Everything is Constitutional Until the Supreme Court Says Otherwise: R. v. Sullivan, 2020 ONCA 333

In R. v. Sullivan, 2020 ONCA 333, the Ontario Court of Appeal considered the effect of declarations of unconstitutionality. This is a difficult area, with stare decisis and res judicata complicating the analysis but I think the Court of Appeal’s analysis is vitiated by a misconception as to the nature of declaratory relief.

The basic problem is this: when a superior court judge in Ontario declares a statutory provision to be unconstitutional, can other superior court judges come to a different conclusion on constitutionality? For the Court of Appeal, Paciocco JA answered “Yes”; I answer “No”.

Section 33.1 of the Criminal Code eliminates the defence of automatism when induced by self-intoxication. Sullivan (and Chan, whose appeal was joined) argued unsuccessfully at trial that s. 33.1 is unconstitutional. The trial judges applied s. 33.1 even though it had previously been declared unconstitutional by a superior court judge in Ontario: R. v. Dunn (1999), 28 C.R. (5th) 295 (Ont. S.C.).

Although Paciocco JA (and Lauwers JA, in concurring reasons) found s. 33.1 to be unconstitutional, he rejected the argument that the Dunn decision bound the trial judges: “the ordinary principles of stare decisis apply…”, which permit judges in courts of coordinate jurisdiction to differ in limited circumstances (at para. 34). In his view,

where a party seeks to rely on a statutory provision that has been declared to be unconstitutional by a superior court judge, a subsequent trial judge should apply that earlier declaration of invalidity and treat the statutory provision as having no force or effect, unless the underlying constitutional issue has been raised by the Crown before them through submissions that the earlier decision is plainly wrong (at para. 38).

With respect, the analysis here is based on a misunderstanding as to the nature of declaratory relief. For Paciocco JA, “declarations made by trial courts are subject to appeal, and if overturned on appeal, will have no effect”, so the only time a declaration can bind superior court and appellate judges is when it has been issued by the Supreme Court of Canada (at para. 35).

This is incorrect. Declaratory relief, once issued by a superior court, is an authoritative statement of the law. As long as relief is not withheld on discretionary grounds or suspended to allow a swift legislative response, it has immediate retrospective and prospective effect. Declarations:

…are available to condemn, in a way that binds all, specific public acts, decisions or legislative provisions as being contrary to law. They are also available to delineate, in a concrete way that binds all, the legal rights, the legal state of affairs or the legal status of parties before the Court when the delineation will have some practical use (Entertainment Software Assoc. v. Society Composers, 2020 FCA 100, at para. 105)

Declarations are issued inter partes but have erga omnes force: Marcotte v. Longueuil (City), 2009 SCC 43, [2009] 3 SCR 65, at para. 28. A court’s order in respect of the validity of a general norm does “not alter the position in law. It will declare what that position is” (R (Ahmed) v. Her Majesty’s Treasury (No. 2), [2010] 2 AC 534, at para. 4). The effect of a declaration of unconstitutionality, just as with a declaration of invalidity of a regulation, is that it no longer exists.

The hallmark of a superior court is the ability to render such declarations: R. v. Lloyd, 2016 SCC 13, [2016] 1 SCR 130, at para. 19. As counsel properly put it to Paciocco JA, a provision declared unconstitutional is so “for all future cases”; “cannot be enforced”; and is “null and void, and is effectively removed from the statute books” (at para. 35). (This view was rejected by L’Heureux-Dubé J in 2747-3174 Québec Inc. v. Quebec (Régie des permis d’alcool), [1996] 3 SCR 919 but in dissent and only in respect of non-constitutional declarations.)

As a result, stare decisis is a red herring. Translated from Latin, the term means to stand by what has been decided. But once a declaration of unconstitutionality has been issued, there is nothing left to stand beside. For similar reasons, res judicata has no application here either (except, perhaps, by analogy): the matter has been decided but not because the requirements of res judicata are met, rather because a declaration with erga omnes force has been issued by a superior court.

Paciocco JA identified the following problem in relation to stare decisis which would arise if the appellants’ arguments about the effect of declarations were accepted:

For example, three superior court judges in succession could find a provision to be constitutional, but the fourth judge’s ruling to the contrary would be the only one to have full force or effect in the province. Unless that fourth decision is appealed, it becomes the law in the province. The Crown can no longer rely on the provision; therefore, decreasing the prospect that the issue of constitutional validity would make it before the provincial appellate court. The development of the law would be driven by coincidence in the sequence of trial level decisions and the fortuity of discretionary decisions about whether to appeal, when it should be determined by the quality of the judicial ruling (at para. 37. On the circumstances in which a superior court judge can depart from a previous decision of another superior court judge, see Holmes v. Jarrett (1993), 68 OR (3d) 667).

But Paciocco JA’s solution to this problem is not effective. On his view, the fourth judge could still come to a contrary conclusion on constitutionality, if convinced that the previous decisions were “plainly wrong” (at para. 38). Equally, on Paciocco JA’s view, three judges could find a provision to be unconstitutional but a fourth judge might find that they were “plainly wrong” to do so and uphold the constitutionality of the provision. Indeed, a subsequent panel of an appellate court could depart from a previous declaration of an earlier panel of the same appellate court.

On my view, by contrast, the fourth judge in Paciocco JA’s hypothetical (or a subsequent appellate panel) could take a different view on the constitutionality of the provision, but only (consistent with stare decisis) where it was of the view that the previous decision or decisions were plainly wrong. If there was a previous decision that the provision was unconstitutional, that would be the end of the matter.

I do not see this as a particularly difficult question. It is true that the practical consequences might be unfortunate. A superior court judge might badly err in granting a declaration of unconstitutionality, with future courts unable to resurrect the invalid provision. But the proper course in such cases is for the province or federal government (which, pursuant to the local rules of court, will be on notice about the constitutional challenge) to appeal the decision. Unless they do, the provision is a dead letter.

The only difficult question I perceive here is that of the territorial scope of a superior court’s declaration of unconstitutionality. On one view, a superior court’s jurisdiction is provincial only, not national. This would certainly ensure that badly erroneous declarations of unconstitutionality only affect one province (and that the condemned provision could be resurrected if the matter was litigated in another province).

On another view (to which I tend to gravitate), the superior courts form part of a unified body of courts with the Supreme Court of Canada at their apex. A declaration of unconstitutionality issued by one is thereby authoritative as to the others. Again, this has nothing to do with stare decisis (and it is well accepted that courts of one province cannot bind the courts of another province) and everything to do with the juridical effect of a declaration of constitutionality.

The result — that a superior court judge in one province can speak authoritatively for the judiciary as a whole — might be inconvenient in some cases but, if so, this is the result of Canada’s federal structure. And, in any event, I prefer this result to the one implicit in Paciocco JA’s analysis, which is that everything is constitutional until the Supreme Court of Canada says otherwise.

With thanks to Maxime St-Hilaire (and various interlocutors on Twitter) for discussion.

This content has been updated on November 19, 2020 at 21:31.