Keeping It Complicated

The Supreme Court’s issues its most originalist decision in years, but pretends it applies a different methodology

Military justice is a somewhat exotic topic; I don’t think my professors mentioned it even once in my time in law school, for instance. The Supreme Court’s decision in R v Stillman, 2019 SCC 40, delivered last week, is concerned with the functioning and limits on the jurisdiction of this parallel justice system. However, it should not only be of interest to the aficionados of this area of the law. Stillman was a relatively rare case where constitutional interpretation is front and centre, and it provides good illustrations of a number of problems with the way we do things on this front.

The issue before the Court was the meaning of the exception to the right to trial by jury guaranteed by section 11(f) of the Canadian Charter of Rights and Freedoms “in the case of an offence under military law tried before a military tribunal”, and specifically of the phrase “military law”. There is no question that specifically military offences created by the Code of Service Discipline that is part of the National Defence Act are “military law”; but what about the ordinary civilian offences (notably those created by the Criminal Code), which are incorporated by references by section 130(1)(a) of the Act? The majority, in an opinion by Justices Moldaver and Brown (with the agreement of Chief Justice Wagner and Justices Abella and Côté) find that these too are “offence[s] under military law”. Justices Karakatsanis and Rowe disagree and dissent.

Both the majority and the dissenting opinion present themselves as applying a purposive approach to the interpretation of section 11(f) of the Charter. However, they do not just differ in the outcomes that they reach. The majority’s professed purposivism shades, into public meaning originalism. The dissent’s has more than a whiff of I have been calling “constitutionalism from the cave”, the substitution by judicial fiat of the constitution that we perhaps ought to have for the one we actually have.


The majority begins by saying, with reference to a well-known passage in R v Big M Drug Mart, [1985] 1 SCR 295, that the provisions of the Charter ― both rights and, it insists, exceptions ―

are to be read purposively, rather than in a technical or legalistic fashion. And, just as courts must take care not to “overshoot” the purpose of a Charter right by giving it an unduly generous interpretation, so too must they be careful not to “undershoot” the purpose of a Charter exception by giving it an unduly narrow interpretation. [22]

The purpose of the right to trial by jury is to protect the accused against the state and also to involve the public in the administration of justice. That of the exception is to preserve the longstanding, separate system of military justice, which serves to maintain discipline and morale in the armed forces. The majority reviews the history, remit, and functioning of this system at considerable length.

Justices Moldaver and Brown then come to the interpretation of the phrase “military law” itself. With reference to Parliamentary debates at the time of the enactment in 1950 of the version of the National Defence Act in force in 1982, they point out that “‘military law’ was understood as ‘the law which governs the members of the army and regulates the conduct of officers and soldiers as such, in peace and war, at home and abroad'” and included “a provision transforming ordinary civil offences into service offences”. [74] They note, further, that the Criminal Code “at the time of the Charter’s enactment defined (and still defines) “‘military law’ as including ‘all laws, regulations or orders relating to the Canadian Forces'”, [75] and point to the Court’s decision in MacKay v The Queen, [1980] 2 SCR 370, where the majority opinion spoke of civilian offences incorporated by reference by the National Defence Act as being part of “military law”. Justices Moldaver and Brown concluded that it is “far more likely that the purpose of the military exception was to recognize and preserve the status quo” than to “reverse[] this longstanding state of affairs”. [78]

Justices Moldaver and Brown go on to reject the argument of the accused persons that the phrase “military law” only refers to purely military offences rather than the civilian ones incorporated by reference in the National Defence Act. To accept this, they say, would be contrary to MacKay and to the text of section 11(f).

They also reject the dissenters’ suggestion that to fall within the purview of “military law” within the meaning of section 11(f) an offence must be sufficiently connected to military service. The majority opinion in MacKay accepted that no special connection was required to make the incorporation by reference of civilian offences a valid exercise of Parliament’s power in section 91(7) of the Constitution Act, 1867, over “Militia, Military and Naval Service, and Defence”. Meanwhile, in R v Moriarity, 2015 SCC 55, [2015] 3 SCR 485, the Court upheld this incorporation against a challenge based on section 7 of the Charter, holding that the subjection of general criminal offences to the military justice system was rationally connected to that system’s purposes. While these cases raised different issues, “there must be coherence among the division of powers analysis, the overbreadth analysis, and the meaning of ‘an offence under military law’ in s. 11(f) of the Charter”. [97] To be sure, tying the scope of the exception in section 11(f) to Parliament’s power in section 91(7) means that Parliament can to some extent determine when the exception applies, but this no different from Parliament enacting criminal law and thereby triggering the application of various rights granted the accused. Besides, the requirement of a sufficient connection to military service is vague, and would cause difficulties in application.


Justices Karakatsanis and Rowe see things very differently. Previous decisions are not dispositive, and the requirement of a connection between the offence and military service is essential to avoid unduly limiting the right to trial by jury and giving Parliament and military prosecutors the ability to shape the contours of this right. Constitutional authority (in terms of division of powers) to enact an offence is not, in itself, a guarantee that the enactment will also comply with the Charter; nor is compliance with one right synonymous with compliance with others. Nor can the exercise of discretion by prosecutors, to bring charges in military court only when appropriate, be a substitute for the judicial enforcement of constitutional rights.

The dissenters appeal to the same passage from Big M setting out the principle of purposive interpretation as the majority, although they warn that exceptions to Charter rights should be approached with caution. The purpose of section 11(f), in their view, is to uphold “the interests of the accused and of society in holding a jury trial when prosecuting serious criminal offences”. [141] These interests must not be undermined allowing trials not sufficiently connected with military service to be held in the military, rather than the civilian, justice system.

Turning to history, Justices Karakatsanis and Rowe point out that the jurisdiction of military courts long remained narrow and was seen as a supplement to that of the civilian courts, only to be resorted to when civilian courts were unavailable. They also refer to MacKay, but to Justice McIntyre’s concurring opinion rather than the majority’s; this concurrence stressed the need for a military connection to bring an offence within the jurisdiction of military courts. This requirement was “adopted by the Court Martial Appeal Court … one year after the Charter, and has been applied with some regularity over the past thirty years”. [164] Justices Karakatsanis and Rowe conclude that “[t]his historical overview … highlights when military courts should have jurisdiction” ― namely “where quick and efficient justice was necessary to uphold discipline”, [166] and not otherwise.

As a result, the possibility that offences committed by persons subject to military justice but which are not sufficiently connected to their military service is an infringement of section 11(f) of the Charter. Justices Karakatsanis and Rowe conclude that this infringement is not justified in a free and democratic society. They go on to find that reading the requirement of connection to the military into section section 130(1)(a) of the National Defence Act in the best remedy in the circumstances.


The majority is right, although its reasoning is unnecessarily complex. The purported purposivism of both opinions obscures what is really going on. As suggested above, the reasons of Justices Moldaver and Brown are, at heart, originalist. The key passage in their opinion is that which discusses the way in which the phrase “military law” had been used by officials, by the Criminal Code, and by the Supreme Court itself, in the decades prior to 1982. Although they do not say so in so many words, Justices Moldaver and Brown thus go a long way towards establishing the public meaning of that phrase at the time of the Charter‘s enactment. Ideally, they would have stopped right there.

The references to the purpose of section 11(f) as a whole or of the military justice exception are superfluous. Purposive analysis may well be a helpful way to undertake constitutional construction ― that is, the development of legal doctrine in areas where constitutional text does not offer sufficient guidance to resolve concrete disputes (for example because the text is vague, or employs terms that appeal to moral or practical reasoning) ― as Randy Barnett and Evan Bernick have suggested. (I summarized and commented on their article here.) But, as Stillman shows, purposivism does not meaningfully contribute to constitutional interpretation ― that is, the activity of ascertaining the meaning of the constitutional text itself. When, as in this case, it is possible to find out what the text means, and to resolve the dispute based on that meaning alone, the speculation that the text was presumably intended to say what it said rather than something else adds nothing to the analysis.

It may be, of course, that the pretense of purposivism is necessary to make originalism palatable to (some of) the current members of the Supreme Court. If so, it might be a reasonable price to pay; but then again, it might not. When Stillman is cited in the future for its unanimous embrace of purposivism, will it be in support of the majority’s empty ― and harmless ― version of the methodology, or of the dissent’s, which consists of emphasizing purposes at the expense of the original meaning of the text?

The dissent starts with a view of how the constitution ought to treat the relationship between civilian and military justice, and insists that this view must become law. It pays little heed to the meaning of the phrase “military law”, reading into it a limitation that is, in its view, desirable, but has no obvious foundation in the constitutional text. While Justices Karakatsanis and Rowe appeal to history, they cherry-pick the record and ignore the crucial period: that immediately preceding the enactment of the Charter. The practice of the previous centuries may be interesting, but it cannot be dispositive given that matters stood very differently by the time the Charter came into being. Still less can the jurisprudence of Canadian military courts in the decades that followed, and its embrace of Justice McIntyre’s concurrence in MacKay, have any bearing of the Charter‘s meaning. The dissent’s use of history appears to be more result-oriented more than a genuine attempt to ascertain “the historical origins of the concepts enshrined” in section 11(f), to borrow Big M‘s language. If this is what purposivism is, then we should run, not walk, away from it.


The reasoning of the Stillman majority is perhaps the most originalist, and specifically public-meaning originalist, in a constitutional case since that of the majority in Caron v Alberta, 2015 SCC 56, [2015] 3 SCR 511. In the meantime, of course, there has been the thoroughly unoriginalist decision in R v Comeau, 2018 SCC 15, [2018] 1 SCR 342. As Benjamin Oliphant and I have written, it would of course be wrong to pretend that the Supreme Court is consistently originalist; but it would also be wrong to deny originalism’s place in Canadian constitutional law. Even seemingly decisive setbacks, like Comeau, are only ever provisional.

It is not just the Court as a whole that is inconsistent; so are individual judges. Every member of the Stillman majority signed onto the Comeau judgment. Justice Wagner, as he then was, and Justices Côté and Abella were the dissenters in Caron, favouring an approach that privileged the supposed intentions of the framers of the provision at issue over its original public meaning. This time they join a majority opinion where original meaning does the heavy lifting. Justice Karakatsanis, by contrast, had co-authored the majority opinion in Caron, but now dissents.

One rather suspects that the judges simply do not give much thought to constitutional interpretation, at least beyond what they see as the needs of individual cases. If this is so, then there is little reason to expect that occasional ― but erratic and not especially well-reasoned ― resort to originalism by the Supreme Court will continue. As Mr. Oliphant and I argued, however, it would be highly desirable if more thought were given to constitutional interpretation, and if the Court went about this task in a more consistent and principled manner.

Lastly, I would be remiss not to add that I am inclined to think that, at the level of policy, the concerns raised by Justices Karakatsanis and Rowe deserve serious consideration. In my comment on Moriarity, I wrote that

there is … a broader question to be asked about the extent to which an institution to which a person belongs ought to be able to discipline that person for behaviour occurring outside the institutional context, for the sake of maintaining “morale,” or harmony, or respect, etc.

I still think so. To be sure, the armed forces are a rather unique sort of institution. Perhaps there is good reason to give them the sort of broad jurisdiction over the actions of their members that, as Stillman holds, the Charter allows. But perhaps not. Yet this is a matter for Parliament to consider. The constitution, on this point, does not constrain it.

Not too Broad

In a decision delivered this morning, the Supreme Court of Canada has upheld the constitutionality of subjecting the members of the Canadian armed forces to the military justice system for all almost offences against acts of Parliament. In R. v. Moriarity, 2015 SCC 55, it ruled that the provisions of the National Defence Act pursuant to which such offences could be tried in the military justice system, regardless of whether the accused was in uniform or on military property when they were committed were not inconsistent with s. 7 of the Canadian Charter of Rights and Freedoms due to overbreadth, as the appellants had argued. In a unanimous decision by Justice Cromwell, the Court held that these provisions were rationally connected to their purpose ― once it was properly understood.

The first issue in a challenge based on s. 7 of the Charter is whether one of the “interests” that it protects ― life, liberty, or security of the person ― is “engaged” at all. Justice Cromwell easily concludes that this is the case here, because the provisions at issue are “part of a scheme through which a person … can be deprived of his or her liberty.” The main question is whether the potential deprivation of liberty pursuant to these provisions would be “in accordance with the principles of fundamental justice.”

Only one principle was said to be offended by the provisions at issue ― that according to which laws must not be overbroad. The Supreme Court’s jurisprudence defines overbreadth as a lack of rational connection between a legal rule’s purpose and its effect. As Justice Cromwell explains, the definition of the purpose is crucial to the overbreadth analysis. A legislative purpose defined too generally, at the level of a value, “will provide no meaningful check on the means employed to achieve it, [because] almost any challenged provision will likely be rationally connected to” it [28]. If, however, “the identified purpose is articulated in too specific terms, then the distinction between ends and means may be lost and the statement of purpose will effectively foreclose any separate inquiry into the connection between them,” [28] ensuring that any law will survive an overbreadth challenge. To steer the middle course between these pitfalls, “[t]he statement of purpose should generally be both precise and succinct.” [29]

All parties agreed that the purpose of making members of the Canadian forces subject to a distinct system of military justice “relates to assuring the discipline, efficiency and morale of the armed forces.” [33] However, the appellants argued that Parliament that the military justice system was only meant to prosecute offences directly connected to the armed forces ― and not those committed by the forces’ members in civilian circumstances. The prosecution, for its part, contended that the military justice system had an additional “public function of punishing specific conduct which threatens public order and welfare.” [47]

The Court rejected both these suggestions. The legislation, it said, was quite clear when it sought to exclude offences committed outside the service from its scope ― as it did for the members of the reserves. Other offences falling within the military’s Code of Service Discipline naturally extend to conduct occurring outside the service. In short, “the overall thrust of the scheme [is] to include offences when committed by an individual subject to the [Code] regardless of what other link may or may not exist between the circumstances of the offence and the military.” Meanwhile, the prosecution’s addition of a broader “public order” purpose would result in confusing the purpose and the effect of the provisions at issue, making the inquiry into the existence of a connection between the two meaningless.

Such a connection existed, the Court concluded. In Justice Cromwell’s words,

Criminal or fraudulent conduct, even when committed in circumstances that are not directly related to military duties, may have an impact on the standard of discipline, efficiency and morale. For instance, the fact that a member of the military has committed an assault in a civil context … may call into question that individual’s capacity to show discipline in a military environment and to respect military authorities. The fact that the offence has occurred outside a military context does not make it irrational to conclude that the prosecution of the offence is related to the discipline, efficiency and morale of the military. [52]

Criminal behaviour by a member of the military, regardless of when and where it occurs, may also erode the trust and the respect others will have towards him or her. As a result, it is at least rational to prosecute offences committed outside the military context through the military justice system.

This seems fair enough, though the Court’s conclusion is asserted more than argued. Admittedly, the bar of rational connection between purpose and effect is a low one. The Court’s decision is narrow. Justice Cromwell flags some issues it does not address (because they were not raised):

the question of the scope of Parliament’s authority to legislate in relation to “Militia, Military and Naval Service, and Defence” under s. 91(7) of the Constitution Act, 1867 and the scope of the exemption of military law from the right to a jury trial guaranteed by s. 11(f) of the Charter are not before us. [30]

But there is also, arguably, a broader question to be asked about the extent to which an institution to which a person belongs ought to be able to discipline that person for behaviour occurring outside the institutional context, for the sake of maintaining “morale,” or harmony, or respect, etc. It is a question I raised in one of my very first posts, which dealt with a challenge to a punishment imposed by a university to students for a series of Facebook posts. I wrote then that

I can see why a university might be interested in what is being said in its lecture halls, or online on forums it maintains (in connection with courses for example). It does have an interest in maintaining a welcoming, respectful learning environment … But does this interest give a university the right to police the conduct of its students off-campus or online?

The Alberta Court of Appeal, which had decided that case, Pridgen v. University of Calgary, 2012 ABCA 139, had not addressed that question. Here, the Supreme Court concluded that it was rational for the military ― a very different institution from a university, of course ― to do so. But again, rationality is a low threshold, and there is little in the court’s decision that would help us answer the question on the merits in other settings.

This is not a criticism, of course. The Court decided the question before it, and did not have to do more. Its decision, like the provisions it considered, was no broader than it had to be.