Day Nine: Maxime St-Hilaire

Associate Professor, Université de Sherbrooke and visiting scholar at SciencePo Paris Law School

Dupond v City of Montreal, [1978] 2 SCR 770

In this case, provisions of a Montreal bylaw (still in force: see Villeneuve c Ville de Montréal, 2018 QCCA  321) allowing the city to temporarily ban an assembly, parade, or other gathering due to public order concerns or anticipated “commotion” withstood a constitutional challenge based on the division of powers. By a 6-3 majority, the Supreme Court managed to rule that what Laskin CJ, dissenting, called a “mini criminal code” didn’t come under the  exclusive federal power over criminal law, but was part of a valid exercise of the  exclusive provincial jurisdiction over local or private matters (Constitution Act, 1867, para 92(16)), as well as of their exclusive powers over municipal institutions para 92(8)), property and civil rights (para. 92(13)), the administration of justice (para 92(14)), and power to impose penalties to enforce provincial legislation (para 92(15)).

I’m not going to say the ruling is absurd, but Beetz J’s majority reasons are weak and epitomize a form of Quebec nationalist bias that is willing to take Quebecers (as individuals) hostage. Whereas para 92(16) plainly reads as a partial residual legislative power, Beetz J’s reasons invert the very logic of all residual legislative power, which is supposed to apply only if the party relying on it has shown that the matter at issue can’t be assigned to any of the positively listed (non-residuary) powers: “Lorsqu’un texte législatif est en soi de nature locale ou privée, il incombe à la partie qui affirme qu’il relève d’une ou de plusieurs catégories de sujets énumérés à l’art. 91 de le prouver”, writes Beetz J, citing the Union St-Jacques de Montréal c. Bélisle, (1874) LR 6 PC 31. Leaving aside the criticism of that strange but common idea that the division of legislative powers would be entirely found in ss 91 and 92 of the Constitution Act, 1867, division of powers analysis must work precisely the other way around.

Another symptom of biased analysis here is the attempt to root the legal norm at issue in a multitude of different heads of legislative power. This mustn’t be conflated with double aspect, areas of shared jurisdiction, or concurrent attribution of legislative power. Normally, once a legal norm’s pith and substance has been identified, the judge must make up his or her mind and link that norm to a single head of legislative power, unless an inherently incidental legislative power such as the provincial penal power provided for at para 92(15) is relevant.

With regard to both the factual and legal contexts, the way in which Beetz J strives to distinguish the provisions at issue from criminal law on the basis that it is directed at “prevention” rather than punishment is entirely unconvincing. So is the manner in which he suggests that freedoms of expression, of the press, of assembly, of association, and of religion are “distinctes et indépendantes de la faculté de tenir des assemblées, des défilés, des attroupements, des manifestations, des processions dans le domaine public d’une ville”.

Beetz J’s reasons in Dupond just don’t stand comparison with Laskin’s CJ, but the decision tends to be celebrated by constitutional lawyers in Quebec as a bulwark against, or moment of resistance to, the centripetal betrayal of the federative compact. The reason I particularly dislike this decision is that I can’t help seeing a disconnect between the reasons given to justify it and the real ones, a hijacking of the applicable law by a form of political bias, and a sacrifice of the protection given the individual by the law on the altar of a more or less conscious constitutional (sub‑)nationalism.

Duplessis used to complain that the Supreme Court is like the Tower of Pisa, always leaning towards the same ― federal ― side. In Dupond, the apex court leaned towards Quebec, and it didn’t look good.

Canada (Human Rights Commission) v Taylor, [1990] 3 SCR 892

In this case, the question whether an infringement of the constitutional freedom of expression by the “quasi constitutional” (see my post on this concept) Canadian Human Rights Act, which protects equality in matters of federal jurisdiction, was justified, was basically treated the same as the question whether an infringement of that very freedom by the Criminal Code was too. Indeed, Taylor and Keegstra were heard one right after the other, handed down on the same day, by identical majorities, and majority and dissenting reasons written by the same judges. They were treated as companion cases for the simple reason that both concerned hate speech.

Yet there’s a difference between infringing freedom of expression through criminal law and doing it through human rights legislation. Criminal law isn’t chiefly about protecting rights. Human rights legislation is. Even when a given human rights statute protects only, or mainly, the right not to be discriminated against, it’s about human rights protection. Unless one believes that s. 33 of the Canadian Charter introduces an absurd hierarchy (according to which linguistic rights of s. 23 would prevail over the right to life of s. 7 or the protection against cruel and unusual treatment of s. 12), the Supreme Court has a responsibility to strive to make constitutional rights and freedoms form a coherent system. Ideally, supreme and constitutional courts should always seek to interpret a right so as to make sense of the other rights, and vice-versa. Hence, they should apply a much higher degree of scrutiny when the infringement of a constitutional right comes from a statute whose pith and substance is the protection of rights. Indeed, I think that s. 1, that is, proportionality, shouldn’t be open to human rights statutes’ infringing on a constitutional right in order to better protect another. Resorting to proportionality in cases where human rights legislation infringes on a constitutional right only contributes to the Supreme Court’s failing to successfully meet the challenge of “rights collision”, which requires a sufficient amount of interpretive delineation, integration, and coherence, notably in order to avoid hierarchization.

New Brunswick Broadcasting Co v Nova Scotia (Speaker of the House of Assembly), [1993] 1 SCR 319

It’s particularly hard for me to be brief on parliamentary privilege, a topic on which I cross-posted a few lines here on Double Aspect, published a little elsewhere, and appeared before the Senate Standing Committee on Rules, Procedure and the Rights of Parliament this past March. I also must disclose that I acted as an alleged expert on global standards in a parliamentary privilege case, Boulerice, until my affidavit was eventually dismissed by the Federal Court of Appeal.

I’ve been holding a grudge against the Supreme Court’s judgment in New Brunswick Broadcasting since I first read it as an LLB student, back in 1997. About a decade later, when I happened to read Kelsen’s 1929 book on democracy (apparently not available in English, though it is in French), I felt a bit relieved, not to say rewarded. Even back in the late 1920s, it was easy to understand that, in 17^th Century Britain, parliamentary privilege’s raison d’être was to protect parliamentarians from courts at a time prior to both judicial independence and responsible government. By the 20^th century, it was readily understandable that parliamentary privilege has to be updated, including in the face of the risk that it would be misused by the parliamentary majority in order to oppress the minority (the opposition), something which constitutional experts in Europe are now generally well aware of. There is no sign of such an understanding in New Brunswick Broadcasting ― nor in the somewhat backpedalling Vaid decision either. Karakatsanis J’s “purposive approach” in Chagnon gives me a little hope (Côté and Brown JJ’s dissenting reasons on this topic are disheartening), but there is still a long way to go.

However, there’s just one funny idea on which I want to focus here: that it somehow makes sense to rely on the fact that the Constitution Act, 1867’s preamble mentions the “desire to be federally united [under] a Constitution similar in Principle to that of the United Kingdom” in order to constitutionalize, that is, to put above legislation, something that in the UK has always been subjected to legislation; that, insofar as Parliament is concerned, s. 18 of the Constitution Act, 1867, while setting limits, clearly makes a matter of federal legislative jurisdiction; and that Fielding v Thomas, [1896] AC 600 (PC) confirmed was a matter of provincial jurisdiction by virtue of former para 92(1) of that same Constitution Act. Well, now that I think about it for the 101^st time, it finally seems to me that stating this funny idea is enough to debunk it.

Doré v Barreau du Québec, 2012 SCC 12, [2012] 1 SCR 395

There has to be some Hell for bad judicial decisions, and a special place in it for Supreme Court’s judgment in Doré v Barreau du Québec. A lot has been said, including in this series, about that decision, by which the Court has allowed itself both not to apply the constitutional bill of rights where it has to (what I have characterized as a denial of constitutional justice) and to apply it where it mustn’t.

In both cases the Court “sort of applies” the Charter, suggesting it makes no difference anyway (see notably McLachlin CJ and Moldaver J’s concurring reasons in Loyola), which isn’t true. For instance, when the “Doré framework” is applied, it’s no longer required, for an infringement to be justified, that it be “prescribed by law”. But there, precisely, lies what I find so shocking about Doré. Section 1 of the Charter, requires that infringements of rights be “prescribed by law” to be justified ― not for the Charter protections to apply at all. How then could the Court conflate the two, to the point of converting a case where the constitutional right at stake probably (because the Supreme Court’s jurisprudence never managed to sort out what “prescribed by law” means) could not be validly infringed (because the infringement wasn’t prescribed by law) into a case where that very constitutional right didn’t apply (even though the possible infringement being prescribed by law never was a condition for the Charter to apply)? A (first year) law student asked me once: “How can it be possible that Supreme Court judges, and moreover unanimously, make such a basic mistake?”, to what I responded: “I have no idea, but I can tell you that this is why I might retire early!” 

Doré clearly is a further symptom of the Supreme Court’s having totally lost the plot on the conditions under which the Charter applies, and trying now to obviate the question, on which I’ve claimed the idea that the SCC’s practice amounts to some “jurisprudence” (that is, to some “law”) is a myth. After Ktunaxa, where the majority just ignored it, we could hope, like Léonid Sirota did, to be witnessing Doré’s Demise. Like a living dead, it came back in the twin TWU cases (here and here), which would for sure have made my list, had 2018 been within the timeframe, and where members of the Court don’t even agree that Doré is still valid law. This is, and has been for a while, the disastrous state of Canadian constitutional law relating to rights and freedoms.

Tsilhqot’in Nation v British Columbia, 2014 SCC44, [2014] 2 SCR 257

There are things one just has to get right. One of them is the first final judgment declaring a constitutional aboriginal title. Yet in Tsilhquot’in, the Supreme Court managed to spoil this historic moment with deeply flawed reasons. When this judgement was handed down, I intended publishing a short enthusiastic post. I ended up publishing a 50-page-long critique.

McLachlin CJ’s unanimous reasons, even on aboriginal title, rely extensively on the 1984 Guerin decision, which, contrary to Calder (1973) and Delgamuukw (1997), was not a landmark decision on aboriginal title ― it dealt with fiduciary duty. Guerin had brought in aboriginal title only in order to give a private law dimension to the relationship between the federal government and the band after a reserve land surrender under the Indian Act, which allowed fiduciary duty to fill a legal vacuum. Guerin wrongly equated reserve land and aboriginal title, an error which was corrected in Osoyoos Indian Band. “It does not matter, in my opinion, that the present case is concerned with the interest of an Indian Band in a reserve rather than with unrecognized aboriginal title in traditional tribal lands,” had written Dickson J. (as he then was) in Guerin. “The appellant argued that, as a matter of law, aboriginal title subsists in a reserve created under the Indian Act. This is clearly incorrect,” wrote Iacobucci J. in Osoyoos Indian Band.

Through this unlikely reliance on Guerin, McLachlin CJ, for the Court, also allowed a hypertrophied version of the fiduciary duty to make a comeback. After Guerin, a collective delirium affected the country’s constitutional and aboriginal law lawyers, who started to spread fiduciary duty on their toasts. It took a while before a few Supreme Court decisions (including Wewaykum and Manitoba Metis Federation) eventually specified that, no, the State doesn’t always have a fiduciary duty towards aboriginal peoples when their rights are at stake, simply because, even then, the State doesn’t always act primarily for the special interest of aboriginal peoples, but sometimes in the public interest at large. By fits and starts, then, the “honour of the Crown” had replaced the Crown’s “fiduciary duty” as a general interpretative principle of s35 CA 1982. But Tsilhquot’in simply ignored all that, as if it had been released in 1989. Did McLachlin CJ have her reasons for the Court written by some geriatric law clerk who had taken his law degree from 1984 to 1987?

There are so many other problems with this judgement that I could not possibly discuss them all here. In particular, I prefer not to address the division of powers aspect of the decision. My doctor told me “Never again”!