to “constrain unilateral federal powers to effect constitutional change.” The amending formula is also animated by the idea that the provinces are equals from a constitutional perspective: “all provinces are given the same rights in the process of amendment.” Of course, the provinces are not strictly equal since section 38 uses population in its formula with the result that the most populous provinces will have a determinative say as compared to the provinces with the smallest populations. That being said, the population formula is itself based on the equality of individuals since it gives greater weight to those provinces with larger populations. For fundamental changes to the Constitution, the unanimity procedure in section 41, which requires the unanimous consent of Parliament and all ten provincial legislatures, is triggered. In short, the amending formulae are designed to promote dialogue, on a more or less equal footing, between the federal government and the provinces with respect to constitutional change. As Ronald Watts has observed, the effort to avoid the section 38 amendment procedure “by reforming the Senate on the sly through the devious use of ordinary legislation constitutes … a non-constitutional and hence ultimately anti-democratic process.”
The Senate Reference is similar to the Secession Reference in its approach to constitutional amendment. Underlying both opinions is a particular understanding of democracy. In the Secession Reference, the Court stated that “Canadians have never accepted that ours is a system of simple majority rule.” Instead, “[o]ur principle of democracy, taken in conjunction with the other constitutional principles discussed here, is richer.” Constitutional provisions can be amended “but only through a process of negotiation which ensures that there is an opportunity for the constitutionally defined rights of all the parties to be respected and reconciled.” The process of negotiation is essentially democratic because it demands that all parties affected by a decision ought to have a say in the decision. The Court described its approach as harmonizing a belief in democracy with a belief in constitutionalism. Constitutional amendment requires “substantial consensus”. In addition, “[by] requiring broad support in the form of an ‘enhanced majority’ to achieve constitutional change, the Constitution ensures that minority interests must be addressed before proposed changes which would affect them may be enacted.”
The benefit to the Court’s approach is that it would prevent unilateral reforms that are anti-democratic. If the federal government were permitted to change democratic institutions unilaterally by simply passing ordinary legislation and by leaving the constitutional text unchanged, then such changes could be beneficial or harmful to democracy. That being said, there is no question that the Court has made it extremely difficult for large-scale institutional reform to take place. Robert Hawkins has referred to the creative solutions in Bill C-7 as “constitutional workarounds”. He argues that the constitutional validity of these workarounds depends on them being advisory or consultative rather than binding on the actions of political actors. Yet, as we have seen, the Court took a different view. Constitutional workarounds may be illegitimate even when they are only advisory or consultative.
The Court has defined what counts as a constitutional amendment such that it now includes items that may once have not been viewed as amounting to constitutional changes. The Court effectively closed the door to those constitutional workarounds that affect what the Court views as an institution’s fundamental nature and role.
III. Consultative Elections and the Role of the Senate
Consultative elections would enable the citizens of each province to have a say in determining the nominees for Senate appointment. The Attorney General of Canada argued that the implementation of consultative elections would not amount to an amendment to the Constitution because neither the constitutional text nor the means of selecting Senators would be altered. Under sections 24 and 32 of the Constitution Act, 1867, Senators are formally appointed by the Governor General. In practice, however, a constitutional convention has evolved whereby the Governor General follows the recommendation of the Prime Minister. Consultative elections could be implemented without changing the text of the Constitution. In the alternative, the Attorney General argued that if the implementation of consultative elections does constitute an amendment, then such an amendment can be unilaterally made by Parliament under section 44.
The Supreme Court rejected both arguments. Consultative elections would fundamentally alter the Constitution’s architecture, and would therefore amount to a constitutional amendment. The Court determined that the Attorney General was privileging “form over substance” for reducing “the notion of constitutional amendment to a matter of whether or not the letter of the constitutional text is modified.” Constitutional interpretation instead requires a broad and purposive approach. Although the constitutional text would not be altered, the “Senate’s fundamental nature and role as a complementary legislative body of sober second thought would be significantly altered.” Consultative elections would also bestow democratic legitimacy on the Senate and would thereby enable it to systematically block the legislation of the House of Commons. The Senate, however, was not designed to be a systematic obstacle to legislation.
The Court determined whether the Constitution’s architecture had been altered by asking whether the Senate’s “fundamental nature and role” had been changed. The language of the Senate’s “fundamental nature and role” is relied on throughout the opinion. There are two observations worth making. The first is that the Court has appointed itself as the institution that is best situated to determine the fundamental nature and role of a given institution in the Canadian constitutional order. The second point is that an institution’s fundamental nature and role is not only open to contestation, but also evolves over time. Yet by determining an institution’s fundamental nature and role, the Court has frozen into place a particular view of the institution, thereby impeding its possibilities for evolution and change.
The Court set forth an originalist argument, observing that the framers of the Constitution Act, 1867 purposefully chose appointment for Senators so that the Senate could “play the specific role of a complementary legislative body of ‘sober second thought.’” The Court quoted from John A. Macdonald’s statements during the Parliamentary debates on Confederation: The Senate would “calmly [consider] the legislation initiated by the popular branch, and [prevent] any hasty or ill considered legislation which may come from that body, but it will never set itself in opposition against the deliberate and understood wishes of the people.”
The Court also referred to its decision in the Upper House Reference for the idea that the framers’ intention was to “make the Senate a thoroughly independent body which could canvass dispassionately the measures of the House of Commons.” According to the Court, the Senate was designed to be independent from the electoral process. Senators were deliberately removed from a partisan political arena and from the requirement to consider the short-term political objectives that such an arena demanded. In addition, the Senate was meant to “be a complementary legislative body, rather than a perennial rival of the House of Commons in the legislative process.” The appointment process meant that “Senators would not have a popular mandate—they would not have the expectations and legitimacy that stem from popular election.” Their role would be one of legislative review rather than as the coequal of the House of Commons.
The Senate’s role as a complementary legislative body, concluded the Court, is part of the architecture of the Constitution Act, 1867. The framers felt no need to specify how a deadlock between the Senate and the House would be resolved because it was assumed that the appointed status of Senators would prevent them from overstepping their role, even though, on the face of it, the Senate has almost as much legislative power as the House of Commons. The appointment of Senators meant that the legislation passed by the House of Commons would usually be adopted except in rare instances in which the Senate would act as a check.
In contrast, the Attorney General argued that the public consultation process would not result in a competing legislative chamber. Senators would “be free to discharge their ‘representational responsibilities as trustees rather than as delegates’” because they would only be appointed for one term. They would be independent by virtue of being free both of constituency and re-election pressures. In addition, there is no change to the “method of selecting Senators” because the Prime Minister’s discretion has not been impaired. If anything, the Prime Minister’s discretion has been tempered by the additional input, which leads to more transparent decision making.
The Court disagreed with the government’s positions. It found that in practice consultative elections would subject Senators to political pressures and would provide them with a popular mandate. Even if they were formally selected from a list of nominees, they would enjoy the “mandate and legitimacy that flow from popular election.” The Court anticipated that Senators would have to develop a campaign platform and make promises to their constituents.
The Court also rejected the Attorney General’s view that these broad changes would not take place because the Prime Minister could simply ignore the results of the consultative elections. Although the Prime Minister could ignore the election results, the purpose of Bills C-20 and C-7 was to use elections to determine the nominees for the Senate. The problem with consultative elections is that they would change “the Senate’s role within our constitutional structure from a complementary legislative body of sober second thought to a legislative body endowed with a popular mandate and democratic legitimacy.”
It is always difficult to predict how an institutional reform such as consultative elections would play out in practice. There are two distinct concerns that are at times conflated in the discussion. The first is whether the Senators would retain their independence from the pressures of popular opinion so that they could bring a “sober second thought” to the legislation they review. It seems reasonable to assume that as long as Senators went through the selection process only once, the pressure for “re-election” would not exist. As such, their independence from popular opinion could be safeguarded and they could therefore fulfill their function of “sober second thought”. Undoubtedly, the pressure to win re-election is only one form of pressure from popular opinion. The Court’s concern was that to gain office, candidates would have to have made promises or taken certain positions which may later affect their independence from popular opinion. Yet it would seem that the re-election pressure is the most significant kind of political pressure that is exerted, with the result that Senators could still bring to bear an independent analysis to the legislation before them.
The second issue is whether the function of the Senate would change as a result of the Senators’ greater democratic legitimacy. The main concern here is one of institutional deadlock between the two chambers of the legislature. In its current configuration, the Senate has used considerable discretion when engaging in the review of legislation even though it has the power to modify legislation. It has tended to use persuasion rather than its veto power to make changes to the law. To the extent the Senate has used its veto power, it has done so infrequently and only when the proposed legislation infringed rights, or harmed a particular region, or amounted to an abuse of legislative power.
Yet a Senate that has greater democratic legitimacy may well take a more muscular approach to its powers. At this point, the Senate’s place within the larger political structure becomes important. As David Smith has argued, reformers must recognize the complementary relationship that exists between the Senate and the House of Commons. If the Senate were elected, then its work and the work of the House of Commons would have to be harmonized. Deadlock between the two popularly elected chambers could become a real possibility. On one view, the Senate would likely unduly complicate and even block the operation of government. On the whole, the concern about institutional deadlock is more pressing than the concern about senatorial independence.
In addition, there is a curious omission in the Court’s description of the Senate’s fundamental nature and role. The Court placed little if any emphasis on the role of the Senate in regional representation. This omission is noteworthy because the Court stated that consultative elections would afford the Senate democratic legitimacy, and this change, suggested the Court, would be too great a departure from the Senate’s fundamental nature and role. But the Senate does enjoy a form of democratic legitimacy that stems from its function as a representative body. The fact that the Senate is a representative body, with the democratic legitimacy that such representation entails, significantly complicates the Court’s assessment that consultative elections are incompatible with the Senate’s fundamental nature and role.
In contrast to the Court’s opinion, many accounts of the Senate emphasize that the framers had at least two if not more objectives. For instance, the Québec Court of Appeal reference on senate reform identified a number of functions of the Senate, including regional representation, the representation of Québec’s Anglophone minority, sober second thought, and the oversight of elected officials. It would appear that in its eagerness to present the Senate as a body that was originally intended to lack democratic legitimacy, the Court significantly downplayed the Senate’s representational function.
The representation of the regions of Canada, however, was arguably the primary function of the Senate. The guarantee of equal regional representation was “essential to concluding the Confederation bargain; no other issue took so long to resolve.” The Maritimes were granted equal regional representation in the Senate because they were outnumbered by the more populous provinces of Ontario and Québec in the House of Commons. The Senate established a rough parity between Upper Canada, Lower Canada, and the Maritimes. The compromise was described by George Brown as follows:
Our Lower Canada friends have agreed to give us representation by population in the Lower House, on the express condition that they shall have equality in the Upper House. On no other condition would we have advanced a step; and for my part, I am quite willing that they should have it. In maintaining the existing sectional boundaries, and handing over the control of local matters to local bodies, we recognize, to a certain extent, diversity of interests, and it was quite natural that the protection for those interests, by equality in the Upper Chamber, should be demanded by the less numerous Provinces.
The important point is that both the Senate and the House of Commons have a representative function: both bodies are designed to represent the people, and represent regions. As David Smith notes, “representation and election are not the same thing.” The fact that Senators are not elected does not mean that they do not represent certain interests in Canada.
The Supreme Court’s discussion in the Upper House Reference provides an instructive contrast on this issue. The Court stated in the Upper House Reference that the “Senate has a vital role as an institution forming part of the federal system.” A “primary purpose of the Senate was … to afford protection to the various sectional interests in Canada in relation to the enactment of federal legislation.” The Court quoted the following words of Sir John A. Macdonald:
In order to protect local interests and to prevent sectional jealousies, it was found requisite that the three great divisions into which British North America is separated, should be represented in the Upper House on the principle of equality. … To the Upper House is to be confided the protection of sectional interests; therefore is it that the three great divisions are there equally represented for the purpose of defending such interests against the combinations of majorities in the Assembly.
It is noteworthy that the Upper House Reference places far more emphasis on the Senate’s role in representing the regions of Canada than its role in providing sober second thought. The judgment observed that “the system of regional representation in the Senate was one of the essential features of that body when it was created. Without it, the fundamental character of the Senate as part of the Canadian federal scheme would be eliminated.” The Court concluded that Parliament may not make changes that would affect the “fundamental features, or essential characteristics of the Senate” because these features were “given to the Senate as a means of ensuring regional and provincial representation in the federal legislative process.”
In the Senate Reference, however, the Court downplayed the representative function of the Senate. It went to great lengths to suggest that consultative elections violated the fundamental role and nature of the Senate by giving it a popular mandate and democratic legitimacy. Although the Senate may lack a popular mandate because the Senators are not elected, it is not the case that the Senate does not have democratic legitimacy. The Court could have said, for example, that it is the fundamental role and function of the Senate to be a representative body. Once it is acknowledged that the Senate does in fact have a representative function, then it is less evident that consultative elections would necessarily undermine the fundamental role and nature of the Senate. If anything, consultative elections would arguably enhance the representative function of the Senate. Another difficulty with the Court’s approach is that it unnecessarily freezes into place an overly narrow view of the Senate’s role and function.
The Court also engaged in a textual analysis of the constitutional provisions at issue. The notion that consultative elections amount to an amendment is also supported by the provisions of the amending formulae in Part V. Section 42(1)(b) of the Constitution Act provides that the general amending procedure found in section 38 applies to those amendments that are in relation to “the method of selecting Senators”. According to the Court, this provision covers the implementation of consultative elections. The phrase “method of selecting Senators” does not apply to only the formal appointment of Senators but also includes the entire selection process. That selection process includes the compilation of a list of candidates from the elections. For this reason, the implementation of consultative elections is subject to the general amending formula without a provincial opt-out as provided for in section 42(1)(b).
One might wonder why the Court did not simply rely on this textual argument, instead of also providing an argument about the Senate’s original purpose. One possibility is that the Court wished to forestall future unilateral change that did comport with the strict wording of the constitutional provisions. According to the Court, consultative elections could only be implemented by following the general amending procedure but without the provincial opt-out right. The Court rejected the government’s alternative position that consultative elections could be implemented under section 44. Section 44 is subject to section 42, and the Court had already determined that section 42(1)(b) applied to changes to the “method of selecting Senators.” Section 44 would not apply to consultative elections since they “would change the Senate’s fundamental nature and role by endowing it with a popular mandate.” The Court effectively ended any possibility of consultative elections without substantial provincial approval.
IV. Senate Abolition, and Senatorial Tenure and Qualifications
The unwritten principle of democracy continued to inform the Court’s positions on senatorial tenure, senatorial qualifications, and the abolition of the Senate. In the words of the Attorney General of the Northwest Territories, “any amendment to the Canadian Constitution must comply with basic democratic consultations.” The more extensive the proposed change, the more rigorous is the expectation of democratic consultation and consent. Senate abolition thus requires the unanimous consent procedure; senatorial term limits require substantial provincial consent under the 7/50 procedure; and the elimination of property and wealth qualifications can be achieved through a combination of the unilateral federal procedure and the special arrangements procedure.
With respect to Senate abolition, the Court rejected the Attorney General’s argument that the general amending procedure applies. The Attorney General claimed that since section 42(1)(b) refers to the “powers” and the “number of members” of the Senate, the abolition of the Senate simply involves taking away the Senate’s powers and members. The Court held that the abolition of the Senate “would fundamentally alter our constitutional architecture—by removing the bicameral form of government that gives shape to the Constitution Act, 1867.” It would also entail an amendment of Part V itself, which requires the unanimous approval of the provinces and Parliament. According to the Court, sections 42(1)(b) and (c) are concerned with Senate reform, not its abolition. Indeed, the reform of the Senate presupposes its existence. The framers of the Constitution Act, 1982 expected there to be continued discussions on Senate reform and they therefore provided a mechanism by which such reforms could be achieved. Neither the wording of the constitutional text nor the historical record supported the idea that section 42 encompassed the abolition of the Senate. The Court emphasized the idea that constitutional amendment includes not only “textual modifications” but also “structural modifications” of the Constitution. The abolition of the Senate would structurally modify Part V given the important role of the Senate in the amending procedures. Without the Senate, the amendment procedure (with the exception of the unilateral provincial procedure) would lack an additional step of review and consideration.
There was no dispute that changes to senatorial tenure amounted to a constitutional amendment. The Attorney General argued that changes to senatorial tenure fell within the scope of the federal government’s unilateral amending power in section 44 because such changes were not expressly contemplated in the language of section 42. In addition, he argued that these changes do not engage the interests of the provinces because the proposed terms track the average length of terms that Senators have served.
While the Court agreed that section 42 does not mention the duration of senatorial terms, it held that amendments not mentioned in section 42 do not automatically fall within the scope of section 44. The unilateral amending procedure in section 44 is limited in its scope and does not encompass every conceivable change to the Senate that is not expressly provided for in another provision of the Part V amending formulae. Instead, section 44, “as an exception to the general procedure, encompasses measures that maintain or change the Senate without altering its fundamental nature and role.” Because the Senate is a “core component” of the federal apparatus, changes that affect its fundamental nature and role cannot be implemented unilaterally by the federal government.
The Court determined that imposing fixed terms does engage the interests of the provinces because it would change the fundamental nature or role of the Senate. Its fundamental nature and role is that of a “complementary legislative body of sober second thought.” Senators’ current security of tenure until retirement at age seventy-five provides Senators with political independence. Fixed terms offer less protection from any consequences that may result from the Senators’ review of legislation. Imposing fixed terms would alter the Senate’s nature and role, and therefore such a change can only be achieved through the general amending formula. Imposing term limits is not a minor change even if the term limits correspond to the average term served by Senators. By contrast, in the Upper House Reference the Court concluded that the mandatory retirement at age seventy-five “did not change the essential character of the Senate.” The Court noted that it was a policy issue as to how long a senatorial term had to be in order to safeguard senatorial independence, and for this reason, the input of the provinces was required.
The Court agreed with the Attorney General that the net worth requirement in section 23(4) of the Constitution Act, 1867, under which Senators must have a personal net worth of at least $4,000, could be repealed through the unilateral amending procedure. The Court found that the removal of the net worth requirement would not affect the Senate’s fundamental role as a complementary chamber of sober second thought, nor would it compromise the independence of Senators. For this reason, the interests of the provinces are not engaged. Indeed, none of the provinces objected to the repeal of this requirement.
The real property qualification in section 23(3), however, requires the consent of Québec’s legislative assembly because section 23(6) allows Senators from Québec to either own the requisite real property in the electoral division for which they are appointed, or to reside in that division. Although the removal of the real property requirement would not alter the fundamental nature and role of the Senate, such an amendment would have to take place according to the special arrangements procedure under section 43 of the Constitution Act, 1982. Québec’s legislature would have to consent to the change.
In the Senate Reference, the Court’s interpretation of the amending procedures is based on a fundamental democratic commitment to consultation and deliberation between and among the relevant stakeholders. Constitutional change cannot take place through unilateral decision making by Parliament even if the proposed reforms improve the democratic calibre of a given institution. The disadvantage to the Court’s approach is that large-scale institutional reform is unlikely to take place given the rigours of the amendment process. The advantage, however, is that the Court has prevented future legislatures from unilaterally changing governing institutions in an anti-democratic direction.
There are, however, drawbacks to the Court’s approach to constitutional interpretation and amendment. The Constitution’s “internal architecture,” claimed the Court, serves as a bar to changes of a fundamental nature of the constitutional order even if such changes are not accompanied by any revisions to the constitutional text. To determine if a proposed amendment changed the Constitution’s internal architecture, the Court asked if the proposed amendment altered the Senate’s fundamental nature and role. In so doing, the Court effectively designated itself as the final authority on the Senate’s function in the constitutional order. In the Senate Reference, the Court downplayed the Senate’s crucial role in representing various regions of Canada. Had the Court acknowledged that representation was a fundamental aspect of the Senate’s function, then it would have been much more difficult to explain why consultative elections undermined the Senate’s nature and role. The Senate has an important representative function even though its members are not elected, and this representative function bestows democratic legitimacy on the Senate. The Court’s approach thus freezes into place a constrained view of the Senate’s role and function.
The Senate Reference did not take place in a jurisprudential vacuum. There are important continuities between the Court’s judgment in the Secession Reference and its opinion in the Senate Reference, most notably a commitment to a democratic process for major constitutional change. Unlike the Secession Reference, though, the Court in the Senate Reference was able to base its holding in the constitutional text of Part V of the Constitution Act, 1982. The Court’s invocation of a constitutional duty to negotiate in the Secession Reference was held by some to rest on a somewhat shaky legal foundation. Patrick Monahan observed that the “Court seems to have taken a purely political obligation and converted it into a legal one.” David Haljan likewise noted that negotiation is a political concept rather than a legal proposition. While it is undoubtedly true that negotiation is required for the continuing compromises necessitated by democracy and federalism, it is not for that reason alone a legal or constitutional duty. By contrast, the Senate Reference rests upon much firmer ground because the Court based its decision on a particular interpretation of the constitutional text. Under this interpretation, constitutional change can only take place through widespread democratic deliberation and decision making.
 Reference Re Senate Reform, 2014 SCC 32 at para 111,  1 SCR 704 [Senate Reference].
 Ibid. Part V of the Constitution Act, 1982, Schedule B to the Canada Act 1982 (UK), 1982, c 11 [Constitution Act, 1982].
 Senate Reference, supra note 1 at para 112.
 For a discussion of the Supreme Court’s approach to democracy, see Yasmin Dawood, “Democracy and the Right to Vote: Rethinking Democratic Rights under the Charter” (2013) 51 Osgoode Hall LJ 251; Yasmin Dawood “Electoral Fairness and the Law of Democracy: A Structural Rights Approach to Judicial Review” (2012) 62 UTLJ 499.
 See ibid at para 111.
 Richard Albert argues, for instance, that the proposed reforms are an example of constitutional amendment by stealth. The government’s strategy was to circumvent the formal procedures for amendment by using irregular procedures that would eventually evolve into an established political practice (and hence, protected as a constitutional convention). See Richard Albert, “Constitutional Amendment by Stealth” (2015) 60:4 McGill LJ 673.
 See Claire Hoy, Nice Work: The Continuing Scandal of Canada’s Senate (Toronto: M&S, 1999) at 34.
 David C Docherty, “The Canadian Senate: Chamber of Sober Reflection or Loony Cousin Best Not Talked About” (2002) 8/:3 J Legislative Studies 27 at 38.
 See CES Franks, “The Senate and Its Reform” (1987) 12:3 Queen’s LJ 454 at 464.
 Ibid at 463, citing Robert MacGregor Dawson, The Government of Canada, 5th ed (Toronto: University of Toronto Press, 1954) at 331.
 See David E Smith, “The Senate of Canada and the Conundrum of Reform” in Jennifer Smith, ed, The Democratic Dilemma: Reforming the Canadian Senate (Montréal: McGill-Queen’s University Press, 2009) 11 at 17 [Smith, “Conundrum of Reform”].
 See David E Smith, “The Canadian Senate: What is To Be Done?” in Christian Leuprecht & Peter H Russell, eds, Essential Readings in Canadian Constitutional Politics (Toronto: University of Toronto Press, 2011) at 57 [Smith, “Canadian Senate”].
 See David E Smith, The Canadian Senate in Bicameral Perspective (Toronto: University of Toronto Press, 2003) at 67 [Smith, “Bicameral Perspective”].
 See Hoy, supra note 7 at 297.
 Docherty, supra note 8 at 31.
 Robert A MacKay, The Unreformed Senate of Canada (Oxford: Oxford University Press, 1926) at 175.
 See Docherty, supra note 8 at 32–33.
 Ibid at 38.
 See Colin Campbell, The Canadian Senate: A Lobby From Within (Toronto: Macmillan of Canada, 1978) at 2.
 Docherty, supra note 8 at 37.
 Hoy, supra note 7 at x.
 Reference Re Senate Reform, 2014 SCC 32 (Factum of the Appellant Attorney General of Canada) at paras 32–41 [AG Factum].
 See ibid at paras 41–45.
 See FA Kunz, The Modern Senate of Canada 1925-1963: A Re-Appraisal (Toronto: University of Toronto Press, 1965).
 See Smith, “Bicameral Perspective”, supra note 13 at 102.
 Daniel Pellerin, “Between Despair and Denial: What to Do about the Canadian Senate” (2005) 11:1 Rev Constitutional Studies 1 at 1.
 Jonathan Lemco, “The Futility of Senate Reform in Canada” (1986) 32:3 Australian J Politics & History 405 at 405.
 AG Factum, supra note 22 at para 129.
 See “Bill C-7: Senate Reform Act” The Canadian Bar Association (2012) at 5.
 Bill C-7, An Act respecting the selection of senators and amending the Constitution Act, 1867 in respect of Senate term limits, 1st Sess, 41st Parl, 2001.
 See AG Factum, supra note 22 at para 2.
 Ibid at para 3.
 See ibid at para 2.
 Ibid at para 93.
 See ibid at para 96, citing Reference Re Secession of Quebec,  2 SCR 217 at para 32, 161 DLR (4th) 385 [Secession Reference].
 AG Factum, supra note 22 at para 96, citing Secession Reference, supra note 37 at para 53.
 AG Factum, supra note 22 at paras 84–85.
 Senate Reference, supra note 1 at para 23, citing Secession Reference, supra note 37 at para 32.
 Senate Reference, supra note 1 at para 26, citing Secession Reference, supra note 37 at para 50.
 See Senate Reference, supra note 1 at para 25.
 See ibid.
 Ibid at para 26.
 See ibid at para 27.
 See ibid.
 Ibid at para 34.
 See ibid at para 37.
 Constitution Act, 1982, supra note 2.
 See Secession Reference, supra note 37 at para 38.
 See ibid at para 41.
 See ibid.
 See ibid at para 43.
 Constitution Act, 1982, supra note 2.
 Reference Re Authority of Parliament in Relation to the Upper House,  1 SCR 54, 102 DLR (3d) 1 [Upper House Reference].
 See Senate Reference, supra note 1 at para 47.
 See ibid at para 48.
 Ibid at para 29. Even prior to patriation, there was a long history of the federal government consulting with and securing the consent of the provinces prior to amending the Constitution even though there was no formal obligation to consult. The Court noted the emergence of a constitutional convention under which amendments that engaged provincial interests required substantial provincial consent. See ibid, citing Reference Re Resolution to Amend the Constitution,  1 SCR 753 at 889–95, 125 DLR (3d) 1.
 Senate Reference, supra note 1 at para 31, citing Patrick J Monahan & Byron Shaw, Constitutional Law, 4th ed (Toronto: Irwin Law, 2013) at 204.
 Senate Reference, supra note 1 at para 31.
 See ibid.
 Ronald L Watts, “Bill C-20: Faulty Procedure and Inadequate Solution” (Testimony before the Legislative Committee on Bill C-20, House of Commons, 7 May 2008)” in Jennifer Smith, ed, The Democratic Dilemma: Reforming the Canadian Senate (Montréal: McGill-Queen’s University Press, 2009) 59 at 60.
 Secession Reference, supra note 37 at para 76.
 See ibid at para 77.
 Robert E Hawkins, “Constitutional Workarounds: Senate Reform and Other Examples” (2010) 89:3 Can Bar Rev 513 at 513, 518–26.
 As Emmett Macfarlane argues, the Court’s decision has created considerable ambiguity about what kinds of reforms or changes to existing practices would amount to a constitutional amendment (see Emmett Macfarlane, “Unsteady Architecture: Ambiguity, the Senate Reference, and the Future of Constitutional Amendment in Canada” (2015) 60:4 McGill LJ 883).
 See Senate Reference, supra note 1 at para 51. The Attorneys Generals of Alberta and Saskatchewan were in favour of consultative elections.
 See ibid.
 See ibid at para 60.
 Ibid at para 52.
 See ibid at para 60.
 See ibid at para 58.
 Ibid at para 56.
 Ibid at para 58, citing Canada, Legislative Assembly, Parliamentary Debates on the Subject of the Confederation of the British North American Provinces, 8th Parl, 3rd Sess (6 February 1865) at 36 (John A Macdonald).
 Ibid at para 57, citing Upper House Reference, supra note 59 at 77. There was some debate about the precedential status of the Upper House Reference. The Attorney General was of the view that section 42 was drafted with knowledge of the judgment, and hence was meant to provide guidance so that the definition of the Senate’s “fundamental” features was not left to the courts to define (AG Factum, supra note 22 at para 82). The opposing view is that the Upper House Reference’s insistence on provincial participation provided the backdrop against which nine out of ten provinces agreed to the amendment procedures in Part V (Reference Re Senate Reform, 2014 SCC 32 (Factum of the appellant Attorney General of British Columbia at para 32)).
 See Senate Reference, supra note 1 at para 57.
 See ibid.
 Ibid at para 58 [emphasis in original].
 See ibid.
 See ibid at para 59.
 See ibid, citing Smith, “Bicameral Perspective”, supra note 13 at 169.
 See AG Factum, supra note 22 at para 134.
 Ibid (citing testimony by Christopher P Manfredi from June 2013 on the possible effects of Bill C-7).
 See ibid.
 Constitution Act, 1982, supra note 2, s 42(1)(b).
 See AG Factum, supra note 22 at para 142.
 See ibid.
 See Senate Reference, supra note 1 at para 61.
 See ibid.
 See ibid at para 62.
 See ibid.
 Ibid at para 63.
 The Court’s opinion engages in the fiction that the Senate does bring sober second thought to bear on questions of legislation, while completely ignoring the significant partisan aspect of Senate voting patterns. Partisan pressure, and not popular pressure, is the real threat to senatorial independence.
 Kunz, supra note 24 at 367.
 Serge Joyal, “Conclusion: The Senate as the Embodiment of the Federal Principle” in Serge Joyal, ed, Protecting Canadian Democracy: The Senate You Never Knew (Montréal: McGill-Queen’s University Press, 2003) 271 at 305.
 David E Smith, “The Improvement of the Senate by Nonconstitutional Means” in Serge Joyal, ed, Protecting Canadian Democracy: The Senate You Never Knew (Montréal: McGill-Queen’s University Press, 2003) 229 at 230.
 See Smith, “Canadian Senate”, supra note 12 at 43.
 See Ian Urquhart, “On Senate Reform” (1991) 3:1 Const Forum Const 67 at 69.
 See Kunz, supra note 24 at 370.
 See Reference Re Bill C-7 Concerning the Reform of the Senate, 2013 QCCA 1807 at 4, 370 DLR (4th) 711.
 See Docherty, supra note 8 at 28.
 Smith, “Conundrum of Reform”, supra note 11 at 14.
 AG Factum, supra note 22 at para 47, citing Janet Ajzenstat et al, eds, Canada’s Founding Debates (Toronto: Stoddart Publishing, 1999) at 286–87.
 Smith, “Bicameral Perspective”, supra note 13 at 85.
 Ibid at 88.
 Upper House Reference, supra note 59 at 66.
 Ibid at 67.
 Ibid at 66–67.
 Ibid at 76.
 Ibid at 78.
 I am not saying that the Court equates the absence of elections with the absence of representativeness or that consultative elections increase all relevant forms of representativeness. Instead, my point is that even if the Court had provided a fuller account of the Senate’s representative and democratic function, it is less obvious that consultative elections are in deep conflict with its fundamental nature and role.
 Constitution Act, 1982, supra note 2.
 See Senate Reference, supra note 1 at para 64.
 Ibid at para 65 [emphasis added].
 See ibid.
 See ibid.
 See ibid at para 53.
 Ibid at para 69.
 Reference Re Senate Reform, 2014 SCC 32 (Factum of the Intervener Attorney General of the Northwest Territories) at para 86–87.
 AG Factum, supra note 22 at para 155.
 Senate Reference, supra note 1 at para 97.
 See ibid.
 See ibid at para 99.
 See ibid at para 101.
 Ibid at para 107.
 See ibid.
 See ibid at para 110.
 See ibid at para 71. The five possibilities of term limits were: “(a) a fixed term of nine years for Senators, as set out in clause 5 of Bill C-7, the Senate Reform Act; (b) a fixed term of ten years or more for Senators; (c) a fixed term of eight years or less for Senators; (d) a fixed term of the life of two or three Parliaments for Senators; (e) a renewable term for Senators as set out in clause 2 of Bill S-4, Constitution Act, 2006 (Senate tenure); (f) limits to the terms for Senators appointed after October 14, 2008 as set out in subclause 4(1) of Bill C-7, the Senate Reform Act; and (g) retrospective limits to the terms for Senators appointed before October 14, 2008” (ibid at para 1).
 See ibid at para 72.
 See ibid.
 See ibid at para 74.
 See ibid at para 75.
 Ibid at para 77.
 See ibid at para 78.
 Ibid at para 79.
 See ibid at para 80.
 See ibid at para 79.
 See ibid at para 80.
 Upper House Reference, supra note 59 at 76–77.
 See Senate Reference, supra note 1 at para 82.
 See ibid at para 86.
 See ibid at para 88.
 See ibid at para 89.
 See ibid at para 91.
 See ibid. Another option would be to specify that the removal of the real property qualification in section 23(3) did not apply to the Senators from Québec (ibid at para 94).
 Patrick J Monahan, “The Public Policy Role of the Supreme Court of Canada in the Secession Reference” (1999) 11 NJCL 65 at 68.
 David P Haljan, “A Constitutional Duty to Negotiate Amendments: Reference Re Secession of Quebec” (1999) 48 ICLQ 447 at 455–56.