Legislation is enacted, amended and repealed over time in response to improvements and changes in social currents. The common law also evolves in the same way as judges pronounce, tweak and elaborate and then occasionally over-rule their previous legal doctrines. A recent search identified over 500 Supreme Court of Canada decisions that had been over-ruled by subsequent panels of the Court, of which 55 over-rulings came in the last 10 years.
Supreme Court of Canada is not Bound by its Own Decisions
As early as 1901, the Supreme Court asserted that it is not bound by its previous decisions. In Burrard Election Case (Duval v. Maxwell), http://canlii.ca/t/ggxht, Gwynne J. said:
Sometimes the Supreme Court needs a do-over.I feel difficulty in concurring in the proposition that it is not competent or proper for this court to reverse a judgment of the court differently constituted if it clearly appear to be erroneous.
This court is not invested with the prerogative of finality as is the House of Lords whose judgments are the law of the land until and unless varied by Parliament. Nor is this court invested with the prerogative of infallibility so as to prevent its seeing error in one of its own judgments.
The modern Court said overturning itself is a step not to be lightly undertaken. It must be convinced on compelling reasons that the precedent was wrongly decided and should be overruled, especially where the precedent was the product of strong Court majorities: Ontario v Fraser (2011, http://canlii.ca/t/fl63q). Occasionally, and counter-intuitively, the Court has over-ruled precedent to generate greater certainty: Minister of Indian Affairs and Northern Development v Ranville (1982, http://canlii.ca/t/1z1cx).
In practice, across all legal subjects, the Court has overruled itself many times. A few of these over-rulings are: R. v Chaulk (1990, http://canlii.ca/t/1fspm); R. v B. (1993, http://canlii.ca/t/1fs50); R. v Robinson (1996, http://canlii.ca/t/1frbh); R. v Salituro (1991, http://canlii.ca/t/1fshg); and Canada v Bedford (2013, http://canlii.ca/t/g2f56).
The Development of Five Legal Doctrines
This article describes how five different legal doctrines were developed over time by the Supreme Court of Canada.
- Collective Bargaining and Freedom of Association
In the 1999 decision of Delisle v Canada (http://canlii.ca/t/1fqn7), the Supreme Court of Canada found that federal legislation, which denied RCMP members employee status to collective bargaining, did not infringe their constitutional freedom of association under section 2(d) of the Charter. The majority 5 – 2 ruling held the Charter right existed independently of any legislation protecting the rights of the RCMP. The members enjoyed freedom of association apart from the legislation.
Stare decisis is the quasi-constitutional principle that the common law “stands decided” when declared by the highest court(s) in the country. Sixteen years later, the same question subjected to the same law, was answered differently by the same Court. In Mounted Police Association of Ontario v Canada (http://canlii.ca/t/gfxx8) a new 6 – 1 majority found the same legislative exclusion to be a violation of section 2(d) of the Charter, one so serious that it could not be justified under section 1.
Although RCMP members participated in a labour relations unit, its Staff Relations Representative Program, the Court agreed this was imposed on them, restricted their choice and interfered with their freedom of association. The Program was part of the RCMP management structure, so it failed to ensure RCMP members were independent from management. The majority of the Court concluded it was “not an association in any meaningful sense, nor a form of exercise of the right to freedom of association”.
As to the Delisle precedent, the Court repeated that overturning its own decisions was not a “step to be lightly taken.” However, it was now taking a broader approach to the right. It said Delisle was decided “before this Court’s shift to a purposive and generous approach to the exercise of freedom of association … the results in Delisle must be revisited.”
- Physician-assisted Suicide
Section 241(b) of the Criminal Code made it a crime for anyone to assist another individual to take their own life for any reason. In 1993 the Supreme Court of Canada ruled 5 – 4 in Rodriguez v British Columbia (http://canlii.ca/t/1frz0) that this criminal restraint on physician-assisted suicide violated Ms. Rodriguez’ Charter guarantees of equality, and of life, liberty and security of the person, but the breaches were justified under section 1 of the Charter. This was not a case of subjecting one to cruel and unusual punishment at the hands of the state. Prohibition of suicide was seen to be a reasonable limit on her Charter freedoms.
Now, the seven-member panel of the Supreme Court of Canada had second thoughts about its Moldowan interpretation. They noted it was widely criticized and did not have precise technical grounding in the legislation. The Court was unanimous in reversing itself 23 years later in Carter v Canada (http://canlii.ca/t/gg5z4). While a generation apart, the Rodriguez and Carter cases were factually similar. Both patients suffered the same degenerative disease and both sought similar liberties, although only the section 7 “life, liberty and security of the person” Charter right was argued in Carter.
The criminal rule against assisting another to die was found to violate the life, liberty and security of a person in a way that could not be justified. The Court concluded that prohibiting physician-assisted death led some individuals to take their own lives sooner due to their fear that they would unable to do so later. The Court acknowledged that the “prohibition deprives some individuals of life.” The prohibition was not minimally impairing and had to be struck out by the Court.
As to the issue of binding precedent, and how it related to the earlier Rodriguez decision, the Court said:
The doctrine that lower courts must follow the decisions of higher courts is fundamental to our legal system. It provides certainty while permitting the orderly development of the law in incremental steps. However, stare decisis is not a straitjacket that condemns the law to stasis. Trial courts may reconsider settled rulings of higher courts in two situations: (1) where a new legal issue is raised; and (2) where there is a change in the circumstances or evidence that “fundamentally shifts the parameters of the debate” . . .
“Both conditions were met in this case. The trial judge explained her decision to revisit Rodriguez by noting the changes in both the legal framework for s. 7 and the evidence on controlling the risk of abuse associated with assisted suicide.”
The argument before the trial judge involved a different legal conception of s. 7 than that prevailing when Rodriguez was decided. In particular, the law relating to the principles of overbreadth and gross disproportionality had materially advanced since Rodriguez. . . . This different question may lead to a different answer. The majority’s consideration of overbreadth under s. 1 suffers from the same defect: see Rodriguez, at p. 614. Finally, the majority in Rodriguez did not consider whether the prohibition was grossly disproportionate.
The matrix of legislative and social facts in this case also differed from the evidence before the Court in Rodriguez. The majority in Rodriguez relied on evidence of (1) the widespread acceptance of a moral or ethical distinction between passive and active euthanasia; (2) the lack of any “halfway measure” that could protect the vulnerable; and (3) the “substantial consensus” in Western countries that a blanket prohibition is necessary to protect against the slippery slope. The record before the trial judge in this case contained evidence that, if accepted, was capable of undermining each of these conclusions.
- Tax Deductions
In 1977, a unanimous five-judge Supreme Court of Canada released the important taxation decision of Moldowan v The Queen (http://canlii.ca/t/1mk9m). It involved interpretation of the federal Income Tax Act relating to deductibility of expenses and losses in hobby farm operations. The provision limits deductible losses “[w]here a taxpayer’s chief source of income for a taxation year is neither farming nor a combination of farming and some other source of income”. In Moldowan, the Court interpreted the provision to allow:
- all deductions where the farm was the taxpayer’s livelihood,
- limited deductions when farming was combined with another income source, and
- no deductions where the operation was a mere hobby farm.
This article describes how five different legal doctrines were developed over time by the Supreme Court of Canada.
Twenty five years later, the Court decided Canada v Craig (2012, http://canlii.ca/t/fs6sb) where the taxpayer was in a very similar situation to the Moldowan case. Craig had an established law practice from which he generated most of his income, as well as a farming business in which he was involved in buying, selling, training and racing of horses. He sought to deduct his losses from the horses.
Now, the seven-member panel of the Supreme Court of Canada had second thoughts about its Moldowan interpretation. It noted the decision was widely criticized and did not have precise technical grounding in the legislation. The unanimous Craig Court wrote:
“It may be that [the lower court] departed from Moldowan because of the extensive criticism of Moldowan. Indeed, Dickson J. himself acknowledged that the section was “an awkwardly worded and intractable section and the source of much debate.” Further, that provision had not come before the Supreme Court for review in the three decades since Moldowan was decided.”
Not surprisingly, the Court went on to conclude it should overrule Moldawan:
“… in making this decision the Supreme Court engages in a balancing exercise between the two important values of correctness and certainty. The Court must ask whether it is preferable to adhere to an incorrect precedent to maintain certainty, or to correct the error.”
Jettisoning Moldowan allowed the Court in Craig to make a fresh interpretation of the provision. The Court had previously read the legislation too narrowly. Farming could be a chief source of income, in combination with Craig’s law practice, so that there should be no limit on the deduction of losses under the Act. The horse‑racing operation was a business, not a personal endeavour or hobby. It was more than a sideline business because Craig invested serious capital and a very significant part of his daily work routine to the farming business.
- Right to Strike
Three Alberta statutes prohibited strikes and imposed mandatory arbitration to resolve collective bargaining disputes. In Reference Re Public Service Employee Relations Act (1987, http://canlii.ca/t/1ftnn), by a 3 – 2 majority, the Supreme Court of Canada decided that the right to strike was not a constitutional right inherent in the Charter.
Twenty eight years later, in Saskatchewan Federation of Labour v Saskatchewan (2015, http://canlii.ca/t/gg40r), the Court (5 – 2) reversed itself on the same legal issue in a different province. There was a constitutional right to strike embedded in the Charter’s section 2 freedom of association, and the Saskatchewan legislation was violating it in a way which could not be justified under section 1. Therefore, it was struck out by the Court.
The Court found a growing trend towards workplace justice in the cases so the right to strike is essential to collective bargaining. As for Supreme Court precedent:
“Given the fundamental shift in the scope of s. 2(d) since the Alberta Reference was decided, the trial judge was entitled to depart from precedent and consider the issue in accordance with this Court’s revitalized interpretation of s. 2(d).”
- Trial Delay
In 1992, an accused person, Morin, argued that her trial date, some 15 months after she had been charged for impaired driving, was set too late. She said this violated her section 11(b) Charter right to a trial within a reasonable time. The Supreme Court of Canada (6 – 1) disagreed, attributing the delay to scarce resources, institutional delay and the actions of the accused, who did not do enough to seek an earlier trial date: R. v Morin (1992, http://canlii.ca/t/1fsc6). The Court said the delay did not have a great consequence to the trial or its fairness.
Trial courts may reconsider settled rulings of higher courts in two situations: (1) where a new legal issue is raised; and (2) where there is a change in the circumstances or evidence that “fundamentally shifts the parameters of the debate” . . .
Twenty four years later, a full, unanimous Supreme Court of Canada reversed itself in R. v Jordan (2016, http://canlii.ca/t/gsds3) where four years had elapsed from the charge to the sentencing hearing. The Morin precedent was now viewed as “too unpredictable, too confusing and too complex” which rationalized delay and prevented the justice system from trying to locate inefficiencies and address them.
The Court set a presumptive ceiling beyond which delay — from the charge to the actual or anticipated end of trial — is presumed to be unreasonable. Trials must be completed within 18 months in provincial courts and within 30 months for trials in superior courts.
Stare decisis is the quasi-constitutional principle that the common law “stands decided” when declared by the highest court(s) in the country. The authority of the Supreme Court of Canada, indeed the orderly functioning of our legal system, depends on all lower courts following its decisions. They must follow Supreme Court decisions regardless of their views of the correctness of those decisions.
When it comes to the Charter, the Court grants leeway to lower courts to deviate from precedent where new legal arguments or issues are raised “as a consequence of significant development in the law, or if there is a change in the circumstances or evidence that fundamentally shifts the parameters of the debate” (2013, Bedford at para 42).
It is an equally serious matter that the Supreme Court be bound by its own previous statements of law. As this article shows, occasionally the Court reverses its own precedents. Many reasons could be posited for those reversals, including changing social norms, turnover of judges, slim majorities in the precedents, undesired impacts and a corresponding sense that the precedent was wrong, an activist yearning to effect change, and new arguments and issues that arise after the precedent decision was made.
Sometimes the Supreme Court needs a do-over. Even then – especially in Charter of Rights cases – a generation later it might want to take yet another look.