What considerations can justify a court overturning a recent constitutional precedent?  This constitutional perennial is once again in the foreground as a Canadian court is asked to revisit a 1993 judgment (R. v. Rodriguez, [1993] 3 SCR 519)) upholding the constitutionality of the criminal prohibition of assisted suicide.  The sole justice from that panel remaining on the Supreme Court of Canada is Chief Justice Beverley McLachlin, who had dissented.

Understandably, the plaintiffs in Carter et al v. Attorney General of Canada (BC Supreme Court, Vancouver Registry No. S112688) to be argued in November 2011, have skirted the question.  The only allegation pleaded that could be relevant to whether a court should overturn Rodriguez is that “(a) significant number of countries now authorize physician-assisted suicide”, although the constitutional significance of foreign practice remains deeply controversial in Canada.

Some further background, then, is needed.  It is often said that constitutional litigation is the only alternative in the face of legislative indifference or inaction.  But in this case the refusal to legislate is, on any account, a considered refusal and not mere inattention.  Since Rodriguez, bills proposing the decriminalization of physician-assisted suicide were introduced in the House of Commons in 1994, 2005, and most recently in 2009, along with a motion calling for a special parliamentary committee to review the criminal prohibition of euthanasia and assisted suicide in 1997.  All of these bills and motions were defeated.  Additionally, a Special Committee of the Senate held hearings in 1995 and in its report recommended that euthanasia and assisted suicide remain criminal offences.  So the circumstances are different from those said to have preceded the Canadian courts’ initial foray into same-sex marriage; that the matter had never been put up for serious debate in any public forum prior to the issuance of the writ.  Nor can it be credibly argued that the legislation in question is the product of animus against any group of persons who lack access to Parliament.

What about changes in public opinion?  The plaintiffs can, if they choose, point to opinion polls reporting comparatively high levels of support from the general public (though not from physicians) for some form of decriminalization.  But Canadian courts are reluctant to let questions of constitutional principle be answered by opinion polls.  WJ Waluchow has argued that courts ought to be wary of majority opinion, and only give effect to what he characterizes as the “community’s constitutional morality”, understood as the “moral norms and convictions to which the community, via its various social forms and practices, has committed itself and that have in some way or other been drawn into the law via the rule of recognition and the law it validates.” (Common Law Theory of Judicial Review (2007), p. 227)  Opinion polls are not a suitable guide for judicial reasoning; courts, on this theory, may only be guided by deep, settled convictions that have been drawn into law.

This brings us, then, to questions of the development of constitutional doctrine.  The Supreme Court of Canada has, in recent years, not been shy about modifying settled Charter of Rights doctrine.  This is often in response to concerns that legal tests that have been adopted have become breeding grounds for confusion and ad hoc decision making.  The career of s.15(1), the Charter’s anti-discrimination provision, has been a poster child for this sort of thing.  Nevertheless, it is rare for the court to directly overturn a previous decision.  It did so recently in Health Services and Support – Facilities Subsector Bargaining Assn. v. British Columbia (2007), where it explicitly overturned a 20 year old precedent that held that the Charter’s guarantee of freedom of association under s. 2(d) does not provide a right to collective bargaining.  It did so by relying (uncharacteristically) on the original intent of the framers and acknowledging that its earlier decision was simply wrong.  The reliance on original intent is highly unusual in Canadian constitutional jurisprudence, and can only be explained by the Court’s anxiety that its departures from precedent appear principled and restrained, and not simply a matter of the Court having changed either its mind or its membership.

The Court will also take changes in international law and the practices of other jurisdictions as capable of justifying a change in domestic constitutional interpretation, but this is highly selective: see Sauvé v. Canada (Chief Electoral Officer) (2002), a prisoner voting case in which McLachlin CJ rejects the relevance of the practices of “self-proclaimed democracies” [para. 41] such as the UK, US, Australia, and New Zealand.

A more common avenue for escaping the strictures of precedent is for the court to formally uphold a precedent, while arguing that new or different facts, once fed into a proportionality test – fast becoming “one test to rule them all” – mandate a different result.  For example, when explaining why its constitutional review of the tobacco advertising restrictions in 2007 would not be governed by a 1995 precedent (RJR-MacDonald v. Canada (AG)), the Court stressed the significance of a different factual matrix:  in 1995, apparently, the Court was unaware that smoking could be addictive or a risk factor in fatal illness (Canada (AG) v. JTI-Macdonald (2007)).  Similarly, in a 2001 case on the constitutionality of extradition to face the death penalty (United States v. Burns, (2001)), the Court explained that ten years earlier (Kindler v. Canada (Minister of Justice), (1991)) it was unaware of the prevalence of wrongful conviction. These explanations, of course, are not always convincing, but they do allow a way around a precedent without admission of a change of heart.

Where does this leave Carter v. AGC?  To succeed in overcoming Rodriguez (given the hurdle placed by the continuing attention of both houses of Parliament), the plaintiffs would have to demonstrate that Rodriguez has been overtaken by developments in Canadian law dealing with the concepts of dignity, sanctity of life, and autonomy and that the experience with physician-assisted suicide in other regimes since 1993 should vitiate any concerns expressed in Rodriguez about the need to maintain the prohibition in the interests of the vulnerable.

On the question of whether there has been robust change to the concept of autonomy in Charter jurisprudence, the plaintiffs face an uphill battle.  There has indeed been an expanded licence for obscenity and indecency, and a minority of judges have held intermittently that the Charter may protect “basic choices going to the core of what it means to enjoy individual dignity and independence” (Godbout v. Longueil (1997)).  But all attempts to date to harness that proposition in support of a constitutional challenge to a criminal law have failed (see, eg, R. v. Malmo-Levine (2003), a Charter challenge to the criminalization of possession of marijuana).

Another question, of course, is the relevance of the contested legacy of the Netherlands and other jurisdictions since 1993.  The statutory framework adopted by the Dutch is considerably different from that adopted in, for example, Oregon.  The impact of each system on such matters as the incidence of non-consensual euthanasia and the availability and quality of palliative care remains has to be carefully assessed.  There is a large and controversial body of scholarship on these and other matters, and much time at trial will have to be given over to expert evidence.

Significantly, however, the plaintiffs in Carter are not proposing the adoption of any legislative framework.  They are simply demanding that the prohibition on assisted suicide be struck down as unconstitutional.  Were such an order granted, it would mean, of course, that there would be no safeguards or guidelines whatsoever to govern the practice of assisted suicide.  A reviewing court would not be in a good position institutionally to choose to adopt a Dutch model or an Oregon model or indeed any model of legislation at all.  The plaintiffs’ expectation, no doubt, is that the court would simply suspend its declaration of invalidity for a year or so to allow for Parliament to craft the necessary safeguards.

And here problems start to metastasize beyond considerations of whether a court is justified in overturning precedent.  A court cannot simply assume that if it strikes down the prohibition, Parliament will be able to draw up a replacement that both permits assisted suicide and satisfies Parliament’s obligation to protect the vulnerable.  Even assuming that such a regime would be possible, a political settlement might not be.  To understand the danger here, one only has to bear in mind the legislative aftermath of the Supreme Court of Canada’s decision to strike down the Criminal Code’s abortion provisions (R. v. Morgentaler (1988)).  The Court’s clear expectation was that Parliament would enact successor legislation, containing some new restrictions that would be applied more evenly across Canada.  But no successor legislation proved politically feasible, and as a result Canada remains without any legislative restrictions on abortion whatsoever.  Among the self-proclaimed democracies it is alone on this.

Once a court ventures into a morally charged debate such as abortion or assisted suicide, it changes the political dynamic in ways that cannot be anticipated.  The possibility of legislative failure has to be borne in mind by any reviewing court.

Bradley W. Miller is an Associate Professor in the Faculty of Law at the University of Western Ontario.