Supreme Court Judgments

  

 

SUPREME COURT OF CANADA

 

Citation: Conseil scolaire francophone de la Colombie-Britannique v. British Columbia, 2020 SCC 13, [2020] 1 S.C.R. 678

Appeal Heard: September 26, 2019 Judgment Rendered: June 12, 2020 Docket: 38332

 

Between:

 

Conseil scolaire francophone de la Colombie-Britannique, Fédération des parents francophones de Colombie-Britannique, Annette Azar-Diehl, Stéphane Perron and Marie‐Nicole Dubois

Appellants

 

and

 

Her Majesty The Queen in Right of the Province of British Columbia and Minister of Education of British Columbia

Respondents

 

- and -

 

Attorney General of Nova Scotia, Attorney General of Prince Edward Island, Attorney General of Saskatchewan, Attorney General of Alberta, Attorney General of Newfoundland and Labrador, Attorney General of the Northwest Territories, Commissioner of Official Languages of Canada, Quebec Community Groups Network, David Asper Centre for Constitutional Rights, Association des juristes d’expression française du Nouveau-Brunswick inc., Association des enseignantes et enseignants francophones du Nouveau-Brunswick inc., Fédération nationale des conseils scolaires francophones, Association des parents de l’école Rose-des-Vents, Association des parents de l’école des Colibris, Canadian Association for Progress in Justice, Société de l’Acadie du Nouveau-Brunswick, Fédération des conseils d’éducation du Nouveau-Brunswick, Assembly of Manitoba Chiefs, Commission nationale des parents francophones, Conseil scolaire francophone provincial de Terre-Neuve-et-Labrador and Canadian Francophonie Research Chair in Language Rights

Interveners

 

 

Official English Translation: Reasons of Wagner C.J.

 

Coram: Wagner C.J. and Abella, Moldaver, Karakatsanis, Côté, Brown, Rowe, Martin and Kasirer JJ.

 

Reasons for Judgment: (paras. 1 to 187)	Wagner C.J. (Abella, Moldaver, Karakatsanis, Côté, Martin and Kasirer JJ. concurring)
Joint Reasons Dissenting in Part: (paras. 188 to 348)	Brown and Rowe JJ.

 

 

 

Conseil scolaire francophone de la Colombie‐Britannique,

Fédération des parents francophones de Colombie‐Britannique,

Annette Azar‐Diehl,

Stéphane Perron and

Marie‐Nicole Dubois                                                                                     Appellants

v.

Her Majesty The Queen in Right of the Province of British Columbia and

Minister of Education of British Columbia                                            Respondents

and

Attorney General of Nova Scotia,

Attorney General of Prince Edward Island,

Attorney General of Saskatchewan,

Attorney General of Alberta,

Attorney General of Newfoundland and Labrador,

Attorney General of the Northwest Territories,

Commissioner of Official Languages of Canada,

Quebec Community Groups Network,

David Asper Centre for Constitutional Rights,

Association des juristes d’expression française du

Nouveau‐Brunswick inc., Association des enseignantes

et enseignants francophones du Nouveau‐Brunswick inc.,

Fédération nationale des conseils scolaires francophones,

Association des parents de l’école Rose‐des‐Vents,

Association des parents de l’école des Colibris,

Canadian Association for Progress in Justice,

Société de l’Acadie du Nouveau‐Brunswick,

Fédération des conseils d’éducation du Nouveau‐Brunswick,

Assembly of Manitoba Chiefs,

Commission nationale des parents francophones,

Conseil scolaire francophone provincial de Terre‐Neuve‐et‐Labrador and

Canadian Francophonie Research Chair in Language Rights                Interveners

Indexed as: Conseil scolaire francophone de la Colombie‐Britannique v. British Columbia

2020 SCC 13

File No.: 38332.

2019: September 26; 2020: June 12.

Present: Wagner C.J. and Abella, Moldaver, Karakatsanis, Côté, Brown, Rowe, Martin and Kasirer JJ.

on appeal from the court of appeal for british columbia

                    Constitutional law — Charter of rights  — Minority language educational rights — Provincial funding of minority language education system — Sliding scale — Substantive equivalence — Justification of infringements — Approach to take in order to situate given number of students on sliding scale so as to determine level of services that must be provided to them — Whether test used to assess quality of educational experience provided to official language minorities varies with number of minority language students — Whether infringements of this right are justified –– Canadian Charter of Rights and Freedoms, ss. 1 , 23 .

                    Constitutional law — Charter of rights  — Remedy — Damages — Trial judge deciding that province had to pay damages to school board to make up deficit it had run because of freeze on funding for school transportation — Whether limited government immunity from damages awards applies to decisions made in accordance with government policies that are found to be contrary to s. 23  of the Charter .

                    The Conseil scolaire francophone de la Colombie‐Britannique (“CSF”) is the sole French‐language school board in British Columbia. Its territory covers the entire province, and it has 37 schools. In June 2010, the CSF, the Fédération des parents francophones de Colombie‐Britannique and three parents who are rights holders under s. 23  of the Charter  (“linguistic minority representatives”) filed a notice of civil claim against the province, submitting that several aspects of the funding of the education system penalized the official language minority and infringed its rights under s. 23  of the Charter . The alleged infringements can be divided into two categories: the first involved systemic claims (among other things, the fact that the CSF had not received an annual grant for building maintenance, the formula used to set priorities for capital projects, a lack of funding for school transportation and a lack of space for cultural activities), and the second involved claims for the purpose of obtaining new schools or improvements to existing schools in 17 communities.

                    The trial judge set out an approach to be taken in order to situate the number of students in a given community on the sliding scale, which serves to determine the level of services to which an official language minority is entitled and is used to decide whether the minority is entitled to a homogeneous school, to educational facilities shared with the majority or to another appropriate solution. Applying her analytical framework, the trial judge issued declarations concerning the right to educational facilities in several communities. She then outlined the test that is to be applied in determining the quality of the educational experience that must be provided to official language minorities. She concluded with respect to several communities that the children of rights holders are entitled to facilities that provide them with an educational experience that is substantively equivalent to the experience of the majority, but she found with respect to other communities that the numbers of children of rights holders warranted not a substantively equivalent educational experience, but one that is proportionately equivalent to the educational experience provided to the majority. Discussing the principles of interpretation that must inform the analysis of infringements of s. 23 under s. 1  of the Charter , the trial judge then concluded that several infringements of the rights holders’ language rights were justified under s. 1. Finally, she concluded that awarding damages would not be appropriate for most of the claims of the linguistic minority representatives, but she found that the freeze on funding for school transportation at a time when the number of students of the linguistic minority was rising constituted an infringement of s. 23, and she awarded $6 million in damages to the CSF. On the other hand, the trial judge declined to award damages to the CSF in compensation for its having been denied the Annual Facilities Grant Rural Factor.

                    The linguistic minority representatives appealed the trial judge’s judgment, arguing that she had made several errors of law in analyzing the alleged infringements of s. 23  of the Charter , including in the approach she had taken in order to situate a given number of students on the sliding scale and in the test she had applied in order to assess the quality of the educational experience provided to official language minorities; in reviewing the justification of the infringements under s. 1; and in granting the remedies being sought. The Court of Appeal dismissed the appeal but allowed the province’s cross appeal and set aside the award of damages for inadequate funding of school transportation.

                    Held (Brown and Rowe JJ. dissenting in part): The appeal should be allowed in part.

                    Per Wagner C.J. and Abella, Moldaver, Karakatsanis, Côté, Martin and Kasirer JJ.: The courts below adopted an inordinately narrow interpretation of s. 23  of the Charter  and its role in the Canadian constitutional order. Section 23 has a remedial purpose related to promoting the development of official language minority communities and changing the status quo. In accordance with an interpretation of that section that takes its remedial purpose fully into account, and in light of the trial judge’s findings of fact, the appeal should be allowed in part.

                    In Mahe v. Alberta, [1990] 1 S.C.R. 342, the Court explained that situating a given number of students on the sliding scale requires that the analysis focus on (1) the services appropriate, in pedagogical terms, for the number of students involved; and (2) the cost of the contemplated services. However, the Court did not provide an exhaustive definition of these two factors. The approach to be taken in order to situate a given number of students on the sliding scale must therefore be clarified.

                    The analysis of the first factor, pedagogical needs, is concerned with whether, in light of the number of students at issue, the level of services proposed by the minority will make it possible to meet all curriculum requirements, that is, those related to the knowledge and skills the students must acquire while in school. The second factor in the analysis, cost, is less important than the first. It comprises the cost of building a new school or launching a program, and the associated operating costs. As a general rule, pedagogical considerations and cost considerations are interlinked and can be assessed simultaneously.

                    The approach to be taken in order to situate a given number of students on the sliding scale is based on the premise that a homogeneous school, that is, a separate facility under the control of the official language minority, is warranted where such a school is available to a comparable number of majority language students. The first step is to determine how many students will eventually avail themselves of the contemplated service on the basis of long‐term projections. That number lies between the known demand and the total number of children of s. 23 rights holders. The burden of proof is on the claimants from the official language minority.

                    At the second step, the court must take a comparative approach in order to determine whether the school contemplated by the minority is appropriate from the standpoint of pedagogy and cost. The approach is intended for the determination of whether the number of students in question from the official language minority is comparable to the numbers of students in the majority language schools. The burden is on the claimants from the official language minority to identify comparator schools. It is necessary to be flexible in determining what constitutes a comparable number. Comparable does not mean identical. If the court finds that the number of minority language students is comparable to the numbers of students in local majority language schools, there is no doubt that the number of minority language students falls at the high end of the sliding scale and that the minority is entitled to a homogeneous school. In other cases, a province-wide comparative exercise is required to ensure fair treatment across the province.

                    The existence of majority language schools that serve a given number of students, regardless of where they are located in the province, supports a presumption that the province considers maintaining those schools to be appropriate from the standpoint of pedagogy and cost, and thus that it is appropriate to create a comparably sized homogeneous school for the minority. The province can rebut this presumption by showing on a balance of probabilities either that the majority language schools used as comparators are not appropriate for that purpose or that the school proposed by the minority is not appropriate from the standpoint of pedagogy or cost.

                    At the third step, the level of services to be provided to the official language minority must be determined. If the court has found at the second step that the number of students is comparable and that the presumption has not been rebutted, that number is at the high end of the sliding scale and the minority is therefore entitled to have its children receive instruction in a homogeneous school. If the result of the province‐wide comparison is that there is no comparable number, the number of minority language students falls below the high end of the sliding scale, that is, at the low end or in the middle. A minority at the lower levels of the scale can qualify for a range of services varying from a few hours of classes in its language to the use and control of premises in a school shared with the majority. In such a situation, the court must show deference to the level of services proposed by the minority language school board in determining whether that level of services is appropriate from the standpoint of pedagogy and cost.

                    When this approach is applied in this case to the claims of the linguistic minority representatives for new schools or for the expansion of existing schools, they are entitled to eight homogeneous schools that were denied by the courts below. The schools in question are warranted by the numbers of minority language students in the communities in question. The trial judge found that the number of minority language students in the communities of Abbotsford (elementary component for children of rights holders in the community of Abbotsford and secondary component for children of rights holders in the communities of the Central Fraser Valley), Burnaby, Northeast Vancouver, East Victoria and West Victoria will in the long term warrant the creation of homogeneous schools. Given that the long‐term projections are the relevant numbers, these communities are therefore entitled to homogeneous schools.

                    For the communities of North Victoria, Whistler, Chilliwack and Pemberton, the trial judge decided on a local basis for comparison even though the comparison must take schools located across the province into account. The appropriate comparative approach requires that the number of students who will eventually avail themselves of the service — 98 for North Victoria, 85 for Whistler, 60 for Chilliwack and 55 for Pemberton — be compared with the numbers of students attending the small schools located across the province that were retained by the trial judge and for which there is no evidence in the record capable of rebutting the presumption that it is appropriate to create a school of comparable size for the minority. Enrolment in these majority language schools ranges from 66 to 73 students. The relevant numbers for North Victoria, Whistler and Chilliwack are comparable to the numbers of students attending these comparator majority language schools. These communities are therefore entitled to homogeneous schools. As for Pemberton, it is difficult to compare the number of students in question there with the numbers of students at the majority language schools located elsewhere in the province that were retained by the trial judge. Given that the available evidence is limited and that additional submissions might be necessary, the question of the level of services warranted by this number of students should therefore be remanded to the court of original jurisdiction for reconsideration.

                    The test used to assess the quality of the educational experience provided to official language minorities does not vary with the number of minority language students. Section 23 gives an official language minority the right to instruction that is equivalent in quality to the instruction provided to the majority. Children of s. 23 rights holders must therefore receive an educational experience that is substantively equivalent to the experience provided to the majority, regardless of the size of the school or program in question. The essentials of the approach from Association des parents de l’école Rose‐des‐vents v. British Columbia (Education), 2015 SCC 21, [2015] 2 S.C.R. 139, which allows for a holistic assessment of the quality of the educational experience provided to the official language minority, do not need to be adapted in a situation in which the schools of the official language minority are small, aside from the fact that a reasonable parent must take into account the inherent characteristics of attendance at a small school. Accordingly, where a minority language school is not comparable in size to nearby majority language schools, what must be considered is whether reasonable parents who are aware of the inherent characteristics of small schools would be deterred from sending their children to a school of the official language minority because the educational experience there is meaningfully inferior to the experience at available majority language schools. Even where the number of students falls at the low end of the sliding scale, such that there is a right to instruction alone, the factors listed in Rose‐des‐vents must be taken into account in assessing the quality of the educational experience from a program of instruction; the right to instruction cannot be entirely severed from the overall educational experience. In the case of a heterogeneous school or a program of instruction, the analysis based on the substantive equivalence test serves to determine whether the instruction over which the minority has control and the facilities to which it has access are of sufficient quality.

                    In light of these comments, the approach adopted by the courts below in this case where the number of students was not comparable to the numbers of majority language students must be rejected, because that approach was based on what was called a proportionality test rather than on that of substantive equivalence. The trial judge’s conclusions are therefore varied to reflect the conclusion that all rights holders whose children attend CSF schools or participate in its programs are entitled to an educational experience that is substantively equivalent to the experience at nearby majority language schools.

                    For the schools in the communities of Nelson, Chilliwack and Mission, the quality of the educational experience must be assessed from the perspective of a reasonable parent who is aware of the inherent characteristics of a small school. When the substantive equivalence test and the proper approach are applied for the CSF school in Nelson, the trial judge’s finding that the educational experience of the minority language students is equivalent to the experience provided to the majority language students should be accepted. As for the CSF school in Chilliwack, a balancing of the advantages and disadvantages shows that the quality of the educational experience provided there is meaningfully inferior to that of the experience at the majority’s schools. This means that the children of rights holders in Chilliwack do not receive an educational experience of the quality guaranteed to them by s. 23  of the Charter . In the case of the CSF school in Mission, the situation is concerning, but the evidence that was adduced is insufficient for the purpose of making the holistic assessment required by the test of a reasonable parent who is aware of the inherent characteristics of a small school. The question of the quality of the educational experience and the impact of the Facility Condition Driver on this situation must therefore be remanded to the court of original jurisdiction.

                    The fact that the province compels the CSF to prioritize the capital projects the latter submits, even in response to infringements of s. 23, does not infringe the right of management guaranteed by s. 23  of the Charter . How much time the province has to remedy the infringements of s. 23 will have to be addressed on a case‐by‐case basis, but the infringements must nonetheless be remedied in a timely fashion.

                    Where an infringement of s. 23 is established, a court must take the approach established in R. v. Oakes, [1986] 1 S.C.R. 103, while applying a particularly stringent justification standard. This very stringent standard is appropriate for three reasons. First, the framers of the Charter  imposed positive obligations on the provincial and territorial governments in s. 23, and these obligations must be fulfilled in a timely fashion in order to avoid the likelihood of assimilation and of a loss of rights. Second, s. 23 is not subject to the notwithstanding clause in s. 33  of the Charter , which reflects the importance attached to this right and the intention of the framers that intrusions on it be strictly circumscribed. Third, s. 23 has an internal limit, the numbers warrant requirement, according to which the exercise of the right for which the section provides will be warranted if there are a sufficient number of students. In adopting this limit, the framers sought to take account of practical considerations, including cost and pedagogical needs, related to the number of students who might benefit from the right in question. Where the government concerned advances a financial argument to justify an infringement of s. 23, the s. 1 analysis will then in some respects duplicate the numbers warrant analysis that has already been completed. For an infringement of s. 23 to be justified under s. 1, it must not therefore be supported by considerations that have already been taken into account at the numbers warrant stage.

                    At the second stage of the approach established in Oakes — proportionality between the effects of the measure that is responsible for limiting the right and the objective that has been identified as important —, it is necessary to take assimilation fully into account as a deleterious effect when the right under s. 23 is infringed. The purpose of s. 23 is not only to ensure the sustainability of the country’s linguistic communities, but also to make it possible for those communities to develop in their own language and culture in the present. In this sense, even though the evidence shows that s. 23 has not been able to counter or slow the process of assimilation, the fact remains that citizens from official language minority communities still have a right to achieve fulfillment in their own language in everyday life. In addition, a court must bear in mind that s. 23 has an individual dimension and that minority language schools have a definite impact on the likelihood of assimilation of French speakers who attend them. Finally, cost savings linked to an infringement of s. 23 cannot be considered a relevant factor in the balancing of the salutary and deleterious effects of the infringing measure.

                    In the case at bar, the courts below erred in ruling that the fair and rational allocation of limited public funds is a pressing and substantial objective that can justify infringements of s. 23 in accordance with the Oakes test. The fair and rational allocation of limited public funds represents the daily business of government. The mission of a government is to manage a limited budget in order to address needs that are, for their part, unlimited. There is accordingly no pressing and substantial objective here that can justify an infringement of rights and freedoms in this case. The justification for the infringements therefore fails at the first stage of the analysis. Without a valid objective, the province cannot justify the infringements of s. 23.

                    As a result, the infringement of s. 23 found by the trial judge on the basis that the CSF had been denied $1.1 million by not having benefited from the Annual Facilities Grant Rural Factor is not justified, and the CSF is entitled to damages in that amount.

                    The limited government immunity from damages awards does not apply to decisions made in accordance with government policies that are found to be contrary to s. 23. Although damages can be awarded against a government where they are an appropriate and just remedy in the circumstances, it may avoid such an award by raising concerns for effective governance, including where a law has been declared to be invalid after the act that caused the infringement. However, the government does not have immunity in relation to government policies that infringe fundamental rights. The possibility of damages being awarded in respect of Charter ‐infringing government policies in this context is unlikely to have a chilling effect on government actions and thereby undermine their effectiveness; on the contrary, it helps ensure that government actions are respectful of fundamental rights. While it is appropriate to give the government immunity in respect of a well‐defined instrument such as a law, the same is not true in respect of undefined instruments with unclear limits, such as government policies.

                    In the case at bar, because the freeze on school transportation funding was a government policy, the trial judge’s order awarding damages for the inadequate funding of school transportation should be restored.

                    Per Brown and Rowe JJ. (dissenting in part): Unlike most Charter  rights, s. 23, which confers the right to minority language education, imposes positive duties on governments to act. It is preventative, remedial, and unifying, and must be interpreted in light of these objectives. The framing of s. 23 as a positive right is particularly significant. The right expresses its own internal limit, the “numbers warrant” criterion, which reflects a carefully struck constitutional bargain between the federal and provincial governments. This numerical threshold ensures that the positive obligations on the provinces are reasonable and reflect what is practical while at the same time providing the appropriate level of services for minority language students. Both the text of s. 23 and its particular nature require courts to give the provision its proper and intended effect, in line with settled jurisprudence. Striking the right balance recognizes that it is possible to breathe life into the s. 23 right, albeit with caution.

                    The analysis to be applied to s. 23 claims follows two main steps. First, a court must determine the level of services warranted by the number of rights holders in a given area (the “numbers warrant” analysis). This entails ascertaining the relevant number of rights holders, then placing that number on a sliding scale of entitlement in order to decide what level of service is warranted.

                    The relevant figure for the purposes of this analysis is the number of persons who will eventually take advantage of the contemplated program or facility. This figure is an estimate that will fall somewhere between the known demand for the service and the total number of persons who potentially could take advantage of the service. The goal is not to establish how many students will take advantage of the facility or program when it is first launched, but rather to forecast how many will do so in the future.

                    After determining the relevant number, the court must then establish, using the sliding scale approach, what level of services is warranted, pedagogically, given the number of minority language students, and the cost of such service. The right to minority language education is internally limited to the services that can be justified, pedagogically and financially, by the number of children of rights holders. To trigger the obligation to publicly fund minority language education, a claimant must demonstrate that this limit is accounted for. The burden of proving all elements of the s. 23 breach rests on the person asserting the breach. This ensures that s. 23’s own internal qualifications and method of internal balancing are fulfilled. The operation of the sliding scale gives effect to this internal limit because the content of the right expands as the numbers increase, thereby ensuring that rights holders receive a level of service that is appropriate to their number. Proper placement on the sliding scale is critical. It ensures that governments will deploy the resources necessary to meet their obligations under s. 23, and that the internal limit to those obligations is accounted for. Furthermore, just as a failure to give effect to the rights conferred by s. 23 can be detrimental to the flourishing of minority languages, improper placement on the scale can also be harmful to minority students. There would be no point, for instance, in having a school for only ten students in an urban centre, as it would deprive the students of the numerous benefits of studying and interacting in larger numbers.

                    To determine the level of service that is warranted for a given group of rights holders, a claimant must first demonstrate, on a balance of probabilities, that the level of service claimed is pedagogically appropriate for the number of children. To assess the pedagogical appropriateness of a given level of service, the existence of majority schools or programs built for similar numbers elsewhere in the province can be a relevant indicator that a homogeneous school or program would be pedagogically appropriate for the number of rights holders’ children. However, the existence of a small school anywhere in the province is not determinative. Courts need to first assess whether the school or program is a relevant comparator, taking into account factors such as whether it is in a rural or urban area, whether it serves a remote or isolated community, whether the school continues to operate at the capacity for which it was built, and whether the school operates as a result of supplementary private funding. As a general rule, because s. 23 calls for publicly funded minority language education, an appropriate comparator would also need to be publicly funded. This approach entails considering the context of schools used in the comparison to ensure that the circumstances are relevant and comparable to the proposed school or service. The assessment of relevant comparators must take into account that there need not be perfect correspondence of pedagogical appropriateness between the minority and the majority language education. At this stage of the analysis, there is no principled reason to constrain comparator schools locally, and the views of school boards are entitled to a measure of deference with respect to particular services among the range of potentially available services that are most pedagogically appropriate below the upper end of the sliding scale, consistent with the principle that minority language groups should have control over those aspects of education which pertain to their language and culture.

                    If successful in showing pedagogical appropriateness, a presumption then arises to the effect that the level of service is also appropriate as to cost. The burden in turn shifts to the province or territory to rebut the presumption, which is suitable given that it is better placed to adduce such evidence. Throughout, the burden for demonstrating pedagogical appropriateness rests with the claimant, who has the onus of establishing a Charter  breach. This approach is consistent with the fact that cost appropriateness is usually subsumed within the assessment of pedagogical appropriateness. It is particularly important to circumscribe the entitlement within the s. 23 analysis given the limited application of s. 1  of the Charter  in cases of s. 23 infringements. Considerations of pedagogy and cost are thus seriously taken into account within the s. 23 analysis itself. Though cost is not usually a factor in determining whether an individual is to be accorded a right under the Charter , in the specific case of s. 23, such a consideration is mandated.

                    Applying a province‐wide presumption of pedagogical and cost appropriateness at the “numbers warrant” stage, as a majority of the Court suggests, affects key elements of the s. 23 analysis and leads to a compression of the middle of the sliding scale. Considerations of pedagogy and cost are effectively withdrawn, regardless of the particular context that may explain the continued relevance of a school elsewhere in the province. A right to a homogeneous school, which is the highest level of entitlement on the scale, is immediately presumed, thereby shifting the claimant’s burden to the province from the outset. How a province could successfully rebut this presumption is unclear, transforming the presumption effectively into a rule. This operates as a fast track to the upper end of the sliding scale, eliminating any middle level. Such an approach is inconsistent with the Court’s past refusal to adopt a view of s. 23 as encompassing only two rights, that is, one with respect to instruction and one with respect to facilities. This view was rejected by the Court in favour of the sliding scale approach, which allows for a progressive increase in entitlement as the number of rights holders increases.

                    At the second step of the s. 23 analysis, courts must determine whether the quality of services granted to the rights holders is substantively equivalent to the quality of services provided to local majority language students (the “substantive equivalence” analysis). A purposive interpretation of s. 23 requires that substantive equivalence apply throughout the sliding scale. This approach recognizes that the quality of official minority language education cannot be meaningfully inferior to that of the majority. The use of a “proportionality” norm at this stage of the s. 23 analysis would mean that the minority’s relative weight to that of the majority will be taken into account not once but twice, each time diminishing the quality and level of the minority’s constitutional entitlement. As such, it must be rejected.

                    A purposive interpretation of s. 23 emphasizes its true purpose of redressing past injustices and providing the official language minority with equal access to high quality education in its own language, in circumstances where community development will be enhanced. A purposive interpretation of s. 23 may be achieved only by applying substantive equivalence as the appropriate norm at the second stage of the s. 23 analysis, irrespective of where a community falls on the sliding scale. The rationale for applying substantive equivalence is further rooted in the broader principle of protecting minority rights, a fundamental underlying principle to the Constitution.

                    The “substantive equivalence” analysis under s. 23 seeks to evaluate the quality of services provided to minority rights holders. The analysis is circumscribed by comparing the quality of the level of services that is warranted for the number of minority language students with the quality of that same level of services provided to neighbouring schools. The analysis must remain global and contextual and must be mindful that instruction cannot be dissociated from the facilities in which it is provided. Multiple factors may be considered, including the quality of instruction, teachers, physical facilities, educational outcomes, extracurricular activities, and travel time. The relevant factors, as dictated by the circumstances of each case, are considered together in assessing whether the overall educational experience is inferior in a way that could discourage rights holders from enrolling their children in a minority language school. These considerations are applied from the standpoint of the reasonable rights holder parent, comparing the minority language school with the local majority schools that represent realistic alternatives for them.

                    The effect of combining a legal presumption of pedagogical and cost appropriateness at the first stage of the s. 23 analysis, as suggested by a majority of the Court, with recourse to substantive equivalence as the proper comparative norm at the second stage, leads to a strained application of substantive equality. This combination will result in the establishment of very small minority language schools being compared to local majority language schools, which are typically larger and equipped with more or better services, thereby lifting local minority groups to the top of the sliding scale. Such an approach departs from the notion of a sliding scale of warranted services that gradually increases based on the number of children that may benefit from them.

                    The analysis under s. 1  of the Charter  provides that the rights and freedoms in the Charter  are subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society. To demonstrably justify a reasonable limit, the state must show that the objective of the proposed measure is pressing and substantial, and that the means by which the measure is achieved are proportionate. The assessment of the objective is a threshold requirement, analyzed without considering the limit’s scope, the means employed or the effects of the measure. The proportionality requirement will be satisfied where the means are rationally connected to the ends; the measure minimally impairs the right at issue; and the salutary effects of the measure outweigh its deleterious effects.

                    The fact that s. 23 is internally limited has an important effect on how the justification analysis should proceed. Rather than affecting the stringency of the s. 1 analysis, as suggested by a majority of the Court, the internal limit of s. 23 results in s. 1 having less to do in practice. If done properly, the analysis under s. 23 will account for many of the considerations that would normally arise under s. 1. Applying cost considerations under the s. 23 analysis does not mean that they will never be relevant under s. 1. Although such instances may be rare in the context of a s. 23  Charter  right, costs may be a pressing and substantial objective under s. 1 where they are linked to other public policy considerations. However, a measure whose sole purpose is financial, and which infringes Charter  rights, can never be justified under s. 1. Decisions about the fair and rational allocation of limited public funds are the bread and butter of government functions and are purely financial. As such, they are not a valid pressing and substantial objective under s. 1.

                    The first and most important remedy for Charter  breaches is a declaration of invalidity. In the context of s. 23, courts must pay particular attention to whether a declaration will be an adequate remedy for a breach. A declaration often strikes the right balance between vindicating Charter  rights and affording governments flexibility to meet their s. 23 obligations. This primary remedy is supplemented by Charter  damages, which are available where appropriate and just, but are not presumed. Trial judges have significant discretion in determining the appropriate remedy for a Charter  breach. However, this discretion is not unfettered, and what is appropriate and just will depend on the circumstances. An award of damages must be fair not only to the claimant whose rights were breached, but also to the state which is required to pay them. Other remedies may also be more responsive to a breach. Under the framework set out in Vancouver (City) v. Ward, 2010 SCC 27, [2010] 2 S.C.R. 28, the first step for assessing Charter  damages is to show a Charter  breach. Second, the claimant must show why damages are an appropriate and just remedy that fulfills one or more of the related functions of compensation, vindication of the right, or deterrence of future breaches. Third, the government may show countervailing factors that render damages inappropriate or unjust, such as the existence of alternative remedies and good governance concerns. Finally, the court determines the quantum of damages.

                    The immunity from Charter  damages set out in Mackin v. New Brunswick (Minister of Finance), 2002 SCC 13, [2002] 1 S.C.R. 405, has been recognized as responding to a good governance concern. According to the Mackin principle, absent conduct that is clearly wrong, in bad faith or an abuse of power, courts will not award damages for the harm suffered as a result of the mere enactment or application of a law that is subsequently declared to be unconstitutional. This confers a limited immunity intended to balance the protection of constitutional rights against the need for effective government. In addition to legislation, Mackin immunity applies to government policies. The Court’s jurisprudence has consistently framed the principle in terms sufficiently broad to cover other instruments. Moreover, there is no principled basis to limit the application of Mackin immunity to legislation, as proposed by a majority of the Court. The question is not about the vehicle of state action but rather the general purpose of the immunity and under what circumstances the state should be liable for damages.

                    However, the rationale underlying Mackin immunity, being the ability to carry out government functions without the threat of damages, does not support its application in the context of s. 23. Normally, governments make regulations and develop policies to carry out their everyday responsibilities. In doing so, they presumably endeavour not to breach Charter  rights. Yet, a government that has breached s. 23 has not carried out its functions as mandated. Unlike most Charter  rights, s. 23 requires action. Legislation must be enacted, policies must be established, and public funds must be spent to give effect to the right. The s. 23 entitlement is particularly vulnerable to inaction, as any delay in implementation can result in assimilation and undermine access to the right itself. An additional barrier of immunity for damages is ill‐suited to the substance of established s. 23 claims. By the time a court reaches the question of remedy, a breach of s. 23 will already have been found. This means that the government has failed to fund minority language education adequately or has unduly delayed in doing so. Accordingly, an exception is warranted such that the immunity does not apply in the context of s. 23. Damages will therefore be appropriate for s. 23 breaches when all the Ward factors are satisfied.

Cases Cited

By Wagner C.J.

                    Applied: Mahe v. Alberta, [1990] 1 S.C.R. 342; Association des parents de l’école Rose‐des‐vents v. British Columbia (Education), 2015 SCC 21, [2015] 2 S.C.R. 139; R. v. Oakes, [1986] 1 S.C.R. 103; Vancouver (City) v. Ward, 2010 SCC 27, [2010] 2 S.C.R. 28; referred to: Hunter v. Southam Inc., [1984] 2 S.C.R. 145; R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295; Law Society of Upper Canada v. Skapinker, [1984] 1 S.C.R. 357; Reference re Secession of Quebec, [1998] 2 S.C.R. 217; Mackell v. Ottawa Separate School Trustees (1915), 34 O.L.R. 335; The Board of Trustees of the Roman Catholic Separate Schools of the City of Ottawa v. Mackell, [1917] A.C. 62; Assn. des Parents Francophones (Colombie‐Britannique) v. British Columbia (1996), 27 B.C.L.R. (3d) 83; Reference re Manitoba Language Rights, [1985] 1 S.C.R. 721; Solski (Tutor of) v. Quebec (Attorney General), 2005 SCC 14, [2005] 1 S.C.R. 201; Arsenault‐Cameron v. Prince Edward Island, 2000 SCC 1, [2000] 1 S.C.R. 3; Doucet‐Boudreau v. Nova Scotia (Minister of Education), 2003 SCC 62, [2003] 3 S.C.R. 3; R. v. Beaulac, [1999] 1 S.C.R. 768; Mackin v. New Brunswick (Minister of Finance), 2002 SCC 13, [2002] 1 S.C.R. 405; Yukon Francophone School Board, Education Area #23 v. Yukon (Attorney General), 2015 SCC 25, [2015] 2 S.C.R. 282; Lavoie v. Nova Scotia (Attorney General) (1989), 91 N.S.R. (2d) 184; Willick v. Willick, [1994] 3 S.C.R. 670; Quebec (Education, Recreation and Sports) v. Nguyen, 2009 SCC 47, [2009] 3 S.C.R. 208; Ford v. Quebec (Attorney General), [1988] 2 S.C.R. 712; Frank v. Canada (Attorney General), 2019 SCC 1, [2019] 1 S.C.R. 3; Sauvé v. Canada (Chief Electoral Officer), 2002 SCC 68, [2002] 3 S.C.R. 519; Newfoundland (Treasury Board) v. N.A.P.E., 2004 SCC 66, [2004] 3 S.C.R. 381; Health Services and Support — Facilities Subsector Bargaining Assn. v. British Columbia, 2007 SCC 27, [2007] 2 S.C.R. 391.

By Brown and Rowe JJ. (dissenting in part)

                    Conseil scolaire francophone de la Colombie‐Britannique v. British Columbia, 2013 SCC 42, [2013] 2 S.C.R. 774; Mahe v. Alberta, [1990] 1 S.C.R. 342; R. v. Lifchus, [1997] 3 S.C.R. 320; Gould v. Yukon Order of Pioneers, [1996] 1 S.C.R. 571; R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295; Mackin v. New Brunswick (Minister of Finance), 2002 SCC 13, [2002] 1 S.C.R. 405; Reference re Secession of Quebec, [1998] 2 S.C.R. 217; British Columbia v. Imperial Tobacco Canada Ltd., 2005 SCC 49, [2005] 2 S.C.R. 473; Caron v. Alberta, 2015 SCC 56, [2015] 3 S.C.R. 511; Arsenault‐Cameron v. Prince Edward Island, 2000 SCC 1, [2000] 1 S.C.R. 3; Arsenault‐Cameron v. Prince Edward Island (1997), 147 Nfld. & P.E.I.R. 308; Arsenault‐Cameron v. Prince Edward Island (1998), 162 Nfld. & P.E.I.R. 329; Association des parents de l’école Rose‐des‐vents v. British Columbia (Education), 2015 SCC 21, [2015] 2 S.C.R. 139; R. v. Jordan, 2016 SCC 27, [2016] 1 S.C.R. 631; Ontario (Attorney General) v. Fraser, 2011 SCC 20, [2011] 2 S.C.R. 3; R. v. Henry, 2005 SCC 76, [2005] 3 S.C.R. 609; Reference re Public Schools Act (Man.), s. 79(3), (4) and (7), [1993] 1 S.C.R. 839; R. v. Beaulac, [1999] 1 S.C.R. 768; Solski (Tutor of) v. Quebec (Attorney General), 2005 SCC 14, [2005] 1 S.C.R. 201; Quebec (Education, Recreation and Sports) v. Nguyen, 2009 SCC 47, [2009] 3 S.C.R. 208; R. v. Oakes, [1986] 1 S.C.R. 103; Frank v. Canada (Attorney General), 2019 SCC 1, [2019] 1 S.C.R. 3; Alberta v. Hutterian Brethren of Wilson Colony, 2009 SCC 37, [2009] 2 S.C.R. 567; Newfoundland (Treasury Board) v. N.A.P.E., 2004 SCC 66, [2004] 3 S.C.R. 381; Reference re Remuneration of Judges of the Provincial Court of Prince Edward Island, [1997] 3 S.C.R. 3; Doucet‐Boudreau v. Nova Scotia (Minister of Education), 2003 SCC 62, [2003] 3 S.C.R. 3; Vancouver (City) v. Ward, 2010 SCC 27, [2010] 2 S.C.R. 28; Henry v. British Columbia (Attorney General), 2015 SCC 24, [2015] 2 S.C.R. 214; Wynberg v. Ontario (2006), 82 O.R. (3d) 561; Carter v. Canada (Attorney General), 2015 SCC 5, [2015] 1 S.C.R. 331; S.A. v. Metro Vancouver Housing Corp., 2019 SCC 4, [2019] 1 S.C.R. 99.

Statutes and Regulations Cited

Act respecting the Board of Trustees of the Roman Catholic Separate Schools of the City of Ottawa, S.O. 1915, c. 45.

Act Respecting the Department of Education, S.M. 1890, c. 37.

Act to repeal so much of an Act of the Third and Fourth Years of Her present Majesty, to re‐unite the Provinces of Upper and Lower Canada, and for the Government of Canada, as relates to the Use of the English Language in Instruments relating to the Legislative Council and Legislative Assembly of the Province of Canada (U.K.), 1848, 11 & 12 Vict., c. 56.

Canadian Charter of Rights and Freedoms , ss. 1 , 7 , 15 , 23 , 24 , 33 .

Common Schools Act 1871, S.N.B. 1871, c. 21, s. 60.

Constitution Act, 1867 , ss. 93 , 133 .

Constitution Act, 1871 (U.K.), 34 & 35 Vict., c. 28, reprinted in R.S.C. 1985, App. II, No. 11, s. 4.

Constitution Act, 1982 , s. 52(1) .

Constitution Amendment, 1998 (Newfoundland Act), SI/98‐25.

Constitution Amendment, 2001 (Newfoundland and Labrador), SI/2001‐117.

Newfoundland Act (U.K.), 12, 13 & 14 Geo. 6, c. 22, reprinted in R.S.C. 1985, App. II, No. 32.

Northwest Territories Act, S.C. 2014, c. 2 [as en. by the Northwest Territories Devolution Act, S.C. 2014, c. 2, s. 2 ], s. 18(1)(o).

Nunavut Act , S.C. 1993, c. 28, s. 23(1) (m).

Of Public Instruction, R.S.N.S. 1864, c. 58.

Public Schools Act, 1896, S.O. 1896, c. 70, s. 76(2).

Public Schools Act, S.M. 1890, c. 38.

School Act, R.S.A. 1922, c. 51, s. 184.

School Act, R.S.B.C. 1996, c. 412, s. 106.3.

School Act, R.S.S. 1909, c. 100, s. 135.

School Ordinance, O.N.W.T. 1901, c. 29, s. 136.

Union Act, 1840 (U.K.), 3 & 4 Vict., c. 35.

Yukon Act , S.C. 2002, c. 7, s. 18(1) (o).

Authors Cited

Behiels, Michael D. Canada’s Francophone Minority Communities: Constitutional Renewal and the Winning of School Governance. Montréal: McGill‐Queen’s University Press, 2004.

Canada. Department of Justice. A Consolidation of the Constitution Acts, 1867 to 1982. Ottawa, 2012.

Canada. House of Commons. House of Commons Debates, vol. 3, 1st Sess., 32nd Parl., October 6, 1980, p. 3286.

Canada. Senate and House of Commons. Minutes of Proceedings and Evidence of the Special Joint Committee of the Senate and of the House of Commons on the Constitution of Canada, No. 38, 1st Sess., 32nd Parl., January 15, 1981, pp. 36‐37.

Canadian Intergovernmental Conference Secretariat. 18th Annual Premiers’ Conference. Statement on Language, Doc. 850‐8/027. St. Andrews, N.B., August 18‐19, 1977.

Canadian Intergovernmental Conference Secretariat. Premiers’ Conference. Communiqué of the Conference, Doc. 850‐9/007. Montréal, February 23, 1978.

Charles, W.H. Understanding Charter Damages: The Judicial Evolution of a Charter Remedy. Toronto: Irwin Law, 2016.

Cook, Ramsay. “Language Policy and the Glossophagic State”, in David Schneiderman, ed., Language and the State: The Law and Politics of Identity. Cowansville, Que.: Yvon Blais, 1991, 73.

Doucet, Michel, Michel Bastarache et Martin Rioux. “Les droits linguistiques: fondements et interprétation”, dans Michel Bastarache et Michel Doucet, dir., Les droits linguistiques au Canada, 3^e éd. Cowansville, Que.: Yvon Blais, 2013, 1.

Giudici, Anja, and Sandra Grizelj. “National unity in cultural diversity: how national and linguistic identities affected Swiss language curricula (1914‐1961)” (2017), 53 Paedagogica Historica 137.

Hogg, Peter W. Constitutional Law of Canada, 5th ed. Supp. Scarborough, Ont.: Thomson/Carswell, 2007 (updated 2019, release 1).

Kymlicka, Will. Multicultural Citizenship: A Liberal Theory of Minority Rights. New York: Oxford University Press, 1995.

Le Dictionnaire Larousse (online: www.larousse.fr), “pédagogie”.

Martel, Angéline. Official Language Minority Education Rights in Canada: From Instruction to Management. Ottawa: Office of the Commissioner of Official Languages, 1991.

Ontario. Department of Education. English‐French Public and Separate Schools. Circular of Instructions. Toronto, 1913.

Ontario. Department of Education. Roman Catholic Separate Schools and English‐French Public and Separate Schools. Circular of Instructions For the School Year September to June, 1912‐1913: Instructions 17. Toronto, 1912.

Oxford English Dictionary (online: www.oed.com), “pedagogy”.

The Official Languages of Canada. A National Understanding: Statement of the Government of Canada on the official languages policy. Ottawa: Minister of Supply and Services, 1977.

Trudeau, Pierre Elliott. The Constitution and the People of Canada. Ottawa: Government of Canada, 1969.

                    APPEAL from a judgment of the British Columbia Court of Appeal (Bauman C.J. and Tysoe and MacKenzie JJ.A.), 2018 BCCA 305, 14 B.C.L.R. (6th) 52, 416 C.R.R. (2d) 278, 425 D.L.R. (4th) 230, [2018] B.C.J. No. 2836 (QL), 2018 CarswellBC 1956 (WL Can.), affirming in part a decision of Russell J., 2016 BCSC 1764, [2016] B.C.J. No. 2007 (QL), 2016 CarswellBC 2685 (WL Can.). Appeal allowed in part, Brown and Rowe JJ. dissenting in part.

                    Robert W. Grant, Q.C., Mark C. Power and Jennifer Klinck, for the appellants.

                    Karrie A. Wolfe and Katherine Webber, for the respondents.

                    Written submission only by Edward A. Gores, Q.C., for the intervener the Attorney General of Nova Scotia.

                    Written submission only by Ruth M. DeMone and Mitchell M. O’Shea, for the intervener the Attorney General of Prince Edward Island.

                    Alan F. Jacobson, for the intervener the Attorney General of Saskatchewan.

                    Randy Steele, for the intervener the Attorney General of Alberta.

                    Written submission only by Barbara G. Barrowman, Q.C., for the intervener the Attorney General of Newfoundland and Labrador.

                    Sarah A. E. Kay, for the intervener the Attorney General of the Northwest Territories.

                    Christine Ruest Norrena, for the intervener the Commissioner of Official Languages of Canada.

                    Marion Sandilands, for the intervener the Quebec Community Groups Network.

                    Kent Roach, for the intervener the David Asper Centre for Constitutional Rights.

                    Érik Labelle Eastaugh, for the interveners Association des juristes d’expression française du Nouveau‐Brunswick inc. and Association des enseignantes et enseignants francophones du Nouveau‐Brunswick inc.

                    Roger J. F. Lepage, for the intervener Fédération nationale des conseils scolaires francophones.

                    Sylvain Rouleau, for the interveners Association des parents de l’école Rose‐des‐Vents and Association des parents de l’école des Colibris.

                    Audrey Boctor, for the intervener the Canadian Association for Progress in Justice.

                    Dominic Caron, for the interveners Société de l’Acadie du Nouveau‐Brunswick and Fédération des conseils d’éducation du Nouveau‐Brunswick.

                    Christian Monnin, for the intervener the Assembly of Manitoba Chiefs.

                    Vincent Larochelle, for the intervener Commission nationale des parents francophones.

                    Andrew Carricato, for the intervener Conseil scolaire francophone provincial de Terre‐Neuve‐et‐Labrador.

                    François Larocque, for the intervener the Canadian Francophonie Research Chair in Language Rights.

English version of the judgment of Wagner C.J. and Abella, Moldaver, Karakatsanis, Côté, Martin and Kasirer JJ. delivered by

                     The Chief Justice —

TABLE OF CONTENTS
Paragraph
I.      Overview	1
II.    Background	4
A.    Principles to Be Applied in Interpreting Section 23	5
B.   Overview of Concepts Specific to Section 23: Sliding Scale and Substantive Equivalence	21
C.    The Appellants and Their Claim	27
III.   Judicial History	30
A.    British Columbia Supreme Court, 2016 BCSC 1764	30
B.    British Columbia Court of Appeal, 2018 BCCA 305, 14 B.C.L.R. (6th) 52	48
IV.   Issues	50
V.    Analysis	51
A.    What Is the Approach to Take in Order to Situate a Given Number of Students on the Sliding Scale?	51
(1)      First Step: Establishing the Number of Students in Question	58
(2)      Second Step: Taking a Comparative Approach in Order to Determine Whether the School Contemplated by the Minority Is Appropriate from the Standpoint of Pedagogy and Cost	61
(3)      Third Step: Determining the Level of Services That Must Be Provided	84
(4)      Summary of the Applicable Approach	90
(5)      Application of the Principles With Respect to the Sliding Scale	94
B.    Does the Test Used to Assess the Quality of Instruction Provided to Official Language Minorities Vary With the Number of Minority Language Students?	104
C.    Does Compelling a School Board to Prioritize Its Capital Projects Infringe Section 23?	139
D.    How Is an Infringement of Section 23 Assessed Under Section 1?	143
E.    Does the Limited Government Immunity From Damages Awards Apply to Decisions Made in Accordance With Government Policies That Are Found to Be Contrary to Section 23?	164
VI.   Disposition	182

I.               Overview

[1]                             A school is much more than just a place to pass on theoretical and practical knowledge. It is also a setting for socialization where students can converse with one another and develop their potential in their own language and, in using it, familiarize themselves with their culture. That is the spirit in which the right to receive instruction in one of Canada’s official languages was elevated to constitutional status by means of s. 23  of the Canadian Charter of Rights and Freedoms  (“Charter ”).

[2]                             This appeal concerns the scope of s. 23 and the interplay between that section and s. 1, as well as between it and the remedial provisions of Canada’s Constitution. The appeal affords an opportunity to identify the approach to be taken in order to determine the level of services that is guaranteed to rights holder parents on the basis of a given number of students, consider the test to be applied in order to determine whether the educational experience of the children of those rights holders is equivalent to the experience provided to the majority, discuss the justification under s. 1 of infringements of language rights, and decide whether damages can be awarded as a remedy in the event of an infringement.

[3]                             The courts below conducted an exhaustive and rigorous analysis of certain of these issues. But I find that they adopted an inordinately narrow interpretation of s. 23 and its role in the Canadian constitutional order. Section 23 has a remedial purpose related to promoting the development of official language minority communities and changing the status quo. In my view, in accordance with an interpretation of that section that takes its remedial purpose fully into account, and in light of the trial judge’s findings of fact, the appeal should be allowed in part.

II.            Background

[4]                             It is well established that Charter  rights must be given a large, liberal and purposive interpretation (Hunter v. Southam Inc., [1984] 2 S.C.R. 145; R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295). In addition, it is essential to place the purpose of the right in question in its linguistic, philosophical and historical contexts (Law Society of Upper Canada v. Skapinker, [1984] 1 S.C.R. 357; Big M Drug Mart Ltd., at p. 344). Before turning to the facts of this appeal, I consider it necessary to review the background to the enactment of s. 23 and the principles that must inform the interpretation of that section.

A.           Principles to Be Applied in Interpreting Section 23

[5]                             I must begin by noting that the question of language is an integral part of Canadian history. It is a dominant theme that is reflected in legislative initiatives from key points in the country’s history (see M. Doucet, M. Bastarache and M. Rioux, “Les droits linguistiques: fondements et interprétation”, in M. Bastarache and M. Doucet, eds., Les droits linguistiques au Canada (3rd ed. 2013), 1, at pp. 30‐52). In language matters, the legislative pendulum has swung back and forth between two conflicting approaches, one based on policies of assimilation, and the other on promoting the development and autonomy of official language communities.

[6]                             In the discussions that preceded the founding of Canada, the framers of the Constitution wanted it to be mandatory that the laws, records and journals of the nascent country be published in both of what are now the official languages: English and French (Constitution Act, 1867 , s. 133 ). Before that, the imperial authorities had imposed English unilingualism under the 1840 Union Act, but they had then abandoned that policy in 1848 owing to opposition on the part of French‐speaking citizens (Union Act, 1840 (U.K.), 3 & 4 Vict., c. 35; An Act to repeal so much of an Act of the Third and Fourth Years of Her present Majesty, to re‐unite the Provinces of Upper and Lower Canada, and for the Government of Canada, as relates to the Use of the English Language in Instruments relating to the Legislative Council and Legislative Assembly of the Province of Canada (U.K.), 1848, 11 & 12 Vict., c. 56; Doucet, Bastarache and Rioux, at p. 33).

[7]                             In adopting the Constitution Act, 1867 , the framers also took an initial step favouring the recognition of language rights in education. At a time when language and religion often went hand in hand, the framers’ purpose in adopting s. 93  of the Constitution Act, 1867  was to give the provinces the exclusive power to make laws in relation to education. This section was intended indirectly to protect French language and culture, because it enabled French‐speaking Quebecers, who were in the minority in the country as a whole but in the majority in their province, to control their education system (Reference re Secession of Quebec, [1998] 2 S.C.R. 217, at para. 38). Section 93 also included provisions whose purpose was to preserve the rights of the Catholic minority in Ontario and the Protestant minority in Quebec in the area of education.

[8]                             The Constitution Act, 1867  did not, however, put an end to the tension that existed between those who advocated a unilingual conception of the country and those who championed a bilingual state. This tension can be explained by, among other things, the fact that at that time, many believed that a true national state should have a shared identity and thus a single common language, and saw education in that single language as a key to creating that shared identity (see, e.g., R. Cook, “Language Policy and the Glossophagic State”, in D. Schneiderman, ed., Language and the State: The Law and Politics of Identity (1991), 73, at pp. 75‐78; A. Giudici and S. Grizelj, “National unity in cultural diversity: how national and linguistic identities affected Swiss language curricula (1914‐1961)” (2017), 53 Paedagogica Historica 137). Beginning in the late 19th century, most of Canada’s provinces and territories therefore adopted legislative initiatives whose effect was to prohibit French-language instruction (see Of Public Instruction, R.S.N.S. 1864, c. 58; see also A. Martel, Official Language Minority Education Rights in Canada: From Instruction to Management (1991), at pp. 164 and 170; An Act Respecting the Department of Education, S.M. 1890, c. 37; The Public Schools Act, S.M. 1890, c. 38; The Common Schools Act 1871, S.N.B. 1871, c. 21, s. 60; The Public Schools Act, 1896, S.O. 1896, c. 70, s. 76(2); Ontario, Department of Education, Roman Catholic Separate Schools and English‐French Public and Separate Schools, Circular of Instructions For the School Year September to June, 1912‐1913: Instructions 17 (1912); and Ontario, Department of Education, English‐French Public and Separate Schools, Circular of Instructions (1913) (collectively, “Regulation 17”); An Act respecting the Board of Trustees of the Roman Catholic Separate Schools of the City of Ottawa, S.O. 1915, c. 45; The School Ordinance, O.N.W.T. 1901, c. 29, s. 136; The School Act, R.S.S. 1909, c. 100, s. 135; The School Act, R.S.A. 1922, c. 51, s. 184).

[9]                             In Ontario, some parents and a Catholic school board — many of whose students were French speakers — challenged the constitutionality of Regulation 17, which prohibited French‐language instruction after the first two years of primary school. They argued that it infringed s. 93  of the Constitution Act, 1867 . The Appellate Division of the Ontario Supreme Court rejected that argument. In his reasons, Garrow J.A. relied in particular on s. 133  of the Constitution Act, 1867  to conclude that the use of any language other than English was merely a concession and not a right, and that the use of French was not protected:

        It is a perfectly natural thing that those of French descent should love their noble language, and even passionately desire to promote, as far as reasonably possible, its perpetuation here. One may even respect a similar sentiment on the part of the Germans, the Italians, and the others settled among us to whom the English is a foreign tongue. But it is not to be ignored or forgotten that, while all are tolerated, the official language of this Province, as of the Empire, is English, and that the official use of any other language is in the nature of a concession and not of a right. This is, I think, well, and indeed in my opinion conclusively, illustrated by the provisions of sec. 133 of the British North America Act . . . .

(Mackell v. Ottawa Separate School Trustees (1915), 34 O.L.R. 335, at p. 343)

[10]                         The Judicial Committee of the Privy Council upheld the Appellate Division’s decision and affirmed that s. 93 did not protect French-language instruction (The Board of Trustees of the Roman Catholic Separate Schools of the City of Ottawa v. Mackell, [1917] A.C. 62, at pp. 70‐72).

[11]                         Seventy years after Regulation 17 was made, the framers of the Charter  adopted s. 23, which enshrines in the Constitution the right of citizens from the country’s English and French linguistic minorities to have their children receive instruction in their language where the number of children so warrants. In adopting that section, they sought “to ensure that vulnerable minority groups are endowed with the institutions and rights necessary to maintain and promote their identities against the assimilative pressures of the majority” (Reference re Secession of Quebec, at para. 74). By doing so, they definitively closed the door on language policies that would prevent instruction in the language of a minority, and chose an approach that favoured the promotion and development of minority language communities across the country.

[12]                         The historical and social context at the root of language rights in education makes clear the unique role of s. 23 in Canada’s constitutional landscape. In an oft‐quoted passage, Dickson C.J. illustrated the section’s importance by stating that it represents a “linchpin in this nation's commitment to the values of bilingualism and biculturalism” (Mahe v. Alberta, [1990] 1 S.C.R. 342, at p. 350). More recently, in Association des parents de l’école Rose‐des‐vents v. British Columbia (Education), 2015 SCC 21, [2015] 2 S.C.R. 139 (“Rose‐des‐vents”), Karakatsanis J. noted that Canada has a bicultural founding character and that its commitment to bilingualism sets it apart among nations (para. 25, citing Assn. des Parents Francophones (Colombie‐Britannique) v. British Columbia (1996), 27 B.C.L.R. (3d) 83 (S.C.), at para. 24).

[13]                         However, the importance of s. 23 is not based solely on its role in the formation of Canada’s identity as a country. The section is also important because of the role it plays in the identity of Canadians as individuals and as members of linguistic communities. Section 23 is intended to preserve culture and language, two core elements of the notions of identity and well‐being of individuals and communities (W. Kymlicka, Multicultural Citizenship: A Liberal Theory of Minority Rights (1995), at p. 89).

[14]                         In Reference re Manitoba Language Rights, [1985] 1 S.C.R. 721, this Court noted, on beginning its analysis, “the essential role that language plays in human existence, development and dignity”, and its importance in “bridg[ing] the gap between isolation and community” (p. 744). In Mahe, the Court stressed the interplay of language and culture, stating that “any broad guarantee of language rights, especially in the context of education, cannot be separated from a concern for the culture associated with the language. Language is more than a mere means of communication, it is part and parcel of the identity and culture of the people speaking it” (p. 362). And in Rose‐des‐vents, at para. 26, the Court endorsed this comment by the Royal Commission on Bilingualism and Biculturalism: “Language and culture are not synonymous, but the vitality of the language is a necessary condition for the complete preservation of a culture” (Report of the Royal Commission on Bilingualism and Biculturalism, Book II, Education (1968), at p. 8).

[15]                         I would add that in conducting the analysis under s. 23, a court must bear in mind that this section has three purposes, as it is at once preventive, remedial and unifying in nature. It is intended not only to prevent the erosion of official language communities, but also to redress past injustices and promote the development of those communities (Solski (Tutor of) v. Quebec (Attorney General), 2005 SCC 14, [2005] 1 S.C.R. 201, at para. 3; Arsenault‐Cameron v. Prince Edward Island, 2000 SCC 1, [2000] 1 S.C.R. 3, at para. 27). Dickson C.J. explained this remedial purpose by reproducing the comment of Kerans J.A. that “the very existence of the section implies the inadequacy of the present regime” (Mahe, at p. 363). In the face of this “inadequacy of the present regime”, s. 23 was thus designed to alter the status quo. Finally, the section also has a unifying purpose in that it accommodates mobility by enabling citizens to move anywhere in the country without fearing that they will have to abandon their language and culture (Solski, at para. 30; House of Commons Debates, vol. 3, 1st Sess., 32nd Parl., October 6, 1980, at p. 3286).

[16]                         To fully achieve its remedial purpose, s. 23 must, however, be implemented vigilantly. As this Court has noted, the likelihood of assimilation and of cultural erosion will increase with each passing school year if nothing is done to prevent them. The result is that the actual effectiveness of s. 23 is particularly vulnerable to government inaction (Doucet‐Boudreau v. Nova Scotia (Minister of Education), 2003 SCC 62, [2003] 3 S.C.R. 3, at para. 29; Rose‐des‐vents, at para. 28). This means that the courts have a crucial role to play, as the framers made them responsible for overseeing the implementation and protection of Charter  rights.

[17]                         I would also point out that, unlike some other rights provisions, s. 23 recognizes rights that are assessed not only in individual terms, but also on a collective level. The rights conferred by s. 23 are individual rights, but they have a collective scope. As this Court stated in Solski, this means that the courts must, in interpreting s. 23, consider each language group’s social context, demographics and history. The courts thus have the delicate task of reconciling the sometimes divergent concerns of French‐speaking minority groups outside Quebec, whose language rights have been acquired at considerable expense, with the particular reality of Quebec’s English‐speaking minority and with how French‐speaking Quebecers, who are in the majority in that province but whose language is in the minority in the country as a whole, perceive their future in Canada (Solski, at para. 5).

[18]                         Finally, I would note that the fact that s. 23 originally resulted from a political compromise cannot on its own justify a restrictive interpretation of the rights for which the section provides. While it is true that this Court has in the past evoked a distinction between language rights resulting from a political compromise and the other rights guaranteed by the Charter , those days are over. The Court made this clear in R. v. Beaulac, [1999] 1 S.C.R. 768, at para. 24:

     Though constitutional language rights result from a political compromise, this is not a characteristic that uniquely applies to such rights. A. Riddell, in “À la recherche du temps perdu: la Cour suprême et l’interprétation des droits linguistiques constitutionnels dans les années 80” (1988), 29 C. de D. 829, at p. 846, underlines that a political compromise also led to the adoption of ss. 7  and 15  of the Charter  and argues, at p. 848, that there is no basis in the constitutional history of Canada for holding that any such political compromises require a restrictive interpretation of constitutional guarantees. I agree that the existence of a political compromise is without consequence with regard to the scope of language rights. [Emphasis added.]

The Court reaffirmed this statement — that the political compromise that gave rise to language rights is without consequence for the scope of those rights — in the context of s. 23 (Arsenault‐Cameron, at para. 27; Doucet‐Boudreau, at para. 27). The result is that the cases that date back to when the Court was equating language rights with a political compromise, which include Mahe, must be considered in light of the subsequent cases in which the Court favoured a liberal interpretation that is consistent with the development of official language communities.

[19]                         To limit the scope of language rights merely because they resulted from a political compromise would represent a dangerous reversion. Many rights that have been granted to Canada’s minorities were dearly won over many years, and it is up to the courts to give full effect to them, and to do so clearly and transparently.

[20]                         I will pause here to point out that although, in this appeal, my colleagues would grant many of the appellants’ claims for relief, they systematically avoid clarifying how s. 23 should be applied so as to enhance the protection of language rights in this country. Our decision in this case must not be limited to making schools available to the appellants as if the case were one of a kind, as it is also necessary to ensure that future claimants are not forced to undertake interminable judicial proceedings in order to have their rights protected, recognized and enforced. To disregard the problems raised by an erroneous interpretation and application of s. 23, in particular the inevitable judicialization and lengthy delays that are characteristically involved in exercising language rights, is to undermine access to justice and could slow Canada’s historical progress toward the ideal being sought in s. 23: that of “giv[ing] effect to the equal partnership of the two official language groups in the context of education” (Arsenault‐Cameron, at para. 26).

B.            Overview of Concepts Specific to Section 23: Sliding Scale and Substantive Equivalence

[21]                         I feel that it will also be helpful to briefly explain two judge‐made concepts that are specific to the interpretation of s. 23: the sliding scale and substantive equivalence. These two concepts were developed to compensate for the silence of s. 23 regarding the level of services and the quality of instruction it guarantees to official language minorities.

[22]                         Under s. 23, the application of the rights of official language minorities depends on there being a sufficient number of children. But the section is silent as to what number would justify the application of the right to instruction and to educational facilities. Section 23(3)(a) provides that the right to instruction in the language of the minority “applies wherever in the province the number of children of citizens who have such a right is sufficient”. It is complemented by s. 23(3)(b), which provides that the right to instruction includes a right to receive that instruction in minority language facilities provided by the government “where the number of those children so warrants”.

[23]                         In Mahe, this Court rejected what was called the “separate rights” approach, according to which s. 23 provides for only two rights: a right to educational facilities where there are a specific number of students and a right only to instruction where the number of students is smaller. The Court held that s. 23 must instead be understood “as encompassing a ‘sliding scale’ of requirement” (p. 366).

[24]                         By virtue of this “sliding scale” concept, s. 23 provides a basis for a range of educational services. The low end of the scale corresponds to the right only to instruction that is provided for in s. 23(3)(a), while the high end corresponds to the “upper level of management and control” provided for in s. 23(3)(b) (Mahe, at p. 370). In other words, at the low end, s. 23 rights holders are entitled to have their children receive instruction in the language of the official language minority, but the extent to which the minority exercises control over the provision of instruction rises with the number of children of rights holders. At the low end of the scale, the minority is entitled only to instruction in its language. In the middle, it might have control over one or more classrooms in a school of the majority or over one part of a school it shares with the majority. It might also have control over the hiring of teaching staff and over certain expenditures. At the high end, the minority has control over separate educational facilities, that is, over a homogeneous school. The number of children of rights holders might also entitle the minority to the management and control of a separate school board. In short, once the minimum threshold of s. 23(3)(a) is crossed, the sliding scale applies to determine the level of services that corresponds to the extent to which the minority will have control over the provision of educational services.

[25]                         Thus, this Court has recognized that s. 23 has an internal limit, the “numbers warrant” requirement. The courts developed the sliding scale concept in order to give substance to that internal limit. Section 23 imposes no constitutional obligation on a government where the number of students in question does not suffice to justify the creation of a minority language program of instruction (Mahe, at p. 367). The right to such a program of instruction corresponds to the low end of the sliding scale, a limit below which the government has no obligation under s. 23. In this way, the courts have recognized that public funds are limited and that governments cannot be required to set up educational facilities for a very small number of students. Where the number of students in question crosses the numbers warrant threshold, however, that number must then be situated on the sliding scale in order to establish the scope of the rights guaranteed to the rights holders by s. 23. This appeal affords us an opportunity to clarify the approach to be taken in order to situate a given number of students on the sliding scale.

[26]                         Section 23 is also silent regarding the quality of the instruction that must be provided to the official language minority. In Rose‐des‐vents, this Court affirmed that an official language minority is entitled to an educational experience that is substantively equivalent to that of the majority. The Court indicated that instruction is not substantively equivalent if a reasonable parent is discouraged from exercising his or her language rights because the minority language school is meaningfully inferior to that of the majority (para. 35). It follows that to assess the quality of instruction, courts must engage in a process of comparing the minority language school with majority language schools that represent realistic alternatives. However, Rose‐des‐vents concerned a situation in which the number of students enrolled at the minority language school was comparable to the numbers of students enrolled at nearby majority language schools. This appeal affords us an opportunity to determine whether the substantive equivalence test from Rose‐des‐vents applies regardless of the number of minority language students in question, or whether the assessment of equivalence must vary with the number of such students.

C.            The Appellants and Their Claim

[27]                         The appellants are the Conseil scolaire francophone de la Colombie‐Britannique (“CSF”), the Fédération des parents francophones de Colombie‐Britannique and three parents who are rights holders under s. 23  of the Charter . The CSF is the sole French‐language school board in British Columbia. It began to exercise its powers in 1997 following legal proceedings under s. 23. At the time of filing of the proceeding in this Court, the CSF’s territory covered the entire province, and it had 37 schools spread over 17 parts of the province.

[28]                         In June 2010, the appellants filed a notice of civil claim in the British Columbia Supreme Court against Her Majesty the Queen in Right of the Province of British Columbia and the Minister of Education of British Columbia (collectively, the “Province”), submitting that several aspects of the funding of the education system penalized the official language minority and infringed its s. 23 rights. The alleged infringements were numerous, and can be divided into two categories. The first, which involved systemic claims, concerned, among other things, the fact that the CSF had not received an annual grant for building maintenance, the formula used to set priorities for capital projects, a lack of funding for school transportation and a lack of space for cultural activities. The second category involved claims by the appellants for the purpose of obtaining new schools or improvements to existing schools in 17 communities.

[29]                         At trial, the appellants were partially successful. They appealed to the British Columbia Court of Appeal. Their appeal was dismissed, while the cross appeal the Province had subsequently filed was allowed.

III.         Judicial History

A.           British Columbia Supreme Court, 2016 BCSC 1764

[30]                         In a comprehensive judgment — the product of an effort of colossal proportions and, I should add, of high quality — Russell J. allowed in part the action of the CSF and the other appellants. To make it easier to understand the trial judge’s reasons, they can be divided into four parts: (1) the approach to be taken in order to situate a given number of students on the sliding scale, (2) the test to be applied in order to assess the quality of the educational experience provided to official language minorities, (3) the factors to be considered in the s. 1 analysis, and (4) the right to be awarded damages. This summary of the judicial history will be limited to those aspects of the decisions of the courts below that are at issue in the appeal to this Court.

[31]                         First, the trial judge set out an approach to be taken in order to situate the number of students in a given community on the sliding scale. She noted that two factors must guide the sliding scale analysis required by Mahe: the appropriate services in pedagogical terms and the cost of the contemplated services.

[32]                         In the trial judge’s view, the analysis required by Mahe is a practical one, which justifies taking a comparative approach. She was of the opinion that both the balancing of pedagogical and cost considerations and the application of the broader principles from that case support this conclusion. She inferred from this that “[t]o determine what is practical for governments to provide in terms of pedagogy and cost, it makes sense to look to what government would provide for a similar number of majority students in the same community” (para. 791 (CanLII)). On the basis of this comparison with nearby schools, the trial judge identified the different thresholds of the sliding scale. She concluded that the high end of the scale is reached if the number of minority language students is comparable to the numbers in nearby majority language schools and that the number of students falls in the middle of the sliding scale if it is high enough to form a class but not high enough to warrant a school of comparable size to those of the majority. As for the low end of the scale, she held that a comparative analysis is not necessary and that there is a right to instruction if the CSF determines that it is warranted by the number of students.

[33]                         The trial judge also noted that the rights held by Canadian citizens under s. 23 vary over time with the number of students. It is possible for the number of students in question to fall at the low end of the sliding scale when a new program begins, but for the program to grow such that its enrolment becomes comparable to that of the majority language schools. If that were to happen, homogeneous facilities could be warranted.

[34]                         The trial judge then applied her analytical framework and drew conclusions with respect to each of the communities in which the appellants alleged that their rights had been infringed in relation to existing schools or school construction projects. She issued declarations concerning the right to educational facilities in several communities. It is unnecessary here to list all the school projects the trial judge approved, because the CSF and the other appellants are not appealing those decisions. However, the CSF and the other appellants are appealing decisions with respect to nine proposals to create schools or to improve existing schools that she denied on the basis that they were not warranted by the number of students. These proposals were for Abbotsford, Burnaby, Chilliwack, Pemberton, Northeast Vancouver, East Victoria, West Victoria, North Victoria and Whistler.

[35]                         The declarations that recognized the appellants’ right to new schools or to the expansion of existing schools did not require the Province to immediately fund the capital projects the appellants were claiming. The trial judge stressed that the cost of these projects was estimated at over $300 million and concluded that requiring the CSF to prioritize its capital projects — including those that were needed in order to remedy infringements of s. 23 — did not violate the CSF’s right of management and control. Rather, this requirement was a way both to recognize the official language minority’s right of management and control and to take into account the impossibility of carrying out all the capital projects simultaneously.

[36]                         Second, the trial judge outlined the test that is to be applied in determining the quality of the educational experience that must be provided to official language minorities. She drew a distinction between minority language schools that are of comparable size to the majority language schools and those that are smaller than them, and concluded that, where the numbers are comparable, rights holders must have facilities that are separate from and equivalent in quality to those of the majority. From a practical standpoint, however, it cannot be expected that where the number of students of a minority language school is not comparable to the numbers at the majority language schools, the former school will provide educational services that are equivalent to those of the latter. The trial judge concluded that in such circumstances, “the minority is not entitled to fully equivalent programmes, amenities and services” (trial reasons, at para. 860), and the appropriate approach is instead one based on proportionality. The relevant facilities and programs must be proportionate to those of nearby majority language schools. Where the number of students of the minority language school is not comparable to those of the majority language schools, the court must ask “whether a reasonable rightsholder would find a minority school to be meaningfully disproportionate to the facilities offered to the majority, based on a local comparison of the global educational experience” (para. 853).

[37]                         The trial judge concluded with respect to several communities that the rights holders are entitled to facilities that provide an educational experience that is substantively equivalent to the experience of the majority. The CSF and the other appellants did not appeal these decisions. However, the trial judge found that the numbers of children of rights holders in Abbotsford, Burnaby, Chilliwack, Kelowna, Nanaimo, Nelson, Penticton, Sechelt, Squamish, Northeast Vancouver, East Victoria, West Victoria, North Victoria and Whistler warranted not a substantively equivalent educational experience, but one that is proportionately equivalent to the educational experience provided to the majority.

[38]                         The trial judge considered the quality of the educational experience in several communities in which the appellants maintained that they were not receiving the experience to which they are entitled. The appellants are appealing her decisions with respect to three of them: the communities of Nelson, Chilliwack and Mission. She found that in Nelson, the minority’s educational experience was comparable to the experience provided to the majority. Regarding Chilliwack, she concluded that the educational experience provided there was superior to the experience to which the community was entitled. As for Mission, she said she was persuaded that the physical education classes provided there were inferior to those provided to the students of majority language schools. The trial judge assumed, without so deciding, that these inferior physical education classes sufficed to establish an infringement of language rights. She concluded that this infringement was caused by one aspect of the Province’s funding system, the “Facility Condition Driver” that is used to set priorities for capital projects. The Facility Condition Driver takes only the condition of buildings into account and is not concerned with the ability of schools to perform their pedagogical function.

[39]                         Third, the trial judge discussed the principles of interpretation that must inform the analysis of infringements of s. 23 under s. 1, and the application of those principles in this case. The court must begin by considering the context, which includes the Province’s economic and budgetary objectives. For this reason, the trial judge rejected the appellants’ contention that the analysis should instead be limited to the need to protect and promote British Columbia’s French‐speaking minority.

[40]                         The trial judge then considered the deference that must be shown to the Province and concluded that a medium level of deference was appropriate in this case. She declined to conclude that the analysis of infringements of s. 23 requires the application of a stringent standard as is the case with infringements of ss. 7  and 15  of the Charter , because s. 23 does not engage the same fundamental considerations.

[41]                         The trial judge also discussed the impact of the cost issue in the justification of an infringement under s. 1. She concluded that cost can, when linked to other non‐financial considerations, constitute a valid objective for the purposes of that section. In the context of this case, she identified the “fair and rational allocation of limited public funds” (trial reasons, at para. 1065) as a pressing and substantial objective.

[42]                         As well, the trial judge discussed the proportionality analysis from R. v. Oakes, [1986] 1 S.C.R. 103. Regarding the question whether the effects of the measure that constituted a limit on the right were proportional to the objective that was found to be important, she stated that the analysis must take both local and systemic effects into account. In her view, an analysis confined to salutary effects at the local level could disguise the fact that, although the CSF’s facilities might be substandard in one community, the system it operates is equal to or better than the systems operated by other school boards.

[43]                         At the local level, the salutary effects include what the funding system provides the rights holders of a given community. At the systemic level, they include what that system provides rights holders across the province. As for what are considered to be deleterious effects, they include, at the local level, the fact that rights holders’ children are not receiving the educational experience to which they are entitled and, at the systemic level, the likelihood of assimilation. However, the trial judge found that the existence of minority language schools will not have a significant impact on the high rate of linguistic assimilation of the minority in British Columbia and that the greater likelihood of assimilation is therefore not a particularly strong deleterious effect.

[44]                         On the basis of this analytical framework, the trial judge concluded that several infringements of the appellants’ language rights were justified under s. 1. Thus, the absence of core facilities in Pemberton, the projected overcrowding at École Victor‐Brodeur in Victoria and its two annexes, the substandard gymnasium in Mission caused by British Columbia’s Facility Condition Driver, and the fact that the CSF had been denied the building maintenance envelope for rural areas (“Annual Facilities Grant Rural Factor”) were infringements that she found to be justified in a free and democratic society.

[45]                         Fourth, the trial judge ruled on damages. She concluded that awarding damages would not be appropriate for most of the appellants’ claims. In her view, the principles from Vancouver (City) v. Ward, 2010 SCC 27, [2010] 2 S.C.R. 28, and Mackin v. New Brunswick (Minister of Finance), 2002 SCC 13, [2002] 1 S.C.R. 405, barred an award of damages.

[46]                         The trial judge considered the limited government immunity established in Mackin to be an initial obstacle to a damages award. She concluded that a government has immunity in respect of decisions based on a law that is subsequently held to be unconstitutional, which is the case here given that several of the claims are grounded in the application of a law respecting capital funding. As well, she concluded in light of the principles from Ward that the possibility of issuing a declaration and the need for the government to be able to make laws without the possibility of being ordered to pay damages were factors that weighed against a monetary award.

[47]                         The trial judge found that the freeze on funding for school transportation at a time when the number of students of the linguistic minority was rising constituted an infringement of s. 23, and she accordingly awarded $6 million in damages to the CSF to make up part of its deficit. In her view, this order would not have a chilling effect on the Province’s policies. On the other hand, the trial judge declined to award $1.1 million in damages to the CSF in compensation for its having been denied the Annual Facilities Grant Rural Factor, as she found that this denial was justified under s. 1.

B.            British Columbia Court of Appeal, 2018 BCCA 305, 14 B.C.L.R. (6th) 52

[48]                         The appellants appealed the trial judge’s judgment, arguing that she had made several errors of law in analyzing the alleged infringements of s. 23  of the Charter , in reviewing the justification of the infringements under s. 1, and in granting the remedies being sought. The Court of Appeal dismissed the appeal, holding that the Charter  did not require the Province to use public funds to finance the schools being claimed. It relied for this on the need for practicality when courts interpret the duties imposed on governments by s. 23. The Court of Appeal held that the trial judge had made no errors either in determining whether s. 23 had been infringed or in her review of the infringements under s. 1.

[49]                         In its cross appeal, the Province submitted that the trial judge had erred in ordering it to pay Charter  damages for failing to adequately fund transportation costs. The Court of Appeal allowed the cross appeal on the basis that the trial judge had not recognized and applied the government’s immunity from being ordered to pay damages that had been established in Mackin. It set aside the award of damages for inadequate funding of school transportation.

IV.         Issues

[50]                         This appeal raises the following issues:

1.         What is the approach to take in order to situate a given number of students on the sliding scale?

2.         Does the test used to assess the quality of instruction provided to official language minorities vary with the number of minority language students?

3.         Does compelling a school board to prioritize its capital projects infringe s. 23?

4.         How is an infringement of s. 23 assessed under s. 1?

5.         Does the limited government immunity from damages awards apply to decisions made in accordance with government policies that are found to be contrary to s. 23?

V.           Analysis

A.           What Is the Approach to Take in Order to Situate a Given Number of Students on the Sliding Scale?

[51]                         This appeal is an opportunity for the Court to clarify the approach to be taken in order to situate a given number of students on the sliding scale. The sliding scale, which serves to determine the level of services to which an official language minority is entitled, is used to decide whether the minority is entitled to a homogeneous school, to educational facilities shared with the majority or to another appropriate solution. It is also used to decide whether the minority is entitled to a separate school board. This Court has stated the principles the courts must apply in this exercise, but has never explained the approach that should be taken.

[52]                         In Mahe, this Court explained that situating a given number of students on the sliding scale requires that the analysis focus on “(1) the services appropriate, in pedagogical terms, for the number of students involved; and (2) the cost of the contemplated services” (p. 384). Obviously, however, given the remedial nature of s. 23, pedagogical requirements will have more weight than cost (p. 385). Mahe did not include an exhaustive definition of these two factors, but their scope can be established by reviewing the case law and terminology.

[53]                         To begin, the word “pedagogy” is defined as “the theory or principles of education; a method of teaching based on such a theory” (Oxford English Dictionary (online)). Similarly, the French word “pédagogie” is defined as [translation] “the whole of the methods used to educate children and adolescents” (Le Dictionnaire Larousse (online)). In Mahe, the Court, per Dickson C.J., stressed “that a threshold number of students is required before certain programmes or facilities can operate effectively” (p. 385). The Court has also referred, in a case in which it was asked to rule on the pedagogical needs factor, to a proposed school’s ability to “mee[t] all curriculum requirements” (Arsenault‐Cameron, at para. 39).

[54]                         Both the definition of “pedagogy” and the concerns expressed by the Court in the past show that this factor relates to the pedagogical viability of the official language minority’s proposal. This factor is concerned with answering the following question: In light of the number of students at issue, will the level of services proposed by the minority make it possible to meet all curriculum requirements, that is, those related to the knowledge and skills the students must acquire while in school?

[55]                         The second factor, cost, is a particular constraint on the application of s. 23, because “it is financially impractical to accord to every group of minority language students, no matter how small, the same services which a large group of s. 23 students are accorded” (Mahe, at p. 385). Financial considerations comprise the cost of building a new school or launching a program, and the associated operating costs. The cost factor takes into account the fact that public funds are limited and must therefore be used wisely. Once the costs associated with the contemplated project are established, the court must ask, in accordance with the Mahe framework, whether they are justified in light of the number of students in question. It is well established that cost is a less important factor than the official language minority’s pedagogical needs.

[56]                         A number of interveners representing or supporting official language minorities have stressed the need to clarify the Mahe framework. It is clear that, because of the lower courts’ interpretation of Mahe and the interminable judicial proceedings that must be initiated in order to assert language rights, the exercise of those rights is too often delayed, if not diminished. The case at bar is a clear example of this. More than ten years has elapsed between the date of filing of the proceedings and this Court’s judgment. As the intervener Canadian Association for Progress in Justice points out, “[t]en years of litigation to determine entitlement is simply not viable” (I.F., at para. 2). Nearly two generations of elementary school students have thus been denied their language rights, and this has contributed to the erosion of British Columbia’s French‐speaking community. As well, the Fédération nationale des conseils scolaires francophones explains why it is necessary to relax the burden a plaintiff must discharge in order to show that a homogeneous school is appropriate. This case clearly illustrates why the assertion that Mahe already provides sufficient guidance for the application of the s. 23 framework cannot be accepted. In my view, it must be explained how the courts are to apply the broad principles from Mahe in order to situate the number of students in question in a given case on the sliding scale. I find that the time has therefore come to set out a straightforward and predictable approach that might even enable rights holders to avoid, to the extent possible, resorting to litigation.

[57]                         To take the pedagogical and cost considerations discussed in Mahe into account, I propose that the following approach be taken in order to situate a given number of students on the sliding scale. This approach is designed to recognize the remedial nature of s. 23 “so as to give effect to the equal partnership of Canada’s two official language groups in the context of education” (Rose‐des‐vents, at para. 27 (emphasis added), citing Arsenault‐Cameron, at para. 26, Mahe, at p. 364). It is based on the premise that a homogeneous school, that is, a separate facility under the control of the official language minority, is warranted where such a school is available to a comparable number of majority language students.

(1)          First Step: Establishing the Number of Students in Question

[58]                         The first step in situating the number of students in question on the sliding scale is to determine how many students will eventually avail themselves of the contemplated service. This is the starting point for the “numbers warrant” analysis. The burden of proof regarding the number of students who will eventually avail themselves of the service is on the claimants from the official language minority. To discharge it, the claimants must adduce evidence that would enable the court to rule on this issue. That evidence might include expert testimony, various statistical tools such as the census, and statistical models that take into account the demographics of the community in question, its geographical location and any other factor that might have an impact on the number of students.

[59]                         Long‐term projections are necessary. As this Court noted in Mahe, “the relevant figure for s. 23 purposes is the number of persons who will eventually take advantage of the contemplated programme or facility” (p. 384). An approach based on a short‐term projection of enrolment rather than on the number of students who will eventually avail themselves of the service would be contrary to s. 23 and to this Court’s case law. The effect of such an approach, which might be said to be “temporal” in nature, would be to place on the minority the burden of asking the government for enhancements to the services provided to it each time the number of students reached a new level on the sliding scale and, if necessary, going to court to assert its rights in this regard. That would not be a desirable result. An approach focused on enrolment over the long term might limit the frequency with which such steps need to be taken.

[60]                          The number of students who will eventually avail themselves of the service lies between the known demand and the total number of children of s. 23 rights holders (Mahe, at p. 384). In my view, it is not necessary in this case to fix a precise number of years as the length of a long‐term projection. On the one hand, such projections must take into account the fact that plans for school construction or expansion projects are prepared on a long‐term basis. On the other hand, the projections must not relate to a future so remote that they do not allow a reliable number to be obtained.

(2)          Second Step: Taking a Comparative Approach in Order to Determine Whether the School Contemplated by the Minority Is Appropriate from the Standpoint of Pedagogy and Cost

[61]                          At the second step, the court must determine whether the school or program proposed by the minority is appropriate from the standpoint of pedagogy and cost for the number of students in question. I favour a simple approach, that is, a comparative approach, which also has the advantage of limiting the need to resort to litigation.

[62]                         This Court has in fact often used comparisons in the context of cases under s. 23. In Mahe, for example, it compared the number of minority language students in Edmonton with the numbers of students of Alberta’s school boards in order to determine whether the minority was entitled to a separate school board. In Arsenault‐Cameron, the Court analyzed the pedagogical viability of a proposed school by comparing the number of minority language students with the numbers of students at several small majority language schools. And in Rose‐des‐vents, it considered the quality of the educational experience by comparing the minority language school’s facilities with those of majority language schools.

[63]                         It is best to take a comparative approach in applying the sliding scale, because it is hard to associate a given number of students with academic standards. It will often be experience and practice that determine whether a given number of students will suffice for a school to function efficiently in light of the province’s curriculum. In this sense, the existence of majority language schools of a similar size represents the best indicator, and in particular the easiest criterion to apply, in order to determine whether a given number of students is sufficient for the achievement of the curriculum’s objectives. It would in fact be hard, for the purpose of showing that a school can satisfy academic standards, to find an argument that is more convincing than the fact that majority language schools of a similar size exist or are maintained.

[64]                         The particular characteristics of a minority language community are not helpful for the purpose of determining whether a proposed school is pedagogically viable. For example, the difference between communities with a strong historical presence and more recently established communities is not relevant, as they all have the same pedagogical needs. A school’s pedagogical viability, that is, its capacity to achieve the objectives of the curriculum, does not vary with a community’s historical nature. Nor does the fact that a community has existed for a long time have any impact on the cost of a proposed school. Moreover, a distinction that implied that long-established official language communities are more legitimate than communities that have been established more recently would be inconsistent not only with the letter but also with the spirit of the Charter . Section 23 guarantees rights to “citizens of Canada” from official language minority communities to which the Charter  applies and requires equal treatment for all.

[65]                         In my view, situating a number of students on the sliding scale requires a province-wide comparative analysis. In Canada, laws relating to education are made at the provincial level as a result of s. 93  of the Constitution Act, 1867 .^[1] Section 23(3)(a) reflects this fact, as it provides that the constitutional right it creates “applies wherever in the province the number of children of citizens who have such a right is sufficient to warrant the provision to them out of public funds of minority language instruction”. This is a first indication that a province‐wide comparison is required.

[66]                         In this Court, the Province maintained that a province‐wide comparison would not be appropriate, because small rural schools are not valid comparators. It argued that such rural schools exist because of the geographical isolation of the students in question. I find that this does not justify systematically removing rural schools from the equation. The cultural isolation of the minority groups to which s. 23 applies is a circumstance that, although different in some respects, resembles, from a sociolinguistic standpoint, the geographical remoteness of certain majority language communities. I acknowledge that the analogy is not perfect. A small rural school is often the only solution when it comes to providing instruction to people living in remote locations, whereas, in the context of linguistic minorities, a homogeneous school, while it is of course one possible solution, is not the only one. It is sometimes possible to provide instruction to small groups either in a heterogeneous school or in a program of instruction. Moreover, financial considerations are not the same in urban and rural areas. The cost of acquiring land can be lower in a village than in a large urban centre.  Nevertheless, official language minority schools, like rural schools, serve to meet the essential educational needs of isolated populations, needs that are distinctive from a geographical or a sociolinguistic standpoint. It therefore cannot be argued that small rural schools must be systematically excluded from the province‐wide comparison. Instead, exceptional cases should be excluded one by one.

[67]                         What is more, I have reservations about an approach in which the comparison would be limited to local schools. The effect of such an approach would be to tie recognition of the minority’s rights to choices made by the majority for its own students. If the Court were to adopt that approach, it would be easier for a minority to obtain homogeneous schools in a city where the majority has chosen to establish small schools than in another city where the majority has chosen to operate bigger schools. In my view, the use of province‐wide data will ensure fair treatment across the province.

[68]                         When a province operates small schools with limited numbers of students in certain parts of its territory, it necessarily believes that they satisfy the pedagogical needs of the students enrolled in them and that they are warranted having regard to the principles governing the sound use of public funds. In Arsenault‐Cameron, this Court drew a positive inference from the fact that there were a number of English‐language schools in the province with fewer than 100 students and that the Minister of Education “was not willing to close them or to say they did not meet the department’s pedagogical standards” (para. 40).

[69]                         I thus find that the existence of majority language schools that serve a given number of students, regardless of where they are located in the province, supports a presumption that it is appropriate from the standpoint of pedagogy and cost to create a comparably sized school for the minority. The province can, however, rebut this presumption by showing on a balance of probabilities either that the majority language schools used as comparators are not appropriate for that purpose or that the school proposed by the minority is not appropriate from the standpoint of pedagogy or cost. I also wish to be clear that the claimants, who are responsible for identifying comparator schools, must endeavour to submit to the court a reasonable number of schools that, to the best of their knowledge, are appropriate comparators in light of the principles set out in these reasons. This will favour judicial economy by ensuring, for example, that courts and parties do not have to analyze hundreds of schools in detail.

[70]                         This approach has the advantage of being both straightforward and consistent with the principles laid down by this Court. It is based on a recognition that pedagogical considerations and cost considerations are interlinked and can be assessed simultaneously. As Dickson C.J. explained, “in most cases pedagogical requirements will prevent the imposition of unrealistic financial demands upon the state” (Mahe, at p. 385). Thus, pedagogical considerations usually encompass those related to cost. In exceptional cases, which will most certainly be rare, the province can still show that a proposed school is inappropriate on the basis either of the students’ pedagogical needs or of cost. The presumption I have just identified is thus consistent with the internal limit of s. 23, as it can be rebutted on the basis that the requirements of one or the other of the considerations from Mahe — pedagogy and cost — are not met.

[71]                         This presumption accounts for the fact that the burden is on the minority to establish the number of students in question and to identify comparator schools. Once this burden has been discharged, the existence of majority language schools of a size similar to that of the proposed minority language school suffices — unless the province produces evidence to the contrary — to establish that the proposed school is pedagogically viable. This interpretation is consistent with this Court’s case law. In Arsenault‐Cameron, the Court held that “there was no evidence that pedagogical concerns could not be met, or that a small school would mean an education that is substandard” (para. 39). The Court thus recognized that the provinces are in the best position to make representations on the pedagogical quality of the majority language schools they operate that might be used for purposes of comparison.

[72]                         Furthermore, the presumption is consistent with the case law to the effect that the provinces are also in a better position than the claimants to provide projections of capital costs and expenditures per student. The minority cannot be expected to be able to evaluate education‐related costs with precision (Lavoie v. Nova Scotia (Attorney General) (1989), 91 N.S.R. (2d) 184, at para. 48).

[73]                         This comparative approach is intended for the determination of whether the number of students of the official language minority is comparable to the numbers of students in the majority language schools. I wish to be clear that comparable does not mean identical. A formalistic approach according to which the number of minority language students must be exactly equal to or greater than the numbers of students in the majority’s schools if the minority language students are to have access to separate and equivalent facilities would have the effect of reinforcing the status quo and would be incompatible with the remedial purpose of s. 23. It is therefore necessary to be flexible in determining what constitutes a comparable number.

[74]                         This flexibility takes into account the importance of homogeneous schools as linguistic islands in a minority environment. In such a context, a homogeneous school is often the only place where all of a person’s activities are carried out in his or her language. As well, it serves as a gathering place for the extended community. Shared facilities cannot fully play these roles, as they make it more difficult to achieve the objectives of s. 23.

[75]                         There are two possible ways for the province to rebut this presumption. It has the choice of arguing (1) that the comparator schools identified by the claimants are not appropriate comparators from the standpoint of pedagogy or cost, or (2) that the school contemplated by the minority does not meet the requirements from the standpoint of pedagogy or cost. I will explain these different possibilities.

[76]                         First, the province can show that the comparator school is not an appropriate comparator. To do so, it might establish that the school in question does not meet the requirements of the provincial curriculum. Here are some examples. A school that serves a student population with very particular needs — disabled students or students in a specialized arts program — is one in which the learning outcomes might justify its having a small number of students. Such schools have very particular educational plans, which explains their departure from the province’s usual standards.

[77]                         The province might also prove that the number of students at a given school has, owing to attrition, fallen to such an extent that the school no longer meets the requirements of the provincial curriculum. However, it would have to show clearly that the school in question no longer meets those requirements. This example is based on the very premise that guided the Court in Arsenault‐Cameron: that if the province maintains this small school, it is because it considers that the school continues to meet the pedagogical requirements. If that is not so, the province must say so clearly.

[78]                         It might also happen that a comparator school is excluded on the basis of financial considerations. This would be true of a school whose operation depends on significant private financing. Such a school is not a valid comparator, because it does not reflect what the province considers to be appropriate education expenditures. It would not suffice here to show that the school has received a few donations. What would have to be shown is that this majority language school would not be viable without private financing. Moreover, a school that serves a particularly isolated location, such as a place that is accessible only by water or air, or a school that serves an isolated location separated from the rest of the province by a road that requires several hours of travel might justify incurring exceptional expenditures for a very small number of students if that is the only way to provide that isolated population with educational services.

[79]                         I wish to be clear that this should not be interpreted as an invitation to conclude that every small majority language school is an exception and that such schools cannot be used as comparators. The idea is to exclude majority language schools whose circumstances are genuinely exceptional.

[80]                         Second, the province can show that the school contemplated by the minority is not appropriate. To do so, it might establish that the service in question would not suffice to achieve the requirements of the provincial curriculum. This can be shown by means of expert testimony or any other evidence the province considers relevant. If, for example, the province wants to show that there are too few students from the official language minority and that this would prevent the socialization objectives of the provincial curriculum from being achieved, it must explain precisely what the objectives of the curriculum are and how the number of students in question would not suffice to achieve them. The province is in the best position to do this, as the education ministry’s experts are the people who develop the curriculum’s standards. As well, it is the provincial education ministry that issues diplomas and ensures that schools meet academic standards. It is therefore entirely appropriate that the ministry be responsible for showing how, even if the contemplated service would be provided to a number of students comparable to the numbers in majority language schools, it would not suffice to satisfy the province’s academic standards.

[81]                         The province can also show that the cost of the service contemplated by the official language minority would represent an unrealistic financial demand (see Mahe, at p. 385). I note that it is perfectly normal for the cost per student of an official language minority school to be higher than that of a majority language school because of the minority language school’s small size. Nevertheless, the province can show that the contemplated project would represent an unrealistic financial demand if the difference between its cost and the average cost of majority language schools of comparable size is significant. For example, the construction of a very small school in the centre of a large urban area where the cost of acquiring land would be prohibitive might represent an unrealistic financial demand.

[82]                         I wish to be clear that caution must be exercised in considering start-up costs. The creation of a school necessarily requires considerable initial costs, but these costs are then amortized, that is, spread over the duration of the building’s use. The exercise of language rights requires, by definition, initial expenditures by the government. As Bastarache J. recognized in Beaulac, “[l]anguage rights . . . can only be enjoyed if the means are provided” (para. 20).

[83]                         Once again, the foregoing is not an invitation to try to show that every minority school project represents an unrealistic financial demand. The purpose of the unrealistic financial demand concept is to avert situations resulting in expenditures that are genuinely disproportionate in comparison with the average expenditures encountered in majority language schools of comparable size. Caution must be exercised in deciding against a school project of the official language minority on the basis of cost, because financial considerations are given less weight than pedagogical considerations when situating a given number of students on the sliding scale.

(3)          Third Step: Determining the Level of Services That Must Be Provided

[84]                         At the third step, the level of services to be provided to the official language minority must be determined. If the court has found at the second step that the number of students is comparable and that the presumption has not been rebutted, that number is at the high end of the sliding scale and the minority is therefore entitled to have its children receive instruction in a homogeneous school.

[85]                         If, on the other hand, the court has found at the second step that the number of minority language students is not comparable, the number then falls below the high end and a homogeneous school is therefore not required. In such circumstances, there is a range of services that can be provided to the minority. These services vary from the provision of certain courses in the minority language to control over a portion of a school shared with the majority. These different levels of services represent a variety of possibilities.

[86]                         The provision of such limited educational services is a reality specific to the minority. Having to share premises or being able to provide only a few classes in their language are realities with which the managers of majority language schools, which are always homogeneous, do not have to deal. As a result, school boards of official language minorities have particular expertise in relation to the services that might be provided to a limited number of students. School boards are in fact favoured vehicles for the concerns of official language minorities. As Dickson C.J. explained in Mahe, although the majority’s intentions may not be bad, “the majority cannot be expected to understand and appreciate all of the diverse ways in which educational practices may influence the language and culture of the minority” (p. 372). This observation justifies recognizing that, when the number of students falls in the lower levels of the sliding scale, the government and, in the event of litigation, the courts must show deference to the school board’s expertise with respect to the appropriate level of services. It is perfectly logical, and even desirable, for the courts to look to the expertise of school boards in such matters given that they have long recognized the importance of the role played by school boards in ensuring that the “special educational needs of the minority” are expressed adequately (Mahe, at p. 373). The CSF itself was in fact created as a result of judicial proceedings (Assn. des Parents Francophones). If no school board exists to represent the interests of the official language minority, governments and courts will have to rely on groups or associations that, or witnesses who, are in a position to express the minority’s special educational needs.

[87]                         At the lower levels of the sliding scale, it is obviously not possible to use the comparative approach I set out above in order to determine what is appropriate from the standpoint of pedagogy and cost. I reiterate that at this stage, the court will have found that the number of minority language students in question is not comparable to the numbers at majority language schools, even at the smallest of those schools. Given that the parties have not proposed a specific approach in this regard and that this case does not turn on such an approach, it would be premature to rule on this issue.

[88]                         I will simply say that, in such a context, governments must be deferential to the proposals made by the school board, and their decisions must be guided by pedagogical and financial considerations, as defined above. However, the cost issue will rarely be decisive, as the sharing of facilities obviously produces economies of scale. The focus must be on whether the contemplated service is able to meet the pedagogical requirements established by the province in a setting that is respectful of the minority’s language and culture. This last factor, the need “to consider the value of linguistic minority education as part of the determination of the services appropriate for the number of students” (Arsenault‐Cameron, at para. 38), is particularly important in cases in which the number of students in question falls in the lower levels of the sliding scale. Where a number of students falls at the high end of the scale and therefore warrants a separate facility, it goes without saying that the educational setting will be respectful of the minority’s language and culture.

[89]                         Thus, if the number of students in question falls in the lower levels of the sliding scale, it is necessary to determine what level of service corresponds to the level of control that would enable the minority to promote its language and culture while at the same time meeting the province’s pedagogical requirements. The greater the control exerted in a school by the minority, the more separate the minority language students will be from those of the majority. The more separate the minority is from the school as a whole, the higher will be the number of minority language students needed in order for the school to be in a position to meet the pedagogical requirements. For example, if the number of minority language students should fall slightly below the high end of the scale, the minority might be entitled to have control over a separate portion of the school — which would permit it to make announcements and post materials in its own language — but still be required to share the gymnasium and the library. It might also have its own school administrators and issue its own report cards. Where, on the other hand, the number of students is lower, meeting pedagogical requirements may — despite the importance of the value of instruction in the language of the minority as a factor to be considered — mean that the appropriate level of service is limited to the provision of certain courses in a shared facility that is entirely managed by the majority.

(4)          Summary of the Applicable Approach

[90]                         In sum, a court that is asked to situate a given number of students on the sliding scale must proceed as follows. The first step is to identify the relevant number of students. To do so, the court must consider the number of students who will eventually avail themselves of the service. At the second step, the court must compare that number with the numbers of students in majority language schools across the province in order to determine what is appropriate from the standpoint of pedagogy and cost. The existence of comparably sized majority language schools, regardless of where they are located in the province, supports a presumption that the province considers maintaining those smaller schools to be appropriate from the standpoint of pedagogy and cost. The province can rebut this presumption by showing that the majority language comparator schools identified by the claimants are not appropriate comparators or that the school proposed by the minority is inappropriate from the standpoint of pedagogy or cost.

[91]                         However, I would stress that if the court finds at the outset that the number of minority language students is comparable to the numbers of students attending local majority language schools, a province‐wide comparison will not be necessary. This ensures that the courts do not engage in an unnecessarily detailed and complex analysis. If the number of minority language students is comparable to the numbers of students in local majority language schools, there is no doubt that the number of minority language students falls at the high end of the sliding scale and that the minority is therefore entitled to a homogeneous school.

[92]                         Once the court has completed a province-wide comparison — where such a comparison is necessary — it moves on to the third step: determining what level of services is appropriate. Where the number of minority language students in question is comparable to the numbers of students attending majority language schools, regardless of where those schools are located in the province (provided of course that the government has not shown that the choice of comparator schools or the proposed service is inappropriate from the standpoint of pedagogy or cost), the number of minority language students falls at the high end of the sliding scale and warrants a homogeneous school.

[93]                         If the result of the province‐wide comparison is that there is no comparable number, the number of minority language students falls below the high end of the sliding scale, that is, at the low end or in the middle. A minority at the lower levels of the scale can qualify for a range of services varying from a few hours of classes in its language to the use and control of premises in a school shared with the majority. These examples are of course not exhaustive. In such a situation, the court must show deference to the level of services proposed by the school board. It must rely on the school board’s expertise in determining whether the proposed level of services is appropriate from the standpoint of pedagogy and cost. The court must consider the importance of receiving instruction in the language of the minority in a setting that is respectful of that minority’s cultural needs. However, the exact approach to be taken in order to situate the number of students in question in the lower levels of the sliding scale will have to be determined in a case in which that specific issue is raised.

(5)          Application of the Principles With Respect to the Sliding Scale

[94]                         In my view, when this approach is applied in the case at bar to the CSF’s claims for new schools or for the expansion of existing schools, the appellants are entitled to eight homogeneous schools that were denied by the courts below. The schools in question are warranted by the numbers of minority language students in the communities in question.

[95]                         The failure to recognize the right to these eight schools resulted from two types of errors. The first type relates to the assessment of the relevant number that would warrant a school under s. 23. In her analysis, the trial judge made short‐term (3‐year) and long‐term (10‐year) projections. But it was the short‐term projections she applied in identifying the level of services to which the official language minority is presently entitled. When she applied the long‐term projections, it was only to determine the minority’s future rights. With respect, that was an error. The relevant number under s. 23 is the number of students who will eventually avail themselves of the service. This error had an impact in relation to five proposed schools being claimed by the appellants, those for the communities of Abbotsford (elementary component for children of rights holders in the community of Abbotsford and secondary component for children of rights holders in the communities of the Central Fraser Valley), Burnaby, Northeast Vancouver, East Victoria and West Victoria.

[96]                         The trial judge found that, in the short term, the number of minority language students in each of these communities falls below the high end of the sliding scale, and that these communities are therefore not presently entitled to homogeneous schools. However, she found that the number of students in each of these communities will in the long term warrant the creation of homogeneous schools (see trial reasons, at paras. 5064 and 5067 (Abbotsford), 5221 (Burnaby), 3805 (Northeast Vancouver), and 4068 (East Victoria and West Victoria)). Given that the trial judge’s long‐term projections are the relevant numbers for the determination of where the number of minority language students should be situated on the sliding scale, I find that these communities fall at the high end of the sliding scale and are entitled to homogeneous schools.

[97]                         The second type of error relates to the basis for comparison used to situate the numbers of minority language students on the sliding scale. The trial judge decided on a local basis for comparison. Unlike the courts below, I have concluded that the comparison must take schools located across the province into account. I will apply the appropriate comparative approach to the communities of North Victoria, Whistler, Chilliwack and Pemberton, where the appellants claim to be entitled to homogeneous schools that were denied by the courts below because of the basis for comparison they used.

[98]                         As I explained above, the first step of the approach to situating the number of students on the sliding scale is to establish the relevant number of students, that is, the number who will eventually avail themselves of the service. In this case, the trial judge found this number to be 98 for North Victoria, 85 for Whistler,^[2] 60 for Chilliwack and 55 for Pemberton.

[99]                         The second step is to compare these numbers with the numbers of students attending schools located across the province in order to determine what is appropriate from the standpoint of pedagogy and cost. According to the evidence in the record, there are 250 majority language schools with fewer than 100 students. Of these, the trial judge retained 11 schools with 100 or fewer students in British Columbia. The schools in question included Madeira Park Elementary, which has 72 students, Deroche Elementary, which has 66, and Dewdney Elementary, which has 73 (trial reasons, at para. 2693). The existence of these schools raises the presumption that it is appropriate from the standpoint of pedagogy and cost to create a school of comparable size for the minority.

[100]                     As I mentioned above, it is possible for the province to rebut this presumption by showing that the schools chosen as comparators are not appropriate for that purpose. In this regard, the trial judge found that a school created to serve the Big White ski resort and one at Port Clements that serves a particularly remote island community were not appropriate comparators. I agree. Schools that were created for a ski resort and for a community serviced only by air or water represent two genuinely exceptional situations that justify excluding them and not considering them as comparators.

[101]                     However, there is no evidence in the record on the basis of which the presumption could be rebutted in relation to the other small schools I mentioned above whose enrolment ranges from 66 to 73 students. What is more, each of these comparator schools, like the schools being proposed, is an elementary school. It is therefore appropriate to compare the relevant numbers of students in North Victoria, Whistler, Chilliwack and Pemberton with the numbers at those schools. I would reiterate that it is necessary to be flexible and that a comparable number does not mean an identical one. In my view, the relevant numbers for North Victoria, Whistler and Chilliwack are comparable to the numbers of students attending comparator majority language schools in British Columbia. They therefore fall at the high end of the sliding scale, and it is appropriate from the standpoint of pedagogy and cost for those students to receive their instruction in homogeneous schools.

[102]                     As for Pemberton, it is in my view difficult to compare the number of students in question there with the numbers of students at the majority language schools located elsewhere in the province that were retained by the trial judge. None of the 11 schools with fewer than 100 students that she retained has approximately 50 students. In spite of this, I find, for two reasons, that it is preferable to remand this question to the court of original jurisdiction for reconsideration. First of all, the available evidence is limited. In retaining the existence of 11 schools with 100 or fewer students in British Columbia, the trial judge did not consider all of the province’s small schools. The appellants’ record contains a list of over 250 schools with fewer than 100 students in British Columbia (A.R., vol. XIX, at pp. 678‐84). The trial judge did not refer to the vast majority of those schools. This omission resulted from the fact that her comparison was mainly a local one. Given that the existence of the small schools in question is crucial to the determination of the rights of the rights holders in Pemberton, it would be preferable to be able to consider complete evidence in light of the principles set out in these reasons. Moreover, if the court were to find that the number of students in question falls below the high end of the sliding scale, additional submissions would then be necessary. As I explained above, where the number of students falls in the middle or at the low end of the sliding scale, deference must be shown to the level of services contemplated by the school board. The appellants did not have the opportunity to make such submissions in this case.

[103]                     I therefore reluctantly conclude that the question of the level of services warranted by the number of students in Pemberton must be remanded to the court of original jurisdiction for reconsideration. I am optimistic that the clarifications provided in these reasons will incite the parties to resolve this matter promptly. And it would of course be open to the appellants to end their action if, in light of these clarifications, they consider that the number of students in question is situated at the appropriate point on the sliding scale and that occupying a heterogeneous facility corresponds to the level of service to which the s. 23 rights holders are accordingly entitled.

B.            Does the Test Used to Assess the Quality of Instruction Provided to Official Language Minorities Vary With the Number of Minority Language Students?

[104]                     Section 23 gives an official language minority the right to an educational experience that is equivalent in quality to the experience provided to the majority. In Rose‐des‐vents, this Court discussed the proper approach for determining what constitutes instruction that is equivalent in quality in a case in which the number of minority language students was comparable to the numbers of students in local majority language schools.

[105]                     The Court stated that the focus must be on substantive equivalence, not on formal equivalence. In other words, it is not enough that public spending per student is the same for the minority and the majority; the quality of the overall educational experience must be meaningfully similar to that of the experience provided to the majority. The Court held that substantive equivalence must be assessed from the perspective of a linguistic minority parent who has to choose between the minority language school and a majority language school. Courts must consider whether reasonable parents would be deterred from sending their children to a school of the official language minority because it is meaningfully inferior to an available majority language school (Rose‐des‐vents, at para. 35).

[106]                     In the case at bar, however, the factual circumstances of certain communities differ from those of the Vancouver West community whose school was at issue in Rose‐des‐vents. In Rose‐des‐vents, the minority language school had 344 students and was overcrowded. It was not a small school. In some of the communities in question in the instant case, the numbers of minority language students are much lower than the numbers of students attending local majority language schools. The Court must therefore determine whether the approach developed in Rose‐des‐vents is applicable in the context of small minority language schools whose size is not comparable to that of the majority language schools.

[107]                     In my view, children of s. 23 rights holders must receive an educational experience that is substantively equivalent to the experience provided to the majority, regardless of the size of the school or program in question. The substantive equivalence test applies everywhere on the sliding scale. The Rose‐des‐vents approach must, however, be applied in light of the specific context of official language minority schools that are not comparable in size to nearby majority language schools. I therefore consider it necessary to offer some guidance on how to assess the quality of the educational experience provided to the minority in the context of small schools. For this purpose, I will begin by identifying the key elements of the Rose‐des‐vents approach that are applicable, and I will then explain how they must be adapted.

[108]                     The approach from Rose‐des‐vents has three key elements. First, the relevant perspective is that of a reasonable parent who must choose between the school of the official language minority and the schools of the majority. As the Court stated, such a reasonable parent cannot expect “to have the ‘very best’ of every aspect of the educational experience” (para. 40). A reasonable parent considers many factors before choosing a school, such as the quality of instruction and of physical facilities (Rose‐des‐vents, at paras. 39‐40). Second, the educational experience does not meet the equivalence requirement if it is meaningfully inferior. This approach thus allows for some variations between minority and majority language schools. Third, minority language schools must be compared with majority language schools that represent a realistic alternative for the parent. The comparison is holistic and contextual, and involves considering all the factors a parent would take into account in choosing a school. I repeat that this approach was developed in a context in which the school in question was comparable in size to the majority language schools. How does this translate to a different context, one in which the minority language schools are smaller?

[109]                     I am of the opinion that the essentials of the approach do not need to be adapted in a situation in which the schools of the official language minority are small, aside from the fact that a reasonable parent must take into account the inherent characteristics of attendance at a small school. Let me explain.

[110]                     First of all, there is no need to adapt the “meaningfully inferior” school concept that was adopted in Rose‐des‐vents (para. 35). Official language minorities are entitled to have their children receive, anywhere in Canada, an educational experience that is substantively equivalent to the experience provided to majority language children. This is based on the principle of equal opportunity. The importance of providing an education of high quality to every child must not be underestimated, as children are “our country’s most important resource, our future” (Willick v. Willick, [1994] 3 S.C.R. 670, at p. 718). Section 23 allows all children, regardless of which official language they study in, to have the same opportunities for advancement in society.

[111]                     I arrive at the same conclusion with regard to the geographical scope of the comparison. The equivalence test must be interpreted in a manner that supports parents in the exercise of their rights. No matter what the size of the school concerned, linguistic minority parents will always have to choose between the minority language school and majority language schools. A comparison must therefore be made between the minority language school, even if it is small, and nearby majority language schools in order to determine whether the educational experience at the minority language school is meaningfully inferior. In this Court, the Province insisted that no comparison can be drawn between a small school and a large majority language school. I disagree. Where a school is not comparable in size to the majority language schools, a reasonable parent must be aware of the inherent characteristics of small schools. If these characteristics are kept in mind, it is quite possible to compare a small school with a larger one.

[112]                     A reasonable parent who is aware of the inherent characteristics of a small school will take into account the same factors as those listed in Rose‐des‐vents in assessing the quality of a school. These factors include quality of instruction, educational outcomes, quality of facilities, extracurricular activities that are offered, travel times, competence of the teachers, and cultural opportunities offered to the official language minority (Rose‐des‐vents, at paras. 39‐40). They are assessed globally in order to determine whether the educational experience at that school is inferior to the experience provided at the majority language schools.

[113]                     Where possible, a court should not undertake an analysis that takes into account only one part or one room of a school, or a single aspect of the educational experience. Even the right to instruction alone, which corresponds to the low end of the sliding scale, cannot be entirely severed from the overall educational experience. As was noted in Mahe, instruction “must take place somewhere and accordingly the right to ‘instruction’ includes an implicit right to be instructed in facilities” (pp. 369‐70). This means that where the number of students in question is not comparable to the numbers at the majority language schools — including where the number of students falls at the low end of the sliding scale — the factors listed in Rose‐des‐vents must be taken into account. In assessing the quality of the educational experience from a program of instruction, for example, the court must not confine itself to an assessment of the quality of instruction and educational outcomes. The quality of the building in which the instruction is provided is also relevant, as are travel times. A parent does not compare each schoolroom or each aspect of the educational experience individually, but assesses the school as a whole.

[114]                     The inherent characteristics of a small school are neutral considerations in the balancing of the factors used to assess the educational experience. They must not be taken into account in the equation. What these characteristics mean is, in particular, that a parent cannot expect there to be the same number of specialized teachers or the same range of facilities as at a big school. In addition, the size of a small school can sometimes require that multiple grades be combined in individual classes in order to attain a sufficient number of students. These are circumstances that a reasonable parent is prepared to accept if his or her child attends a small school.

[115]                     I would stress that in a country like Canada where the educational system is adequately funded, parents are entitled to expect an educational experience of high quality regardless of the size of their children’s school. For example, a school whose teachers are not properly trained cannot provide a substantively equivalent educational experience. The same is true of a small school whose building is in a state of repair clearly inferior to that of the majority language schools. It is also very hard to imagine how a school that has no gymnasium or outdoor play area for students can provide an educational experience that is substantively equivalent to the experience at majority language schools that do have such facilities. To me, these factors are minimum standards below which an educational experience cannot be said to be substantively equivalent.

[116]                     In short, the Rose‐des‐vents approach continues to apply where the number of students at a minority language school is comparable to the numbers of students attending nearby majority language schools. Where a minority language school is not comparable in size to nearby majority language schools, a judge must consider whether reasonable parents who are aware of the inherent characteristics of small schools would be deterred from sending their children to a school of the official language minority because the educational experience there is meaningfully inferior to the experience at available majority language schools.

[117]                     In light of these comments, the approach proposed by the Province and adopted by the courts below must be rejected. Relying on what she called a proportionality test rather than on that of substantive equivalence, the trial judge found that where the number of minority language students is not comparable to the numbers of majority language students, the equivalence of the educational experience must be assessed by considering “whether a reasonable rightsholder would find a minority school to be meaningfully disproportionate to the facilities offered to the majority, based on a local comparison of the global educational experience” (trial reasons, at para. 853 (emphasis added)).

[118]                     The effect of applying this proportionality test is to endorse an educational experience of inferior quality for official language minorities, with the degree of inferiority being determined by the size of the minority language school in comparison with that of the majority language school. Such an outcome is incompatible with the remedial purpose of s. 23. As this Court noted in Arsenault‐Cameron, the interpretation of s. 23 must be based on “the true purpose of redressing past injustices and providing the official language minority with equal access to high quality education in its own language, in circumstances where community development will be enhanced” (para. 27 (emphasis added)). I am aware that the trial judge did not mean to reduce the assessment of equivalence to a simple mathematical calculation according to which, for example, 100 students attending a minority language school would be entitled to an educational experience whose quality is half that of the experience provided to 200 students at a majority language school. Nevertheless, I do not see how this test could be applied without indeed becoming a simple mathematical calculation. I am especially concerned about what signal the application of such a test would send regarding the quality of the educational experience that must be provided to official language minorities.

[119]                     In addition, the concept of proportionality rests on the premise that the quality of a school depends primarily on the size of its student body. I cannot subscribe to this premise. There are a great many small majority language schools across the country, and they do not necessarily provide an educational experience that is proportionately inferior in quality to the experience provided by larger schools. Provincial and territorial governments ensure that, throughout their respective territories, all the students for whom they are responsible receive an educational experience that is substantively equivalent regardless of the size of the school they attend.

[120]                     In the case at bar, I am of the view that all children of rights holders who attend CSF schools or participate in its programs are entitled to an educational experience that is substantively equivalent to the experience of the majority. All communities served by the CSF are therefore entitled to schools or programs of instruction in which the educational experience is substantively equivalent to the experience at nearby majority language schools, and the trial judge’s conclusions must be varied to reflect this.

[121]                     Before applying the principles and clarifications set out above to the facts of this case, I would like to clarify the relationship between the position of a given number of students on the sliding scale and the assessment of the quality of the educational experience on the basis of the substantive equivalence test. I note that the assessment of the quality of the educational experience does not make it possible to change the position of a given number of students on the sliding scale in order to raise or lower the level of services the number warrants. These are two different stages that must be distinguished clearly.

[122]                     At the first stage, the application of the sliding scale concept makes it possible to establish the required level of services, that is, the range of services which varies from a program of instruction alone to a homogeneous school. The level of services reflects the extent of the control the minority is entitled to have over the provision of instruction. This concept of control clearly illustrates the differences between the levels of the sliding scale. At both homogeneous and heterogeneous schools, instruction is provided in a building that contains, to a greater or lesser extent, specialized facilities. At the high end of the sliding scale, where the official language minority is entitled to a homogeneous school, that minority has control over every aspect of the premises and of the provision of instruction. In the middle of the scale, at a heterogeneous school, for example, the minority has control over the provision of instruction but over only some aspects of the building, and must negotiate access to certain facilities, such as the gymnasium, with the majority.

[123]                     At the second stage, the analysis based on the substantive equivalence test is applied in order to make a holistic assessment of the quality of the educational experience provided to the official language minority. This analysis does not make it possible to move a given number of students higher or lower on the sliding scale. Rather, its purpose is to determine whether a reasonable parent would be deterred from exercising his or her language rights because the educational experience provided to the minority is meaningfully inferior to the experience provided to the majority. In the case of a homogeneous school, the analysis serves to determine whether the instruction and the facilities over which the minority has control are of sufficient quality. In the case of a heterogeneous school or a program of instruction, it serves to determine whether the instruction over which the minority has control and the facilities to which it has access are of sufficient quality.

[124]                     I will now discuss more specifically three schools in respect of which the appellants argue that the trial judge erred in assessing what constitutes an educational experience equivalent to that of the majority. The schools in question are in the communities of Nelson, Chilliwack and Mission. I note that these schools are small in comparison with nearby majority language schools. The quality of the educational experience provided in these schools must, as I mentioned above, be assessed from the perspective of a reasonable parent who is aware of the inherent characteristics of a small school. I will also discuss how the Facility Condition Driver used by the Province to prioritize its capital projects affects the CSF’s ability to provide high quality educational services.

[125]                     The appellants note that the school in Nelson (École des Sentiers‐alpins) does not have the facilities needed to offer a middle school enrichment program. It has only one specialized space at its disposal, an art classroom. One of the schools used as comparators, which does, however, have a much higher enrolment, has rooms for woodworking, sewing and cooking classes. The other school used as a comparator has no specialized spaces and does not even have a gymnasium or a library. The trial judge found that, because the facilities of the minority language school in Nelson are superior to those of one of the two comparator schools, the educational experience of the minority language students is equivalent to the experience provided to the majority language students.

[126]                     I agree. The appellants’ complaints are based on the absence of certain specialized workspaces for middle school students. Even on applying the substantive equivalence test and the proper approach I discussed above, I am of the view that a reasonable parent who is aware of the inherent characteristics of small schools would conclude that the absence of rooms for woodworking, sewing and cooking classes does not make the educational experience meaningfully inferior to the experience of students at the majority language schools. The limited number of specialized workspaces is an inherent characteristic of a small school. A reasonable parent would see this limitation as a neutral factor in his or her balancing exercise for the purpose of determining whether the educational experience is substantively equivalent to that of the majority. It is true that there is a nearby majority language school that has such facilities, but it is much larger than the minority language school in question. Moreover, one majority language school that is comparable in size has almost no specialized facilities. Absent arguments other than the lack of certain specialized workspaces, a reasonable parent could not conclude that the minority receives educational services inferior to those of the majority and be deterred as a result from exercising his or her rights under s. 23  of the Charter .

[127]                     The appellants also argue that the educational experience at the school in Chilliwack (École La Vérendrye) is meaningfully inferior to the experience at nearby majority language schools. First, the French‐language school does not have a proper gymnasium, but instead rents one in a community hall. The logistics for getting to the hall are complex and involve a number of steps. In addition, the gymnasium is too small for sports like basketball, and it is so cold in the room that jackets sometimes have to be worn there. The students do not use the gymnasium at the beginning of the week, because they would then have to deal with disorderly premises and the smell of alcohol as a result of community activities held in it on the weekend.

[128]                     The situation is quite different at the local majority language schools. Each of the local schools used as comparators has a gymnasium owned by the English-language school board. In almost every case, the gymnasium is located inside the school building, with the exception of one school where the gymnasium is in a separate building but is accessible by a covered walkway. As well, the gymnasiums at the local majority language schools are larger. Their average size is 390 square metres, whereas the one used by students of the French‐language school is 280 square metres.

[129]                     Second, the appellants complain about the quality of the library. The French-language school in Chilliwack does not have a real library. Books are instead stored in a portable classroom. Moving between the classrooms and the library is difficult. Owing to a lack of space, part of the library is also used to store sports equipment. A comparison between this library and those of the local majority language schools shows that the latter libraries are larger, with the exception of two that are smaller.

[130]                     The location of the school, the lack of space and the quality of the infrastructures also present problems. The school is in a rural area on the outskirts of Chilliwack. It is located on a busy industrial road with frequent and noisy heavy vehicle traffic, and is also surrounded by agricultural land on which activities causing strong, unpleasant odours take place. As the trial judge acknowledged, the fact that the school is not centrally located in the community means that it is not easily accessible to all of the area’s students, and also limits its ability to serve as a gathering place for the French‐speaking community.

[131]                     In addition, the quality of the building is greatly inferior to that of the nearby schools because of its age and its state of disrepair. The trial judge also noted that almost all of the students use school transportation and that the average ride time is twice as long as that of students attending the comparator schools. Despite the high proportion of minority language students who depend on school transportation, the buses are unable to use the school’s parking lot because of its small size.

[132]                     The trial judge weighed these disadvantages against the advantages of this school, noting that it provides services of high quality in French and that its classes are smaller than those of the schools with which it was compared. For example, the average kindergarten class has 14 students, whereas those of the majority language schools have 20. She also noted that the school offers a technological environment of superior quality.

[133]                     After balancing all of these factors, the trial judge concluded that the rights holders receive more than what they are entitled to on the basis of the number of students in question. In my view, this outcome illustrates the shortcomings of reasoning based on a proportional approach. For instance, the trial judge noted that if the size of the minority language school’s parking lot is assessed in proportion to the number of its students, it can be seen that this school has more parking spaces than any of the majority language schools used as comparators. With respect, this conclusion is puzzling given that this otherwise favourable proportion does not alter the fact that the buses that almost all the minority language students must use to travel to the school are unable to enter the parking lot. Similarly, the trial judge found that the gymnasium and the library are larger than the corresponding facilities of most of the majority language schools if they are analyzed on a proportional basis. The effect of the application of this proportionality‐based approach is to turn a gymnasium where most sports cannot be played and a library set up in a cramped portable classroom into facilities that are considered to be of sufficient quality.

[134]                     My view is instead that a balancing of the advantages and disadvantages of the French‐language school in Chilliwack shows that the quality of the educational experience provided there is meaningfully inferior to that of the experience at the majority’s schools. A reasonable parent, even one who is aware of the inherent characteristics of small schools, would likely be deterred from enrolling his or her child at this school. The children of rights holders in Chilliwack do not receive an educational experience of the quality guaranteed to them by s. 23  of the Charter .

[135]                     The appellants also maintain that the school in Mission (École des Deux‐rives) provides an educational experience that is inferior to the experience provided in the majority language schools. The problem that requires our intervention with respect to the school in Mission differs from the one relating to the school in Chilliwack. The trial judge found that the gymnasium used by the students of École des Deux‐rives is deficient in several ways. It is half the size of the gymnasiums at the nearby majority language schools. Its size is 166 square metres, whereas that of the comparator schools’ gymnasiums is 373 square metres. Even if the small class sizes at École des Deux‐rives are taken into account, this school’s gymnasium “underperforms [that of] every majority school”, which “makes it difficult to deliver core physical education services”, particularly for older students (trial reasons, at paras. 4957‐58). The gymnasium is so cramped that physical education periods must be divided up among students in the same class. And when they do have access to the gymnasium, students must restrain their movements for safety reasons, which limits the attainment of learning outcomes.

[136]                     The trial judge did not answer the question whether this constitutes an infringement of s. 23, however. She noted that it is instead necessary to assess the educational experience that is provided from a global perspective, as opposed to analyzing a single aspect of that experience. She “assume[d], without deciding, that a substandard physical education experience is sufficient to ground a breach of s. 23” (para. 4963). The trial judge also stated that one factor that had materially contributed to the infringement of s. 23 with respect to the minority language school in Mission — an infringement that was found to be justified under s. 1, however — was the Facility Condition Driver used by the Province to prioritize the funding of capital projects, because it does not take students’ pedagogical needs into account (trial reasons, at paras. 4990 and 5003; C.A. reasons, at paras. 246 and 253). The Facility Condition Driver takes only the life cycle of buildings into account, and is not concerned with a building’s ability to meet students’ pedagogical needs. The appellants submit that the Facility Condition Driver prevents them from providing an education that is substantively equivalent in Mission.

[137]                     The Court of Appeal upheld the trial judge’s decision and added that there was no infringement of s. 23, because the appellants had not introduced sufficient evidence to enable the trial judge to assess the global educational experience. The Court of Appeal also concluded that the evidence concerning the Facility Condition Driver was insufficient to support a finding that there was an infringement.

[138]                     The situation in Mission is concerning. The physical education classes provided at École des Deux‐rives are far from ideal. Unfortunately, as the Court of Appeal observed, the evidence that was adduced is insufficient for the purpose of making the holistic assessment required by the test I discussed above of a reasonable parent who is aware of the inherent characteristics of a small school. I also note that the evidence that was presented concerning the impact of the Facility Condition Driver is limited. I therefore conclude, reluctantly once again given the delays that have already occurred in this entire case, that the question of the quality of the educational experience in Mission and the impact of the Facility Condition Driver on this situation must be remanded to the court of original jurisdiction to be reassessed on the basis of the principles set out in these reasons.

C.            Does Compelling a School Board to Prioritize Its Capital Projects Infringe Section 23?

[139]                     The appellants contend that the Province may not, as a solution for clear infringements of s. 23  of the Charter , compel the CSF to prioritize the capital projects the latter submits. This prioritization obliges the CSF to accept that some infringements of s. 23 will persist longer than others. The Province counters that the government has a limited budget and that it is up not to the Province but to the CSF to prioritize projects in exercising its right of management. In the current situation, the Province would have to be a sort of mind reader when it comes to the CSF’s projects. I agree with the Province on this point.

[140]                     In my view, the prioritization required by the Province does not infringe the right of management guaranteed by s. 23  of the Charter . On the contrary, the Province is furthering the exercise of that right in asking the CSF to indicate where funds should be invested on a priority basis.

[141]                     The real question instead relates to the timeframe for a remedy. In other words, the appellants are asking this: How much time does the Province have to remedy the infringements of s. 23? This will have to be addressed on a case‐by‐case basis. The Province’s ability to pay and the CSF’s ability to manage multiple projects will have to be taken into account.

[142]                     The infringements must nonetheless be remedied in a timely fashion. This requirement is in keeping with the unique and distinct nature of s. 23. Section 23 rights are particularly vulnerable to foot‐dragging by public authorities because of the “numbers warrant” requirement in that section. The force of assimilation is such that the number of children of rights holders could fall irreversibly below the number needed to warrant the provision of services in a linguistic minority community while the authorities delay fulfilling their constitutional obligations (Doucet‐Boudreau, at para. 29). As this Court recently noted, “there is a critical need both for vigilant implementation of s. 23 rights, and for timely compliance in remedying violations” (Rose‐des‐vents, at para. 28).

D.           How Is an Infringement of Section 23 Assessed Under Section 1?

[143]                     This Court has held that it is possible to justify a limit on s. 23 language rights under s. 1  of the Charter (Quebec (Education, Recreation and Sports) v. Nguyen, 2009 SCC 47, [2009] 3 S.C.R. 208, at para. 37; Rose‐des‐vents, at para. 49; Ford v. Quebec (Attorney General), [1988] 2 S.C.R. 712, at pp. 772‐73). Where an infringement of s. 23 is established, a court must therefore take the approach established in Oakes for determining whether the infringement of a Charter  right can be justified in a free and democratic society. This Court has never yet determined that an infringement of s. 23 was justified under s. 1.

[144]                     In the instant case, the British Columbia Supreme Court, whose judgment was affirmed by the Court of Appeal, was of the view that the fair and rational allocation of public funds is a pressing and substantial objective that can justify an infringement of s. 23 in accordance with the Oakes test. Both of the courts below also held that assimilation is not a significant deleterious effect of the infringing measure, because minority language schools can only slow the inevitable process of assimilation in British Columbia. As well, the courts below found that the savings resulting from an infringement of s. 23 can be considered a salutary effect of the infringing measure. This analytical framework was applied to all the infringements of s. 23 alleged by the appellants, and it served to justify a number of those infringements. It is therefore necessary to clarify the applicable framework for determining whether an infringement of s. 23 is justified, as well as the role of financial considerations and linguistic assimilation in this framework.

[145]                     It will be helpful at this point to reiterate the criteria from Oakes before applying them in the specific context of s. 23. Oakes requires that two criteria be met in order to justify an infringing measure. First, the objective of the measure must be an important one. More specifically, the measure must “relate to concerns which are pressing and substantial in a free and democratic society before it can be characterized as sufficiently important” (pp. 138-39). This Court stated that such a standard must be high in order to ensure “that objectives which are trivial or discordant with the principles integral to a free and democratic society do not gain s. 1 protection” (p. 138).

[146]                     Second, once the importance of the objective is established, the party seeking to justify the infringement must show that the means chosen to attain the objective are reasonable. This second stage involves the application of a proportionality test that has three branches: (1) the measure must be rationally connected to the objective, (2) the means chosen should impair the right or freedom in question as little as possible, and (3) there must be a proportionality between the effects of the measure that is responsible for limiting the right or freedom and the objective that has been identified as important.

[147]                     In my opinion, three factors weigh in favour of applying a particularly stringent standard for justifying an infringement of the right to minority language instruction. First, the framers of the Charter  imposed positive obligations on the provincial and territorial governments in s. 23. Under s. 23, governments must provide public funding for minority language instruction where the number of children so warrants. They must fulfill these obligations in a timely fashion in order to avoid the likelihood of assimilation and of a loss of rights (Doucet‐Boudreau, at para. 29; Rose‐des‐vents, at para. 28). The adoption of a flexible approach to the justification of an infringement of s. 23 could jeopardize the section’s remedial purpose, the importance of which I explained above.

[148]                     Second, s. 23 is not subject to the notwithstanding clause in s. 33  of the Charter . The decision in this regard reflects the importance attached to this right by the framers of the Charter  as well as their intention that intrusions on it be strictly circumscribed. In Frank v. Canada (Attorney General), 2019 SCC 1, [2019] 1 S.C.R. 3, which concerned the right to vote of Canadians residing abroad, I reiterated McLachlin C.J.’s statement in Sauvé v. Canada (Chief Electoral Officer), 2002 SCC 68, [2002] 3 S.C.R. 519, that the framers had signalled the special importance of that right by excluding it from the scope of the notwithstanding clause. I added that, because of this exemption, any intrusions on the right are to be reviewed on the basis of a stringent justification standard (Frank, at para. 25; Sauvé, at paras. 11 and 14). This also applies in the context of s. 23.

[149]                     What s. 23 does is to protect an official language minority from the effects of decisions of the majority in the area of education by granting the minority a certain autonomy in relation to its education system. The history of the relationship between the majority and the minority in this area shows that the minority’s interests are not well served if it does not have some control over the management and funding of its schools. By excluding s. 23 from the scope of the notwithstanding clause, the framers of the Charter  sought to prevent the majority from being able to shirk its constitutional obligations and thus avert a return to the time when the minority was unable to develop in its own language and culture.

[150]                     Third, s. 23 has an internal limit, the numbers warrant requirement, according to which the exercise of the right for which the section provides will be warranted if there are a sufficient number of children, that is, of students. In adopting this limit, the framers sought to take account of practical considerations, including cost and pedagogical needs, related to the number of students who might benefit from the right in question. Where s. 23 is found to have been infringed, the government concerned will often advance a financial argument to justify the infringement. In such a case, the s. 1 analysis will then in some respects duplicate the numbers warrant analysis the court has already completed, because if there are enough students in a given case to exceed the numbers warrant threshold, the court will already have balanced the considerations related to cost and pedagogical needs in the first analysis. Balancing them again in the s. 1 analysis should normally lead to the same result. It would make no sense if considerations that justify the exercise of the right at one stage could also justify its infringement at a second stage. For an infringement of s. 23 to be justified under s. 1, it must not therefore be supported by considerations that have been taken into account at the numbers warrant stage.

[151]                     For the foregoing reasons, I am of the view that s. 23 is one of the Charter  provisions whose infringement is especially difficult to justify. I conclude that any intrusions on s. 23 must therefore be analyzed and justified on the basis of a very stringent standard.

[152]                     That being said, the Court must determine whether the fair and rational allocation of limited public funds can be a pressing and substantial objective in the s. 23 context. The courts below relied mainly on Newfoundland (Treasury Board) v. N.A.P.E., 2004 SCC 66, [2004] 3 S.C.R. 381, to hold that this objective is pressing and substantial. But as this Court has pointed out, “[t]o the extent that the objective of the law was to cut costs, that objective is suspect as a pressing and substantial objective under the authority in N.A.P.E.” (Health Services and Support — Facilities Subsector Bargaining Assn. v. British Columbia, 2007 SCC 27, [2007] 2 S.C.R. 391, at para. 147). In this context, the courts must therefore display strong scepticism in determining whether the objective is pressing and substantial (Health Services, at para. 147, quoting N.A.P.E., at para. 72).

[153]                     In my view, the courts below erred in ruling that “the fair and rational allocation of limited public funds” is a pressing and substantial objective in the case at bar. Public funds are limited by definition. Every government allocates its funds among its various programs on the basis of certain scales, and as fairly as possible. If merely adding the words “fair and rational” to the word “allocation” sufficed to transform the allocation of public funds into a pressing and substantial objective, it would be disconcertingly easy for any government to intrude on fundamental rights. I cannot accept such a result. The fair and rational allocation of limited public funds represents the daily business of government. The mission of a government is to manage a limited budget in order to address needs that are, for their part, unlimited. This is not a pressing and substantial objective that can justify an infringement of rights and freedoms. Treating this role as such an objective would lead society down a slippery slope and would risk watering down the scope of the Charter . I would add that, from a practical standpoint, the appropriateness of such an objective would be nearly impossible to verify.

[154]                     Given my conclusions on the existence of a pressing and substantial objective, it will not be necessary to consider the proportionality test, and in particular the rational connection and minimal impairment branches. However, I will offer some comments on proportionality between the effects of the measure that is responsible for limiting the right and the objective that has been identified as important.

[155]                     Regarding the proportionality stage, the trial judge did not consider the high rate of assimilation of French‐speaking students to be a particularly strong deleterious factor. She justified this conclusion by stating that the evidence showed that the rate of assimilation of French speakers is high in the province and that the number of schools of the official language minority does not have a significant impact on this.

[156]                     Yet it is clearly recognized that one of the purposes of s. 23 is to counter the assimilation of official language minority communities. In light of this purpose, it defies logic to suggest that the high rate of assimilation of French speakers in British Columbia is a deleterious effect that is not particularly strong. On the contrary, it is my view that such evidence attests to that community’s linguistic fragility, to its vulnerability. This means that it is necessary to be all the more careful in reviewing any infringement of s. 23 and to take assimilation fully into account as a deleterious effect when the right under that section is infringed.

[157]                     I am therefore of the opinion that the purpose of s. 23 is not only to ensure the sustainability of the country’s linguistic communities, which is a concern focused on the future, but also to make it possible for those communities to develop in their own language and culture, a concern focused on the present. In this sense, even though the evidence before the trial court shows that s. 23 has not been able to counter, or even slow, the process of assimilation in British Columbia, the fact remains that French‐speaking citizens in that province have a right to achieve fulfillment in their own language in everyday life. The context in which the right to minority language instruction is exercised in that province may be a difficult one, but under no circumstances should this interfere with the exercise of that right.

[158]                     I would add that, in analyzing the justification for an infringement of the right guaranteed by s. 23  of the Charter , a court must bear in mind that while it is true that s. 23 has a collective dimension, it also has an individual dimension, because the holders of the right in question are, first and foremost, individuals. It is difficult to assess the impact of minority language schools on the assimilation process at the collective level. However, these schools have a definite impact, at the individual level, on the life paths and likelihood of assimilation of French speakers who attend them. From an individual perspective, an infringement of s. 23 increases the likelihood of assimilation for those who are unable to attend a school of the official language minority.

[159]                     The trial judge found at the proportionality stage that the savings resulting from the infringements of s. 23 represented a salutary effect in the sense contemplated in Oakes. I cannot agree with that finding. Savings that result from an infringement of s. 23 cannot serve to justify the infringement. Such reasoning obscures the very purpose of s. 23, which is to protect the minority from effects deleterious to it that are caused by budgetary choices of the majority. Cost savings linked to an infringement of s. 23 cannot be considered a relevant factor in the balancing of the salutary and deleterious effects of the infringing measure.

[160]                     The application of these principles in this case leads me to conclude that the infringements of the appellants’ s. 23 rights are not justified.

[161]                     I concluded above that the Province has infringed the appellants’ rights in Abbotsford (elementary component for children of rights holders in the community of Abbotsford and secondary component for children of rights holders in the communities of the Central Fraser Valley), Burnaby, Chilliwack, Northeast Vancouver, East Victoria, West Victoria, North Victoria and Whistler. These infringements stem from a lack of appropriate educational services and from a deficiency in the quality of the educational experience to which the appellants are entitled under s. 23.

[162]                     In her reasons, the trial judge found that there were infringements of s. 23 in the communities of Pemberton and Victoria. Those infringements resulted, in Pemberton, from inadequate access to the core facilities required in order to provide an educational program and, in Victoria, from projected overcrowding at École Victor‐Brodeur. The trial judge also found that the appellants’ fundamental rights were infringed because the Province had denied the CSF an amount for building maintenance, the Annual Facilities Grant Rural Factor. However, the courts below held that each of these infringements was justified under s. 1. In my opinion, they were wrong.

[163]                     I am of the view that neither the infringements of s. 23 that I identify in these reasons nor the ones found to have occurred by the courts below can be justified in a free and democratic society. As I explained above, the first stage in the analysis of a justification for an infringement of s. 23 is the identification of the pressing and substantial objective being pursued by the legislature in question. In this case, the courts below found, for each of the infringements, that the objective being pursued was the fair and rational allocation of public funds. I have concluded that such an objective cannot be a pressing and substantial objective in the sense contemplated in Oakes. The justification for the infringements therefore fails at the first stage of the analysis. Without a valid objective, the Province cannot justify such infringements of s. 23  of the Charter .

E.            Does the Limited Government Immunity From Damages Awards Apply to Decisions Made in Accordance With Government Policies That Are Found to Be Contrary to Section 23?

[164]                     Governments have a limited immunity from damages awards in relation to infringements of the Charter . This immunity does not apply in the case of conduct that is “clearly wrong, in bad faith or an abuse of power” (Ward, at para. 39, quoting Mackin, at para 78). In the instant case, no one claims that the Province acted in bad faith. The issue is instead whether the immunity applies to decisions made in accordance with government policies.

[165]                     The limited immunity governments have can be raised when there are countervailing considerations against the payment of damages. Such considerations include, in particular, the existence of alternative remedies and concerns for good governance (Ward, at para. 33). The case at bar requires the Court to determine whether this limited immunity applies where an infringement results from a decision made in accordance with a government policy.

[166]                     The trial judge awarded damages to the CSF as a remedy for the infringement of s. 23 resulting from the inadequate funding of school transportation. The Court of Appeal, relying on Ward and Mackin, reversed the trial judge’s decision and cited the government’s limited immunity in this regard. It held that concerns for good governance justify affording governments immunity for decisions made in accordance with any type of government policy that is subsequently found to constitute an infringement. In my view, this was an error. There is no government immunity for decisions made in accordance with government policies.

[167]                     In Ward, this Court explained the approach to be taken in order to determine whether an award of damages is an appropriate remedy for an infringement of the Charter . A brief review of that approach will help to situate the limited government immunity in its proper context. The first step identified in Ward is to establish a Charter  infringement. The second is to determine whether damages would serve a useful function or purpose. The functions of compensation, vindication and deterrence can satisfy this requirement. At the third step, the government may raise its limited immunity to oppose a damages award by citing considerations such as the existence of alternative remedies and concerns for good governance. These are not the only possible considerations, as others may be identified over time. The fourth step is to determine the quantum of the damages.

[168]                     This Court made it clear that there will be situations in which good governance concerns justify requiring that a minimum threshold of gravity be crossed before damages are awarded (Ward, at para. 39). One such situation had arisen in Mackin, in which a claimant sought damages for a government’s acts pursuant to a law that had been duly enacted but was subsequently found to violate the Charter . In that context, the Court concluded that the possibility of a damages award could have a chilling effect on the work of those who make laws and those who enforce them, owing to a fear of being held liable. Such an outcome would be unacceptable, because the legislature and those who enforce laws must be able to perform their functions without fear of reprisals. This means that there must be a minimum threshold of gravity and that, absent conduct that is “clearly wrong, in bad faith or an abuse of power”, damages may not be awarded under s. 24(1)  for acts carried out pursuant to a law that is subsequently declared to be invalid and of no force or effect under s. 52(1)  of the Constitution Act, 1982  (Ward, at para. 39, quoting Mackin, at para. 78).

[169]                     The Province argues that Ward extended the scope of the government immunity to include decisions made by a government in accordance with its policies. This Court noted at para. 40 of that case that “the state must be afforded some immunity from liability in damages resulting from the conduct of certain functions that only the state can perform. Legislative and policy‐making functions are one such area of state activity. The immunity is justified because the law does not wish to chill the exercise of policy‐making discretion.” Considered in its context, the concept of policy making relates to government policies that are based on laws. The Province’s argument in this regard must fail. Only one situation in which the limited immunity applies was recognized in Ward, that of government decisions made under laws that were duly enacted but were subsequently declared to be invalid.

[170]                     The Court did not, however, rule out the possibility that other situations might eventually justify the recognition of a limited immunity on grounds of good governance (Ward, at para. 42). This raises the question whether the possibility of damages being awarded in relation to decisions made by a government in accordance with its own policy could undermine good governance. In my view, awarding such a remedy in this context is unlikely to have a chilling effect on government actions and thereby undermine their effectiveness.

[171]                     On the contrary, the possibility of damages being awarded in respect of Charter ‐infringing government policies helps ensure that government actions are respectful of fundamental rights.

[172]                     An overly broad application of the limited government immunity could in fact have undesirable effects. It would permit a government to avoid liability for damages simply by showing that its unlawful actions are authorized by its policies.

[173]                     This concern is accentuated by the fact that government policies do not usually result from a public process and that the “government policy” concept has not been defined. Does it concern any form of directives or guidelines issued by the government? If so, the immunity the Court is being asked to recognize would be very broad. In contrast, a law is an easily identifiable instrument. It is the product of a vote taken by a legislative body. The preparation of a law is a transparent public process that is central to the democratic process. It is therefore appropriate to give the government, in respect of a well‐defined instrument such as a law, an immunity that it is nonetheless inadvisable to give it for undefined instruments with unclear limits, such as government policies. The latter approach would reduce the chances of obtaining access to justice and an appropriate and just remedy for individuals whose rights have been infringed. The effect of expanding the scope of the government immunity like this is that bringing an action for damages in response to a Charter  infringement would become illusory.

[174]                     In Ward, at para. 38, McLachlin C.J. in fact warned against an overly broad interpretation of good governance concerns that would have the effect of precluding any type of remedy:

Good governance concerns may take different forms. At one extreme, it may be argued that any award of s. 24(1) damages will always have a chilling effect on government conduct, and hence will impact negatively on good governance. The logical conclusion of this argument is that s. 24(1) damages would never be appropriate. Clearly, this is not what the Constitution intends. Moreover, insofar as s. 24(1) damages deter Charter  breaches, they promote good governance. Compliance with Charter  standards is a foundational principle of good governance.

[175]                     To extend the government immunity to all decisions made in accordance with government policies amounts to adopting the “extreme” form of good governance concerns that was explicitly rejected in Ward.

[176]                     Moreover, the government immunity for acts carried out under laws that were duly enacted is justified by the importance of the legislative process. In Mackin, at para. 78, Gonthier J. endorsed these remarks of Dussault and Borgeat concerning the liability of governments in civil matters, and noted that they apply in the Charter  context:

     In our parliamentary system of government, Parliament or a legislature of a province cannot be held liable for anything it does in exercising its legislative powers. The law is the source of duty, as much for citizens as for the Administration, and while a wrong and damaging failure to respect the law may for anyone raise a liability, it is hard to imagine that either Parliament or a legislature can as the lawmaker be held accountable for harm caused to an individual following the enactment of legislation. [Footnotes omitted.]

(R. Dussault and L. Borgeat, Administrative Law: A Treatise (2nd ed. 1990), vol. 5, at p. 177)

[177]                     In this passage, the limited government immunity is justified by the fact that the law is the “source” of duty for the government. The enactment of laws is the fundamental role of legislatures, and the courts must not act so as to have a chilling effect on the legislatures’ actions in this regard. When the legislative branch enacts a law, it confers powers on the executive branch. In contrast, a minister’s decisions respecting school transportation are not a “source” of duty for the government in the same way as a law. When the executive branch adopts a government policy, it confers powers on itself. In light of this distinction, there is good reason not to extend the limited government immunity to government policies.

[178]                     In my view, there is no need in this case to identify every act that might be considered to be carried out under a law. For example, this appeal does not raise the question whether the government has a limited immunity when it makes orders in council or regulations pursuant to laws that have been duly enacted, and I will take no position on that question.

[179]                     In short, the general rule continues to be that damages can be awarded against a government where they are an appropriate and just remedy in the circumstances. However, the government may avoid such an award by raising concerns for effective governance. One situation in which it may oppose the payment of damages is where a law has been declared to be invalid after the act that caused the infringement, but it does not have immunity in relation to government policies that infringe fundamental rights.

[180]                     In the case at bar, the trial judge awarded $6 million in damages for chronic underfunding of the CSF’s school transportation system between 2002‐3 and 2011‐12. The Province had frozen funding for school transportation during that period, which coincided with a significant increase in the CSF’s enrolment, and this resulted in a deficit of between $6 and $14 million. However, the Court of Appeal reversed the decision to award damages on the basis of the limited government immunity and of the fact that damages are not an appropriate remedy absent conduct that is wrong. Because I have concluded that the limited government immunity does not apply in respect of government policies, and because the freeze on school transportation funding was a government policy, I would set aside the Court of Appeal’s conclusion and restore the trial judge’s order awarding damages for the inadequate funding of school transportation.

[181]                     The trial judge also found that the CSF had been denied $1.1 million by not having benefited from the Annual Facilities Grant Rural Factor. She held that this was an infringement of s. 23 but that the infringement was justified under s. 1. Given that I concluded above that this infringement was not justified, the appellants are therefore entitled to $1.1 million in damages on the basis that insufficient funding was granted for rural minority language schools. Since the parties are not disputing the quantum of the damages awarded by the trial judge, there is no reason not to accept it.

VI.         Disposition

[182]                     I would allow the appeal in part and set aside the judgment of the Court of Appeal.

[183]                     I would issue the following declarations in favour of rights holders under s. 23  of the Charter  who live in the communities listed below; the effect of these declarations is to replace the corresponding declarations issued by the trial judge or to add to her declarations.

(a)         Rights holders in the Abbotsford area are entitled to have their elementary school age children receive instruction at a homogeneous minority language school with space for 85 elementary school students (or such other number as the parties agree to) that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools (this declaration replaces the trial judge’s declaration 13 (see trial reasons, at para. 6834 m))). The lack of minority language educational facilities in Abbotsford prevents the CSF from providing a global educational experience that is substantively equivalent to the experience at nearby majority language elementary schools (this declaration and the corresponding declaration for secondary school age children replace the trial judge’s declaration 14 (see trial reasons, at para. 6834 n))).

(b)         Rights holders in the Central Fraser Valley area (Abbotsford, Mission and Chilliwack) are entitled to have their secondary school age children receive instruction at a homogeneous minority language school that is a combined elementary and secondary school, that has space for 120 secondary school age students (or such other number as the parties agree to) and that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools (this declaration replaces the trial judge’s declaration 12 (see trial reasons, at para. 6834 l))). The lack of minority language educational facilities for secondary school age students in the Central Fraser Valley prevents the CSF from providing a global educational experience that is substantively equivalent to the experience at nearby majority language secondary schools (this declaration and the corresponding declaration for elementary school age children replace the trial judge’s declaration 14 (see trial reasons, at para. 6834 n))).

(c)         Rights holders living in the proposed Burnaby catchment area are entitled to have their elementary school age children receive instruction at a homogeneous minority language school with space for 175 students (or such other number as the parties agree to) that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools (this declaration replaces the trial judge’s declaration 15 (see trial reasons, at para. 6834 o))).

(d)         Rights holders in the Chilliwack area are entitled to have their elementary school age children receive instruction at a homogeneous minority language school with space for 60 students (or such other number as the parties agree to) that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools. The educational facilities presently available to École La Vérendrye do not allow the CSF to provide a substantively equivalent educational experience.

(e)         Children of rights holders in the Kelowna area are entitled to an educational experience that is substantively equivalent to the experience at nearby majority language schools.

(f)          Children of rights holders in the Mission area are entitled to an educational experience that is substantively equivalent to the experience at nearby majority language schools.

(g)         Children of rights holders in the Nanaimo area are entitled to an educational experience that is substantively equivalent to the experience at nearby majority language schools.

(h)         Children of rights holders in the Pemberton area are entitled to an educational experience that is substantively equivalent to the experience at nearby majority language schools. The educational facilities presently available to École La Vallée in Pemberton do not allow the CSF to provide a substantively equivalent educational experience.

(i)           Rights holders living in the catchment area of École Entre‐lacs in Penticton are entitled to have their elementary and middle school age children receive instruction at a homogeneous minority language school with space for 175 students (or such other number as the parties agree to) that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools. The educational facilities presently available to École Entre-lacs do not allow the CSF to provide a substantively equivalent educational experience (this declaration replaces the trial judge’s declarations 7 and 8 (see trial reasons, at paras. 6834 g) and h))).

(j)           Rights holders living in the catchment area of École du Pacifique in Sechelt are entitled to have their elementary school age children receive instruction at a homogeneous minority language school with space for 90 students (or such other number as the parties agree to) that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools. The educational facilities presently available to École du Pacifique do not allow the CSF to provide a substantively equivalent educational experience (this declaration replaces the trial judge’s declarations 5 and 6 (see trial reasons, at paras. 6834 e) and f))).

(k)         Rights holders in the Squamish area are entitled to have their elementary school age children receive instruction at a homogeneous minority language school with space for 135 students (or such other number as the parties agree to) that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools (this declaration replaces the trial judge’s declarations 2 and 3 (see trial reasons, at paras. 6834 b) and c))).

(l)           Rights holders living in the proposed catchment area for the Northeast Vancouver elementary school project are entitled to have their elementary school age children receive instruction at a homogeneous minority language school with space for 270 students (or such other number as the parties agree to) that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools (this declaration replaces the trial judge’s declaration 11 (see trial reasons, at para. 6834 k))).

(m)       Rights holders living in the proposed East Victoria catchment area are entitled to have their elementary school age children receive instruction at a homogeneous minority language school with space for 275 students (or such other number as the parties agree to) that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools.

(n)         Rights holders living in the proposed North Victoria catchment area are entitled to have their elementary school age children receive instruction at a homogeneous minority language school with space for 98 students (or such other number as the parties agree to) that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools.

(o)         Rights holders living in the proposed West Victoria catchment area are entitled to have their elementary school age children receive instruction at a homogeneous minority language school with space for 299 students (or such other number as the parties agree to) that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools.

(p)         Rights holders in the Whistler area are entitled to have their elementary school age children receive instruction at a homogeneous minority language school with space for 85 students (or such other number as the parties agree to) that provides an educational experience that is substantively equivalent to the experience at nearby majority language schools.

[184]                     I would remand the following questions to the court of original jurisdiction for reconsideration:

(a)         the question of the educational experience at École des Deux‐rives in Mission and the impact of the Facility Condition Driver on the CSF’s ability to provide a substantively equivalent education; and

(b)         the question of the level of services warranted by the number of students in the Pemberton area.

[185]                     I would also restore the trial judge’s order concerning school transportation, and I order the respondents to pay $6 million in Charter  damages to the CSF over a period of 10 years in respect of the inadequate funding of school transportation from 2002‐3 to 2011‐12.

[186]                     I also order the respondents to pay $1.1 million in damages to the CSF to compensate it for the amount it was denied in respect of the Annual Facilities Grant Rural Factor.

[187]                     With costs to the appellants throughout.

The following are the reasons delivered by

                     Brown and Rowe JJ. (dissenting in part) —

I.               Introduction

[188]                     The fundamental importance of bilingualism to Canadian society was affirmed by this Court in Conseil scolaire francophone de la Colombie‐Britannique v. British Columbia, 2013 SCC 42, [2013] 2 S.C.R. 774, at para. 106. Similarly, in Mahe v. Alberta, [1990] 1 S.C.R. 342, Dickson C.J. described the “vital role of education in preserving and encouraging linguistic and cultural vitality” as the “linchpin” of Canada’s commitment to bilingualism and biculturalism: p. 350. And the promotion and preservation of bilingualism through education also connotes a commitment to official minority language rights. To borrow from L’Heureux‐Dubé J.’s felicitous metaphors in Lifchus, bilingualism and minority language rights are “forever as closely linked as Romeo with Juliet or Oberon with Titania and they must be presented together as a unit”: R. v. Lifchus, [1997] 3 S.C.R. 320, at para. 27.

[189]                     As our colleague notes, it is impossible to separate the history of Canada from the history of its official languages: paras. 5‐12. Language and culture are inextricably linked to the Canadian identity. This is one of the reasons that bilingualism and the protection of official linguistic minorities have long been, and continue to be, of profound concern to many Canadians. The entrenchment of these two concepts in the Constitution speaks volumes about their importance.

[190]                     Unlike most rights under the Charter , s. 23  imposes positive duties on governments to act. Its framing as a positive right is particularly significant in this regard. As highlighted by the Chief Justice, the objective of s. 23 is threefold, being at once preventative, remedial, and unifying: para. 15. In order to give s. 23 its full effect, courts must interpret it in light of that objective. But in doing so, they must be mindful of the way in which s. 23 has been framed. In particular, they must recognize that it expresses its own internal limit, the “numbers warrant” criterion ⸺ the product of a carefully struck constitutional bargain, which expressly calls upon courts, and has been interpreted as calling upon courts, to apply considerations of pedagogy and costs. As a result, a thorough limitations analysis is required under s. 23 itself, which will normally leave s. 1  of the Charter  with a lesser role.

[191]                     Like the Chief Justice, we are of the view that a purposive interpretation of s. 23 requires that substantive equivalence apply throughout the sliding scale of entitlement in recognition that the quality of official minority language education cannot be meaningfully inferior to that of the majority. We therefore reject the lower courts’ conclusion that “proportionality” is the applicable standard in the middle or lower ends of the sliding scale. We recognize that the lower courts did not intend to diminish the importance of s. 23 in applying a “proportionality” standard; however, in our view, the use of such a norm risks depriving rights holders of their lawful entitlement under s. 23.

[192]                     We also agree with the Chief Justice that no infringement of s. 23 arises based on the prioritization requirement: paras. 139-40.

[193]                     Despite our substantial agreement with the Chief Justice, this area of our law is fraught with questions on which reasonably held differences can and will arise. We stress that our positions are not far apart. While we agree with him on goals, we disagree on how they are best achieved. To our mind, there are important reasons for maintaining a more conventional approach in line with settled jurisprudence. As we will explain below, our differences stem principally from the Chief Justice’s use of a province‐wide presumption of pedagogical and cost appropriateness in the “numbers warrant” analysis. This presumption affects key elements of the s. 23 analysis, leading to a compression of the middle of the scale and to a strained application of the substantive equivalence test. Indeed, our colleague’s approach departs from the settled approach to the sliding scale, which has been foundational to s. 23 since Mahe. The new presumption leads to an application of the “numbers warrant” requirement that fails to give proper effect, in our respectful view, to s. 23’s important internal limitation. While our colleague acknowledges this internal limit, he does not give effect to it in the manner that it has been given effect in our jurisprudence to date. We are not persuaded that such a departure from earlier jurisprudence is warranted.

[194]                     We are in full accord with the Chief Justice that Charter  rights must be given a broad, liberal, and purposive interpretation. But, doing so includes giving proper effect to a key part of s. 23, specifically, the “numbers warrant” requirement. Care must be had to give to s. 23, as to any right, only that scope and content that its text can bear: Gould v. Yukon Order of Pioneers, [1996] 1 S.C.R. 571, at paras. 13 and 49-50. As this Court has said, “the Charter  was not enacted in a vacuum”, and its guarantees ought to be understood in their “proper linguistic, philosophic and historical contex[t]” so as not to “overshoot the actual purpose of the right or freedom in question”: R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295, at p. 344.

[195]                     As for the analysis under s. 1  of the Charter , we agree with the Chief Justice that the courts below erred in accepting that the “fair and rational allocation of limited public funds” was a valid pressing and substantial objective, albeit for slightly different reasons. We also agree that the courts below erred in discounting the serious effects of assimilation in characterizing heightened assimilation as not particularly deleterious. To the contrary, s. 23 must serve as a bulwark against assimilation.

[196]                     The Chief Justice also concludes that the immunity from Charter  damages set out in Mackin v. New Brunswick (Minister of Finance), 2002 SCC 13, [2002] 1 S.C.R. 405, does not extend to government policies. Again, with respect, we must disagree. This Court’s jurisprudence has consistently framed the principle in terms sufficiently broad to cover policies, and there is no principled reason to distinguish between legislation and other instruments, notably regulations and policies. Indeed, such regulations and policies can be made only where they are authorized by legislation. Thus, the distinction between instruments is artificial. However, an exception to this general immunity is warranted in the case of s. 23, which is a positive right that requires the expenditure of public funds and is particularly susceptible to government inaction. Thus, we find ourselves in agreement with our colleague that Charter  damages are properly awarded in the circumstances of this case, but differ as to the general statement of the law with respect to Charter  damages in other contexts.

[197]                     Finally, as grounds for revisiting Mahe, the Chief Justice raises concerns about access to justice and long delays: paras. 20 and 56. But the delay in this case resulted not from a lack of clarity in the law, but from the scope and magnitude of the claim. As we highlight below (at para. 200), the appellants effectively challenged the provision of minority language education throughout the entire province of British Columbia. Notwithstanding what can only be described as the remarkable efforts of the trial judge, the adjudication of such a claim would inevitably consume a substantial amount of time. More fundamentally, however, this appeal is concerned not with access to justice but with the content of that justice. It remains our view that while significant remedial measures have been shown to be warranted in British Columbia, that is not a sound basis to change the rules that will henceforth operate not only in that province but also in the other nine provinces and three territories. We are unpersuaded that circumstances requiring specific remedies in British Columbia constitute a sufficient basis to reframe the general operation of s. 23 for every jurisdiction in the country. Thus, as set out below, our differences rarely relate to what is needed on the facts of this case and ordinarily relate to whether a more general reframing of the rules is needed.

II.            Facts and Decisions Below

[198]                     Like the Chief Justice, we note that this case deals with the interaction of s. 23 and other constitutional provisions: para. 2. The main issue in this case is, however, more specific, pertaining to how the s. 23 analysis applies when the number of students is at the middle or lower ends of the sliding scale. Indeed, the courts below, the parties in this appeal, and several intervenors framed the issue as determining how the s. 23 analysis applies when the numbers are not at the high end of the scale. In particular, we must consider whether substantive equivalence applies when the number of rights holders’ children does not warrant the creation of a homogeneous school.

[199]                     This issue is somewhat obscured in the Chief Justice’s analysis, as the new approach to s. 23 tends to cause the middle of the sliding scale to disappear. While he frames the question as being how to place the relevant number on the sliding scale, our precedents (notably Mahe) already state the general principles dealing with this issue, directing us on how to draw from them. Some refinement may well be in order. That often arises as one is called on to apply settled principles in the context of specific cases. But, in our respectful view, our colleague goes beyond refinement in the application of settled principles and instead substitutes new principles, notably in the form of a province‐wide presumption of pedagogical and cost appropriateness.

[200]                     While we agree with the Chief Justice’s account of the facts, we highlight certain facts that demonstrate the exceptional complexity of this case. The trial lasted 238 days (over a span of approximately three years) and led to one of the longest ⸺ if not the longest ⸺ trial decision in Canadian history (2016 BCSC 1764). The trial judge heard from over 40 lay witnesses and 13 experts. She was presented with over 1,600 exhibits and over one thousand pages of written argument. There were two categories of claims: (1) systemic or global claims; and (2) community claims. In all but one community claim, the appellants sought the maximum entitlement of a new homogeneous school. Finally, the appellant Conseil scolaire francophone de la Colombie‐Britannique (“CSF”) estimated that the total cost of implementing its claimed entitlements would be over $300 million, not including the cost of site acquisition. As noted by the trial judge, in 2013 the CSF had requested $350 million in capital funding, which was roughly equivalent to the total capital funding distributed to all of British Columbia’s school districts in that year. We can readily agree that the whole dispute took far too long, but in part that is because the appellants put virtually everything in issue and took maximalist positions on every issue. That is their right, but the fact that the litigation has taken a very long time is in no small measure a function of the wide scope of what is claimed.

III.         Analysis

A.           Historical Relevance of Section 23

[201]                     At paragraphs 5 to 12 of his reasons, the Chief Justice provides a historical overview of language rights in education in Canada in order to fully appreciate their role in guiding the s. 23 analysis. This historical overview illustrates that s. 23, in particular the limiting criterion of “where numbers warrant”, reflects a bargain struck between the federal and provincial governments. This bargain is relevant because, as we will see, it is mirrored in the limitation contained in the text of s. 23. The issue is therefore not whether s. 23 ought to be given a broad versus restrictive interpretation as the Chief Justice suggests (at paras. 18‐19), but rather about giving s. 23 its proper and intended effect.

[202]                     Minority language education rights have long been considered as worthy of constitutional protection. In a 1969 White Paper, Prime Minister Trudeau proposed to extend language guarantees and to include them in a charter. The paper proposed, specifically:

4. The Charter  should also recognize and guarantee with respect to the English and French language.

. . .

(e) the right of the individual to have English or French as his main language of instruction in publicly supported schools in areas where the language of instruction of his choice is the language of instruction of choice of a sufficient number of persons to justify the provision of the necessary facilities. [Emphasis added.]

(P. E. Trudeau, The Constitution and the People of Canada (1969), at pp. 54 and 58)

[203]                     Given the general provincial reticence about a proposed charter, and given also that language rights pertaining to education fell within the legislative authority of the provinces, early proposals regarding minority education were expressly omitted from the Victoria Charter of 1971: M. D. Behiels, Canada’s Francophone Minority Communities: Constitutional Renewal and the Winning of School Governance (2004), at p. 23. Although a Special Joint Committee of the Senate and of the House of Commons established after the failure of the Victoria Charter recommended constitutional entrenchment of a right to minority language instruction, the federal government chose to focus on language policy within matters of federal jurisdiction ⸺ for instance, in the federal public service: Behiels, at pp. 23‐24 and 33.

[204]                     A shift would later occur. In a 1977 policy statement, the federal government stated a more ambitious and broader language policy. Acknowledging that “this involve[d] matters of provincial jurisdiction” (“Preface” by P. E. Trudeau in A National Understanding: Statement of the Government of Canada on the official languages policy (1977), at p. 12), the federal government would seek “the active support of the provinces, which controlled crucial areas such as education, the courts, social and health services, and culture”: Behiels, at p. 33.

[205]                     That year, following the 18th Annual Premiers’ Conference in St. Andrews, New Brunswick, the premiers issued a statement on language, agreeing to “make their best efforts to provide instruction in education in English and French wherever numbers warrant”: Canadian Intergovernmental Conference Secretariat, 18th Annual Premiers’ Conference, Statement on Language, Doc. 850‐8/027 (August 18‐19, 1977) (emphasis added). The Council of Education Ministers was also asked to report on the state of minority language education in the provinces. Its report was ratified at a special conference in Montreal in February 1978, at which the Premiers resolved that “[e]ach child of the French‐speaking or English‐speaking minority is entitled to an education in his or her language in the primary or the secondary schools in each province wherever numbers warrant”: Canadian Intergovernmental Conference Secretariat, Premiers’ Conference, Communiqué of the Conference, Doc. 850‐9/007 (February 23, 1978) (emphasis added); Behiels, at p. 41.

[206]                     The draft version of what eventually became s. 23 was subject to scrutiny and debate before the Special Joint Committee of the Senate and of the House of Commons on the Constitution of Canada. In one notable exchange, then‐Minister of Justice and Attorney General of Canada Jean Chrétien, responding to New Brunswick Member of Parliament Eymard Corbin’s concerns about the term “where numbers warrant”, referred back to the agreements of the premiers as follows:

     Mr. Chrétien: Mr. Corbin, you have very clearly expressed the reason why we have maintained that phrase. Those were the terms used by the Provincial Premiers at the time of the 1978 Agreement in Montreal.

. . .

     Perhaps we could have dropped this expression and the courts would have reached the same decisions, but on this issue we felt it was appropriate not to exceed the words chosen by the Provincial Premiers.

     Mr. Corbin: Mr. Minister, what distresses me in all this . . .

     Mr. Chrétien: In fact, the oddest and the most surprising thing is that Mr. Hatfield, your Premier of New Brunswick, does not like the expression “where numbers warrant”, but he was a party to the Montreal agreement.

     Mr. Corbin: Yes, especially since the first agreement was reached at St. Andrews.

     Mr. Chrétien: In St. Andrews, New Brunswick. [Emphasis added.]

(Minutes of Proceedings and Evidence of the Special Joint Committee of the Senate and of the House of Commons on the Constitution of Canada, No. 38, 1st Sess., 32nd Parl., January 15, 1981, at pp. 36-37)

[207]                     Given that the Charter  required provincial support — s. 23 even more so, being a provision drafted by the federal government that imposed positive obligations on the provinces in an area over which they had exclusive legislative authority — it makes sense that the words were chosen in a manner that reflected the concerns of the provinces. Indeed, it is unsurprising. The Charter , after all, records a political agreement, whose contents reflect the negotiation and compromise that was necessary to secure the support for entrenching it in the Constitution.

[208]                     In this case, the use of the expression “numbers warrant” reflects a bargain, and records what the provinces accepted. It imposes a numerical threshold and ensures that the obligations on the provinces are reasonable and reflect what is practical. This limit was expressed within the text of s. 23 itself. This Court has confirmed the primacy of the written text of the Constitution on various occasions: see, e.g., Reference re Secession of Quebec, [1998] 2 S.C.R. 217, at para. 53 (“Secession Reference”); British Columbia v. Imperial Tobacco Canada Ltd., 2005 SCC 49, [2005] 2 S.C.R. 473, at para. 65; Caron v. Alberta, 2015 SCC 56, [2015] 3 S.C.R. 511, at para. 36.

[209]                     This Court has been mindful of both the text of s. 23 and its particular nature. In Mahe, this Court struck a balance by recognizing that it is possible to “breathe life” into the s. 23 right, but at the same time do so with caution: p. 365.

B.            The Analytical Framework of Section 23  of the Charter 

[210]                     The analysis to be applied to s. 23 claims follows two main steps. First, a court must determine the level of services warranted by the number of rights holders in a given area (the “numbers warrant” analysis). This entails ascertaining the “relevant” number of rights holders, then placing that number on a sliding scale of entitlement in order to decide what level of service is warranted. Secondly, courts must determine whether the quality of services granted to the rights holders is substantively equivalent to the quality of services provided to local majority language students (the “substantive equivalence” analysis).

(1)          The “Numbers Warrant” Analysis

[211]                     The analysis begins, then, by determining the relevant number. In Mahe, this Court stated that “the relevant figure for s. 23 purposes is the number of persons who will eventually take advantage of the contemplated programme or facility”: p. 384 (emphasis added). It added that this figure will fall somewhere between “the known demand for the service and the total number of persons who potentially could take advantage of the service”: ibid. We will refer to those two parameters as the “known demand” and the “potential demand”.

[212]                     This straightforward, estimates‐based approach has since been affirmed by this Court. In Arsenault‐Cameron v. Prince Edward Island, 2000 SCC 1, [2000] 1 S.C.R. 3, the trial judge had accepted evidence that approximately 155 children of s. 23 rights holders lived in the relevant area (the potential demand) and that this number could rise to 306 in the course of several years. He had also considered a survey indicating that the parents of 49 children were definitely interested in sending their children to a French school in Summerside and that this number could rise to 64 in the next three years (the known demand): Arsenault‐Cameron v. Prince Edward Island (1997), 147 Nfld. & P.E.I.R. 308 (P.E.I.S.C. (T.D.)), at para. 52; Arsenault‐Cameron v. Prince Edward Island (1998), 162 Nfld. & P.E.I.R. 329 (P.E.I.S.C. (App. Div.)), at para. 58. The Appeal Division reversed the judgment of the Trial Division, reasoning in part that only the known demand for the first three years of the program should frame the numbers analysis: ibid. This Court allowed the appeal, affirming not only the methodology we have described, but directing that it ought to be applied using present‐day estimates, both for known demand and potential demand. Relying on the evidence accepted by the trial judge, the Court held that the relevant number fell “somewhere between 49 and 155”: para. 59.

[213]                     This is not to say, however, that a court cannot rely on more sophisticated methods to supplement this basic approach. For example, there may be reliable statistical or actuarial evidence put before the trial judge projecting what the known demand and potential demand will be in the future, which might allow for more precision in identifying the numerical range within which the relevant number should fall. Indeed, Mahe (at p. 384) suggests that such evidence may be taken into account, as does this Court’s statement in Arsenault‐Cameron that “[t]he potential demand for services could be determined by inferring that the established demand would increase after the services actually became available”: para. 59.

[214]                     Whatever the degree of precision in the evidence, the important point is that the purpose of this assessment is to estimate the number of students that will eventually take advantage of the proposed program or facility. And it is an estimate. The goal is not to establish how many students will take advantage of the facility or program when it is first launched, but rather to try to forecast how many will do so in the future. We therefore agree with the Chief Justice when he says “it is not necessary in this case to fix a precise number of years as the length of a long-term projection”: para. 60. We also agree that the projection should take account of the long‐term planning of such projects, as well as the necessity to use estimates that remain reliable. Assessing how far into the future the projection can go should also be done on a case‐by‐case basis, and will depend on the quality and reliability of the evidence presented to the trial judge.

[215]                     After determining the relevant number, the court must then establish, using the sliding scale approach, what level of services is warranted, pedagogically, given the number of minority language students, and considering the cost of such service: Mahe, at p. 384. Here, our colleague proposes that a legal presumption of pedagogical and cost appropriateness should apply whenever a school of comparable size exists anywhere in the province: paras. 65 and 68-69. This is an innovation for which no adequate basis has been shown, in our view.

[216]                     As our colleague recognizes, the right to minority language education pursuant to s. 23, on its express terms and on this Court’s jurisprudence, is internally limited to the services that can be justified, pedagogically and financially, by the number of children of rights holders: paras. 25 and 150. To trigger the obligation to publicly fund minority language education, a claimant must demonstrate that this limit is accounted for. As is the general rule, “the burden of proving all elements of the breach of a Charter  right rests on the person asserting the breach”: P. W. Hogg, Constitutional Law of Canada (5th ed. Supp.), at p. 38‐7. This ensures that s. 23’s “own internal qualifications and its own method of internal balancing” are fulfilled: Mahe, at p. 369. The operation of the sliding scale gives effect to this internal limit: the content of the right expands as the numbers increase, thereby ensuring that rights holders receive a level of service that is appropriate to their number: Mahe, at p. 366. Significantly, this internal limit also takes account of certain contextual factors, such as the local or urban setting of the program or facility: Mahe, at p. 385; Association des parents de l’école Rose‐des‐vents v. British Columbia (Education), 2015 SCC 21, [2015] 2 S.C.R. 139, at para. 47. This is not fully recognized, we would suggest, by our colleague: para. 64.

[217]                     Proper placement on the sliding scale is critical for several reasons. First, making sure that the level of services accorded to minority language groups conforms to what the numbers warrant ensures not only that governments will deploy the resources necessary to meet their obligations under s. 23, but that the internal limit to those obligations is also accounted for. Governments are therefore mandated to do “whatever is practical in the situation to preserve and promote minority language education”: Mahe, at p. 367 (emphasis added). This Court has recognized, for example, that there is little that governments can do “in the case of a solitary, isolated minority language student”: ibid.

[218]                     The second reason that proper placement is crucial relates to the first point: just as a failure to give effect to the rights conferred by s. 23 can be detrimental to the flourishing of minority languages, improper placement on the scale — that is, not respecting the internal limit of s. 23 — can also be harmful to minority students. As Dickson C.J. explained in Mahe, not only would such a situation be impractical, it would also undermine the educational experience of minority rights holders:

There is no point, for example, in having a school for only ten students in an urban centre. The students would be deprived of the numerous benefits which can only be achieved through studying and interacting with larger numbers of students. [Emphasis added; p. 385.]

In short, properly determining the entitlement by taking into account the internal limit is a means of preserving s. 23’s internal limit, while at the same time providing the appropriate level of services for minority language students.

[219]                     The legal presumption proposed by the Chief Justice, however, has the effect of displacing the “numbers warrant” analysis. Considerations of pedagogy and cost are effectively withdrawn at the moment a school with comparable numbers can be found anywhere in the province, regardless of the particular context that may explain the continued existence of such a school. A right to a homogeneous school — the highest level of entitlement on the scale — is immediately presumed, thereby shifting the claimant’s burden to the province essentially from the outset. How a province could successfully rebut this presumption is unclear. Our colleague views minority language schools as somewhat akin to majority language schools that serve geographically isolated communities. The rebuttal of his presumption appears to be limited to “genuinely exceptional” cases such as when the school can be reached only by plane or ferry or otherwise requires significant travel to service a remote area: paras. 78-79. Thus, the presumption is, effectively, a rule: there will always be a homogeneous school, rather than one that shares facilities. And this will apply irrespective of why a small majority language school is maintained — including the fact that it may be the only way to provide students with instruction, indeed the only way for the community to continue to exist. While we are not in a position to make any calculation, one can safely assume that the cost implications of this will be quite considerable across ten provinces and three territories.

[220]                     As pointed out by the Chief Justice, this approach to determining the level of services is simple and predictable: para. 56. But simplicity and predictability are not necessarily virtues when they are achieved at the expense of other important considerations. Here, a claimant need simply show that there is somewhere in the province or territory a comparably sized majority language school; it would then follow that there is a (presumptive) entitlement to a similarly sized minority language school, without regard to other variables. Predictability may be thus achieved, but at the expense of having proper regard to relevant and important context; indeed, this Court has stated that “a number of complex and subtle factors must be taken into account beyond simply counting the number of students” in undertaking this analysis: Mahe, at p. 386 (emphasis added).

[221]                     We also respectfully suggest that the simplicity of the new approach is more apparent than real. While claimants can readily identify small schools in various parts of a province or territory, it will then be for the government of that jurisdiction to present evidence of each and every one of them in order to rebut the presumption ⸺ an unavoidably lengthy and cumbersome process. In that regard, the Chief Justice’s attempts to circumscribe the number of majority schools put forward as comparators to those that are truly comparable falls shorts of meeting his lofty goal. At para. 69, he writes: “. . . the claimants, who are responsible for identifying comparator schools, must endeavour to submit to the court a reasonable number of schools that, to the best of their knowledge, are appropriate comparators in light of the principles set out in these reasons” (emphasis added). With respect, given that it is the Chief Justice’s position throughout his reasons that the presumption may be rebutted only in “genuinely exceptional” cases, we are doubtful that claimants will refrain from invoking large numbers of the smallest majority schools across the province regardless of whether they are appropriate comparators to the situation the linguistic minority faces in an attempt to maximize gains.

[222]                     Nor is it evident what is to happen where the Province successfully rebuts the presumption (see our colleague’s reasons at para. 90.) We would suggest that one turns again to the type of contextual analysis described in Mahe, but it is not clear.

[223]                     Relatedly, as the new test has the practical effect of shifting the onus of proof, serious concerns can arise regarding trial fairness in the application of the new test. The Chief Justice states that “there is no evidence in the record on the basis of which the presumption could be rebutted in relation to the other small schools I mentioned above whose enrolment ranges from 66 to 73 students”: para. 101. That may well be accurate, but, as the Province did not know at trial that it had a presumption to rebut, it did not have as a practical matter the opportunity to seek to meet its (new) evidentiary burden. This is made more problematic by the circumstances in Pemberton, where “[t]he appellants did not have the opportunity to make such submissions in this case”: para. 102. If the appellants are to be given the opportunity to supplement their submissions, must the same opportunity not be accorded to the Province? And, why only Pemberton and not other communities? We understand the desire to bring finality to these questions, but fairness must not be lost sight of in that pursuit.

[224]                     The legal presumption operates as a fast track to reach the upper end of the sliding scale. One might ask whether the new approach would be read as replacing the sliding scale with an express elevator. This seems to us to be inconsistent with this Court’s refusal to adopt a view of s. 23 as encompassing only two rights (one with respect to instruction and one with respect to facilities) — an approach that the Court rejected in favour of the sliding scale approach, which allows for a progressive increase in entitlement as the number of rights holders increases: Mahe, at p. 366.

[225]                     The Chief Justice says that, under his approach, there is still space for the middle of the scale. We do not share his view. The legal presumption that is proposed is engaged as soon as a school serving a similar number of students can be found anywhere in the province, regardless of context, and favours granting the highest level of entitlement in any such case. This would essentially eliminate any middle level to the scale.

[226]                     Departing significantly from a precedent of our Court requires compelling reasons: R. v. Jordan, 2016 SCC 27, [2016] 1 S.C.R. 631, at para. 45, quoting Ontario (Attorney General) v. Fraser, 2011 SCC 20, [2011] 2 S.C.R. 3, at para. 56, and R. v. Henry, 2005 SCC 76, [2005] 3 S.C.R. 609, at para. 44. With respect, our colleague does depart from precedent; in this, we differ: paras. 71-72. We have already noted the extraordinary complexity of this case. There is no basis to conclude that s. 23 cases are usually met with abnormal delays, that there is a culture of complacency with regards to s. 23 rights, or that the test developed by this Court in Mahe and other cases is maladjusted to the reality of claimants; in this, again we differ from our colleague (paras. 20 and 56).

[227]                     How, then, does a court determine the level of service that is warranted for a given group of rights holders? We suggest a two‐step approach that draws substantially from our precedents, from as early as Mahe to as recently as Rose‐des‐vents. First, the claimant must demonstrate, on a balance of probabilities, that the level of service claimed is pedagogically appropriate for the number of children. If successful, a presumption arises to the effect that the level of service is also appropriate as to cost. The burden then shifts to the province or territory to rebut the presumption of cost appropriateness, which is a suitable burden given that it is better placed to adduce such evidence. Throughout, however, the burden for demonstrating pedagogical appropriateness properly rests with the claimant ⸺ who, after all, has the onus of establishing a Charter  breach.

[228]                     To assess the pedagogical appropriateness of a given level of service, the existence of majority schools or programs built for similar numbers elsewhere in the province can be a relevant indicator that a homogeneous school or program would be pedagogically appropriate for the number of rights holders’ children. Indeed, we agree that the existence of a school or program of a similar size is a useful indicator that this level of service is pedagogically appropriate for the number of students, provided that the school or program exists in a context that is relevant and somewhat comparable. We also agree that if the number of minority students is comparable to majority students locally, the minority will necessarily fall at the high end of the scale as they exist in the same context, and there will therefore be no need to look province‐wide (para. 91).

[229]                     However, we do not view the existence of a small school anywhere in the province as effectively determinative. Rather than presuming that a school or program elsewhere in the province demonstrates pedagogical appropriateness without qualification, the court needs first to assess whether the school or program is a relevant comparator, taking into account factors such as whether it is in a rural or urban area, whether it serves a remote or isolated community, whether the school continues to operate at the capacity for which it was built (including whether it is facing being phased out because of declining enrolment), and whether the school operates as a result of supplementary private funding.

[230]                     In considering factors such as whether a majority school exists in a rural setting, we are not proposing that such schools be “systematically excluded”, as the Chief Justice suggests (at para. 66); rather, our approach entails considering the context of schools used in the comparison to ensure that the circumstances are relevant and comparable to the proposed school or service. Further, as a general rule, because s. 23 calls for publicly funded minority language education, an appropriate comparator would also need to be publicly funded. At the same time, the assessment of relevant comparators must take into account that there need not be perfect correspondence of pedagogical appropriateness between the minority and the majority language education: “. . . it is important to consider the value of linguistic minority education as part of the determination of the services appropriate for the number of students [and to ensure that the] pedagogical requirements established to address the needs of the majority language students [are not] used to trump cultural and linguistic concerns appropriate for the minority language students”: Arsenault‐Cameron, at para. 38; Rose‐des‐vents, at para. 31. Other factors might therefore be relevant to assessing comparability depending on the circumstances of each case, including the particular situation of the minority community. For example, it would be inappropriate and unrealistic to rely on schools such as Crawford Bay Elementary/Secondary, a “state‐of‐the‐art facility” that serves an isolated community that can only be reached by ferry and one for which the Crawford Bay community fundraised and made a sizeable financial contribution to the project: trial reasons, at paras. 2890‐92 (CanLII).

[231]                     We remain unpersuaded of the Chief Justice’s view that the fact that 250 schools in British Columbia have fewer than 100 students, per se, is a sufficient basis to conclude the pedagogical and cost appropriateness of a standalone, homogeneous facility for a similar number of rights holders: para. 99. A careful review of the evidence shows that some of these “schools” are not what is commonly understood as a school: a number of them are virtual/online educational programs (e.g. Kootenay‐Columbia Virtual School; On‐Line Learning Centre; YouLearn.ca‐CE; YouLearn.ca‐DL; YouLearn.ca‐ALT); others are continuing education programs (e.g. Continuing Ed SD 28; SD 33 Continuing Ed; Continuing Ed SD 36; and Continuing Ed SD 38); and others are alternative education programs (e.g. Kootenay‐Columbia Learning Centre; Open Doors Alternate Education; and Crossroads Alternate School). It is equally unrealistic to rely on such “schools” as the basis of a presumption that rights holders are entitled to standalone facilities. For comparison to be helpful in the context of s. 23, one must look at realities that are reasonably comparable. This is why it is crucial to contextualize the comparison at the outset, rather than moving this work into rebutting a presumption.

[232]                     We note that, at this stage of the analysis, there is no principled reason to constrain comparator schools locally. At the second stage of the analysis, which looks to equivalence, the analysis is conditioned by the neighbouring schools to which parents could realistically send their children if they were deterred from enrolling them at the minority language school: Rose‐des‐vents, at para. 35. This logic, however, does not hold with respect to what is pedagogically or cost appropriate at the first main step. It follows that, while we disagree with the creation of a province‐wide presumption of pedagogical and cost appropriateness, we share the Chief Justice’s view that, when comparison becomes relevant to assess the appropriate level of services to which a community is entitled, the court can legitimately look at schools and programs elsewhere in the province. We add that to recognize comparison as a useful tool does not limit litigants from raising other ways of assessing pedagogical appropriateness.

[233]                     As a last point on pedagogical appropriateness, we agree with the Chief Justice that, for communities falling below the upper end of the sliding scale, the views of school boards are entitled to a measure of deference with respect to particular services among the range of potentially available services that are most pedagogically appropriate (para. 86). Indeed, according deference in this context is in line with and builds on this Court’s reasoning in Mahe that to “satisfy [s. 23’s] purpose . . . the minority language group [should] have control over those aspects of education which pertain to or have an effect upon their language and culture”: p. 375. We add that in according deference, courts must remain mindful that “[t]he welfare of the students, and thus indirectly the purposes of s. 23, demands that programmes and facilities which are inappropriate for the number of students involved should not be required”: Mahe, at p. 385 (emphasis added).

[234]                     We now turn to considerations of cost. In our view, a finding that the level of services is pedagogically appropriate for the number of children should lead to a presumption that the level of services is also appropriate in respect of cost. The province or territory can rebut this presumption by demonstrating that it is financially impractical to accord this level of services to the minority rights holders.

[235]                     This approach is consistent with, and draws from this Court’s direction in Mahe, reiterated in Rose‐des‐vents, that cost appropriateness is usually subsumed within the assessment of pedagogical appropriateness. As the Court has stated, “in most cases pedagogical requirements will prevent the imposition of unrealistic financial demands upon the state” (Mahe, at p. 385) and “costs will usually be subsumed within pedagogical needs in determining what level of service the numbers warrant” (Rose‐des‐vents, at para. 47). It is also consistent with the practical reality that provinces and territories are better placed to bring evidence regarding the financial feasibility of providing for a particular level of services out of public funds.

[236]                     We pause here to note the importance of circumscribing the entitlement within the s. 23 analysis, something that is consistent with both the text of the provision and this Court’s jurisprudence. It is particularly important to do so given what we will later explain is s. 1’s limited application in cases of s. 23 infringements. It follows that considerations of pedagogy and cost are seriously taken into account within the s. 23 analysis itself. As the Court stated in Mahe, “[c]ost, the second factor, is not usually explicitly taken into account in determining whether or not an individual is to be accorded a right under the Charter . In the case of s. 23, however, such a consideration is mandated”: p. 385 (emphasis added). When costs are subsumed within a serious consideration of pedagogical appropriateness, they are considered effectively, as mandated by the Constitution.

(2)          The “Substantive Equivalence” Analysis

[237]                     This appeal raises the question of whether substantive equivalence is the norm that ought to apply at the second stage of s. 23 analysis throughout the sliding scale, or whether a different norm should be adopted whenever a community falls at its middle or lower ends. The courts below suggested that “proportionality” should govern the analysis in those instances. We disagree.

[238]                     As discussed previously, s. 23, like many other Charter  provisions, is the result of a careful compromise that aims to recognize our country’s historic and continuing official language diversity. As this Court held in Arsenault‐Cameron, in order for s. 23 to “give effect to the equal partnership of the two official language groups in the context of education”, courts must give s. 23 a “purposive interpretation” that emphasizes its “true purpose of redressing past injustices and providing the official language minority with equal access to high quality education in its own language, in circumstances where community development will be enhanced”: paras. 26-27, quoting Reference re Public Schools Act (Man.), s. 79(3), (4) and (7), [1993] 1 S.C.R. 839, at p. 849 (“Manitoba Reference”), and R. v. Beaulac, [1991] 1 S.C.R. 768, at para. 25.

[239]                     As the Court has affirmed, this purposive interpretation entails applying two general principles. First, the answers to the questions under the s. 23 analysis “should ideally be guided by that which will most effectively encourage the flourishing and preservation of the French‐language minority in the province. Secondly, the right should be construed remedially, in recognition of previous injustices that have gone unredressed and which have required the entrenchment of protection for minority language rights”: Manitoba Reference, at p. 850; see also Rose‐des‐vents, at para. 32; Beaulac, at para. 25; Secession Reference, at para. 80; Arsenault‐Cameron, at para. 27.

[240]                     In light of those two general principles, and keeping in mind that the second stage of the analysis is concerned with the quality of educational services, we are of the view that a purposive interpretation of s. 23 may be achieved only by applying substantive equivalence as the appropriate norm at this stage of the analysis, irrespective of where a community falls on the sliding scale. Indeed, as this Court emphasized in Beaulac:

Equality does not have a lesser meaning in matters of language. . . . This Court has recognized that substantive equality is the correct norm to apply in Canadian law [and that]

. . .

[l]anguage rights must in all cases be interpreted purposively, in a manner consistent with the preservation and development of official language communities in Canada . . . [First emphasis added; second emphasis in original; paras. 22 and 25.]

[241]                     This rationale is further rooted in the broader principle of protecting minority rights, which is a fundamental underlying principle to the Constitution, carrying a powerful normative force that is binding upon both courts and governments: Secession Reference, at paras. 32, 54 and 79‐82. In Solski (Tutor of) v. Quebec (Attorney General), 2005 SCC 14, [2005] 1 S.C.R. 201, this Court emphasized “the serious difficulties resulting from the rate of assimilation of French‐speaking minority groups outside Quebec, whose current language rights were acquired only recently, at considerable expense and with great difficulty” and noted that courts have a responsibility to be “sensitive to the future of each language community” when interpreting minority language rights: para. 5. In light of the foregoing, substantive equivalence is necessarily the norm that must apply to address the needs of linguistic minorities given the past injustices they have suffered.

[242]                     Whereas our conclusion on this point aligns with that of the Chief Justice (at para. 107), we are of the view that his approach, taken as a whole, does not hold true to the s. 23 analysis as set out in the jurisprudence. As described above, under his approach at the first step (the “numbers warrant” analysis), a minority language group is presumptively entitled to a homogeneous school whenever a majority language school with a similar number of students exists anywhere in the province. The problems that follow from this become clear when both steps ⸺ the “numbers warrant” analysis and the “substantive equivalence” analysis ⸺ are examined together.

[243]                     As we have explained, the “substantive equivalence” analysis under s. 23 test seeks to evaluate the quality of services provided to minority rights holders: Rose‐des‐vents, at paras. 30, 33, 38 and 40; Mahe, at p. 378. Multiple factors may be considered including, but not limited to, the quality of instruction, teachers, physical facilities, educational outcomes, extracurricular activities, and travel time: Rose‐des‐vents, at paras. 38‐40. As the Court has explained, “[t]he relevant factors are considered together in assessing whether the overall educational experience is inferior in a way that could discourage rights holders from enrolling their children in a minority language school”, and “the extent to which any given factor will represent a live issue in assessing equivalence will be dictated by the circumstances of each case”: Rose‐des‐vents, at para. 39. These considerations are to be applied from the standpoint of the reasonable rights holder parent, comparing the minority language school with the local majority schools that represent realistic alternatives for them: Rose‐des‐vents, at paras. 35‐37. Here, the comparison is local because the only realistic alternative for a parent is to send their child to a neighbouring majority language school.

[244]                     And here the problem with the Chief Justice’s reasoning becomes apparent. His use of a province‐wide presumption under the first step of the analysis will result in the establishment of very small minority language schools that, at the second step of the analysis, will be compared to local majority language schools. The latter will typically be larger, and equipped with more or better services (including facilities) than the small majority schools located elsewhere in the province which were used as comparators under the first step of his analysis. Putting the two steps together thus reveals that our colleague’s approach is bound to result in a small population of students suddenly being entitled to essentially the same services as a much larger local majority school despite its relatively small size, while smaller majority schools receive relatively inferior services. This “express elevator” phenomenon, whereby the level of services owed to a local minority group is lifted to the top of the scale, is a substantial and, in our respectful view, unwise departure from this Court’s jurisprudence developed under s. 23. It departs from the notion of a scale of warranted services that gradually increases based on the number of children that may benefit from them. As noted, we agree with our colleague that substantive equivalence is the comparative norm that applies throughout the sliding scale. Where we differ is the effect of combining his proposed presumption at step 1 (with which we disagree) with his recourse to substantive equivalence as the proper comparative norm at step 2 (with which we agree).

[245]                     When assessing the substantive equivalence of the services provided, the analysis is circumscribed by comparing the quality of the level of services that is warranted for minority language students with the quality of that same level of services provided to neighbouring schools. In other words, when only instruction is warranted, the “substantive equivalence” analysis will assess whether the quality of instruction of the minority is comparable to the quality of instruction of the majority, taking into account the nature and quality of the facilities necessary to provide instruction. Similarly, when access to certain facilities is warranted as part of the level of service to which rights holders are entitled, the quality of those facilities and the quality of similar facilities granted to the majority locally will be included in assessing the substantive equivalence of the global educational experience provided to them. On the other hand, when a particular service is not warranted (e.g. non‐core facilities), the quality of that service will not be taken into account in the “substantive equivalence” analysis. This is particularly important to bear in mind where elementary and secondary education are provided within the same homogeneous facility. Indeed, one must not lose sight of where either group of rights holders stands on the sliding scale. If elementary‐age students fall at the upper end of the scale and are entitled to homogeneous facilities but secondary‐age students fall at the middle of the scale and are entitled only to instruction with access to core facilities, dispensing both elementary and secondary education within the same homogeneous facility does not give rise to higher entitlement for the secondary‐age students (e.g. both will be entitled to a gymnasium, but it might be that no football field is provided for the secondary‐age students). The quality of the services received by secondary‐age students in this scenario must reflect their position on the sliding scale. In conducting their analyses, courts must be careful not to adopt a formalistic approach and must consider the quality of the level of services globally: Rose‐des‐vents, at para. 35.

[246]                     In summary, the “substantive equivalence” analysis is delimited by the level of services to which the community is entitled, but this does not lead to an itemized evaluation of services and facilities; the analysis must remain global and contextual and must be mindful that instruction cannot be dissociated from the facilities in which it is provided: Rose‐des‐vents, at para. 39; Mahe, at pp. 369‐70.

[247]                     This approach enables the substantive equivalence norm to apply at any level of the scale. Indeed, even where the level of service to which the minority is entitled falls at the bottom or the middle of the sliding scale, it remains possible to assess the substantive equivalence of that level of service with neighbouring schools. For instance, where the number of rights holders does not call for a homogeneous school but rather for instruction with access to core facilities, the quality of that instruction and of those facilities will need to be of a substantively equivalent quality to that available to the majority.

[248]                     It is also important to keep in mind that reasonable parents will be conscious of the particularities of a smaller school or program. We agree with the Chief Justice’s statement that “a judge must consider whether reasonable parents who are aware of the inherent characteristics of small schools would be deterred from sending their children to a school of the official language minority because the educational experience there is meaningfully inferior to the experience at available majority language schools”: para. 116 (emphasis in original). We would, however, extend his emphasis on the size of the school to the size of the program, a necessary nuance to account for instances where the entitlement falls in the middle of the scale, that is, where the entitlement would not give rise to a homogeneous school. Indeed, it would be only reasonable for parents to consider all the specific characteristics of a minority language school in determining whether the experience is meaningfully inferior to an available majority language school. It bears emphasizing that substantive equivalence (being “substantive”) does not operate on formalistic grounds. As the Court stated in Rose‐des‐vents, “the education provided [to minority and majority linguistic communities] need not be identical”: para. 31.

[249]                     On a final note, we emphasize that we reject the lower courts’ reliance on “proportionality” as the norm to assess the quality of minority language education compared to that of majority language communities. In our view, the retention of any concern for proportionality at the second step of the s. 23 analysis means that the minority’s relative weight to that of the majority will be taken into account not once but twice, each time diminishing the quality and level of the minority’s constitutional entitlement. This is not to say that, by applying substantive equivalence, school facilities must be the same size as those of the majority or that minority communities should have the same number of programs. (Indeed, concerns about the level of services are addressed within the earlier “numbers warrant” analysis). Rather, substantive equivalence denotes that, once it has been established that rights holders are entitled to a certain level of services, the quality of such services should be substantively equivalent to what is granted to the majority in the same locality.

C.            Section 1

[250]                     Although this Court has yet to recognize an instance in which s. 1  of the Charter  might operate to save a measure that would otherwise infringe s. 23, this Court has recognized that it is possible: Quebec (Education, Recreation and Sports) v. Nguyen, 2009 SCC 47, [2009] 3 S.C.R. 208, at para. 37.

[251]                     The approach to s. 1 is well known. Section 1 provides that the rights and freedoms in the Charter  are subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society. To demonstrably justify a reasonable limit, the state must satisfy the Oakes test: the objective of the proposed measure must be pressing and substantial, and the means by which the measure is achieved must be proportionate: R. v. Oakes, [1986] 1 S.C.R. 103. The assessment of the objective is a threshold requirement, analyzed without considering the limit’s scope, the means employed or the effects of the measure: Frank v. Canada (Attorney General), 2019 SCC 1, [2019] 1 S.C.R. 3, at para. 38. The proportionality requirement will be satisfied where (i) the means are rationally connected to the ends, (ii) the measure minimally impairs the right at issue, and (iii) the salutary effects of the measure outweigh its deleterious effects: Oakes, at pp. 138‐40; Alberta v. Hutterian Brethren of Wilson Colony, 2009 SCC 37, [2009] 2 S.C.R. 567, at paras. 76‐77.

[252]                     Writing for this Court in Mahe, Dickson C.J. was clear that s. 23 serves a remedial purpose and “was designed to correct, on a national scale, the progressive erosion of minority official language groups”: p. 364. For this reason, he said it ought to receive a broad interpretation whereby courts “breathe life” into the expressed purpose of this section. Noting, however, the unusual nature of the right — that is, that it confers an entitlement that puts positive obligations on provincial or territorial governments to either “alter or develop major institutional structures” — he emphasized that caution in the interpretation of s. 23 is equally advisable: p. 365.

[253]                     One way of giving effect to a broad interpretation while maintaining caution is by not reading out the internal limit that qualifies the content of the right, as set out in s. 23 itself — the “numbers warrant” limitation. The fact that s. 23 is internally limited has an important effect on how the justification analysis should proceed. Indeed, this internal limit was recognized by the drafters as serving a similar function as s. 1. As explained by the Minister of Justice at the time, Minister Chrétien:

     Mr. Corbin

. . .

     That is why I find the words “where numbers warrant” offensive.

     Mr. Chrétien: Mr. Corbin, you have very clearly expressed the reason why we have maintained that phrase. Those were the terms used by the Provincial Premiers at the time of the 1978 Agreement in Montreal.

     You know just how delicate the issue is. Why did they use those words? To establish a criteria of reasonableness; you say yourself that in every instance French speakers have been more than reasonable.

     The same criteria exists at the beginning of the charter, in Clause 1. There were some who did not want any restrictions imposed. The restriction which does exist is a criteria of reasonableness. Some may argue, and I tend to agree with them, that the expression “where numbers warrant” is nothing more than an expression of reasonableness; the expression could be dropped, and the legal situation would not change. [Emphasis added.]

(Minutes of Proceedings and Evidence of the Special Joint Committee of the Senate and of the House of Commons on the Constitution of Canada, at pp. 36-37)

[254]                     As can be seen, s. 23 does not create, and was intended not to create, an absolute right. Nor is its text susceptible to an interpretation that is so broad that its content becomes infinite, leaving all of the work in defining its scope to be done in a s. 1 analysis. Rather, s. 23 contains an internal limit that must be given effect, and not swept into a s. 1 analysis.

[255]                     The Chief Justice approaches matters differently.  In his view, this internal limit affects the stringency with which the Oakes test is applied under s. 1, noting that it is one factor that supports “a particularly stringent standard for justifying an infringement of the right to minority language instruction”: para. 147. In our view, rather than affecting the stringency of the s. 1 analysis, the nature of s. 23 and the analysis which it calls for simply mean that in practice, s. 1 will have less to do. If done properly, the thorough analysis called for under s. 23 will account for many of the considerations that would normally arise under s. 1.

[256]                     Therefore, we agree with the Chief Justice that cost considerations will largely be taken into account in s. 23 itself: “the s. 1 analysis will then in some respects duplicate the numbers warrant analysis the court has already completed, because if there are enough students in a given case to exceed the numbers warrant threshold, the court will already have balanced the considerations related to cost and pedagogical needs in the first analysis”: para. 150.

[257]                     By applying cost considerations under the s. 23 analysis, we do not mean to suggest that such considerations will never be relevant under s. 1. Although such instances may be rare, to the extent that they are not taken into account in the s. 23 analysis, they may well be considered under s. 1 if they are linked to other considerations. This was established in Newfoundland (Treasury Board) v. N.A.P.E., 2004 SCC 66, [2004] 3 S.C.R. 381. Although the Chief Justice mentions N.A.P.E. (at para. 152), in our view it is worth discussing the case in more detail, given that it is a source of disagreement between the parties.

[258]                     In N.A.P.E., this Court considered the role of cost in the s. 1 analysis — specifically, whether financial considerations can constitute a “pressing and substantial” objective. Binnie J., for the Court, reviewed the case law on cost considerations under s. 1 in detail: paras. 63-71. In particular, he noted the holding in Reference re Remuneration of Judges of the Provincial Court of Prince Edward Island, [1997] 3 S.C.R. 3 (“PEI Provincial Court Judges Reference”), that “a measure whose sole purpose is financial, and which infringes Charter  rights, can never be justified under s. 1”: N.A.P.E., at para. 63, quoting PEI Provincial Court Judges Reference, at para. 284. In contrast, the case law established that “financial considerations wrapped up with other public policy considerations could qualify as sufficiently important objectives under s. 1”: N.A.P.E., at paras. 66‐69 (emphasis in original). In sum, the “‘sole purpose’ test” required considering whether a measure was solely financial or whether it was linked to other considerations: N.A.P.E., at para. 71. Binnie J. concluded:

The result of all this, it seems to me, is that courts will continue to look with strong scepticism at attempts to justify infringements of Charter  rights on the basis of budgetary constraints. To do otherwise would devalue the Charter  because there are always budgetary constraints and there are always other pressing government priorities. Nevertheless, the courts cannot close their eyes to the periodic occurrence of financial emergencies when measures must be taken to juggle priorities to see a government through the crisis. [Emphasis in original; para. 72.]

[259]                     Therefore, while we agree with the Chief Justice that courts must be skeptical of objectives that aim to reduce costs (at para. 152), we highlight that N.A.P.E. says more than that: it establishes that costs may be a pressing and substantial objective where they are linked to other public policy considerations.

[260]                     In any event, on the facts of this case, we agree with the Chief Justice’s conclusion that the “fair and rational allocation of limited public funds” is not a pressing and substantial objective as required by N.A.P.E.: para. 153. As he explains, decisions about allocating funds are the bread and butter of government functions and are purely financial. Further, adding “fair and rational” does not transform this objective into something pressing and substantial: ibid. Therefore, the Province has failed to identify a valid pressing and substantial objective in this case. In holding otherwise, the courts below erred. This objective is solely financial and thus fails to meet the threshold of pressing and substantial.

[261]                     Even though the Province’s failure to identify a pressing and substantial objective in justifying its breaches of s. 23 is sufficient to dispose of the analysis under s. 1, we would add the following comments about the lower courts’ approach to the final stage of the Oakes test, proportionality.

[262]                     In particular, we note that the trial judge relied on the evidence of experts who agreed that the assimilation rate of Francophones in British Columbia is high, approximately 70 percent. And while recognizing the remedial purpose of s. 23 and its role in combatting assimilation, the trial judge nevertheless factored this high rate of assimilation into her proportionality analysis. With respect to both the systemic and community claims, she saw the fact that assimilation was almost a foregone conclusion and that the provision of minority language instruction or facilities could only delay the inevitable as “not . . . a particularly strong deleterious effect”: para. 2148.

[263]                     In our respectful view, this evidence was not properly considered. By way of comparison, in Big M Drug Mart, the government sought to justify a provision of the Lord’s Day Act, R.S.C. 1970, c. L.‐13, alleged to violate s. 2( a )  of the Charter  on the ground that “the choice of the day of rest adhered to by the Christian majority is the most practical”: p. 352. The Court found this argument to be “fundamentally repugnant” (at p. 352) because the justification was fundamentally at odds with what s. 2( a )  seeks to protect: freedom of religion. In our view, be it at the stage of considering the objective or the means, the same logic applies: it is repugnant to invoke a justificatory criterion that is antithetical to the right itself. In this case, it was repugnant to the purpose of s. 23 to state that because assimilation was inevitable, heightened assimilation was not particularly deleterious.

[264]                     In fact, the heightened rate of assimilation in British Columbia ought to have led to the opposite inference. Evidence of a high rate of assimilation should be a spur to action. As this Court explained in Doucet‐Boudreau v. Nova Scotia (Minister of Education), 2003 SCC 62, [2003] 3 S.C.R. 3, as assimilation increases, the “numbers might cease to ‘warrant’”, thereby causing the cultural erosion s. 23 was designed to prevent and, if tolerated, allowing governments to avoid their obligations under s. 23: para. 29. In light of this, the high rate of assimilation facing Francophones in British Columbia is better described as gravely deleterious.

D.           Charter Damages

[265]                     Like our colleague, we conclude that Charter  damages are not precluded in this case. However, we arrive at that conclusion for different reasons. In our view, the immunity set out in Mackin generally applies to government policies. In this we differ from our colleague, as he would confine the immunity to legislation, albeit he also states that it is unnecessary to consider whether it applies to instruments such as decrees or regulations: para. 178.

[266]                     Framing the inquiry as requiring the Court to determine whether applying the immunity to other instruments will have a chilling effect on governance (at para. 170) misapprehends the rationale for Mackin immunity as a condition precedent for its application. These are two different things, which ought not to be confused. The rationale for requiring a heightened threshold of gravity in order to be entitled to damages is to not put a chill on good governance, which requires that “public officials carry out their duties under valid statutes without fear of liability in the event that the statute is later struck down”: Vancouver (City) v. Ward, 2010 SCC 27, [2010] 2 S.C.R. 28, at para. 41. But the application of Mackin immunity is not contingent on a demonstration that allowing for damages will have a chilling effect on government in each particular case.

[267]                     In our assessment, the same rationale for granting immunity when legislation is later found to be unconstitutional applies to policies or other measures later found unconstitutional. The question is not about the vehicle of state action but rather under what circumstances the state should be liable for damages.

[268]                     All this said, and while Mackin immunity generally applies to regulations and policies, an exception is warranted for s. 23 given the particular nature of that right. Nonetheless, before awarding Charter  damages for a s. 23 breach, a court must thoroughly undertake the analysis called for in Ward, lending particular attention to the existence and viability of alternative remedies.

(1)          Standing

[269]                     Charter  damages are available when they will “meaningfully vindicate the rights and freedoms of the claimants”: Ward, at para. 20 (emphasis added). The appellants in this case consist of both s. 23 rights holders (that is, parents) and the CSF. This raises the question of whether the CSF is a proper “claimant” for the purposes of damages. As this issue was not argued before us, we do not propose to address it in detail. We note, however, that it did arise before the lower courts, with the trial judge assuming without deciding that the appellants had standing (at para. 1131) and the Court of Appeal declining to decide the matter given that it was not raised at trial or on appeal: 2018 BCCA 305, 14 B.C.L.R. (6th) 52, at para. 306. As the respondents did not object to standing and there is no indication the rights holders object to the CSF acting on their behalf, we will likewise assume (without deciding) that the CSF has standing for Charter  damages.

(2)          The Framework for Charter Damages

[270]                     In Ward, this Court established the framework for determining the availability of Charter  damages. Specifically, McLachlin C.J. explained that the “first and most important remedy” for Charter  breaches is a declaration of invalidity under s. 52(1)  of the Constitution Act, 1982 ; this primary remedy is “supplemented” by s. 24  of the Charter : para. 1. Charter  damages will be available under s. 24(1)  where they are “appropriate and just”: para. 4. Trial judges have significant discretion in determining the appropriate remedy under s. 24(1) ; however, this discretion, while broad, is not unfettered, and what is “appropriate and just” depends on the circumstances: paras. 17‐19. An award of damages must be “fair not only to the claimant whose rights were breached, but to the state which is required to pay them”: para. 21. Other remedies may also be more responsive to a breach: ibid.

[271]                     McLachlin C.J. then set out a four‐step framework for assessing Charter  damages: para. 4. The first step, of course, is to show a Charter  breach. Secondly, the claimant must show why damages are a just and appropriate remedy that fulfills one or more of the related functions of compensation, vindication of the right, or deterrence of future breaches. Thirdly, the government may show “countervailing factors [that] defeat the functional considerations that support a damage award and render damages inappropriate or unjust”: para. 4. Finally, the court determines the quantum of damages.

[272]                     In the context of s. 23, the first step, establishing a breach, is undertaken through the analysis we set out above. The other three steps, to which we now turn, should be approached mindful of the particularities of s. 23.

(a)           Determining if Damages Are an Appropriate and Just Remedy

[273]                     At this stage, the court considers whether the purposes of compensation, vindication, and/or deterrence will be served by an award of damages: Ward, at para. 25. The purpose of compensation is “usually the most prominent” and “recognizes that breach of an individual’s Charter  rights may cause personal loss which should be remedied”: ibid. This purpose will likely be particularly important in the context of s. 23, given that the right requires the expenditure of public funds; if this does not occur, rights holders should be compensated in appropriate circumstances. The purposes of vindication and deterrence may also assume particular significance where governments had an incentive to underfund or delay in adequately funding minority language education.

(b)          Determining if There Are Any “Countervailing Factors” to Awarding Damages

[274]                     At this step, the government can raise considerations that render damages inappropriate or unjust: Ward, at para. 33. The category of countervailing factors is not closed but includes (1) the existence of alternative remedies; and (2) “good governance” concerns: ibid. The arguments in this case have largely centered on Mackin immunity, which the Court in Ward recognized was a good governance concern: para. 39. Nonetheless, both alternative remedies and Mackin immunity raise relevant considerations in the context of s. 23.

(i)            Existence of Alternative Remedies

[275]                     This factor recognizes that the purpose of an award of damages can often be addressed through other remedies, including declarations: Ward, at paras. 34 and 37. It bears repeating that this Court views a remedy under s. 52(1) as the “first and most important remedy” for Charter  breaches, which may be “supplemented” by remedies under s. 24: Ward, at para. 1. From this, it follows that courts must give due consideration to whether a declaration may be an adequate alternative remedy. Indeed, declarations are a common remedy in the context of s. 23 as “there is a tradition in Canada of state actors taking Charter  declarations seriously”: Rose‐des‐vents, para. 65. Recognizing the importance and strength of this remedy, the Court in both Mahe and Rose‐des‐vents held that a declaration was an appropriate remedy: Mahe, at pp. 392‐93; Rose‐des‐vents, at para. 82.

[276]                     This is not to suggest that a claimant must have exhausted all other avenues for recourse before seeking Charter  damages. The point is that the government must show that other remedies are available that would sufficiently address the breach: Ward, para. 35.

[277]                     In the context of s. 23, courts must pay particular attention to whether a declaration will be an adequate remedy for a breach. A declaration will often strike the right balance between vindicating Charter  rights and affording governments flexibility to meet their s. 23 obligations. As this Court emphasized as early as Mahe, there are many ways in which governments can fulfil their s. 23 obligations; they need room for trial and error:

It is not possible to give an exact description of what is required in every case in order to ensure that the minority language group has control over those aspects of minority language education which pertain to or have an effect upon minority language and culture. Imposing a specific form of educational system in the multitude of different circumstances which exist across Canada would be unrealistic and self‐defeating. . . . It is up to the public authorities to satisfy these general requirements. Where there are alternative ways of satisfying the requirements, the public authorities may choose the means of fulfilling their duties. 

. . .

As the Attorney General for Ontario submits, the government should have the widest possible discretion in selecting the institutional means by which its s. 23 obligations are to be met; the courts should be loath to interfere and impose what will be necessarily procrustean standards, unless that discretion is not exercised at all, or is exercised in such a way as to deny a constitutional right.  [Emphasis added; pp. 376 and 393.]

[278]                     The Court returned to this idea in the Manitoba Reference, noting that “[t]his Court should be loath, however, to detail what legislation the Government of Manitoba must enact in order to meet its constitutional obligations”: p. 860 (emphasis deleted). Therefore, a declaration will often be an appropriate remedy as it will indicate that a breach has occurred and identify the level of entitlement without dictating the particular modalities by which the breach is to be remedied.

(ii)         Good Governance Concerns

[279]                     Good governance concerns can also be countervailing factors to awarding damages in the context of s. 23. As Moldaver J. noted for this Court in Henry v. British Columbia (Attorney General), 2015 SCC 24, [2015] 2 S.C.R. 214, “good governance concerns” were not defined in Ward. Moldaver J. saw the phrase as “serv[ing] as a compendious term for the policy factors that will justify restricting the state’s exposure to civil liability”: para. 39.

[280]                     As noted, Mackin immunity has been recognized as responding to a “[g]ood governance concer[n]”: Ward, at paras. 38‐39. The arguments before us centred on the application of Mackin immunity to policies. We will address this matter shortly, but first, we turn to the last Ward factor as applied to s. 23 — that is, the quantum of damages.

(c)           Quantum

[281]                     When quantifying damages, the same purposes underlying Charter  damages (compensation, vindication, and deterrence) inform the quantum: Ward, at para. 47. An award of damages must be a meaningful response to the seriousness of the breach and deter future breaches: para. 54. A court should keep in mind that the breach is “an independent wrong, worthy of compensation in its own right”: para. 55.

[282]                     At the same time, an award of Charter  damages must be fair to both the claimant and the state: Ward, at para. 53. A court must take into account “the public interest in good governance, the danger of deterring governments from undertaking beneficial new policies and programs, and the need to avoid diverting large sums of funds from public programs to private interests”: ibid.

[283]                     Deference is owed to both the trial judge’s choice of remedy and assessment of quantum: Doucet‐Boudreau, at para. 87; Ward, at para. 73.

(3)          Mackin Immunity Applies to Policies

[284]                     The application of Mackin immunity to government policies is a live issue in the present case. According to the Mackin principle, “absent conduct that is clearly wrong, in bad faith or an abuse of power, the courts will not award damages for the harm suffered as a result of the mere enactment or application of a law that is subsequently declared to be unconstitutional”: Mackin, at para. 78. This confers a “limited immunity” intended to balance the protection of constitutional rights against the need for effective government. In other words, it shows whether the remedy is appropriate and just in the circumstances and ensures that “the effectiveness and efficiency of government action [is not] excessively constrained”: Mackin, at para. 79. As the Court reiterated in Ward, the principle recognizes that “the state must be afforded some immunity from liability in damages resulting from the conduct of certain functions that only the state can perform”: para. 40. The rationale underlying the immunity is that because “duly enacted laws should be enforced until declared invalid,” a “minimum threshold of gravity” is needed to interfere with good governance: Ward, at para. 39.

[285]                     Before this Court, the appellants have argued that Mackin immunity applies only to legislation — not to policies. The Chief Justice shares this view. We respectfully disagree. We say that the immunity generally applies to policies. Indeed, while our colleague sees this issue as never before having been considered (at para. 169), this Court’s jurisprudence has consistently framed this principle in broad terms that encompass policies. Moreover, the same rationale is engaged regardless of the vehicle for governmental action. We say respectfully that this jurisprudence should be viewed as dispositive of the broad issue of whether Mackin applies to policies. As we will discuss, however, an exception is warranted for s. 23.

[286]                     In Mackin itself, the Court formulated the immunity in undeniably broad terms:

     According to a general rule of public law, absent conduct that is clearly wrong, in bad faith or an abuse of power, the courts will not award damages for the harm suffered as a result of the mere enactment or application of a law that is subsequently declared to be unconstitutional. In other words “[i]nvalidity of governmental action, without more, clearly should not be a basis for liability for harm caused by the action”. In the legal sense, therefore, both public officials and legislative bodies enjoy limited immunity against actions in civil liability based on the fact that a legislative instrument is invalid. [Emphasis added; emphasis in original deleted; citations omitted; para. 78.]

[287]                     This formulation contemplates more than simple legislation. In particular, the references to “application of a law,” “invalidity of governmental action,” and “legislative instrument” suggest that the principle applies to governmental measures other than legislation, and done in furtherance of legislation — which typically entails the formulation, adoption, and implementation of policies. We also highlight the reference to “both public officials and legislative bodies”, which plainly encompasses acts of both the legislatures and other public officials. Indeed, this has been identified as a key holding from the case: W. H. Charles, Understanding Charter Damages: The Judicial Evolution of a Charter Remedy (2016), at p. 61.

[288]                     Any lingering doubt on this point was dispelled in Ward, which summarized the Mackin principle as follows:

The Mackin principle recognizes that the state must be afforded some immunity from liability in damages resulting from the conduct of certain functions that only the state can perform. Legislative and policy‐making functions are one such area of state activity. The immunity is justified because the law does not wish to chill the exercise of policy‐making discretion. [Emphasis added; para. 40.]

Notably, Ward expressly includes policy‐making, as distinct from enacting legislation, among those “functions” in respect of which “the state must be afforded some immunity from liability in damages”.

[289]                     McLachlin C.J. continued, affirming for the Court that the immunity may apply in situations beyond “[s]tate conduct pursuant to a valid statute”, referring to the broader language of “state agents . . . doing what is required for effective governance”:

     State conduct pursuant to a valid statute may not be the only situation in which the state might seek to show that s. 24(1) damages would deter state agents from doing what is required for effective governance . . . . It may be that in the future other situations may be recognized where the appropriateness of s. 24(1) damages could be negated on grounds of effective governance.

     Such concerns may find expression, as the law in this area matures, in various defences to s. 24(1) claims. Mackin established a defence of immunity for state action under valid statutes subsequently declared invalid, unless the state conduct is “clearly wrong, in bad faith or an abuse of power” (para. 78). If and when other concerns under the rubric of effective governance emerge, these may be expected to give rise to analogous public law defences. By analogy to Mackin and the private law, where the state establishes that s. 24(1) damages raise governance concerns, it would seem a minimum threshold, such as clear disregard for the claimant’s Charter  rights, may be appropriate. Different situations may call for different thresholds, as is the case at private law. [Emphasis added; paras. 42‐43.]

[290]                     McLachlin C.J.’s explanation that different thresholds may apply for different situations is crucial. The applicability of Mackin immunity is not properly determined by applying hard and fast rules. It may apply in many contexts, but that is not to say that it will necessarily apply with the same force.

[291]                     Henry (2015) is consistent with our view of the matter. There, the Court described the Mackin principle as applying to “state action taken pursuant to a law”; in other words, it applied to “actions taken in good faith under a law [state actors] believed to be valid”: para. 42 (emphasis added). These descriptions, too, plainly contemplate the possibility of immunity from Charter  damages for measures other than the enactment of legislation.

[292]                     Finally, the Court of Appeal for Ontario dealt directly with the question in Wynberg v. Ontario (2006), 82 O.R. (3d) 561 (C.A.). And it saw “no principled basis” to limit the application of Mackin to legislation, noting that Mackin itself used broad language such as “exercise of . . . powers” and “government action”: para. 194. The court reasoned that one of the primary functions of governments is advancing society through new policies and programs and that “[p]otential liability for damages creates the risk of interfering with effective governance by deterring governments from creating new policies and programs”: para. 196.

[293]                     Clearly, then, the jurisprudence supports Mackin immunity’s application to government policies. Indeed, we agree with the respondents that “[t]here is no principled basis to protect government from liability only when it enacts a law that is later found to be unconstitutional, but not when it implements a policy that serves the same normative function”: R.F., at para. 166. Ultimately, the debate over the reach of Mackin immunity confuses the vehicle of government action with the substance. Sometimes legislation is required to give effect to a policy decision; other times, legislation may already provide for delegation that allows for the development of regulations or policies that can achieve the same purpose. Mackin immunity is not concerned with the vehicle of government action but rather with effective governance. The question is when governments should face liability in carrying out their functions.

[294]                     We note that the Chief Justice has not accounted for this line of case law, except to comment on the statement in Ward referring to “legislative and policymaking functions” as follows:

Considered in its context, the concept of policy making relates to government policies that are based on laws. . . . Only one situation in which the limited immunity applies was recognized in Ward, that of government decisions made under laws that were duly enacted but were subsequently declared to be invalid. [para. 169.]

With respect, we do not find this persuasive. As our review of the case law hon Mackin immunity shows, the Court has consistently framed the immunity in sufficiently broad terms so as to encompass policies. Further, and whereas our colleague notes that applying the immunity to policies would lead to the undesirable effect that “[i]t would permit a government to avoid liability for damages simply by showing that its unlawful actions are authorized by its policies” (at para. 172), the same can be said of how the Chief Justice views the immunity: to avoid liability for Charter  damages, the state need only formalize its preferred policy in legislation. This, however, puts form before substance, by simply shifting the analysis to how the state will take a given action rather than preventing it from taking that action. Again, the focus should not be on the vehicle of state action but rather the purpose of the immunity. Whether the state acts through legislation, regulations, or policies, the rationale behind the immunity is that the state should be able to carry out its functions without the threat of damages, absent some threshold of misconduct. Policymaking is clearly a key state function.

[295]                     We also note the Chief Justice’s concern that it is difficult to define what a policy is, whereas it is much easier to identify legislation: para. 173. In our view, it is not necessary to exhaustively define a “policy”: as we have reiterated, the immunity is not concerned so much with the vehicle of government action but rather the purpose of the immunity. Although we conclude below that an exception from Mackin immunity is warranted for s. 23, we note that the policies at issue here are clearly of the kind that would normally attract the immunity (see the discussion of the transportation breach and Annual Facility Grant (“AFG”) Rural Factor, below at paras. 330-39). Both policies apply across the province while providing supplements to particular boards; they are the kind of instrument used by the Province to fulfil its obligations arising, inter alia, under s. 106.3 of the School Act, R.S.B.C. 1996, c. 412. They therefore engage good governance concerns as they are indisputably a means of funding education throughout the province and fulfilling statutory obligations.

[296]                     Finally, the Chief Justice considers that recognizing that the immunity applies to policies would contravene the following statement by McLachlin C.J. in Ward:

Another consideration that may negate the appropriateness of s. 24(1) damages is concern for effective governance. Good governance concerns may take different forms. At one extreme, it may be argued that any award of s. 24(1) damages will always have a chilling effect on government conduct, and hence will impact negatively on good governance. The logical conclusion of this argument is that s. 24(1) damages would never be appropriate. Clearly, this is not what the Constitution intends. Moreover, insofar as s. 24(1) damages deter Charter  breaches, they promote good governance. Compliance with Charter  standards is a foundational principle of good governance. [Emphasis added; para. 38.]

On our reading, McLachlin C.J. resolved that observation in the next paragraph of Ward, when she described Mackin immunity: see also Charles, at p. 88. This immunity is not, after all, absolute: it does not apply where there is a “minimum threshold of gravity,” that is, behaviour that is “clearly wrong, in bad faith or an abuse of power.” In other words, para. 38 of Ward (cited by our colleague at para. 174) rejects the extreme that an award of damages will always have a chilling effect on governance, while para. 39 explains that awarding damages in situations where a minimum threshold of gravity is present will not have a chilling effect. Recognizing that Mackin immunity applies to policies will not displace the carve‐out for situations involving a minimum threshold of gravity; therefore, it is consistent with the idea that damages will not always have a chilling effect.

(4)          An Exception Is Warranted for Section 23

[297]                     We have concluded that Mackin immunity is applicable to policies. However, the rationale underlying the immunity, including its application to policies, does not sit well in the context of s. 23. In our view, an exception is warranted such that the immunity does not apply in the context of s. 23.

[298]                     Thus far, the Mackin principle, and “good governance concerns” in general, has not been considered in the context of a positive right. Normally, governments make regulations and develop major policies to carry out their everyday responsibilities; in doing so, they presumably endeavour not to breach Charter  rights. The rationale for Mackin immunity as applied to these normal activities is that governments should be able to carry out their functions with the assurance that they will not be held liable for damages unless a threshold of misconduct has been met.

[299]                     In contrast, s. 23 is a right that requires action: unlike most Charter  rights, which direct that the state shall not act in a particular way, s. 23 by its terms requires that the state shall act. Legislation must be enacted, policies must be established, and public funds must be spent to give effect to the right. This is because s. 23 is a particular kind of Charter  right: it imposes “positive obligations on government to alter or develop institutional structures”: Mahe, at p. 365. Section 23 is also a remedial provision “designed to remedy an existing problem in Canada, and hence to alter the status quo”: p. 363. Courts must not be afraid to “breathe life” into the provision or “avoid implementing the possible novel remedies needed to achieve that purpose”: p. 365; see also Doucet‐Boudreau, at para. 87. Further, as s. 23 requires the expenditure of public funds, it is especially vulnerable to inaction:

One distinctive feature of s. 23 is that it is particularly vulnerable to government inaction or delay. Delay in implementing this entitlement or in addressing s. 23 violations can result in assimilation and can undermine access to the right itself. As this Court has noted before, for every school year that governments do not meet their obligations under s. 23, there is an increased likelihood of assimilation and cultural erosion (Doucet‐Boudreau, at para. 29). Left neglected, the right to minority language education could be lost altogether in a given community. Thus, there is a critical need both for vigilant implementation of s. 23 rights, and for timely compliance in remedying violations. [Emphasis added.]

(Rose‐des‐vents, at para. 28)

[300]                     By the time a court reaches the question of remedy, a breach of s. 23 will already have been found. This means that the government has failed to fund minority language education adequately or has unduly delayed in doing so. Such a conclusion may not be reconciled in the context of a right that requires the expenditure of funds and risks allowing the right to become meaningless or hollow if government action takes too long. As a result, an additional barrier of immunity for damages is ill‐suited to the substance of established s. 23 claims. Indeed, the rationale underlying Mackin immunity — the ability to carry out government functions without the threat of damages — does not support its application here, since a government that has breached s. 23 has not carried out its functions as mandated by that provision.

[301]                     We hasten to add that a thorough analysis of all the Ward factors is crucial before awarding damages. This remedy will not be presumed: as discussed, a court must ensure that an award of damages will satisfy the purposes of compensation, deterrence, and/or vindication. Considering the flexibility that governments require to carry out their s. 23 obligations, an award of damages will not necessarily fulfil one of these purposes in every case. The court should also pay special attention to whether an adequate alternative remedy would be sufficient to address the breach. It bears repeating that declarations under s. 52(1)  of the Constitution Act, 1982 , are the “first and most important remedy” for Charter  breaches (Ward, at para. 1) and that this Court’s s. 23 case law confirms that a declaration will usually be sufficient in this context (Mahe, at pp. 392‐93; Rose‐des‐vents, at para. 82). The court must also properly quantify damages, keeping in mind that an award of damages must be fair to both the claimant and the state: Ward, at para. 20. Damages will be appropriate only when all the Ward factors are satisfied.

E.            Application to This Case

[302]                     Having set out the proper approach to the analysis under s. 23, the proper analysis under s. 1 , and the Charter  damages framework, we now turn to their application to the facts of this appeal. In some instances, we come to a similar conclusion as the Chief Justice, but for different reasons. Meanwhile, some of our colleague’s conclusions illustrate the problems that we have identified with his approach.

(1)          Section 23 Analysis

(a)           The “Numbers Warrant” Analysis

[303]                     In the case before us, the trial judge erred in our respectful view in assessing the relevant number in several catchment areas. In these areas, she lowered the relevant number based upon her findings that “when the CSF starts a new school to divide a catchment area, the programme tends to grow gradually, adding grades and growing cohorts over time” and that “the number of children will warrant different facilities and amenities as the [project] grows”: para. 3795. Hence, she relied too heavily on current demand, giving paramount weight to the number of persons who could currently take advantage, rather than the number of persons who would eventually take advantage, of the contemplated program or facility.

[304]                     This error is apparent from a survey of her factual findings. For instance, in the Northeast Vancouver Catchment Area, the trial judge found the known demand to be 194 elementary‐age students (at para. 3779) and the potential demand to be 320 (at para. 3773), before determining that the number relevant to the “numbers warrant” analysis should fall approximately between 25 and 45 students (at paras. 3797 and 3806), even though she acknowledged that about 270 students would eventually take advantage of the proposed facility. This lower range reflects her focus on short‐term estimates and was accordingly well below the two parameters that ought to have delineated her analysis: the known demand and potential demand. In all, this error affected her analysis in seven catchment areas: the Proposed Abbotsford Elementary Catchment Area,^[3] the Proposed Fraser Valley Secondary Catchment Area,^[4] the Proposed Burnaby Catchment Area,^[5] the Northeast Vancouver Catchment Area,^[6] the East Victoria Catchment Area,^[7] the North Victoria Catchment Area,^[8] and the West Victoria Catchment Area.^[9] This error also caused her to misplace these communities on the sliding scale of entitlements, as we discuss below.

[305]                     While she used the wrong number for s. 23 purposes in these areas, the trial judge also estimated the number of students who would eventually take advantage of the program in each case. This is the number that should have been applied to determine these communities’ proper placement on the sliding scale.

[306]                     For the Proposed Burnaby Catchment Area, the Northeast Vancouver Elementary Catchment Area, the East Victoria Catchment Area, and the West Victoria Catchment Area, the trial judge acknowledged that, using the number of children who would eventually take advantage of the contemplated facilities and program, these communities would all fall at the high end of the sliding scale and would be entitled to distinct, homogeneous facilities that offer a global educational experience equivalent to what is offered at small majority schools in the community where the rights holders live: paras. 3805, 4068 and 5221. This is indeed the level of service to which these communities are entitled under s. 23  of the Charter .

[307]                     For the Proposed Abbotsford Elementary Catchment Area and the Proposed Fraser Valley Secondary Catchment Area, the CSF proposed to build a homogeneous elementary/secondary school in Abbotsford that would serve children from both these catchment areas. The trial judge acknowledged that, even if she had used the number of children who would eventually take advantage of the contemplated facilities and programs, she would still have concluded that their numbers were insufficient to warrant distinct homogeneous schools for each of these groups, placing them at the middle to high end of the sliding scale: paras. 5064, 5066 and 5067. She noted that the CSF was owed deference in its determination of what was pedagogically appropriate and that other smaller elementary/secondary schools existed across the province; with these considerations in mind, she concluded that the numbers of elementary‐age children in Abbotsford and of secondary‐age children in the Proposed Fraser Valley Secondary Catchment Area who would eventually take advantage of the proposed facilities could together warrant a homogeneous school with core facilities that are proportionate to those at majority schools in light of the proposed school’s size: paras. 5064 and 5066.

[308]                     In this particular case, we see no reason to dispute the trial judge’s finding that deference was warranted towards the CSF’s view that it is most pedagogically appropriate for these children to share a school with other rights holders rather than with majority language children. This falls squarely in line with the proposition in Mahe that the “minority language group [should] have control over those aspects of education which pertain to or have an effect upon their language and culture”: p. 375. It follows that rights holder parents of elementary‐age children in Abbotsford and secondary‐age children in the Proposed Fraser Valley Secondary Catchment Area are entitled to a homogeneous elementary/secondary school in Abbotsford that offers a global educational experience equivalent to what is offered at small majority schools in the community where they live. To be sure, these two groups each fall, on their own, in the middle to high end of the sliding scale and are therefore not entitled to two homogeneous schools; however, given the trial judge’s finding with regards to the deference owed to the CSF, they can, together, be entitled to a distinct, homogeneous elementary/secondary school.

[309]                     As for the North Victoria Catchment Area, the trial judge acknowledged that, using the number of children who would eventually take advantage of the contemplated facilities and program, this community would fall in the middle of the sliding scale and be entitled to instruction in a series of classrooms with access to facilities proportionate to those afforded to the majority: para. 4263. To make that determination, she relied mostly on the size of local majority schools, which were built for much greater numbers: para. 4070. As we have explained, however, there is no principled reason to constrain comparator schools locally when assessing whether a level of service is pedagogically appropriate for the number of children. Rather, the existence of other schools elsewhere in the province with similar numbers can be a relevant factor when determining where a group falls on the sliding scale.

[310]                     In our view, had the trial judge properly considered comparable majority schools elsewhere in the province when assessing the North Victoria claim, she would have found that this community falls on the higher end of the sliding scale, warranting a homogeneous facility that provides students with a global educational experience that is substantively equivalent to the situation at local majority schools. A number of factors lead us to this conclusion. Among those factors is the existence of a majority school that was built for and operates at a similar capacity in a similar urban setting, Connaught Heights Elementary School,^[10] which the trial judge considered as part of her analysis of the Burnaby claim: paras. 5168 and 5218-19. We also note that the trial judge considered homogeneous facilities to be pedagogically appropriate for similar numbers of students in communities such as Nelson and Sechelt: paras. 2697, 2901 and 2902-03. In our view, it is pedagogically appropriate for elementary‐age children in the Proposed North Victoria Catchment Area to receive a minority language education in distinct, homogeneous facilities that offer an experience that is substantively equivalent to local majority schools.

[311]                     A second error invoked by the appellants relates to the placement on the sliding scale of the communities in the Whistler Catchment Area (Elementary), the Pemberton Catchment Area, and the École Élémentaire La Vérendrye Catchment Area (“Chilliwack (Elementary)”). They argue that the trial judge did not properly consider the evidence to the effect that a number of small majority language schools exist elsewhere in the province: A.F., at para. 41. For these three communities, the trial judge found that the number of rights holders’ children — 85 in Whistler (Elementary), 60 in Chilliwack (Elementary), and 55 in Pemberton — did not warrant the construction of new homogeneous schools and held instead that they fell in the middle of the sliding scale, warranting instruction with access to core facilities: paras. 2208, 2344 and 4766. We agree. While the trial judge considered local majority schools as comparators at this stage of the analysis, she also considered smaller schools elsewhere in the province but found that they could not be used as proper comparators in Whistler (Elementary), Pemberton, and Chilliwack (Elementary). Indeed, she stated that “[t]he Province rarely builds schools to that capacity. Where it has, the school was built to serve an isolated and remote community; a new school was the only practical way of providing those children with an education”: paras. 2206 and 2342; see also para. 4764.

[312]                     As we explained, the existence of majority schools or programs with similar numbers elsewhere in the province can be a relevant factor to take into account, but it is not determinative in determining whether a contemplated program or facility is pedagogically appropriate for the number of students. The trial judge properly excluded majority schools with similar numbers as these schools existed in a different, incomparable contexts to the situation in these three catchment areas. She further found that, in the context of these communities, providing minority language education in homogeneous schools “would deprive [the children] of the pedagogical benefit of interacting with large populations, and would not be cost‐effective in light of the size of comparator schools”: para. 2207; see also paras. 2343 and 4765. We see no reason to disturb her finding in that regard.

(b)          The “Substantive Equivalence” Analysis

[313]                     We now turn to the second stage of the s. 23 analysis. As highlighted above, a purposive interpretation of s. 23 entails applying the substantive equivalence norm throughout the sliding scale. As such, the courts below erred in applying a “proportionality” standard. After a careful review, we are of the view that the trial judge’s use of proportionality had an impact on her conclusions regarding the following catchment areas: the Squamish Catchment Area (Elementary), the Sechelt Catchment Area, the Penticton Catchment Area, the Pemberton Catchment Area, the Whistler Catchment Area (Elementary), the current École de L’Anse‐au‐sable Catchment Area (“Kelowna (Secondary)”), the Nanaimo Catchment Area (Secondary), and Chilliwack (Elementary). However, for her analyses in the Nelson Catchment Area (Elementary) and the Nanaimo Catchment Area (Elementary), we are satisfied that the trial judge properly applied substantive equivalence.

[314]                     As noted, the trial judge erred in applying “proportionality” in the catchment areas for Squamish (Elementary), Sechelt, Penticton, and Pemberton. Remedying her mistake proves simple in these communities. After reviewing the trial judge’s analysis regarding these communities, we are satisfied that she rightly concluded that rights holders were entitled to homogeneous facilities in the cases of Squamish (Elementary), Sechelt and Penticton, and that rights holders were entitled to access to core facilities in the case of Pemberton. As such, her conclusions that s. 23 was breached in these communities holds true and we will simply modify her conclusions to specify that rights holders are entitled to a substantively equivalent experience, rather than a “proportionate” one.

[315]                     The matter is slightly more complex when it comes to the Whistler (Elementary), Kelowna (Secondary), Nanaimo (Secondary), and the Chilliwack (Elementary) catchment areas. In these cases, the trial judge concluded that there was no breach of s. 23. She made that determination based on the “proportionality” standard.

[316]                     With respect to Whistler (Elementary), the trial judge found that the experience at École Élémentaire La Passerelle was “proportionate” to that at local majority schools. As discussed above, she erred in applying proportionality as the appropriate norm. The trial judge identified two comparator schools: Spring Creek Elementary School and Myrtle Philip Elementary School. Her comparative analysis focused heavily on the former. Nevertheless, her reasoning shows that she engaged in comparison with both. In our view, the findings of fact made by the trial judge support a conclusion that the elementary students attending École Élémentaire La Passerelle are receiving a global educational experience that is substantively equivalent to that of the majority.^[11]

[317]                     As to Kelowna (Secondary), the trial judge concluded that a “reasonable rightsholder parent would find that the global educational experience afforded to secondary students is proportionate to the number of students that can be expected to enrol in a programme”: para. 4506. Again, the reliance upon proportionality as the standard to assess quality is erroneous. In any event, we are also of the view that in this case, the findings of fact made by the trial judge support the conclusion that the French‐language students attending the secondary program in Kelowna are receiving a global education experience that is substantively equivalent to that of the majority. Accordingly, we do not find a breach.^[12]

[318]                     Regarding Nanaimo (Secondary), the trial judge also erred by applying the “proportionality” standard. Although she did not conduct a comparative analysis regarding the global experience of the Nanaimo Francophone Secondary Programme students with that of local majority comparator schools, we have sufficient facts to conclude that Nanaimo secondary students are manifestly not receiving the level of services to which they are entitled. As noted by the trial judge, there is currently only one teacher for the secondary program despite the program having 50 students between five grades. We note that the trial judge found the relevant number to be 70, thus placing them at the middle of the scale. Nanaimo Francophone Secondary Programme students take only two courses in French and may take math and science with French immersion students only if those courses are not full; in other words, priority is given to majority students to access those courses. Therefore, not only are the s. 23 rights holders not receiving most of their instruction in the language of the minority, but they are also put in a situation where their Francophone experience may be impeded by the preference given to majority students. Given that the number of secondary students in Nanaimo falls at the middle of the sliding scale and that they are entitled to receive a minority language education with access to core facilities that provides a substantively equivalent experience, we conclude that the current global experience, regardless of the other factors normally assessed, is subpar.

[319]                     Considering the matter of Chilliwack (Elementary), the trial judge again erred in applying the “proportionality” standard. While she correctly compared the same level of services at majority schools, she did not rule on whether the quality of services offered in Chilliwack was substantively equivalent. Despite noting that the facilities were far from ideal, she concluded that “rightsholders in Chilliwack are receiving more than what the numbers warrant”: para. 4842. This is so because rights holders in Chilliwack currently have access to a homogeneous school when their numbers warrant only “instruction with access to the core facilities required to provide minority language education”: para. 4766. Given that the number of students in Chilliwack falls in the middle to low end of the sliding scale, we are of the view that its rights holders under s. 23  of the Charter  are entitled to have their elementary‐age children receive a minority language education with access to core facilities that provide an experience that is substantively equivalent to local majority schools. The existence of a homogeneous school in Chilliwack cannot dictate our analysis as to what the numbers warrant or be used to judicially obtain more than what s. 23 warrants. While rights holders in Chilliwack are receiving a higher level of services than what their numbers warrant, we must still determine whether the quality of the services to which they are entitled (that is, education with access to core facilities) is substantively equivalent to that same level of services offered at local majority schools. While the trial judge refrained from making conclusions on this point, we are of the view that the evidence on the record supports a finding that the global experience at École Élémentaire La Vérendrye is substantively inferior to that of local majority schools, notably due to its lack of access to core facilities for physical education.^[13]

[320]                     The situation for the Proposed Mission Catchment Area requires a more thorough examination. The only pleaded deficiency raised at École Élémentaire Des Deux‐rives was the inadequacy of the gym. The trial judge’s findings make clear that the gym was, indeed, inadequate: it is about half the size of a basketball court, measuring 166 m² compared to average of 373 m² in surrounding schools. Even considering the small class sizes, the gym “underperforms every majority school”; “makes it difficult to deliver core physical education services”, especially for older students; and raises concerns for safety and meeting learning outcomes: paras. 4957‐58. The trial judge was unsure whether a single deficiency was sufficient for a breach, noting that the question is whether the global educational experience, and not one aspect of the experience met the appropriate standard. She assumed, without deciding, that it was sufficient. She held that if there was a breach, it was caused by the Building Condition Driver (“BCD”), discussed below, but was justified by s. 1. The Court of Appeal upheld this conclusion and added, in the alternative, that there was no breach of s. 23 as the appellants had failed to provide enough evidence to allow the trial judge to find that the global experience was inferior.

[321]                     The difficulty with Mission arises, as the Chief Justice observes, because claimants must provide sufficient evidence to assess the global educational experience: para. 138. As this Court explained in Rose‐des‐vents:

. . . no school is likely to be considered by all parents to be equal or better than its neighbours in every respect. The comparative exercise must be alive to the varied factors that reasonable parents use to assess equivalence. The fact that a given school is deficient in one area does not mean that it lacks equivalence in an overall sense. In particular, both quality of instruction and facilities can represent important elements of comparison. . . .

Thus, the comparative exercise is contextual and holistic, accounting for not only physical facilities, but also quality of instruction, educational outcomes, extracurricular activities, and travel times, to name a few factors. [Emphasis added; paras. 38-39.]

However, as to the treatment of each individual factor, the Court observed:

Of course, the extent to which any given factor will represent a live issue in assessing equivalence will be dictated by the circumstances of each case. The relevant factors are considered together in assessing whether the overall educational experience is inferior in a way that could discourage rights holders from enrolling their children in a minority language school. [Emphasis added; para. 39.]

Thus, while a court must consider the global experience at a school, certain factors might assume particular significance, depending on the circumstances of the case. In other words, an aspect of an educational experience might be so deficient that a reasonable rights holder, even aware of the particularities of a small school, might consider the global educational experience to be meaningfully inferior. We hasten to add that this assessment is necessarily fact‐specific. Different considerations would apply, for example, when a certain facility is necessary to meet the curricular requirements.

[322]                      The difficulty here is that the appellants’ evidence centred solely on the gymnasium, as they acknowledge. We disagree with their position that it was up to the Province to present evidence demonstrating the superior aspects of the school; the appellants had the burden of showing that the global experience at École Élémentaire Des Deux‐rives was not substantively equivalent.

[323]                      At the same time, we share the Chief Justice’s concern that the litigation has taken over 10 years. Further, the findings of the trial judge demonstrate that the gymnasium is wholly inadequate, to the point that it raises safety concerns. Given these particular circumstances, we are content to consider the situation as though evidence were presented on the educational experience as a whole. This approach preserves the test from Rose‐des‐vents focusing on the global educational experience. In fairness to the Province, we will assume that all other aspects of the experience are substantively equivalent. The question is therefore whether a reasonable rights holder would consider that the global educational experience at École Élémentaire Des Deux‐rives is meaningfully inferior, assuming all aspects other than the gymnasium are substantively equivalent.

[324]                     In our view, a reasonable rights holder would consider the global experience to be meaningfully inferior. A parent would expect that a minority language school could provide, at a minimum, physical education courses that occur in a safe environment and satisfy the curriculum. This is not to say that the rights holder would expect the best physical education courses or a variety of sports teams. We also note that not all schools will necessarily have a gym in the building; students might access a gym in another school or a community centre, for example. The point is that the quality of the physical education facilities to which students have access must, at a minimum, allow schools to provide physical education courses safely and in a way that achieves required curricular learning objectives. The evidence is clear that the gym at École Élémentaire Des Deux‐rives is so inferior that teachers have difficulty meeting even these basic requirements. In our view, this supports a finding of a breach of s. 23.

[325]                     As for the BCD, however, we agree with the Chief Justice (at para. 138) that the record and arguments on this point are, unfortunately, insufficient to allow us to decide whether a breach arises. But whereas the Chief Justice would remit this matter to the trial judge, we view the insufficiency of the record and of the appellants’ arguments here as determinative.

[326]                     The BCD is a “Capital Driver” used by the Ministry of Education to prioritize capital projects. It measures the remaining economic life of buildings but does not take into account whether a space meets the educational needs of the school. The appellants argue that the BCD breaches s. 23 given its failure to consider building functionality. They point to Mission as an example of the problems with the BCD.

[327]                     Before this Court, the appellants’ arguments on this point were cursory: A.F., at para. 112. Indeed, the Court of Appeal noted that the appellants’ arguments on the BCD were unclear, even after they were asked to provide a table setting out the remedies sought: para. 243. It accordingly limited its analysis to Mission: ibid. We are not prepared to find a breach of s. 23 on such a vague and unelaborated basis.

[328]                     Finally, we turn to the Central Victoria Catchment Area, which must be considered in light of the CSF’s proposal to divide the current École Victor‐Brodeur Catchment Area into four new ones, which would be known as the East Victoria, West Victoria, Central Victoria, and North Victoria Catchment Areas. Because the errors we have described above led her to conclude that homogeneous schools were not warranted in the near future in East, West, and North Victoria, the trial judge focused on the current situation at École Victor-Brodeur, which she considered would persist until the numbers eventually warranted homogeneous schools. Given our findings that rights holders are entitled to homogeneous schools in these three catchment areas, however, it follows that the situation at École Victor‐Brodeur must be considered in the context of this new division. Therefore, the trial judge’s finding that the numbers at École Victor‐Brodeur will ultimately exceed its capacity is no longer true, as students who currently attend that school will attend the other schools in East, West, and North Victoria. Moreover, the problem of lengthy transportation times will be resolved for those students who currently bus from those areas. It follows that her conclusion of breach of s. 23 at École Victor‐Brodeur is no longer sustainable. Furthermore, the trial judge’s findings about the current situation at École Victor‐Brodeur demonstrate that there is currently no breach at that school: the overcrowding has been resolved, the experience is substantively equivalent, and the school is equipped to meet its present and reasonably foreseeable needs for space.

(2)          Section 1 Analysis

[329]                     We now turn to the infringements that the trial judge found were justified under s. 1. As we have explained, cost considerations can play a role in the s. 1 analysis, provided they are linked to other considerations. In the case at bar, however, the pressing and substantial objective that was identified — the fair and rational allocation of limited public funds — is not a valid objective under s. 1. The lack of a valid pressing and substantial objective is determinative of the s. 1 analyses. It follows that none of the identified breaches are justified.

[330]                     We also find an unjustified breach of s. 23 regarding the Annual Facilities Grant (“AFG”) Rural Factor. This credit, which is used to assist rural schools with facility upgrade funding, did not apply to the CSF initially. The Ministry contemplated applying it to the CSF in 2009, but it failed to do so until 2012‐13. As a result, CSF incurred a shortfall of about $1 million. The trial judge held that the delay was due to economic circumstances and the political consequences of reallocating funds away from the majority. She found a breach of s. 23, but held that it was justified. In doing so, she relied on an invalid pressing and substantial objective. Before this Court, only the trial judge’s s. 1 analysis is challenged; her finding of a breach is not. There is no reason to reconsider her s. 23 analysis. As for s. 1, her reliance on an invalid pressing and substantial objective is determinative. Therefore, the breach is not justified.

(3)          Remedy

[331]                     As we have explained, most of the breaches we have identified can be remedied by declarations. The remedy of Charter  damages has, however, been raised in respect of the AFG Rural Factor, which we have just discussed, and the transportation funding breach. 

[332]                     The transportation funding breach arose as a result of changes to the Ministry’s transportation funding system. The Ministry provided supplements to some school boards in recognition of their unique circumstances. The CSF received the Transportation and Housing Supplement. The Ministry froze that supplement between 2001 and 2012, with one increase in 2010‐11. Notably, the Ministry froze the CSF’s supplement at 1999 levels despite being aware as early as 1999 that the system was not responding to the CSF’s transportation needs. During the freeze, the CSF’s enrolment grew, and its actual transportation costs exceeded the supplement. In 2006, the Ministry implemented a 15 percent Francophone Supplement. Although the supplement was helpful, the CSF still incurred a deficit of between $6 million and $14 million.

[333]                     Drawing on Arsenault‐Cameron, the trial judge recognized that transportation funding is key in the context of s. 23, as long travel times can deter enrolment in a minority language school. She also noted that the right to determine appropriate travel times falls largely within the minority board’s jurisdiction. She held that the transportation funding freeze had caused a breach of s. 23 as it prevented the CSF from properly carrying out its mandate. This breach was not justified.

[334]                     Applying the Ward framework with due consideration of the s. 23 context, we conclude that an award of damages is appropriate in relation to both the AFG Rural Factor and the transportation breach.

[335]                     The first step of the Ward analysis requires establishing a breach of s. 23. We have concluded that the AFG Rural Factor breach was not justified. Further, the Province does not contest the transportation breach itself, focusing instead on Mackin immunity. We see no reason to reconsider the trial judge’s finding of an unjustified breach. Step 1 is therefore satisfied in both cases.

[336]                     The second step of Ward considers whether an award of damages would satisfy the purposes of compensation, vindication, and/or deterrence. This step is also satisfied. An award of damages would satisfy the purpose of compensation in both cases as the CSF incurred quantifiable financial losses. It would also deter governments from unduly delaying funding of minority language education: in this case, the delay in applying the AFG Rural Factor to the CSF (some 10 years) and in implementing the Francophone Supplement (some 7 years) was simply unreasonable. While some kinds of delay may be justifiable, the cited reasons here were purely financial and political. Just as purely financial considerations cannot justify a breach of s. 23 under s. 1, they cannot be used to avoid a damages award when a breach has been found. Similarly, governments cannot avoid damages based on concerns about political consequences, which appears to have been the case with the AFG Rural Factor. Governments must be deterred from using such reasons to justify a breach or a delay in funding. It bears repeating that the entrenchment of minority rights in a constitution is meant to ensure that minorities are not subjected to the whims of the majority: Secession Reference, at para. 74. Accepting that a government could delay acting due to a fear of negative political consequences would contradict the very purpose of s. 23.

[337]                     The third step of Ward looks to countervailing factors. As Mackin immunity does not apply in the context of s. 23, it is not necessary to consider its application here. However, a court must give due consideration to whether there is an adequate alternative remedy — especially whether a declaration would be sufficient. The Province argues that a declaration would be sufficient in the case of the transportation breach as it would provide legal and practical guidance. We are not persuaded that a declaration is sufficient in either case. The AFG Rural Factor now applies to the CSF, and the Francophone Supplement is addressing CSF’s current transportation needs. Further, the Transportation and Housing Supplement no longer exists. In these circumstances, a declaration would not provide practical guidance because the relevant policies that caused the s. 23 breaches are no longer in place and the Ministry has rectified both situations. Further, a declaration would not address the financial losses that the CSF incurred as a result of these two s. 23 breaches.

[338]                     The final step of Ward looks to the quantum of damages. Here, the trial judge’s assessment of quantum is entitled to deference: Ward, at paras. 17 and 73. The amount of damages must be fair to both the claimant and the state: Ward, at para. 20. The trial judge noted that, had she found a breach of s. 23 in relation to the AFG Rural Factor, she would have awarded $1.1 million in damages. This figure was based on her finding that the CSF had incurred a shortfall of around $1 million. As for the transportation breach, the trial judge found that the CSF’s deficit was between $6 million and $14 million; however, the deficit was partially caused by the CSF’s inefficiencies. She accordingly awarded $6 million in damages, holding that the Province should not be responsible for the entire deficit. The trial judge’s assessment of quantum in both instances was reasonable and in keeping with the direction in Ward that damages must be fair both to the claimant and the state. We see no reason to interfere with her assessment in either case.

[339]                      We therefore conclude, based on a thorough review of the Ward factors, that damages are appropriate for both the AFG Rural Factor and transportation breaches.

F.             Special Costs

[340]                     In the courts below, the parties bore their own costs. Before this Court, the appellants seek special costs or, if they are not successful, their costs throughout. The respondents argue that the appellants have not met the test for special costs and request that each party bear its own costs. The Chief Justice awards the appellants costs throughout but has not addressed special costs.

[341]                     We agree with the respondents that the appellants have not met the test for special costs. As this Court explained in Carter v. Canada (Attorney General), 2015 SCC 5, [2015] 1 S.C.R. 331, a two‐step test applies for special costs:

First, the case must involve matters of public interest that are truly exceptional. It is not enough that the issues raised have not previously been resolved or that they transcend the individual interests of the successful litigant: they must also have a significant and widespread societal impact. Second, in addition to showing that they have no personal, proprietary or pecuniary interest in the litigation that would justify the proceedings on economic grounds, the plaintiffs must show that it would not have been possible to effectively pursue the litigation in question with private funding. In those rare cases, it will be contrary to the interests of justice to ask the individual litigants (or, more likely, pro bono counsel) to bear the majority of the financial burden associated with pursuing the claim. [Emphasis added; para. 140.]

[342]                     The Court further noted that “only those costs that are shown to be reasonable and prudent will be covered by the award”: para. 142.

[343]                     At the first stage of the Carter test, we agree with the appellants that this case raises issues of public importance. They have failed to satisfy the second step, however, which requires that they “have no personal, proprietary or pecuniary interest in the litigation that would justify the proceedings on economic grounds”: Carter, at para. 140. We agree with the respondents that the appellants have a pecuniary interest in the litigation given that they seek Charter  damages: S.A. v. Metro Vancouver Housing Corp., 2019 SCC 4, [2019] 1 S.C.R. 99, at para. 70. Further, they have not shown that they would have been unable to fund this litigation without private funding; indeed, they do not even address this requirement in their submissions.

[344]                     We also agree with the respondents that this case is distinguishable from Rose‐des‐vents. First, the relief sought here is much broader given the various declarations and the requests for Charter  damages. In contrast, in Rose‐des‐vents, the parents sought “a simple declaration of a lack of equivalence under s. 23, without seeking a positive remedy at the outset”: para. 88. Further, the trial judge in Rose‐des‐vents had awarded special costs, whereas the trial judge here declined to do so. Given the highly discretionary nature of costs, an appeal court should be loath to interfere absent a clear error.

IV.         Conclusion

[345]                      We would allow the appeal in part, and set aside the judgment of the Court of Appeal, for the reasons set out above. We would make the following declarations:

a)            Rights holders under s. 23  of the Charter  living in the Proposed Burnaby Catchment Area are entitled to have their elementary‐age children receive a minority language education in homogeneous facilities with space for 175 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary schools;

b)            Rights holders under s. 23  of the Charter  living in the Northeast Vancouver Catchment Area are entitled to have their elementary‐age children receive a minority language education in homogeneous facilities with space for 270 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary schools;

c)            Rights holders under s. 23  of the Charter  living in the East Victoria Catchment Area are entitled to have their elementary‐age children receive a minority language education in homogeneous facilities with space for 275 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary schools;

d)            Rights holders under s. 23  of the Charter  living in the West Victoria Catchment Area are entitled to have their elementary‐age children receive a minority language education in homogeneous facilities with space for 299 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary schools;

e)               Rights holders under s. 23  of the Charter  living in the Proposed Abbotsford Elementary Catchment Area with regards to elementary‐age children and in the Proposed Fraser Valley Secondary Catchment Area with regards to secondary‐age children are entitled to receive a minority language education in a homogeneous elementary/secondary school in Abbotsford with space for 85 elementary‐age students and 120 secondary‐age students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary and secondary schools respectively, for that same level of service;

f)             Rights holders under s. 23  of the Charter  living in the North Victoria Catchment Area are entitled to have their elementary‐age children receive a minority language education in homogeneous facilities with space for 98 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary schools;

g)            Rights holders under s. 23  of the Charter  living in the Squamish Catchment Area are entitled to have their elementary‐age children receive a minority language education in homogeneous facilities with space for 135 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary schools;

h)            Rights holders under s. 23  of the Charter  living in the Sechelt Catchment Area are entitled to have their elementary‐age children receive a minority language education in homogeneous facilities with space for 90 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary schools;

i)             Rights holders under s. 23  of the Charter  living in the Penticton Catchment Area are entitled to have their elementary‐age and middle school‐age children receive a minority language education in homogeneous facilities with space for 175 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary and middle schools;

j)             Rights holders under s. 23  of the Charter  living in the Pemberton Catchment Area are entitled to have their elementary‐age children receive a minority language education with access to core facilities with space for 55 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary schools for that same level of service;

k)            The services currently provided to rights holders living in the Pemberton Catchment Area do not allow the CSF to offer a global educational experience that is substantively equivalent to that offered at local majority language elementary schools for that same level of service;

l)             Rights holders under s. 23  of the Charter  living in the Whistler Catchment Area are entitled to have their elementary‐age children receive a minority language education with access to core facilities with space for 85 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary schools for that same level of service;

m)         Rights holders under s. 23  of the Charter  living in the current École de L’Anse‐au‐sable Catchment Area (Kelowna) are entitled to have their secondary‐age children receive a minority language education with access to core facilities with space for 80 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language secondary schools for that same level of service;

n)            Rights holders under s. 23  of the Charter  living in the Nanaimo Catchment Area are entitled to have their secondary‐age children receive a minority language education with access to core facilities with space for 70 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language secondary schools for that same level of service;

o)            The services currently provided to rights holders living in the Nanaimo Catchment Area do not allow the CSF to offer a global educational experience that is substantively equivalent to that offered at local majority language secondary schools for that same level of service;

p)            Rights holders under s. 23  of the Charter  living in the École Élémentaire La Vérendrye Catchment Area (Chilliwack) are entitled to have their elementary‐age children receive a minority language education with access to core facilities with space for 60 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary schools for that same level of service;

q)            The services currently provided to rights holders living in the École Élémentaire La Vérendrye Catchment Area (Chilliwack) do not allow the CSF to offer a global educational experience that is substantively equivalent to that offered at local majority elementary schools for that same level of service;

r)             Rights holders under s. 23  of the Charter  who live in the Proposed Mission Catchment Area are entitled to have their elementary‐age children receive a minority language education in homogeneous facilities with space for 65 to 100 students (or such other numbers as the parties agree to) that provide them with a global educational experience of a quality that is substantively equivalent to the educational experience offered at local majority language elementary schools; and

s)            The physical education facilities currently provided to rights holders living in the Proposed Mission Catchment Area do not allow the CSF to offer a global educational experience that is substantively equivalent to that offered in local majority elementary schools.

[346]                      The trial judge’s damages award for the transportation funding breach is reinstated, and the Court orders the respondents to pay $6 million in damages to the appellants over a period of ten years, to compensate for the chronic underfunding of its transportation system between 2002‐03 and 2011‐12.

[347]                      We would also order the respondents to pay damages of $1.1 million for the breach of s. 23 arising from the Province’s application of the AFG Rural Factor.

[348]                      The appellants are entitled to their costs in this Court and in the courts below.

                    Appeal allowed in part with costs, Brown and Rowe JJ. dissenting in part.

                    Solicitors for the appellants: Gall Legge Grant Zwack, Vancouver; Juristes Power, Vancouver.

                    Solicitor for the respondents: Ministry of Attorney General Legal Services Branch, Victoria.

                    Solicitor for the intervener the Attorney General of Nova Scotia: Attorney General of Nova Scotia, Halifax.

                    Solicitor for the intervener the Attorney General of Prince Edward Island: Department of Justice and Public Safety, Charlottetown.

                    Solicitor for the intervener the Attorney General of Saskatchewan: Attorney General of Saskatchewan, Regina.

                    Solicitor for the intervener the Attorney General of Alberta: Attorney General of Alberta, Edmonton.

                    Solicitor for the intervener the Attorney General of Newfoundland and Labrador: Department of Justice and Public Safety, St. John’s.

                    Solicitor for the intervener the Attorney General of the Northwest Territories: Department of Justice, Yellowknife.

                    Solicitor for the intervener the Commissioner of Official Languages of Canada: Office of the Commissioner of Official Languages in Canada, Gatineau.

                    Solicitors for the intervener the Quebec Community Groups Network: Conway Baxter Wilson, Ottawa.

                    Solicitor for the intervener the David Asper Centre for Constitutional Rights: University of Toronto, Toronto.

                    Solicitor for the interveners Association des juristes d’expression française du Nouveau‐Brunswick inc. and Association des enseignantes et enseignants francophones du Nouveau‐Brunswick inc.: Université de Moncton, Moncton.

                    Solicitors for the intervener Fédération nationale des conseils scolaires francophones: Miller Thomson, Regina.

                         Solicitors for the interveners Association des parents de l’école Rose‐des‐Vents and Association des parents de l’école des Colibris: Nicolas M. Rouleau Professional Corporation, Toronto.

                         Solicitors for the intervener the Canadian Association for Progress in Justice: IMK, Montréal.

                    Solicitors for the interveners Société de l’Acadie du Nouveau‐Brunswick and Fédération des conseils d’éducation du Nouveau‐Brunswick: Pink Larkin, Fredericton.

                   Solicitor for the intervener the Assembly of Manitoba Chiefs: Public Interest Law Centre, Winnipeg.

                    Solicitors for the intervener Commission nationale des parents francophones: Larochelle Law, Whitehorse.

                    Solicitors for the intervener Conseil scolaire francophone provincial de Terre‐Neuve‐et‐Labrador: Lidstone & Company, Vancouver.

                    Solicitor for the intervener the Canadian Francophonie Research Chair in Language Rights: University of Ottawa, Ottawa.