By: Lisa Silver
PDF Version: If Not Now, When?
Case Commented On: R v Natomagan, 2022 ABCA 48 (CanLII)
The opening paragraphs of the recent Alberta Court of Appeal decision in R v Natomagan, 2022 ABCA 48 (CanLII), belie the significance of the decision. It commences like many other appellate sentencing decisions, setting out the lower Court’s ruling by focusing on a narrow ground of appeal. In this case, that ground encompasses the Crown appeal against the imposition of a determinate rather than an indeterminate sentence for a designated dangerous offender. By paragraph 3, the Court has shown its hand and finds the sentencing judge “applied the wrong legal standard.” By paragraph 5, the Court allows the appeal and imposes an indeterminate sentence. So far, as expected. But it is in the next paragraph where the decision steps out of the ordinary and becomes a case to read closely, thoroughly, and with interest. There, the Court raises concerns with the “unfettered reliance” on the use of “actuarial risk assessment tools” in determining custodial options for Indigenous offenders within the criminal justice system (at para 6). The Court directly connects these biased risk assessment tools to the overrepresentation of Indigenous offenders in the carceral system (at paras 7 to 13). Finally, the Court provides a well-placed caution requiring judges to make informed decisions in using these tools (at para 141). Despite this warning and well-placed concern, the Court, as foreshadowed by the opening paragraphs, reverts to the usual by finding the offender, Ashton Natomagan, to be an “intractable risk to the public” (at para 137). This means the biased and discriminatory risk assessment tools did not impact the ultimate finding that he was a danger, requiring an indeterminate sentence (at paras 137 to 138). This disconnect between law and reality is a continuing theme in the criminal justice experience of Indigenous offenders. Although this decision is a positive step in recognizing wrongs and attempting to ameliorate injustices, more must be done now to change the future outcomes for Indigenous offenders like Ashton.
Analyzing the Tools & Enunciating the Principles
A good portion of the decision analyzes the record and the evidence heard at the sentencing hearing. According to the Court, the experts at trial agreed that the predictions made by these risk assessment tools were “somewhat less reliable when applied to Indigenous offenders” (at para 10). It must be noted that the sentencing judge, Justice Terry Clackson, acknowledged risk assessment tools can “overstate” risk when “not adjusted adequately” for Indigenous offenders (R v Natomagan, 2019 ABQB 943 (CanLII) at para 37). Despite this concern, Justice Clackson found the tools have predictive value (at para 37). Risk assessment tools consider several factors, such as “childhood instability, substance abuse, suicidal ideation, employment, and number of criminal convictions” (at para 10 ABCA), all of which are prevalent in the life experiences of Indigenous offenders.
The tools score the offender based on these high-risk factors and then, based on actuarial data of previous offenders, give a percentage of offenders within that scoring category who are likely to re-offend (at para 10). These tools, with their predictive powers, do not account for the overrepresentation of Indigenous offenders with these high-risk factors, nor do the tools calibrate for “discriminatory effects in the input factors”, which produce biased results (at para 11). Moreover, although the discriminatory effects and inaccurate results of these tools have been recognized, according to the Court, the discriminatory effects have not been “corrected or quantified” (at para 11). The court conducted their own examination of the “actuarial assessment methodology” in finding it was deeply flawed by overestimating the risk posed by Indigenous offenders, failing to account for historical discrimination, and perpetuating systemic (at paras 13, 35 – 53, 102 – 104). Indeed, much of the predictive value of risk assessment tools as applied to Indigenous offenders is founded on statistical inflation by virtue of the disproportionate number of Indigenous people charged resulting from over-policing and over-charging. This in turn creates a disproportionate measure of recidivism rates by Indigenous people in comparison to non-Indigenous offenders (at para 110).
Why use these tools at all? The assessments purport to ensure that the sentencing judge does not base the need to protect the public purely on past behaviour (at para 94). Moreover, these tools connect to the objective of the regime, which is to prevent future public harms. Additionally, both case law and statute have embedded the use of risk assessment tools into the dangerous offender hearing process (see e.g., R v Boutilier, 2017 SCC 64 (CanLII) at paras 39 to 44 and s 752.1 of the Criminal Code, RSC 1985, c C-46 (Code)).
Consistent with the sentencing judge’s findings and statutory requirements, the Court did not suggest that the risk assessment tools are useless and ought not to be used when applied to Indigenous offenders. Rather, the Court views risk assessment tools as valid when used properly. That proper use, according to the Court, must recognize the discriminatory effects of such tools when applied to Indigenous offenders. This recognition requires the sentencing judge to make decisions informed by this finding. To be informed is to conduct a “thorough” “evidence-based inquiry” on the offender’s background and circumstances to determine prospective risk (at para 98 & 120). As part of that inquiry, the judge must understand any “inaccuracy flowing” from racialized discrimination in the assessment methodology and must “minimize” the impact (at paras 117, 122 & 124). But it is not just the judge’s task. Accordingly, “all criminal justice system participants should take reasonable steps to address systemic biases against Indigenous people head-on” (at para 122). This requires scrutiny of the “roles played and tools used by all decision-makers who influence the deprivation of Indigenous offenders through any means, including sentencing, placement and parole” (at para 123).
Application to Ashton Natomagan
For Ashton, there was no real issue with the dangerous offender designation. His life reads like a nightmare full of abuse, suicide and imprisonment (at para 9). It also is a life harmed by racism both in and out of the justice system as he was subjected to “over-policing, over-charging” and onerous bail conditions (at para 9). His crimes are heinous and disturbing involving physical and sexual violence against women. Justice Clackson was aware of all of this and scrupulously applied the law in the area (see Natomagan, ABQB at paras 88 to 90). Even so, the Court found the sentencing judge erred in imposing an indeterminate sentence, considering the finding Ashton was a very high risk of offending (at para 88). But, Justice Clackson, in imposing a 20-year determinate sentence, found Ashton in an age range where “burnout” (a phenomenon wherein aging reduces risk, at para 84) could possibly affect his risk of re-offending (Natomagan, ABQB at para 93). Moreover, such a lengthy sentence might see a “new paradigm” for the treatment of Indigenous offenders (Natomagan, ABQB at para 93). Finally, Justice Clackson found Ashton personally “suffered a great deal as a result of Canada’s mistreatment of Indigenous persons”, and his “criminality is a product of those injustices” (Natomagan, ABQB at para 94). As Ashton did not willingly choose this path, “his moral culpability is, therefore not that of one who chooses the life of a violent recidivist” (Natomagan, ABQB at para 94). Taking his personal circumstances into account with the systemic over-representation of Indigenous people in jails, as well as the expert evidence that Ashton would have “real benefit from long-term isolation and gradual socialization under the direction and with the support of Elders,” a determinate sentence was appropriate (Natomagan, ABQB at paras 95 – 97).
For Justice Clackson, the reduced moral culpability was an important factor in imposing a determinate sentence. This aspect of Justice Clackson’s reasoning, specifically the moral culpability finding, was not directly discussed by the Court. What was directly discussed was the moral responsibility “non-indigenous Canadians had in creating” Ashton’s risk (Natomagan, ABQB at para 98 and Natomagan, ABCA at 90). The Court found Justice Clackson erred in taking this into account while acknowledging that “undeniably Canada’s legacy of colonialism contributed” to Ashton’s life experiences (at para 90). This acknowledgement did not change the fact Ashton was a risk to society and therefore ought not impact the sentencing decision (at para 90). It should be noted that in the context of historical and societal wrongs against Black Nova Scotian offenders, Justice Derrick in R v Anderson, 2021 NSCA 62, found sentencing principles should be “informed by society’s role in undermining the offender’s prospects as pro-social and law-abiding citizen” (at para 159). This powerful sentiment is one that has a place in sentencing an Indigenous offender, even a high-risk one like Ashton, who “we will never know … might have been … had his life experiences – including his experiences in the criminal justice system – been less harsh” (Natomagan, ABCA at para 139).
The Court did however place reduced weight on the risk assessments due to its “lower predictive value” owing to the “social, economic, and historical factors” Indigenous offenders disproportionately experience (at para 134). The Court also commented on the lack of information emanating from Ashton himself as he refused to participate in clinical interviews (at para 133). There was no discussion as to whether this might have occurred because of discrimination and bias in the assessment process itself.
More Needs to be Done Now
No doubt it is vitally important that informed decision-making occurs in these contexts. It is equally important that the court give direction on this informed approach. But is that enough? If the court finds the tools, although “prone to overestimating the risk posed by Indigenous offenders by failing to consider or to account for past discrimination, thereby potentially contributing to custodial over-representation” must still be considered, then detailed guidelines and integrative training on how to use such biased tools in assessing an Indigenous offender would give added weight and meaning to the direction (at para 13). This case was an opportunity for the court to not only call out bias and racism but also to put some appellate grit behind these concepts.
The court does see the need for an evidence-based approach, meaning evidence on the inaccuracy or frailties of the assessment tool should be heard. Certainly, such factual knowledge will assist the judge. But evidence-based informed decision-making can only be useful if that evidence can really assist in understanding the impact of racial injustices on a person. In fact, the court emphasizes the difficulties in parsing what in these assessments are systemic racial discrimination and what can be attributed to an individual’s “countless complex experiences” (at para 118). The evidence may be there, but it is the interpretation of that evidence that requires more guidance and more structured responses from the court. A sentencing judge in a dangerous offender hearing faces an onerous and delicate task in ensuring the public is adequately protected while also ensuring the least restrictive sentence needed to provide that protection is imposed. Predictive sentencing is playing a high-stakes game of chance for both the public and the offender. The stakes are magnified when it is an Indigenous offender whose path has been impacted by societal wrongs.
Interestingly, the Court connects Parliament’s “instruction” in s 718.2(e) of the Code to “pay particular attention” to the circumstances of the Indigenous offender to the need for accurate risk assessments, which inform dangerous offender sentencing principles (at para 13). Considering the Court’s findings that risk assessment tools are prone to overestimate risk, then it is surprising that these tools, as they are presently used, are to be considered at all. The tools, as the Court suggests, are not infallible (at para 12), but in the case of Indigenous offenders, they are harmful and potentially wrong. To leave this to weight (the ultimate probative value of the evidence) seems to minimize the problems with the use of these tools. Rather than “limitations”, the reliability and relevancy of these tools are brought into question (at para 12). In R v Jones, 1994 CanLII 85, [1994] 2 SCR 229, the admissibility of psychiatric evidence at the dangerous offender hearing was in issue. According to the majority decision of Justice Charles Gonthier, at the sentencing stage, strict rules of evidence do not apply even though the Crown must prove disputed facts beyond a reasonable doubt. In a dangerous offender hearing, “if the sentencing judge is to obtain the accurate assessment of the offender that is necessary to develop an appropriate sentence, he will have to have at his disposal the broadest possible range of information” (at 291, emphasis added). Although this suggests the risk assessment evidence is best left to the judge’s discretion and weight determinations, if accuracy is the goal, then the prejudicial effect of admitting evidence gleaned from discriminatory risk assessment tools could outweigh their probative value even at the sentencing stage.
Of course, the barrier to such an aggressive stance as suggested above is the statutory need for assessments. This can and should be remedied by amendments to the Code, similar to the recent amendments to the bail provisions under s 493.2 requiring “special attention” be given to the circumstances of Indigenous peoples in making a bail decision. Although s 718.2(e) requires Gladue and Ipeelee principles apply to the dangerous offender regime, there is an opportunity to make explicit in the dangerous and long-term offenders section of the Code (see Part XXVI) that risk assessments are biased against Indigenous people and must either be used with caution or be rejected if required. The new section could also permit other kinds of evidence to be accepted in place of such risk assessment tools, which are more tailored to the Indigenous experience and would reflect Indigenous culture and practices. This kind of modification can also inspire further opportunities for targeted and meaningful treatment of Indigenous offenders in carceral institutions and in the community.
In the end, this decision is an important and vital step towards reconciliation, but it stops. Indigenous offenders need more programs and better tools to assist in their rehabilitation and to provide a better future outside of custody. Criminal justice participants, like lawyers and judges, need specific guidelines and training on how to account for indigenous historical wrongs and those wrongs that manifest themselves personally on the offender, as in Ashton’s case. This is a decision that comes close to a watershed moment in Indigenous overrepresentation in the justice system yet fails to provide any real solutions for Indigenous offenders like Ashton, who are caught inextricably in the system. The conclusionary paragraph on Ashton’s situation is telling. There, the Court connects systemic racism in the justice system to Ashton’s dire personal circumstances (at para 139). Equally telling is the sense of despair as the Court acknowledges “[w]e will never know who Mr. Natomagan might have been, or whether he would have committed the same offences, had his life experiences—including his experiences in the criminal justice system—been less harsh” (at para 139). Sadly, for the Court this connection cannot change the brutal truth that Ashton’s own personal circumstances divulge no “reasonable expectation that a determinate sentence will adequately protect the public” (at para 141). For Ashton the recognition of racism in risk assessment tools is a pyrrhic victory where everyone in society is the loser.
Although the Court in this decision makes a detailed analysis of the discriminatory potential of risk assessment tools and directs sentencing judges to heed the unjust application of such tools, the Court offers no solution to the situation posed by the Indigenous offender before them. As in R v Gladue, 1999 CanLII 679, [1999] 1 SCR 688 the principles enunciated, although enlightened and progressive, would not impact the sentence imposed. In Ashton Natomagan’s case, the risk he poses to society if released is significant and cannot be mitigated through historical wrongs, both collective and personal. Rather than the determinate sentence wrongly imposed, Ashton Natomagan must serve an indeterminate sentence, which may hold him in custody for life. This is not a solution to the problem of over-incarceration of Indigenous peoples, but it is the reality.
The Natomagan decision raises the question of how systemic change can happen when legal principles and precedent can only bend so far. It also speaks a stark truth that for offenders like Ashton Natomagan, the future is bleak indeed without an overhaul of not just the way Indigenous offenders are sentenced but also in the way society treats Indigenous people. The result of this decision is that we need our government to implement better treatment options and stronger statutory directions. We also need our Judges to have the tools and training required to make real, concrete change. This involves a hard look at risk assessment tools, not just a recognition that they are inherently biased. More must be done now to break the cycle for Ashton and those like him.
This post may be cited as: Lisa Silver, “If Not Now, When?” (March 16, 2022), online: ABlawg, http://ablawg.ca/wp-content/uploads/2022/03/Blog_LS_If_Not_Now.pdf
To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca
Follow us on Twitter @ABlawg