a) The first restatement: Loyola High School v Quebec

The first Supreme Court case after Doré to deal with a rights-limiting administrative decision was Loyola High School v Quebec.Footnote ³⁵ Loyola is a Catholic high school that had applied to the Quebec Minister of Education, Recreation and Sports for an exemption from the requirement to offer a mandatory non-denominational program in ‘Ethics and Religious Culture’(ERC).Footnote ³⁶ Loyola had proposed an alternative to the ERC program, which the statutory framework authorised the Minister to approve if it was ‘equivalent’ to the ERC. The Minister did not find Loyola’s proposed alternative to be equivalent, however, concluding that it was faith-based rather than secular and amounted to a departure from the objectives of the ERC.Footnote ³⁷ Loyola sought judicial review on the basis that the decision unjustifiably infringed the right to freedom of religion.

Abella J wrote the majority judgment, with McLachlin CJ and Moldaver J writing a minority judgment concurring in part. Having begun the judgment with a reminder that the Court had “eschew[ed]” the Oakes formula in favour of a “robust proportionality analysis consistent with administrative law principles,” the majority’s analysis unsurprisingly focused on the balance the Minister struck between the ERC program’s objectives and the right to freedom of religion.Footnote ³⁸

The majority noted that the ERC, for its part, was intended to foster diversity and inculcate respect for all religions and peoples. The program explored the elements of religion without taking a denominational or dogmatic angle. It encouraged students to reflect on the values and norms central to different religious groups and to interact respectfully with members of other faiths.Footnote ³⁹

The Minister’s explanation for refusing to exempt Loyola’s alternative program was that it did not take a neutral perspective in offering instruction in the Catholic religion. In both the majority and the minority’s view, this was not a justifiable limit on Loyola’s religious freedom. For Abella J and the majority, requiring the school “to teach Catholicism from a neutral perspective” imposed a heavy burden on freedom of religion with little advancement of the ERC program’s statutory objectives.Footnote ⁴⁰ Moreover, the Minister’s reasoning did not give any weight to the implications that her decision held for religious freedom: “There is, in short, no balancing of freedom of religion in relation to the statutory objectives.”Footnote ⁴¹

The majority did uphold the part of the Minister’s decision that insisted that Loyola teach non-Catholic spiritual worldviews from a neutral perspective. To allow Loyola to teach other ethical frameworks through the lens of Catholic dogma would run the risk of belittling non-Catholic belief systems. This, the majority said, would undermine the very objectives of the ERC program. On the other hand, requiring Loyola’s teachers to offer instruction on non-Catholic religious worldviews from a neutral perspective imposes only minor limitations on the right to freedom of religion.Footnote ⁴² In this respect, the Minister’s decision did strike an appropriate balance between the right to freedom of religion and the program’s objectives.

McLachlin CJ and Moldaver J appear to take a markedly different analytical approach than the majority, even though the judgments are not that different in the result. Most glaring in the minority’s reasons for judgment is the almost complete failure to mention Doré. The minority refers to Doré once in its judgment, and only then in its summary of the matter’s procedural history.Footnote ⁴³ The minority does not look to Doré for guidance in assessing the justifiability of the Minister’s decision.

On the contrary, the minority hews quite closely to the approach set out in Oakes, assessing the justifiability of the limitation by asking whether the Minister’s decision limited the right as little as reasonably possible. Finding that the Minister understood the ‘equivalency’ requirement of the exemption regulation to require schools to propose alternative programs that essentially replicated the content of the ERC program, the minority concluded that the Minister not only rendered nugatory the purposes of the exemption but limited the right to religious freedom extensively. A more flexible interpretation of the regulation, allowing more deviation from the ERC program, would still align with the objectives of the program without interfering with rights as substantially. “In short, the Minister’s decision was not minimally impairing.”Footnote ⁴⁴

While the minority is quite clear in the extent to which it relies on the minimal impairment or less restrictive means inquiry, the majority’s approach—ostensibly an application of the Doré approach—is riddled with less restrictive means analysis. Consider the majority’s gloss on Doré in the following paragraphs:

The majority’s comments here are confusing, following so soon after the Court’s unanimous decision in Doré. They suggest that minimal impairment does have a role to play in assessing the justifiability of rights-limiting administrative decisions after all, and that the inquiry into minimal impairment and proportionality in the strict sense are related rather than separate.

I suggest in Part 5 that this is the correct understanding of the relationship between minimal impairment and proportionality in the strict sense, even if the majority judgment in Loyola and the unanimous judgment in Doré do not explain that relationship satisfactorily. I argue below that the inquiry into minimal impairment, by itself, provides an answer to the question of a limitation’s justifiability only under one condition. Spoiler alert: that condition is when the less restrictive alternatives are just as effective in achieving the pressing and substantial goals as the impugned, more restrictive option. In all other cases, where the less restrictive alternatives are also less effective than the more restrictive options, the analysis depends on one of the proportionality-centric modes of reasoning.

Even if I am right about the integrated and interdependent nature of minimal impairment and proportionality in the strict sense as a matter of theory, it does nothing to clarify what Doré and Loyola want administrative decision makers to actually do as a matter of doctrine. Are they to rely on the balancing test alone? Or proceed through the last two stages of Oakes? Or conduct a new kind of balancing inquiry that somehow blends both of those inquiries?

There is some guidance to be found, mercifully, in a close reading of the first of two Supreme Court decisions from 2018: Law Society of British Columbia v Trinity Western University.Footnote ⁴⁷ This case illustrates how interest balancing works.

b) Interest balancing in Trinity Western University v Law Society of British Columbia

As in Loyola, the litigation between Trinity Western University (TWU) and the Law Society of British Columbia (LSBC) concerned a conflict between the freedom of religion and an administrative decision maker’s exercise of discretion in pursuing statutory objectives. TWU is a private evangelical Christian post-secondary institution that has wanted to open a law school for many years. In order to enrol in the LSBC’s bar admission program, applicants must show proof of academic qualification. Graduates of law schools that are approved by the LSBC are automatically taken to have met this requirement.

TWU therefore applied to the LSBC for approval of its proposed law school. After some prevarications and reversals, the LSBC ultimately decided not to approve TWU’s law school. The main reason for this decision was that, at the time, TWU required all students to agree to a code of conduct (the “Community Covenant Agreement”) that, among other things, prohibits “sexual intimacy that violates the sacredness of marriage between a man and a woman.”Footnote ⁴⁸

In this regard, TWU had effectively adopted an admissions policy that discriminated against members of the LGBTQ community. Approving a law school that discriminates in this fashion, the LSBC reasoned, would undermine the statutory objectives it is mandated to pursue: i.e., upholding and protecting the public interest by preserving and protecting the rights and interests of all persons, and ensuring the integrity of lawyers.Footnote ⁴⁹

A majority of the Court upheld the LSBC’s decision as a reasonable exercise of administrative discretion, finding that it struck a proportionate balance between statutory objectives and Charter rights. McLachlin CJ concurred in separate reasons. Rowe J concurred in the result, finding there to be no limitation of religious freedom. Brown and Coté JJ dissented. I focus here on the approach taken in the majority judgment and on some of McLachlin CJ’s comments.

In contrast to Loyola, the majority does not appear to rely heavily on the inquiry into less restrictive means or minimal impairment. Rather, it goes directly to the balancing inquiry and considers whether the extent of the limitation on TWU’s religious freedom rights is justified by the benefits that refusing to approve TWU’s law school will deliver.

In performing this balancing analysis, the majority concluded that the LSBC’s decision did not limit TWU religious freedom to a significant extent. First, the LSBC did not refuse flat-out to approve TWU’s proposed law school. Rather, it rejected the proposal for a law school that included a discriminatory admissions policy. It remained open to TWU to make the Community Covenant optional for students rather than mandatory, which would soften a great deal of the discriminatory impact of TWU’s commitment to community values.Footnote ⁵⁰ More to the point, adherence to the Covenant is not absolutely required by the faith to which the TWU community subscribes.Footnote ⁵¹

Second, studying law in an environment where the community’s religious beliefs are pervasive, as they would be where all students are required to conduct themselves in accordance with the Covenant, was preferable to members of the community, but not necessary for their spiritual growth.Footnote ⁵²

On the other hand, the majority found that the refusal to approve TWU’s law school, at least as long as it insisted on a mandatory Covenant, was aligned with the LSBC’s statutory objectives in two important ways. First, it maintained “equal access to and diversity in the legal profession.”Footnote ⁵³ Because members of the LGBTQ community are unlikely to want to attend an institution hostile to their identity, the mandatory Covenant effectively restricted admission to the law school to persons who do not identify as LGBTQ.Footnote ⁵⁴

Second, if members of the LGBTQ community did decide to attend TWU and to agree to the terms of the Covenant, this would come at considerable personal cost: “Attending TWU’s law school would mean that LGBTQ students would have to deny a crucial component of their identity in the most private and personal of spaces for three years in order to receive a legal education.” ^{Footnote 55}

Refusing to accredit TWU’s law school furthered the LSBC’s statutory mandate to preserve rights and freedoms, the majority held, by preventing the risk of serious harm to those members of the LGBTQ community who might have chosen to attend TWU.

In the majority’s analysis, the harms of approving TWU’s law school were significant, and outweighed the relatively minor limitations on the TWU community’s religious freedom rights. Put differently, the benefits gained by not approving TWU’s law school (i.e., promoting equal access to law schools and reducing the risk of harm to members of a minority community) far outweighed the deleterious consequences of doing so (i.e., minor limitations on the right to religious freedom). This is a prime example of the mode of reasoning that has been called ‘interest balancing’.Footnote ⁵⁶

McLachlin CJ also engaged in interest-balancing reasoning in her judgment, although for her the analysis was slightly different. She found on the one hand that the limitation of the TWU community’s religious freedom rights was serious. On the other hand, for the LSBC to have approved TWU’s proposal would have amounted to an organ of state condoning discrimination against members of the LGBTQ community. There is great value, the Chief Justice went on, in organs of state sending the unambiguous message that the Canadian government will never condone private discrimination. The LSBC’s refusal to accredit TWU’s law school sent this message, McLachlin CJ concluded, and the benefits of sending this message were great enough in her view to justify even a serious limitation of TWU’s religious freedom rights.Footnote ⁵⁷

Both the majority and McLachlin CJ adopted interest balancing as the mode of reasoning in Trinity Western. By considering whether the benefits achieved by limiting a right are greater than those that would be achieved by upholding the right, decision makers and courts of review can come to a conclusion about which course of action is constitutionally appropriate. This is an example of proportionality analysis in the strict sense, or balancing as the final stage of the Oakes analysis envisages it. Indeed, in its early section 1 jurisprudence, the Supreme Court indicated that a proper application of the final stage of the Oakes analysis would involve a contextualised weighing up of salutary benefits and deleterious consequences, rather than a comparative assessment of the abstract value of Charter rights and broader social goals.Footnote ⁵⁸

The most relevant aspect of interest balancing as a mode of reasoning—and indeed the way the majority and McLachlin CJ applied it here—is that it does not concern itself at all with whether there are less restrictive alternatives. The conclusion stands on its own logic. In this regard, note that both the majority and McLachlin CJ acknowledged that the less restrictive means inquiry was of no assistance in this case. As the majority put it, “the LSBC was faced with only two options—to approve or reject TWU’s proposed law school.”Footnote ⁵⁹ One of these options was obviously less restrictive than the other, but approving TWU’s law school only because it would have been less restrictive to do so misses the point entirely. The real question is not which option is less restrictive, but whether the option that does limit rights but achieves statutory objectives is justified.

In other words, the answer to the minimal impairment stage of the Oakes test in these situations will always be that there are, in fact, no less restrictive means by which the decision maker can pursue the relevant statutory objectives. It is worth quoting McLachlin CJ at length on this point:

What the Trinity Western judgments leave us with is the germ of an idea that an inquiry into the availability of less restrictive means is just not useful, at least in some situations. And in situations where this question does not lead us any closer to an answer, decision makers and courts must rely on a different mode of reasoning. This is not to say that the minimal impairment test is universally useless. Rather, it is to suggest that in some situations it will not help a decision maker to determine whether or explain why a rights-limiting decision is justified or not.

I take up this idea in Part 5 below, where I sketch out the theory and the doctrine of a three-headed approach to administrative decision making where Charter rights are engaged, suggesting more precisely when minimal impairment analysis provides a determinative answer to the question of a limitation’s justifiability. Before doing that, I return to the second area of confusion that Doré begat: the relationship between Charter rights and Charter values. Bringing clarity to this relationship is relevant to developing the third mode of analysis I present here.

a) Values in interest balancing

The Court has already articulated at least one task for Charter values, which is to guide the assessment of how serious or extensive a rights limitation is. Having a sense of the seriousness or extent of a rights infringement is critical to interest balancing as a mode of reasoning about the justifiability of rights limitations, because we have to know how serious a rights limitation is if we are to weigh its adverse consequences against the benefits it produces. This is especially useful in those situations where less restrictive means analysis turns out to be of little assistance to a decision maker.

In R v Keegstra, a case involving restrictions on hate speech, the Court noted that the seriousness of a rights limitation depends on how closely the prohibited conduct is connected to the core of a right.Footnote ⁶⁵ Elaborating on this idea in RJR-MacDonald v Canada, the Court explained that each right serves a set of “even more fundamental values.”Footnote ⁶⁶ When an activity or form of conduct does not promote those values, it cannot be said to lie close to the core of the relevant right.Footnote ⁶⁷ Prohibiting or restricting such conduct or activities will thus result in rights limitations that are not significant.

Although Keegstra and RJR-MacDonald dealt with legislative limits to Charter rights, Charter values do the same work in the context of administrative rights limitations. The majority put the point quite clearly in Trinity Western, tracing a line of reasoning through its administrative law jurisprudence back to at least the 2009 decision in Hutterian Brethren:Footnote ⁶⁸

Charter values, on this approach, are neither goods to be protected in the same way as Charter rights nor goods to be weighed against Charter rights in the balancing analysis. Rather, Charter values inform and sustain the content and scope of rights. In the first place, by identifying the broader and more fundamental values that Charter rights protect and promote, courts and administrative decision makers can assess whether restrictions on conduct or activities amount to serious, moderate, or minor limitations on rights. In the second place, seeing values and rights as tied closely together in this way alleviates the concern Côté and Brown JJ expressed in Trinity Western that Charter rights face some kind of threat from Charter values.

b) Values-advancing reasoning

A second way in which Charter values can inform an administrative decision maker’s limitations analysis is for the decision maker to ask, not which of the available options impairs rights less, but which realises Charter values more fully. To illustrate the point, think of those cases where an administrative decision maker is required by their empowering statute to work towards the achievement of a particular statutory objective.

If one course of action open to the decision maker would promote the achievement of that objective but would at the same time impose limits on a Charter right, the decision maker will have to determine if limiting the right is justified. In Trinity Western, there were only two options before the decision maker, and there was nothing to be gained by asking which of them impaired rights less: one of them obviously did. But knowing that a course of action that does not advance statutory objectives is less impairing of rights tells us nothing about the constitutionality of the course of action that does advance those objectives.

In Trinity Western, then, the majority and McLachlin CJ relied on interest balancing instead of minimal impairment as the mode of reasoning. In both sets of reasons, the judges considered how the LSBC’s decision balanced the adverse consequences of limiting TWU’s right to religious freedom on one hand, and the benefits of a government institution acting to protect the interests of members of the LGBTQ community on the other.

The difference in the two sets of reasons, however, highlights one of the shortcomings of interest balancing as a mode of reasoning. For the majority, it was easy enough to explain the balance between the limitation and its benefits, because they saw the rights limitation to be minor but understood the benefits achieved by limiting the right to be significant. The benefits outweighed the harms. To use the visual metaphor so often relied on in this area of jurisprudence, weighing the limitation’s significant benefits against its minor impact on rights tipped the scales in the favour of the limitation.

In contrast, the mechanics of McLachlin CJ’s logic are less obvious. Although she also saw great value in the LSBC protecting the LGBTQ community against discrimination in legal education, she found the limitation of TWU’s religious freedom rights to be extensive and serious rather than minimal. If it is intuitively right, or at least intelligible, for significant benefits to justify minor limitations on rights, interest balancing (and the balancing metaphor that accompanies it) does not offer any analytical clarity when significant benefits and serious rights limitations are weighed against each other. The scales, surely, are even. Once the Chief Justice had concluded that the extent of the rights limitation in Trinity Western was serious and that the benefits produced were significant, it is not clear how she nevertheless found the benefits to outweigh the costs.

If simply weighing up adverse consequences and salutary benefits does not produce an obvious or intelligible answer, judges and administrative decision makers must do more to explain their decisions. In R v KRJ, a majority of the Court (per Karakatsanis J) emphasised the importance of clear and accessible reasoning at the final stage of the Oakes analysis:

Although this passage extols the virtue of explanation, it does not actually describe a mode of reasoning that can supply those explanations. Happily, interest balancing is not the only mode of reasoning on which decision makers and judges can rely when inquiries into minimal impairment are of no help. What I call values-advancing reasoning encourages officials to consider which of the options available to them advances or promotes Charter values better, without having to engage in the kind of intuitive, subjective, and ad hoc weighing of costs and benefits that interest balancing sometimes involves. Indeed, Doré itself foreshadowed this analytical approach by enjoining decision makers to “ask how the Charter value at issue will best be protected in view of the statutory objectives.”Footnote ⁷¹

An example of this mode of reasoning can be found in the second of the Supreme Court’s 2018 decisions dealing with a rights-limiting administrative decision: Groia v Law Society of Upper Canada. Footnote ⁷² The facts of this matter are similar to those in Doré, in that they involve a law society’s imposition of disciplinary sanctions on a lawyer and a complaint by that lawyer that the sanctions unjustifiably limited his Charter right to freedom of expression.

In this case, the Law Society of Upper Canada’s disciplinary bodies found that in defending a criminal accused at trial, Mr. Groia’s spirited and vigorous oral arguments had crossed the line into incivility. Specifically, Mr. Groia had frequently accused the Ontario Securities Commission’s counsel, the prosecutor in the case, of abuse of process.Footnote ⁷³ The Law Society’s Appeal Panel confirmed the initial Hearing Panel’s finding that Mr. Groia had committed professional misconduct, suspended him from practice for one month and made a costs award against him of $200,000.

The Law Society is mandated by statute to advance the public interest, the cause of justice, and the rule of law, by regulating the practice of law in Ontario.Footnote ⁷⁴ Key to achieving these objectives is the setting and enforcing of standards of professional conduct, including a requirement of civility during criminal trials. In a majority judgment, Moldaver J explained that incivility can undermine the fairness and integrity of judicial proceedings in at least four ways.Footnote ⁷⁵

In the first place, “overly aggressive, sarcastic or demeaning courtroom language may lead triers of fact, be they judge or jury, to view the lawyer—and therefore the client’s case—unfavourably.”Footnote ⁷⁶ Second, incivility can distract lawyers, judges, and juries from the factual and legal merits of the case being litigated. It does not serve the interests of justice if a trial judge is preoccupied with policing lawyers’ conduct instead of focusing on the evidence and legal questions. Third, incivility and fractiousness may increase stress for complainants or witnesses, for whom a trial is already a stressful experience, compromising the probative value of their testimony or undermining their willingness to participate at all. Finally, incivility in the form of vitriol, sarcasm, and baseless allegations of impropriety may undermine public confidence in the administration of justice.

At the same time, lawyers’ obligations to conduct themselves with civility cannot be allowed to compromise their duty of resolute advocacy. Indeed, the Supreme Court has stated that a lawyer’s commitment to their client’s cause is a principle of fundamental justice,Footnote ⁷⁷ and the Charter right to fair trial implies that criminal defence lawyers be free to advocate vigorously on behalf of their clients.Footnote ⁷⁸

In an adversarial criminal justice system, the defence lawyer’s commitment to their client’s cause and the duty of resolute advocacy take on particular importance. Vigorous and fearless advocacy ensures that a criminal accused can mount a full defence and challenge all aspects of the case against them. The implication of this demand for resolute advocacy is that the speech of lawyers in criminal trials must not lightly be limited. Defence counsel must not fear recrimination or discipline at the hands of the law society for doing what the criminal trial process and the Charter demand in the first place.

The balance the Law Society was required to strike here lay between Mr. Groia’s right—and indeed his duty—to speak freely in defence of his client, and the requirement of civility during criminal trials. Both of the goods in the balance, however—civility in trial proceedings and freedom of speech in legal representation—serve the same underlying Charter value: namely, that criminal trial procedures are as fair and truth-seeking as they can be. Criminal trials in which defence counsel are free to mount resolute and fearless cases on the behalf of their clients, and that are conducted with civility, will tend to allow for factual and legal issues to be properly ventilated and considered by judges and juries.

Where the question arises as to whether it is justifiable to limit a lawyer’s freedom of speech in order to promote the objective of civility, a decision maker will benefit from adopting a values-advancing mode of reasoning and asking which course of action will best promote the underlying value. Indeed, values-advancing reasoning explains the majority’s decision in this case even though there is some analytical work that an inquiry into the availability of less restrictive means can do as well. The Law Society’s Appeal Panel could have considered, for example, whether there was a less severe punishment that it could have imposed on Mr. Groia.

But considering whether there are less restrictive punishments is, of course, consequent on a finding that someone has committed a punishable offence in the first place. The primary question for the Appeal Panel, and the question to which values-advancing reasoning is most applicable, is whether sanctioning Mr. Groia’s conduct as uncivil or affirming it as an example of resolute advocacy would better promote the underlying constitutional value.

Following Doré, the majority framed the analysis as an exercise in balancing.Footnote ⁷⁹ In conducting the exercise, the majority paid close attention to the ‘core values’ of freedom of expression identified in Keegstra and RJR-MacDonald, particularly the search for truth. A finding of professional misconduct is more likely to strike a proportionate balance between rights and statutory objectives, the majority reasoned, where the impugned conduct “lies far from the core values of lawyers’ expressive freedom.”Footnote ⁸⁰ On the other hand, “[r]easonable criticism advances the interests of justice by holding other players accountable.”Footnote ⁸¹

In other words, when a lawyer’s conduct at trial is on the line of incivility but does not help in the search for truth, it does not advance the values that a lawyer’s right to expressive freedom at trial is meant to serve. In that case, a lawyer’s incivility may in fact undermine the search for truth for all the reasons that Moldaver J outlined. The balance between rights and statutory objectives here is to be found by considering whether allowing the lawyer’s conduct or sanctioning it will better promote the integrity of the criminal justice system by promoting the search for truth. This is values-advancing reasoning.

As it turned out, the Court decided in Mr. Groia’s favour and overturned the Appeal Panel’s finding as unreasonable.Footnote ⁸² Key to the Appeal Panel’s consideration of the case was the fact that Mr. Groia’s allegations of abuse of process had been based on a mistaken understanding of relevant points of law. However, as the Appeal Panel noted and the Supreme Court emphasised, Mr. Groia made these errors in good faith: his sincerely held views on these points of law simply turned out to be wrong.

The Supreme Court concluded that the Appeal Panel had not found a proportionate balance between the statutory objective of legal professionalism and right to free speech during trial. It found the Appeal Panel’s responses to Mr. Groia’s conduct at trial to be excessive: “Branding a lawyer as uncivil for nothing more than advancing good faith allegations of impropriety that stem from a sincerely held legal mistake is a highly excessive and unwarranted response.” ^{Footnote 83}

On its own, this is not a particularly helpful comment. It merely asserts that the Appeal Panel’s decision was excessive, without explaining why. We can dig a little deeper into the Court’s reasoning, though, to see that the question it is really trying to answer is whether the Appeal Panel had properly considered whether, and how, its decision would promote the underlying value of a fair and truth-seeking criminal justice process.

In this respect, the Court pointed out that the Appeal Panel appeared not to consider the broader consequences of branding Mr. Groia’s conduct uncivil and punishing him for it:

It is not difficult to see how the Court uses values-advancing reasoning here, even if it does not specifically identify that as the operative mode of reasoning. The Court’s logic is that the Appeal Panel’s decision undermines rather than advances the search for truth at trial by discouraging lawyers from speaking up when they believe, sincerely, that procedures are abusive or unfair. This is what leads the Court to conclude that the decision was unreasonable, not the mere fact that it was excessive.

Indeed, it is no longer sufficient for decision makers or courts of review to blandly assert that the limitation of a right is excessive or that it is not. The culture of justification that Vavilov affirms, and the obligation to explain difficult value judgments that KRJ describes, require decision makers and courts to provide an intelligible and transparent explanation for why a decision is excessive. Values-advancing reasoning allows officials to do precisely this.

a) The discomfort with Doré

Much of the concern with Doré’s approach to rights analysis revolves around the idea that an administrative-law based reasonableness review—even a ‘robust’ one—will underpower rights review. This criticism has come from both academic commentators and judges, and is succinctly summarised by Lauwers JA in a majority judgment in the Ontario Court of Appeal:

The general tenor of this concern is a mistrust of administrative decision makers’ competence to properly balance Charter rights and statutory obligations. And perhaps it is unrealistic to expect what Lauwers JA calls a ‘line decision maker’ to have an understanding of Charter rights, proportionality analysis, and human rights theory as sophisticated as his own.

But the fact that line decision makers are not human rights theorists does not preclude them from being able to reason through the justifiability of a rights limitation, nor from explaining their conclusions on those questions. As the UK Supreme Court said in Denbigh (speaking then as the Judicial Committee of the House of Lords), line decision makers will sometimes be called on to make decisions that affect human rights but they “cannot be expected to make such decisions with textbooks on human rights law at their elbows.”Footnote ⁸⁷

At the same time, the recipients of line decision makers’ decisions tend not to be human rights theorists either. The explanation for a decision must be tailored to the audience to which it is given. A decision written in dense legal language is no more intelligible to a layperson than a decision that is poorly explained. Decision makers must be sensitive to the context in which they render decisions, just as much as courts of review must be sensitive to the context in which decision makers operate. Sometimes, as Vavilov points out, a common-sense explanation of an administrative decision—including, presumably, a rights-limiting decision—is exactly what the context requires:

The Supreme Court has been quite clear, on the one hand, that administrative justice need not always look like judicial justice. Doré’s critics have proposed on the other hand that when Charter rights are involved, administrative justice must look like judicial justice. Decisions that engage Charter rights, they contend, must be correct. Moreover, the critics continue, Oakes and the inquiry into minimal impairment provide the framework for making correct decisions, while the Doré approach guarantees only reasonable decisions.Footnote ⁸⁹ What Doré’s critics demand most loudly, perhaps, is that administrative decisions affecting Charter rights be judicially reviewed on the standard of correctness rather than on the standard of reasonableness.

Doré’s critics, then, take heart in Vavilov’s formalism. The new standard of review test adopted in Vavilov categorically requires that all constitutional questions be reviewed on the standard of correctness. Since the question of whether a Charter right has been justifiably limited appears very much to be a constitutional question, it seems that there is no room left even for Doré’s robust reasonableness review of Charter rights-limiting administrative decisions.Footnote ⁹⁰

This argument misses the point, I think. Even if, after Vavilov, courts are supposed to review Charter-affecting administrative decisions on the standard of correctness, that has absolutely no impact on what Doré instructs administrative decision makers to do. Doré provides administrative decision makers with a range of modes of reasoning by which to inquire into whether or not a limit on a Charter right is justifiable, and explain the outcome to the subjects of their decision. Whether that decision is ‘correct’ or ‘reasonable’ is completely irrelevant. All that matters to the subject of the decision is that it be justified, and all that matters to the decision maker is that they are able to explain their decision to the people it affects.

Sometimes the best and most accessible explanation is given by proceeding through all the stages of Oakes, sometimes by interest balancing, and sometimes by values-advancing reasoning. What courts do on review is neither here nor there to decision makers’ efforts to explain their decisions as well as they can. And indeed, the more clearly and transparently decision makers are able to explain their decisions, the less likely the subjects of those decisions are to seek judicial review.

Moreover, the line between correctness and reasonableness when considering whether a rights limitation has been justified by the decision maker is, after Vavilov and the emphasis it places on a culture of justification, something of a red herring. Vavilov’s account of the difference between correctness review and reasonableness review makes nonsense of the idea that the latter is stricter or more exacting than the former. All that separates the standards of review after Vavilov is whose reasoning demonstrates that the outcome is justified: the administrative decision maker’s or the court’s. The majority in Vavilov (Abella J concurring on this point) said,

Correctness review, it follows, involves a court of review considering for itself which outcomes of a decision-making process can be justified. There is no higher standard of justification here: whether it is a court coming to its own conclusion (on the standard of correctness), or a court reviewing a decision maker’s explanation for their conclusion (on the standard of reasonableness), an acceptable decision in either case is one that can be justified in an intelligible and transparent way. And in that regard Doré and the jurisprudence that follows continue to be relevant because they set out the full range of modes of reasoning by which any official—an administrative decision maker or a court of review—can consider the justifiability of a Charter rights limitation.

Doré’s critics are right, of course, in suggesting that the Oakes formula offers one way for administrative decision makers to explain their decisions about rights limitations. But insisting that Oakes is the only way to do so, and thereby rejecting Doré’s approach, puts the critics in the odd position of having to explain what Doré is actually about. In a valiant and understandable attempt to make sense of Doré’s vexing conflation of rights and values, the critics propose that judicial review should be bifurcated: where only Charter values are engaged it may be sufficient for decision makers to follow Doré’s approach and forego the minimal impairment analysis. Where Charter rights are limited, however, decision makers must resort to the more scrupulous analysis involved in the full Oakes formula, including an inquiry into minimal impairment.Footnote ⁹²

Mark Mancini suggests something along these lines in calling for a “stricter reasonableness review” where administrative decisions engage Charter rights.Footnote ⁹³ Without explicitly calling for a full Oakes analysis, Mancini recommends that the Supreme Court “could impose explicit reasoning requirements” where Charter rights are in play, arguing that Charter rights could be understood as elements of the legal constraints that bind administrative decision makers under Vavilov’s conception of reasonableness.Footnote ⁹⁴ “Reasoning about constitutional rights in this way,” Mancini concludes, “would require more than what Doré currently prescribes.”Footnote ⁹⁵

It is certainly true, as I acknowledge above, that Doré leaves much to be desired. It does not offer much in the way of guidance to administrators as to how to reason through whether a limitation is justified in those cases where less restrictive means analysis is not helpful. To that extent I agree with Mancini that reasoning about rights requires more than what Doré offers, at least explicitly. Values-advancing reasoning and interest balancing, therefore, are modes of reasoning that supply Doré’s omission in this respect. In what follows below, I explain more precisely why Doré’s critics are wrong to think that Oakes is the only option, emphasising instead the value of a three-headed approach to justification.

b) The doctrinal mistake: the dangers of rights objectivism

The critics’ first mistake is doctrinal. Doré itself makes clear that the approach it introduces is only meant to be used in quite narrow circumstances—that is, in the context of adjudicated decisions. The unanimous Court was careful to emphasise this, as is clear from the opening few paragraphs of the judgment. Doré’s case, the Court noted, “raise[d] squarely the issue of how to protect Charter guarantees and the values they reflect in the context of adjudicated administrative decisions.” ^{Footnote 96}

Confronting the question of whether the presence of a Charter issue required the adoption of the Oakes test instead of the more traditional administrative law test for reasonableness, the Court went on:

The decisions in Doré, Groia, and Trinity Western highlight that the Court is alive to the distinctiveness of adjudicated administrative decision making. In each of those cases, the decision maker was faced with just two options from which to choose. Each decision maker had to decide whether to take the decision that would advance statutory objectives, even though doing so would limit rights, or whether to refuse to take that decision because the rights limitation was unjustifiable.

This situation is quite different to that in which a legislature finds itself. Once the government has identified an important goal towards which to work, the legislature is free to construct any number of policy frameworks, from the ground up, through which to pursue that objective. Where there is a multitude of options and the resulting legislative decision will affect a diverse range of people in numerous ways, it makes sense to ask which of those options impairs Charter rights the least.

Not all administrative decisions are adjudicative, though. On the contrary, some are legislative. The common law of procedural fairness has long recognised that administrative decisions range in nature from more adjudicative to more legislative, and treats them differently. Where an administrative decision is legislative—for example where a public utilities regulator considers whether to approve the rates a telephone operator has set,Footnote ⁹⁸ or where a municipality passes a zoning by-law that applies generally to a cityFootnote ⁹⁹ —the common-law duty of procedural fairness may not even arise.Footnote ¹⁰⁰

On the other hand, when an administrative decision looks like a court’s decision, the procedures by which it is made should approximate those of a court. Consider this classic statement from the Supreme Court’s important 1999 decision in Baker v Canada:

This all goes to show that the common law is already quite willing to treat adjudicative and legislative administrative decisions differently. The recognition of this difference underlies the unanimous Court’s approach in Doré. When an administrative decision maker is seized of a legislative decision and must choose between the same kind of policy options as a legislature when considering and debating a bill, they may well benefit from considering which of those options is least restrictive of Charter rights.

But where the decision is adjudicative, where the decision maker has only a limited number of options open to them and these affect only a small number of people in foreseeable ways, asking which of those options is less restrictive is just not very useful. The confinement of the Doré approach to adjudicated decisions is important because it recognises and leaves room for less restrictive means analysis in other, more legislative contexts, where that inquiry will actually help the decision maker to reason through the justifiability of the available options.

Doré’s critics, nonetheless, insist on a kind of rights objectivism that demands all rights limitations be objectively proven to be minimally impairing. Only if a decision maker can demonstrate that a decision is the least restrictive way to achieve statutory objectives, they argue, can it be concluded that the decision is justifiable. The critics admit of no other form of explanation or argument that a rights limitation might be justified. When objective proof of minimal impairment is unavailable, the critics add, no rights limitation can be said to be justifiable.

But think again about Trinity Western: the LSBC concluded that refusing to accredit TWU’s law school would achieve the statutory objectives of promoting the public interest, while approving it would not. There was only one way for the LSBC to further its statutory objective here, even though that was not the least restrictive option available to it. To ask about minimal impairment, when the minimally impairing option fails to further statutory objectives, is the wrong question to ask.

The correct question—and the one that the culture of justification requires us to ask—is whether the option that does further statutory objectives but also limits rights is justified. The risk that the critics run in insisting that decision makers follow only the objectively least restrictive course, is in closing off even the possibility of explaining why rights limitations in adjudicated decisions might be justified by the advances they make in achieving statutory objectives. Indeed, that undermines the entire purpose of section 1 to the extent that it prevents administrative decision makers from fulfilling statutory objectives by taking decisions they would otherwise be able to explain as justified limitations on rights.

It is important to realise, then, that Doré does not close the door on less restrictive means analysis or on the rest of the Oakes analysis. Quite the opposite, Doré indicates that Oakes still has its place in Charter-engaging administrative decision making. The Doré approach—which as I have explained here can take the form of either interest balancing or values-advancing reasoning—stands alongside the full Oakes analysis as one mode of reasoning about the justifiability of rights limitations. More to the point, the Doré approach is particularly useful in precisely those situations where less restrictive means analysis does not offer intelligible answers. This is an important doctrinal distinction that, once appreciated, shields Doré from a great deal of the abuse it has taken.

c) The theoretical mistake: how proportionality in the strict sense works behind the scenes

Rights objectivism’s second mistake is theoretical, and it picks up where the doctrinal mistake leaves off. Even in the context of a legislative administrative decision, the minimal impairment inquiry can answer the question of justifiability only under one condition: namely, when all of the options open to the decision maker are equally effective in achieving the relevant statutory objectives.

Put simply, the course of action that impairs rights least, but which achieves the statutory objective as effectively as any other alternative, is plainly preferable. It is not justifiable to limit rights more than necessary to achieve a statutory objective, so when one option is more restrictive but no more effective, it cannot be justifiable to select it.

A recent Ontario Court of Appeal decision about changes to election law is illustrative.Footnote ¹⁰² The issue in Working Families was whether the limitation imposed on freedom of expression by extending the period of restricted political spending before a provincial election from six months to twelve months was justified. The legislative history was important here, because the Ontario government had already amended the legislation once before. In those earlier amendments, however, the government had left the restricted period at six months and its own experts had concluded “that a 6-month period of pre-writ regulation was reasonable.”Footnote ¹⁰³

The evidence suggested, then, that a six-month restricted period was just as effective in achieving the goals of fairer and more equitable elections as a twelve-month period. But as Morgan J noted, the government had nonetheless selected the more restrictive of these equally effective options. That, by itself, was enough for him to conclude that the limitation was not justified under section 1:

If a less restrictive alternative is also less effective in achieving the statutory objective, however, determining that it is less restrictive is not the end of the story. It must now be asked, with respect to both the less restrictive but less effective alternative and the more restrictive but more effective option, whether the extent of the limitation is justified by the benefits that it produces. The balance between the adverse and salutary consequences of each option is different in each case, and a consideration of the proportionality of that balance is required in both cases. Lebel J pointed this out in Hutterian Brethren:

What the critics miss here is that balancing—either interest balancing or values-advancing reasoning—turns out to do a great deal of the analytical lifting even in those cases where an inquiry into less restrictive alternatives is analytically helpful. It is true that minimal impairment analysis provides a very good way to explain rights-limiting decisions in some limited circumstances, but it is not useful, by itself, in most other situations. In the situations where minimal impairment analysis cannot provide an objectively determinative answer to the question of whether a limitation is justified—as it did in Working Families—a decision maker must resort to persuasion, argument, and reasons: exactly what Vavilov’s culture of justification requires.

Of course, it is difficult to make an argument that a course of action which limits a right more than others, but realises a valuable goal more fully than others, strikes the best available balance between rights and broader social goals. For that matter, it is equally difficult to argue that a less restrictive but less effective course of action is more compliant with the Charter. But these are the difficult arguments that Vavilov’s culture of justification requires decision makers to make. As the Court said in KRJ, it is important that these difficult value judgments about the constitutionality of rights limitations be made transparent and accessible.Footnote ¹⁰⁶

Seen in the light of Vavilov, the Doré approach (interest balancing or values-advancing reasoning) and the Oakes formula (including minimal impairment analysis) are just different ways to make these value judgments explicit. The Court’s approach in Vavilov embraces the importance of persuasion and argument. What matters is not that administrative decisions comport with some idea of objective moral truth, but that the people affected by an administrative decision understand why an official made the decision they did.Footnote ¹⁰⁷