Securing the Presumption of Innocence in Prison Disciplinary Proceedings
John Howard Society of Saskatchewan v Saskatchewan (Attorney General)
In a 2019 article, Canadian prison law scholar Lisa Kerr referred to the prison as a “black box in punishment theory.” Kerr was critiquing the disconnect between theories of punishment and the everyday realities of the prison. She explained that the conditions of confinement have a material impact on the severity of a sentence, and argued that scholars and courts ought to grapple more directly with how prison sentences are experienced by those serving them.
I have always found the black box to be an apt metaphor in describing the legal realities of the prison. Once an offender is sentenced, they enter a stage of the process that most Crown attorneys, defence lawyers and judges – people who are otherwise experts in arrest, detention, pre-trial matters, the trial process and sentencing – know very little about. Although the number of “prison lawyers” in Canada is growing, it remains a small bar. And while some provinces have committed additional resources to prison law issues in recent years (Legal Aid Ontario developed a prison law strategy in 2019, and in 2024 a new prison law clinic was launched at the University of Manitoba funded by Legal Aid Manitoba), access to legal assistance for prisoners remains inadequate. This is especially so in provincial facilities, where individuals are on remand or are serving shorter sentences. This lack of access to counsel means that for many prisoners, Charter and other legal rights are experienced as contingent rather than guaranteed.
Viewed against this backdrop, the Supreme Court of Canada’s recent decision in John Howard Society of Saskatchewan v Government of Saskatchewan (Attorney General for Saskatchewan is notable. There, a majority of the Court concluded that Saskatchewan’s use of the balance of probabilities standard in deciding major disciplinary offence cases in provincial prisons violated section 11(d) of the Charter.[1] The majority also held that this standard infringed the residual protection provided to the presumption of innocence by section 7. Three dissenting justices, led by Justice Côté, concluded that the province was not constitutionally obligated to adopt the beyond a reasonable doubt standard for this category of disciplinary offence.
In the courts below, the John Howard Society had based its arguments exclusively on section 7 of the Charter. However, before the Supreme Court, it argued that section 11(d) was also engaged. The major obstacle to the section 11(d) argument was the Supreme Court’s 1990 decision in R v Shubley. In that case, a majority of the Court concluded that prison disciplinary proceedings were not criminal proceedings, nor did they carry true penal consequences. According to R v Wigglesworth, this meant that disciplinary offences were not “offences” within the meaning of section 11, and that the section was therefore not engaged.
Writing for the majority in John Howard, Chief Justice Wagner concluded that judicial developments since Shubley meant that the decision now “rest[ed] on eroded legal foundations.” In particular, Shubley’s overly “formalistic” analysis led the majority to wrongly conclude that prison disciplinary proceedings did not have “true penal consequences,” despite the fact that they could increase both the time an individual spent in prison as well as the severity of the conditions of their incarceration.
Having concluded that Shubley should be reconsidered, Chief Justice Wagner re-examined the application of the Wigglesworth criteria:
In Wigglesworth, this Court developed two tests for determining which “offences” will trigger the protections of s. 11. First, s. 11 can be invoked when the proceedings at issue are “criminal in nature” (p. 559). Proceedings of this kind are “intended to promote public order and welfare within a public sphere of activity” and stand in contrast to “private, domestic or disciplinary matters which are regulatory, protective or corrective” (p. 560). Second, s. 11 can be invoked when the proceedings may lead to the imposition of “true penal consequences” (p. 561). Such consequences include “imprisonment” or a “fine which by its magnitude would appear to be imposed for the purpose of redressing the wrong done to society at large rather than for the maintenance of internal discipline” (ibid.).
Did the fact that a conviction for a prison disciplinary offence could result in administrative segregation or loss of earned remission constitute “imprisonment” within the meaning of Wigglesworth? “Adopting [a] functional definition of imprisonment,” the answer was yes:
Imprisonment under Wigglesworth’s true penal consequence test must therefore include state-imposed sanctions that, in light of their attributes, represent a deprivation of liberty at least as severe as that resulting from an initial sentence of imprisonment. This approach is necessary to ensure that the state cannot simply label forms of imprisonment with euphemisms in order to circumvent the application of s. 11 of the Charter.
The majority concluded that administrative segregation and loss of earned remission were “sanctions that impose a constraint on an individual’s freedom of movement and segregate them from others to a degree at least equivalent to that of a sentence of incarceration.” For this reason, the presumption of innocence was triggered, as was the higher burden of proof.
The majority also concluded that the regulation ran afoul of section 7. The section 7 protection of the presumption of innocence is triggered where “the state (a) accuses an individual of moral wrongdoing and (b) seeks to punish the individual with severe liberty-depriving consequences for such wrongdoing.” Both elements were satisfied in this case. The regulation at issue thus infringed sections 11(d) and 7, and could not be saved under section one.
A Growing Judicial Trend
The John Howard case is consistent with a growing judicial willingness to pry open the black box and supervise how the state treats prisoners in their care. Indeed, one of the more striking features of the majority’s decision is its narration of the history of judicial treatment of the prison. Chief Justice Wagner explained that “The conditions of imprisonment have traditionally been understood as falling within the purview of correctional institutions, not courts ... This abstentionist approach to the conditions of imprisonment emerged in part from the common law’s historical view that a person who had been convicted of an offence and sentenced to prison was “devoid of rights.”” Over time, however, that position has shifted, and the opposite presumption has taken hold: “a person confined to prison retains all of his civil rights, other than those expressly or impliedly taken from him by law” (John Howard, quoting Solosky v The Queen).
This evolution in judicial thinking is reflected in a number of cases decided in the past few decades, including two recent provincial court of appeal decisions (here and here) dealing with the practice of administrative segregation, and a growing number of habeas corpus cases successfully challenging prison conditions. Taken together, these cases represent a positive development in Canadian law, one which seeks to establish minimum standards of dignified treatment in the prison context.
In theory, the statements in John Howard establish a powerful foundation for future Charter challenges in the prison context. However, it bears noting that both the John Howard case and the administrative segregation cases were brought by organizations that advocate on behalf of prisoners. There remain considerable hurdles to prisoners bringing these types of challenges forward without the support of such organizations.
The Critics
Some individuals will be disappointed about the outcome in John Howard. Correctional institutions are likely to view the decision as rendering routine prison disciplinary proceedings more cumbersome, and as an impediment to maintaining order and security in the prison.
These are ultimately empirical claims. In my view, it is inevitable that the higher standard of proof will introduce some added complexity into disciplinary proceedings. At the same time, the beyond a reasonable doubt standard is already being employed in federal correctional institutions. Moreover, the majority in John Howard concluded that this standard applies to serious disciplinary offences rather than to all offences. This suggests that the concerns about complexity and maintaining order and security can be managed.
As the majority’s reasoning emphasizes, moreover, it is clear that, at least with respect to some punishments handed out following a conviction for a disciplinary offence, the punishment does have the effect of lengthening the period of imprisonment or rendering it more severe. To treat such punishments as merely administrative in nature would be to ignore the reality of how they operate.
Approach to Charter Interpretation
Finally, since I have been keeping track of how the Supreme Court articulates the basic rules of Charter interpretation in its opinions (see previous blog post here), I note that the majority in John Howard makes some very important comments about the need for courts to take a generous approach to Charter interpretation. Indeed, the majority explained that an “overly formalistic” approach to Charter interpretation can and has provided a basis for overturning prior Supreme Court precedent:
When Shubley is situated within the full constellation of this Court’s subsequent case law on Charter interpretation, it is clear that its reasoning in applying the true penal consequence test rested on a formalistic method of interpretation that has since been consistently eschewed.
This Court has overturned precedents that adopted an overly formalistic method of Charter interpretation before (see, e.g., R. v. Beaulac, [1999] 1 S.C.R. 768, at paras. 16-25; Health Services and Support — Facilities Subsector Bargaining Assn. v. British Columbia, 2007 SCC 27, [2007] 2 S.C.R. 391, at para. 30; Quebec (Attorney General) v. A, 2013 SCC 5, [2013] 1 S.C.R. 61, at paras. 338-47, per Abella J., dissenting, but not on this point).
Legal interpretation becomes formalistic when there is excessive adherence to matters of form at the expense of substance. When formalism is adopted, legal interpretation becomes a “mechanical and sterile categorization process” that undermines the law’s capacity to achieve its underlying purpose and ignores how understandings of legal concepts must adapt in light of varying social contexts (see R. v. Turpin, [1989] 1 S.C.R. 1296, at p. 1332).
Avoidance of formalism takes on heightened importance in constitutional interpretation because a constitution’s “function is to provide a continuing framework for the legitimate exercise of governmental power and, when joined by a Bill or a Charter of Rights, for the unremitting protection of individual rights and liberties” (Hunter v. Southam Inc., [1984] 2 S.C.R. 145, at p. 155). For this reason, this Court has insisted that the interpretation of a Charter right be “a generous rather than a legalistic one, aimed at fulfilling the purpose of the guarantee” (R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295, at p. 344).
The absence of any mention of Quebec Inc. or City of Toronto – two 5-4 decisions in which the Court took what is looking increasingly like a brief detour into a more textual approach to Charter interpretation – is notable. As in the recent Power decision, the majority’s opinion seems to signal a preference for the older, better established precedents on the basic approach to Charter interpretation.
* I am grateful to Leo Russomanno, Jeremy Opolsky and Andrew Bernstein for conversations about this case, and to the wonderful students in uOttawa’s winter 2024 Supreme Court Seminar and Osgoode’s 2025 Criminal Law and the Charter course for similar discussions.
[1] Section 11(d) provides that “Any person charged with an offence has the right (d) to be presumed innocent until proven guilty according to law in a fair and public hearing by an independent and impartial tribunal.”


I'm sympathetic to the end result but, it seems like the "generous approach" to interpretation does a lot of heavy lifting in getting there.
Particularly with respect to the loss of earned remission, it seems to me that the loss of liberty occurred at trial when the person received the sentence in accordance with the principles of fundamental justice. Not lowering the sentence is no further deprivation of liberty.