Calls for consociational representation for FMCs

The adoption of the Charter of Rights and Freedoms in 1982 provided francophones with a number of explicit consociational guarantees. The Charter's sections 16 to 23 address official-language protections. Building on these protections, courts have progressively expanded the polity-based education and healthcare-management rights of FMCs. According to Foucher, this jurisprudence has in effect affirmed FMCs’ right “to live in their own language” (Reference Foucher and Wallot2005: 146).

Theorists, meanwhile, have explored whether FMCs are owed, or indeed already enjoy, constitutionally protected cultural self-rule—what has been termed “non-territorial autonomy” (Chouinard, Reference Chouinard2014; Elkins, Reference Elkins1992; Nieguth, Reference Nieguth2009; Poirier, Reference Poirier, Thériault, Gilbert and Cardinal2008, Reference Poirier2012). Representation is often viewed as a corollary of such autonomy (Kymlicka, Reference Kymlicka1995: 32). Indeed, certain of the above authors (Elkins, Reference Elkins1992: 16), as well as others (Leger-Haskell, Reference Leger-Haskell2009: Magnet, Reference Magnet1995), have proposed apportioning polity-based representation to FMCs. Francophone advocacy groups have at times pressed for such representation. For example, during the debate over the Charlottetown Accord, groups recommended that one senator from each province represent the official-language minority of that province (Kymlicka, Reference Kymlicka and Seidle1993: 62).

At the same time, traditionally francophone districts have increasingly come under threat. Thrice recently, FMCs in New Brunswick and Nova Scotia, drawing in part on language rights unique to “national groups,” have challenged electoral maps that submerged them into anglophone districts in the name of voter parity. The ensuing cases were all decided in favour of the FMCs. Yet the rulings, and the legislative and scholarly discussions flowing therefrom, lacked theoretical clarity. As I will show, despite appeals by FMCs for representation that flows from their consociational status as a “national group,” courts have failed to say whether such rights exist.

Calls for consociational representation for Indigenous peoples

Indigenous peoples, too, are among Canada's “national groups.” Prior to colonization they were sovereign; in recent decades they have called for internal self-determination (Coulthard, Reference Coulthard2014: 64). Indigenous rights and protections are variously said to be rooted in natural rights, historic proclamations and treaties, international law, and in modern Canadian political, constitutional and jurisprudential developments. In the Constitution Act, 1982 and the Charter of Rights and Freedoms, two key Indigenous rights were recognized. Section 35 of the Constitution Act affirmed certain Indigenous “existing rights,” including (per federal government and court interpretations) the “inherent right of self-government.” Meanwhile, section 25 of the Charter often called the “non-derogation clause,” anticipated clashes between individual and Indigenous rights, buffering—perhaps even blocking—diminution of the latter (Arbour, Reference Arbour2003).

Some scholars have suggested the “inherent right of self-government” carries with it a corollary right to representation in public government (Schouls, Reference Schouls1996: 739). Others have suggested guaranteed representation is owed to Indigenous peoples as a consequence of their cession of sovereignty in the same way British Columbia and Newfoundland acquired seats in Parliament in exchange for joining Canada (Knight, Reference Knight2001: 1108). At least one scholar has proposed that certain historic treaties may guarantee Indigenous representation (Ladner, Reference Ladner1997). Finally, some thinkers suggest Indigenous peoples are owed power in Parliament because of their unique constitutional status as fiduciary dependents (Royal Commission on Electoral Reform, 1991: 182).

During the 1990s, numerous plans for guaranteed Indigenous representation were drafted. In 1991, the Royal Commission on Electoral Reform proposed creating Aboriginal Electoral Districts (1991: 182). In 1992, the Charlottetown Accord included provisions for Indigenous representation in Parliament. In 1995, the Liberal government's “Inherent Right Policy” urged “specific guarantees” of Indigenous representation in public government. In 1996, the Royal Commission on Aboriginal Peoples suggested Indigenous self-government might include “sharing power in joint governmental institutions, with guaranteed representation for the nations and peoples involved” (1996: 106). In the 1990s, Quebec, New Brunswick, Nova Scotia and the Northwest Territories explored, but did not implement, guaranteed Indigenous representation (Niemczak and Jutras, Reference Niemczak and Jutras2008).

Today, in three provinces, electoral boundaries laws give Indigenous peoples specific consideration. In Alberta, the presence of “an Indian reserve or a Metis settlement” is among multiple factors that, taken together, qualify up to four districts for “exceptional” departure from voter parity. In Newfoundland and Labrador, one district enjoys special exemption from parity largely on the grounds “that persons of Aboriginal descent form the majority.” In Ontario, “representation of Indigenous people” is among the reasons two low-population, heavily Indigenous districts were formed in 2017.

Beyond these narrow exceptions, polity-based Indigenous representation has gone unimplemented, and discussions surrounding it have faded. This is in part due to hesitations among Indigenous peoples to adopt alien institutions (White, Reference White1993b) or legitimize colonial rule (Knight, Reference Knight2001: 192). But it is also due to vigorous non-Indigenous opposition to, and lack of legal clarity surrounding, the integration of consociational Indigenous representation with existing representational rights of Canadians as individuals (Schouls, Reference Schouls1996: 748).

This clash was exemplified by the first and only charter challenge to confront Indigenous consociational representation. In Friends of Democracy v. Northwest Territories, the court in effect condemned the notion that Indigenous power sharing could permissibly compromise the voting rights of Canadians at large. As this article will show, the reasoning of the court sits uncomfortably with liberal theory, and, indeed, is at odds with a seminal subsequent ruling relating to Indigenous section 25 and 35 rights.

The individualism principle in apportionment

In liberal political theory, individualism is the principle that the irreducible rights-bearing unit is the individual (Kymlicka, Reference Kymlicka1989: 140). Per this principle, the state should not reward, punish or prescriptively categorize individuals on the basis of group affiliations (for example, race, class, gender) but rather should treat them in a manner that is “difference blind.” Assessing apportionment in the light of individualism, then, involves determining whether a districting scheme is “difference blind” versus whether (and in what way) it subsumes individuals into groups.

As individualism hinges on “blindness,” a maximally liberal representational scheme might be expected to take no note of voters’ affiliations. Yet in many electoral systems, including the first-past-the-post system of Canada, this would be illogical and even intolerable. As Karlan observes, “The instrumental purpose of voting—having one's preferences taken into account in choosing public officials—necessarily involves aggregating the votes of individuals to achieve a collective outcome” (Reference Karlan1993: 249). Moreover, for voting to have meaning, apportionment must aggregate not just any electors but those who share politically salient interests.

Achieving meaningful aggregation requires deliberate departure from difference blindness, making apportionment a rare instance where liberalism embraces difference-conscious lawmaking. Hence, few apportionment schemes are individualistic on their face. Indeed, some that are ostensibly individualistic have been judged unconstitutional precisely because they fail to provide power to, and thus abridge the rights of, voters of certain groups (Issacharoff et al., Reference Issacharoff, Karlan and Pildes2007: 538). For example, in the United States, courts have found that the politically salient interests of residents of racial minority neighbourhoods are denied when their electoral preferences are drowned out via citywide at-large apportionment.

It must be emphasized, however, that liberal tolerance for grouping voters for the purpose of districting is distinct from assigning rights to groups themselves (Gerken, Reference Gerken2001). The US Supreme Court emphatically denied in Shaw v. Hunt that the “right to an undiluted vote … belongs to the minority as a group and not to its individual members. It does not” (1996: §917). Instead, aggregating voters who share politically salient interests is said to provide each voter with a meaningful vote. Hence, in the above scenario, it is not the racial minority neighbourhood that bears rights, but the individual residents therein.

It must also be noted that, though liberalism embraces aggregation for purposes of representation, it may condemn certain types of aggregations. This hostility typically relates to the kind of group being recognized and, sometimes, to the overtness of that recognition. The most common method of aggregating voters is by territory, which requires eschewing “blindness” only so voters may be grouped based on the commonality of where they live. Liberalism's acceptance of geographic aggregation is underscored by its embrace of the traditional districting principles of “contiguity” and “compactness.”

As well, representation may be apportioned to voters who form a “community of interest.” Liberal justice usually condones, and may even insist upon, grouping voters by community-of-interest-related factors that correspond easily with proximity, such as socioeconomic level, cultural heritage, employment type or municipal residence. Such aggregations lead to districts that are, for instance, predominantly blue collar or Italian-American or composed of military personnel or limited to residents of a specific city. More controversial are groupings that hinge on immutable, politically divisive traits like race. In the US, the legality of so-called “affirmative racial gerrymanders” has been hotly contested, especially when such gerrymanders defy geographic compactness, resulting in odd-shaped districts.

In Canada, apportionment questions relating to liberal individualism have provoked legislative, though not constitutional, controversy. As representatives in Canada are elected from geographic districts, voters must be aggregated by proximity. Moreover, in affirming such districting principles as “community of interest” and “minority representation” (Courtney, Reference Courtney2001: 159), courts have confirmed that the Charter rejects “difference blindness” in favour of aggregating individuals by commonalities beyond mere proximity. The limits of such aggregation are contested. As Pal notes, boundary makers have long wrangled over the proper definition of “community of interest,” disagreeing as to whether groupings based on race and ethnicity are desirable or, conversely, intolerable (Reference Pal2015: 258). Courts have had little to say on this matter (Courtney, Reference Courtney2001: 168). Even less judicially clear is whether “community of interest” aggregation may, or indeed must, compromise other districting values, such as those that I will discuss next, related to egalitarianism.

The egalitarianism principle in apportionment

In liberal theory, egalitarianism holds that all individuals are moral equals and should be treated as such, enjoying legal parity vis-à-vis one another (Kymlicka, Reference Kymlicka1989: 140). Assessing apportionment in the light of egalitarianism requires determining whether an electoral map treats individuals as equals or whether (and to what degree) it instead overrepresents some and underrepresents others.

As liberalism rests on the political equality of individuals, then a maximally liberal apportionment scheme would provide representation that is “equal.” But as Pitkin famously observed (Reference Pitkin1967), representation is a concept understood in multifarious ways. How one understands it affects whether one feels it has been apportioned equally. I suggest there are many dimensions of representational egalitarianism, of which three are relevant here: formal equality, substantive equality and “community of interest” equality.

Formal equality is said to result when apportionment adheres to representation by population. Under strict “rep by pop,” representatives are elected by and/or represent equal numbers—of people, citizens, qualified voters, or some other subset of individuals. This practice purportedly gives electors equal “power” or provides constituents equal “weight.” Where districts are not equipopulous, they are “malapportioned.” Individuals in districts with a greater population than average are “underrepresented” and their voting power “diluted.” Individuals in low-population districts are in turn “overrepresented.”

Of course, formal egalitarianism is not the only way representation may be “equal.” Another way is via “substantive egalitarianism,” valuing equality of outcome. In Canada, this value is captured in the concept of “effective representation.” In Dixon v. British Columbia, the BC Supreme Court identified two essential functions of representation: the “legislative role,” performed when legislators cast votes, and the “ombudsperson role,” where representatives act as liaisons between constituents and government (1989: 29). The ombudsperson role is often said to be unusually difficult in certain types of districts, such as those that are geographically large or remote. Egalitarian liberals may thus insist that voters in large or remote districts be numerically overrepresented. In Canada, such overrepresentation is common. It eschews formalistic parity in favour of “substantive” parity, in which it is not the “weight” or “power” of voters that is equal but the “effectiveness” of representation they receive.

A third dimension of representational egalitarianism is “community of interest” equality. As noted previously, voters who share politically salient concerns form communities of interest. Where such communities are split between multiple electoral districts (“cracked”), or drowned within a larger district (“stacked”), voters’ ability to elect their favoured candidate, and thus to have their politically salient interests heard, may be thwarted. They consequently suffer unequal treatment vis-à-vis members of unimpaired communities of interest. As Dixon observes, apportionment that achieves perfect numeric parity and yet preferences the politics of voters belonging to only certain interest groups is nonetheless inegalitarian (Reference Dixon1968: 272). As with substantive egalitarianism, community of interest egalitarianism may in Canada permit, or even require, non-equipopulous districting (Stephanopolous, Reference Stephanopolous2013: 816). Some scholars have speculated that the Charter not only allows but may even mandate creation of districts for small but distinct communities of interest (Knight, Reference Knight2001: 1109).

Unlike apportionment questions relating to individualism, questions concerning egalitarianism have in Canada been legally contentious. Departure from formal egalitarianism was once considerable. Adoption of the Charter resulted in malapportionment challenges, including the landmark Reference re Provincial Electoral Boundaries (Saskatchewan), better known as the Carter case. The Supreme Court of Canada's sole apportionment decision, Carter held that the right to vote guarantees “not equality of voting power per se, but the right to ‘effective representation.’” While “relative parity” is the principal requirement of effective representation, such representation must also take into account substantive equality and must consider the “effective representation” of members of various sorts of aggregations, including communities of interest.

As a consequence of Carter, Canada eschews the strict US standard of “one person, one vote.” Formal equality and meaningful aggregation are balanced. Still, subsequent lower-court decisions have affirmed a rough guideline for permissible deviation from parity of +/−25 per cent. (Parliament and the majority of Canada's provinces employ this standard. A few provinces have tighter guidelines.) Under the +/−25 per cent standard, “effective representation” may be pursued through the formation of districts that vary in population as much as 25 per cent above or below average. Courts have suggested deviations beyond this limit are highly suspect, except in “exceptional circumstances.”

The universalism principle in apportionment

Disputes involving individualism and egalitarianism will be familiar to students of voting rights. In Canada, conflicts over apportionment typically relate to one or the other of these principles. However, apportionment must be considered in an additional dimension, informed by universalism. This is because, like all federal, consociational or otherwise “compound” states, Canada is not “universal.”

Universalism, according to Gray, is the liberal principle “affirming the moral unity of the human species and according a secondary importance to specific historic associations and cultural forms” (Reference Gray1995: xii). Universalism holds that peoples (nations, linguistic groups, religious groups and so on) are not politically primordial; they do not bear rights qua group. Rather, the sole rights-bearing collectivity, and thus the only proper demos, is the whole. “Universal” political systems consider their citizens to form a single, indivisible polity. Asch calls universalism “the true ‘one person-one vote’ orientation to democracy” (Reference Asch1990: 94).

In theory, of course, universalism precludes statehood. But universalism may be seen to stand at odds with another cherished political principle, self-determination. Self-determination holds that peoples may freely choose their political status. For a plethora of reasons, there has come to exist what Margalit and Raz call a “core consensus” supporting the right to self-determination (Reference Margalit and Raz1990: 439). Liberals are key members of this consensus; despite universalist convictions they typically accept and even champion certain expressions of self-determination (MacMillan, Reference MacMillan1998: 127). Self-determination, in turn, gives rise to non-universalism, of both an external and internal variety.

State formation is the clearest manifestation of external self-determination. Many states, particularly those comprising a single self-determining people, constitute themselves so representation attaches solely to individuals. Other states, constituted by multiple collaborating polities, choose “compound” arrangements, which preserve for each polity a measure of internal self-determination. Federal non-universalism provides shared and self-rule to polities encapsulated in territorial units; in the subspecies of “multinational” federalism, such as that of Canada, certain encapsulated polities are peoples, who thereby enjoy de facto internal self-determination. Consociational non-universalism, meanwhile, provides direct and de jure internal self-determination to peoples even when they are territorially diffuse. Consociations may feature elaborate “grand coalitions” of ethnic blocs, as in Lebanon, or may involve more limited ethnonational power sharing, as in Canada.

When polities join to constitute a compound state, be it federal, consociational or otherwise, they must determine how to subdivide power. This requires confronting a thorny puzzle of democratic theory. Democracy is “the people deciding,” but how should they decide? Paradoxically, the “how” of democracy cannot be resolved democratically. Issacharoff calls this a “first order” dilemma, as it involves defining the powers of the self-determining demoi upon which democracy is based (Reference Issacharoff2008: 232). “First order” dilemmas are ideally resolved before, or at least rendered moot by, the instantiation of the state. In the classic US case, power was divided pre-politically, with framers striking an agreement and etching it into an inviolable charter.

Only after “first order” questions are dealt with can framers formalize what may be considered “second order” rights, rights that address how power should be distributed to individuals within polities. “Second order” rights, of course, include difference-blind and egalitarian apportionment. It may thus be said that decisions related to the principle of universalism are “constitutionally prior” to those concerning individualism and egalitarianism. As the case of Canadian federalism makes clear, “first order” apportionment produces knock-on effects on “second order” principles. The fact that Newfoundland and British Columbia have the same number of senators violates egalitarianism, by providing Newfoundlanders nine times more senatorial clout per capita than British Columbians.

While these knock-on effects are not without controversy, theorists seem to view them as more tolerable than those generated by other non-universal forms, such as consociation. Consociation is not only inegalitarian but also flies in the face of individualism (McCrudden and O'Leary, Reference McCrudden and O'Leary2013: 35; Steiner, Reference Steiner1990: 1551; Wippman, Reference Wippman1998: 231). Consociation, like federalism, accords power to non-equipopulous groups, but unlike federalism, it assigns individuals to those groups based not on where they live but on who they are, their ethnicity, religion, language and so forth. In this way, consociation is distinctively and unapologetically difference-conscious. Hence, classical liberals such as Barry (Reference Barry1975) seem especially critical of consociation.

Regardless of such opinions, when non-universal structures are constitutionally entrenched, their knock-on effects are ipso facto constitutional. The overweighting of small provinces in Canada's federal scheme, or the overrepresentation of Quebec on the Supreme Court, is effectively beyond challenge. Likewise, the world's best-known consociation has thus far proven legally immune. Belgium's power-sharing arrangement apportions representation directly to French and Walloon polities, producing consequent distortions of individual and egalitarian rights. Twice, lawsuits challenging these distortions have come before the European Court of Human Rights; both suits were unsuccessful. In effect, the court deemed these distortions unavoidable “second order” results of a non-universal scheme that was constitutionally prior (McCrudden and O'Leary, Reference McCrudden and O'Leary2013).

Far more vexing are cases of emergent non-universalism (Pildes, Reference Pildes and Choudhry2008: 173). Such cases, though common, are not well theorized (Requejo and Nagel, Reference Requejo and Nagel2017: 9). They arise where new claimants to the right of self-determination emerge in states that are already constituted. States that reject emergent claims out of hand exhibit a double standard: As Van Dyke observes, “[In] a multinational state, it is as inappropriate to think of majority rule as it would be in the world as a whole” (Reference Van Dyke1985: 172).

Yet addressing such emergent demands requires engaging with even thornier challenges than those resolved by constitutional framers. This is for two reasons. First, emergent claims arise where rules concerning the “how” of democracy are already entrenched. In Canada, all the seats at the power-sharing table were long ago assigned to individuals and federal polities. Where occupants possess seats by right, they are likely to jealously guard them (Requejo and Nagel, Reference Requejo and Nagel2017: 14). Emergent polities, such as Indigenous peoples and FMCs, do not enjoy the luxury of pressing their claims pre-politically, in the vacuum of a constitutional convention. They must instead jockey for space at a table that is already full.

The second reason emergent claims are tricky is because they problematize not just the “how” of power sharing, but also the “who.” If democracy is “the people deciding,” and the “people” are not universal, then who are the peoples? Are emergent claimants legitimate rights-bearers, or pretenders to the throne? The aforementioned “core consensus” on the right of self-determination is, Margalit and Raz note, “but the eye of a raging storm concerning the precise definition of the right, its content, its bearers, and the proper means of its implementation” (Reference Margalit and Raz1990: 439). Thinkers feud endlessly not merely over whether this or that group deserves this or that degree of sovereignty, but over what principles should guide the investigation. Theory, then, like democracy, cannot resolve the “who” of power sharing. Still, I think we can carve it down to size.

Canada, it seems, features three categories of emergent power-sharing claimants. The first are groups who are owed self-determination—by their constitutional status, international law, or otherwise—but wish not to share power within the public institutions of the state. This is perhaps because they deem the state illegitimate or seek secession or demand power sharing confederally (with the state, not within it) or wish merely for internal autonomy without shared rule. At various times, for various reasons, certain Indigenous and francophone groups have identified with this category. I cannot assess the validity of their secessionist, autonomist or decolonial demands. What I think is certain is that, if they wish not to share power within Canada, they should not be forced to do so.

The second category of emergent power-sharing claimants are the inverse of the first: groups that do wish to share power but are clearly not owed it. Such a group was at the centre of the famous US apportionment case Reynolds v. Sims. There, the Supreme Court condemned Alabama's practice of assigning state senate seats by county rather than population. Chief Justice Warren, deeming analogies to the special case of the federal Senate “inapposite,” blasted the notion that counties are rights-bearing polities. He stated, “Political subdivisions of states … never have been considered as sovereign entities” (1964: 377). Similarly, in Canada's aforementioned Carter case, the court's minority condemned the non-universalism of Saskatchewan's “strict quota of urban and rural ridings” in which the latter was protected qua polity. (The majority found the Saskatchewan map sufficiently egalitarian and thus, in effect, declared concerns about non-universalism moot.)

The final emergent claimants in Canada are the ones relevant to this article: those who wish to share power within the state and seem morally and/or legally owed it. As noted previously, FMC claimants might be owed internal self-determination based on constitutional guarantees; Indigenous claimants might be owed it on constitutional grounds as well as due to natural rights, historic proclamations and treaties, and international law. Yet “who” questions remain. If Indigenous peoples are owed self-determination, are they owed it collectively, or are First Nations, Metis and Inuit owed it separately? Are FMCs one polity, or are Acadians distinct from Ontario francophones? Also, might additional groups be valid claimants? Do African-Nova Scotians deserve internal self-determination? Do Doukhobors? These are questions I must leave for further study (or legal challenge). My purpose, after all, is not to prove that “national groups” are owed power sharing, but merely to show that if they are, and if they want it, their claims should be approached through the lens of non-universalism. This, I will now show, has not been done.

Adjudication of Indigenous representation

In the 1999 case Friends of Democracy v. Northwest Territories, the Northwest Territories’ Indigenous-majority legislature was accused of violating the section 3 voting rights of residents in the predominantly non-Indigenous city of Yellowknife. There, several electoral districts were severely underrepresented, one exceeding parity by +152 per cent. The territorial government defended the scheme, citing “substantive” and “community of interest” needs of districts outside Yellowknife. Indigenous interveners, meanwhile, presented a non-universal defense. They maintained that the existing balance of ethnonational power in the territory was protected by sections 35 of the constitution and 25 of the Charter, rights that would be violated by providing additional seats to Yellowknife. In effect, the interveners argued that Indigenous peoples, as a consociating “national group,” possessed a non-universal right to a fixed share of representation in the Northwest Territories government. Without such a right, they feared their homeland would be swamped and democratically dominated by “settlers.”

The court disagreed. Ruling in Yellowknife's favour, it decreed that the legislature's apportionment scheme was impermissibly inegalitarian. It expressed doubt that section 35 provides a guarantee of non-universal Indigenous representation, and, moreover, questioned whether section 25 in fact trumps individual voting rights. Analyzed through the lens of liberal theory, the court in effect ruled that “second order” egalitarianism (requiring that districts be of relatively equal size) trumps “first order” non-universalism (requiring that Indigenous rights not be “derogated from”). Unsurprisingly, the ruling was controversial. It precipitated a “constitutional crisis” in the Northwest Territories (Northwest Territories, 1999a: 66), with the territory's umbrella First Nations organization calling for Ottawa to dissolve the territorial legislature—a move that would thwart “settler” takeover. It also ended longstanding efforts to devise a territorial constitution that would formally enshrine ethnonational power sharing (Northwest Territories, 1999b: 4).

At least one legal scholar deemed the Friends interpretation of section 25 “the only possible exception” to the common judicial understanding of the non-derogation clause (Morse, Reference Morse2002: 421). Indeed, the Friends interpretation was soon rejected in Campbell v. British Columbia. Handed down by the British Columbia Supreme Court in 2000, Campbell has been called Canada's most significant case involving Indigenous peoples and section 3 (Morse, Reference Morse2002: 394). In it, applicants accused the Nisga'a First Nation self-government agreement of abridging charter-protected voting rights of non-Nisga'a. The court affirmed that this was true—but, in effect, it held that such “second-order” abridgement was protected under the “first-order” non-derogation guarantees of section 25 (Isaac, Reference Isaac2002: 444).

While Campbell involved Indigenous self-rule rather than shared rule, the decision suggests that, if guaranteed Indigenous representation is indeed a right under section 35, the “first order” shield of section 25 should protect that right from charges of epiphenomenal section 3 inegalitarianism. This view presents a challenge to much of the scholarly and legal discourse on guaranteed Indigenous representation. Again, as Schouls (Reference Schouls1996) has observed, attempts to apportion polity-based representation to Indigenous peoples have foundered due to incompatibility with egalitarian representational requirements. Arguably, the “second order” cart has been placed before the “first order” horse. Ladner discerned this perverse circumstance when she noted that most “studies have focused on integrating Aboriginal representation within the existing electoral scheme. … If they had examined guaranteed representation as a pre-existing right … they might have arrived at different, more consistent conclusions” (Reference Ladner1997: 86).

Adjudication of FMC representation

Charter cases have also arisen concerning non-universal representation rights of francophones. The first case, Raîche v. Canada, was a 2004 challenge in which FMC voters in New Brunswick protested a federal apportionment plan that, to increase parity, transferred them from a francophone-majority district to an anglophone one. The Federal Court of Canada agreed, determining that the boundaries commission had erred in two respects. First, the commission violated the Electoral Boundaries Readjustment Act by placing too much importance on parity and too little on preserving “communities of interest.” Second, it violated the Official Languages Act, which requires that the federal government “enhanc[e] the vitality of the English and French linguistic minority communities.”

In discussing both errors, the court called the aggrieved francophones a “community of interest.” It is not clear this description fits. Again, in apportionment, a “community of interest” comprises voters who share politically salient interests, aggregated to provide each voter with a voice. Yet the protection of francophone representation under the Official Languages Act does not operate in this manner. The protection attaches not to individuals but group—and not just any groups, but exclusively to two “national groups.” Similarly, much of the scholarly discussion surrounding Raîche has focused on whether boundary commissions must accord greater import to preserving “communities of interest” (Pal, Reference Pal2015: 253). While this may be so, insufficient attention has been paid to Raîche’s “first order” implications. In Raîche, the Official Languages Act provided New Brunswick FMCs with a similar sort of consociational protection that, in Campbell, section 25 provided to the Nisga'a: a shield, buffering “national groups” from charges of individual voting-rights abuse. Raîche in part succeeded because FMCs are not a run-of-the-mill “community of interest” whose aggregation and overrepresentation merely facilitate “effective representation” (per Carter), but rather are a distinct polity whose right to representation flows from, and is guarded by, Canada's antecedent non-universalism.

Since Raîche, two more court cases have explored alleged violations of FMC representational rights. L'Association francophone des municipalités du Nouveau Brunswick et al. v. New Brunswick challenged a 2013 New Brunswick provincial redistricting map that increased formal parity by submerging several FMCs into anglo districts. The plaintiffs claimed the plan breached multiple charter provisions, including section 3 (“effective representation”) and section 16.1, requiring that New Brunswick promote the equal “status, rights and privileges” of anglo and francophone communities (New Brunswick Court of Queen's Bench, 2014: 5). The suit was settled out of court, with the province amending its apportionment legislation. Henceforth, New Brunswick boundaries commissions “shall consider the effective representation of the English and French linguistic communities in complying with section 3” (New Brunswick, 2014). In the case of other “communities of interest,” meanwhile, commissions merely “may” depart from voter parity. Arguably, then, the province has singled out official-language minorities as deserving particular attention when balancing “second order” values of parity versus “community of interest.” Yet it remains unclear whether francophones, qua polity, are recognized as bearing “first order” rights.

The most recent case, decided in January 2017, was Reference re the Final Report of the Electoral Boundaries Commission, arising from Nova Scotia's 2012 provincial reapportionment. There, the legislature had tasked the boundaries commission with insuring that no district exceeded parity beyond +/−25 per cent. This constraint compelled the commission to erase all three of the province's significantly overrepresented FMC “protected seats,” which had existed for 20 years. The Nova Scotia attorney general submitted a reference to the provincial court of appeals inquiring about the constitutionality of the new map. Nova Scotia's francophone association intervened, blasting the province for treating French speakers as no different from other communities of interest. The interveners suggested section 3 of the Charter must be interpreted in the context of other charter provisions, such as sections 16 through 23, protecting official-language minority communities.

In its ruling, the court held that the province had indeed breached section 3, but for reasons related to process, not content. The court maintained that the constraint placed on the boundaries commission had prevented it from weighing parity against other constitutional requirements, including providing effective representation to “communities of interest.” Having condemned this process, the court did not bother to assess the constitutionality of the resulting map. It thus did not explore whether the new boundaries violated the rights of FMCs, either as a run-of-the-mill “community of interest” or as a non-universal polity.

It can thus be seen that, since Raîche, francophones have demanded protection of their representation on the grounds that FMCs bear polity-based rights as a “national group.” It can also be seen that courts and legislators have not acknowledged or fully responded to these appeals, instead conceiving of FMCs as a “community of interest” that at best is owed distinctive “second order” attention. It therefore appears that, in the eyes of the court, the rights of FMCs may diverge in degree, though not in kind, from those of non-”national groups.”