SCC Continues to Roll Back Harper-Era Criminal Code Amendments in R. v. Ndhlovu

In its recent decision, R v Ndhlovu, 2022 SCC 38 [Ndhlovu], the Supreme Court of Canada [“SCC”] struck down two provisions of the Criminal Code, RSC 1985, c C-46 [“the Code”] relating to registration of offenders convicted of designated sexual offences in the national sex offender registry. In its 5-4 ruling, the majority held that ss. 490.012 and 490.013(2.1) of the Code unjustifiably violated s. 7 of the Charter, which guarantees the right to life, liberty, and security of the person. 

The national sex offender registry was created by the Sex Offender Information Registration Act, SC 2004, c 10, [“SOIRA”]. Section 490.012 of the Code requires courts to order offenders convicted of designated sexual offences to be registered for a duration determined by the severity of the offence. In 2011, the Harper government added s. 490.013(2.1), which requires courts to order mandatory lifetime registration of offenders convicted of multiple sexual offences and removed judicial discretion to impose SOIRA orders. The amendments also removed the requirement for police to show reasonable grounds before consulting the registry.

With this ruling, the SCC continues to roll back Code provisions introduced by the Conservative government which imposed harsher, one-size-fits-all sentences for a range of offences.

Content warning: The following section includes details of sexual assault that some readers may find distressing.

Facts of the Case

In 2011, then 19-year-old Mr. Ndhlovu groped two women at a house party in Alberta. Later, he digitally penetrated one of the women while she was asleep. He continued and tried to remove her bra, even after she woke up and motioned for him to stop. He fled after she pushed him and repeatedly told him to stop.

In 2015, Mr. Ndhlovu pled guilty to two counts of sexual assault against the two complainants.   

Judicial History

The sentencing judge, Justice Moen, found that Mr. Ndhlovu was unlikely to reoffend. Despite the fact that the two convictions of sexual assault arose out of a single incident, the impugned provisions subjected Mr. Ndhlovu to mandatory lifetime registration in the national sex offender registry. Mr. Ndhlovu challenged  the impugned provisions following sentencing, claiming violations of ss. 7 and 12 of the Charter.

Justice Moen found that the impugned provisions violated the principles of fundamental justice, and therefore, breached s. 7 rights. The Court held that the provisions were overbroad, as the blanket nature of the registration requirement would inevitably capture offenders with little to no recidivism risk. This would not advance SOIRA’s purpose of aiding police investigations of sex crimes. Justice Moen also found that the provisions were grossly disproportionate, given the significant burdens of compliance on the offender, despite already having served their sentence. 

The judge ruled that the provisions could not be justified under s.1 of the Charter. While the public interest objective of the provision was pressing and substantial, she found that there was no evidence that eliminating judicial discretion is rationally connected to the provisions’ objective. Relying heavily on the Crown expert, she held that lifetime registration did not minimally impair offenders’ rights. Given the lack of evidence to support claims about the registry’s effectiveness, the deleterious effects of the provisions outweighed its salutary effects.

A majority at the Court of Appeal of Alberta [“ABCA”] allowed the Crown’s appeal, holding that the provisions did not violate the Charter. The ABCA ruled that the provisions were neither overbroad nor grossly disproportionate, because  all convicted sex offenders have an increased propensity to reoffend. It further held Parliament was free to infer an increased recidivism risk from multiple convictions of a sexual offence.

Justice Khullar, in her dissent, found that the provisions were overbroad. She agreed with the sentencing judge’s holding that the negative impact on offenders’ liberty interests outweighed its benefits, and as such, the s.7 infringements could not be justified.

On appeal, the SCC sought to clarify the following: 

  1. Do the impugned provisions unjustifiably violate s. 7 of the Charter? 
  2. If the impugned provisions cannot be justified, what is the correct remedy?

The Majority Decision

A majority of the SCC found that the provisions interfered with offenders’ liberty, given the onerous reporting obligations that could last decades or a lifetime, under threat of imprisonment, fines, or both.

Writing for the majority, Justices Karakatsanis and Martin agreed with the sentencing judge that both provisions were overbroad. In the absence of judicial discretion and case-by-case risk assessment, the provisions capture offenders who are not at an increased risk of recidivism. Moreover, the Court noted that offenders who commit multiple offences in the same incident are not associated with an increased recidivism risk. The majority also held that neither provision could be justified under s.1, for substantially the same reasons detailed by the sentencing judge. 

The SCC declared both provisions invalid, with immediate and retroactive effect in the case of lifetime registration orders under s. 490.013(2.1). In the case of s. 490.012, the Court suspended the declaration of invalidity for 12 months and applied it prospectively. 

The dissent agreed with the majority on the unconstitutionality of s. 490.013(2.1), but found that s. 490.012 was constitutional. Writing for the dissent, Justice Brown held that given the uncertainty in determining recidivism risk, Parliament was justified in requiring mandatory registration. 


Reducing a person to the language of risk

What is striking in this case is the preoccupation with risk assessment in criminal justice administration. This fails to account for non-retributivist criminal justice goals, replacing concern for offenders’ rehabilitation with making ex-ante predictions about their future actions. 

For the majority, prior conviction is only one of multiple variables that determine risk of recidivism. Therefore, relying on this factor as a proxy for risk is insufficient. Moreover, the majority cited that ten percent of sex offenders are not at an enhanced risk of recidivism at the time of sentencing, compared to the general population of offenders. The dissent, for its part, argued that even offenders at low risk of reoffending relative to the general criminal population are at greater risk of committing another sexual offence. Therefore, in the face of uncertainty, Parliament was right to cast a wide net.

Proponents may argue that risk-based decision-making serves the rule of law by being more objective and consistent than ad hoc assessments of a person’s capacity and willingness to learn from their mistakes. But such arguments fail to appreciate the extent to which normative values permeate risk assessment tools, while simultaneously obscuring them from scrutiny. What appears at face-value as numbers and statistics inevitably contain assumptions about highly contested values and questions of law and policy, that are concealed by its outward scientific objectivity. 

Overemphasis on risk undermines non-retributivist goals of the criminal justice system like rehabilitation of the offender, by reducing a person’s capacity to change to a generalized statistic. Even if one concedes the objectivity of risk-based law and policy, we lose something important in the unfettered drive towards removing human error in decision-making—the humanity required to help offenders confront the consequences of their actions and take necessary steps to meaningfully change their behaviour.  

Even when SOIRA orders are measured against different values like efficiency and effectiveness, it hardly fares better. Decades after  its enactment, the Crown’s own evidence failed to show how SOIRA orders have helped law enforcement investigate and prevent sexual offences. The investigator administering the registry in Edmonton testified she had only received 15 requests from police to access the registry in the 2 years she had been overseeing it. It is untenable that such significant infringements of liberty can be justified for such minuscule (potential) gains.  

State monitoring – routine or onerous?

The majority considered in detail the scope, frequency, and duration of the reporting obligations, before rightly concluding they were exacting and onerous. They include, among other things, random compliance checks, disclosure of primary and secondary residences, notification of dates in which the offender intends to be away from their residence, and addresses of every location at which they intend to stay during this time.

It is shocking, therefore, that both the ABCA and the SCC’s dissent describe these requirements as being akin to reporting obligations that “routinely occur as part of everyday life” (Ndhlovu, para 165). The dissent even goes so far as to say “describ[ing] these conditions as onerous strips that term of all meaning” (para 165). Yet, it is hard to imagine anyone who is not under police custody, serving a prison sentence, or on parole being subjected to such intrusive state monitoring, as offenders complying with SOIRA orders. Onerous is not only a reasonable description of the requirements of SOIRA orders, it is disingenuous to liken it to the ordinary cost of contemporary life.

The dissent further justifies its understatement of the cost of compliance by arguing that it is a “registry,” (para 165, emphasis in original) and as such is only effective when it is comprehensive. Regardless of the blameworthiness of the crime, offering up such utilitarian rationales for serious infringements of fundamental freedoms is insidious, to say the least. It is disappointing that the dissent makes this argument to justify universal approaches to criminal justice, all to enable little more than administrative convenience and enforcement practicality for the police.

Judicial discretion and rule of law

To their credit, the dissent points to significant concerns related to judicial discretion, which can invite troubling claims from some judges about who is and is not a “real sex offender” and what kinds of assault are “serious enough” to warrant serious consequences. Enhanced discretion inevitably carries the risk of arbitrariness, which in turn undermines rule of law values. 

To be clear, recognizing that there are degrees of severity within the category of sexual offences is no excuse for undermining complainants’ claims. This is especially so given the prevalence of sexual abuse and assault in Canada, compounded by discriminatory attitudes toward women, people with physical and mental disabilities, LGBTQ+ groups, and people from racialized and economically vulnerable communities.

However, these concerns do not justify the impugned provisions. Firstly, it is doubtful that mandatory registration for those who are unlikely to or will never reoffend will protect victims of sexual violence. Secondly, as the majority in Ndhlovu notes, the SCC’s own jurisprudence has established that the over-inclusion of even one person runs afoul of s. 7 rights, and practicality is no answer to overbreadth (para 78). This is especially relevant given that in passing the impugned provisions, Parliament not only eliminated judicial discretion, it also chose not to provide tailored guidance to courts when enforcing the provisions, that may have helped to overcome its concerns about under-inclusion.

The majority, therefore, rightly notes that there is a middle ground between mandatory registration and rampant discretion that could retain the purported benefits of a sex offender registry, without egregiously infringing Charter rights. In this case, Parliament simply chose not to take it.

Potential further litigation?

It is worth noting an interesting point made by the dissent, that the majority’s description of SOIRA orders poses doctrinal difficulties. According to the dissent, the majority’s ruling may imply that SOIRA orders constitute a form of punishment, and as such fall within Charter protections (Ndhlovu, para 167).  

If so, SOIRA orders would be subject to s.11(i) of the Charter. This section gives offenders the right to benefit from the lesser punishment, in cases where punishment for an offence has been changed in the time between committing the offence and sentencing. This would mean that offenders convicted of historic sexual offences, committed prior to the establishment of the registry, could avoid being registered altogether. 

Existing jurisprudence has avoided this issue, characterizing SOIRA orders as a consequence of conviction rather than punishment. But if the dissent is right, this may not be the last constitutional challenge against SOIRA orders.  

Azka Anees

Azka Anees is a second-year JD student at Osgoode Hall Law School. In addition to contributing to, Azka works on the Osgoode Hall Law Journal as a Senior Editor. Her wide-ranging research interests include Charter jurisprudence, philosophy of law, and the intersection of law and literature. Azka holds a BA in Politics, Philosophy and Economics from Exeter University and an MSc in Philosophy and Public Policy from the London School of Economics. In her free time, she enjoys reading literary fiction, watching character-driven TV shows, and baking from scratch (the quest to achieve the Ghirardelli box brownie level of perfection continues!).

You may also like...

Join the conversation

Loading Facebook Comments ...