R. v. Suter, 2018 SCC 34

SUPREME COURT OF CANADA Appeal Heard: October 11, 2017 Judgment Rendered: June 29, 2018 Docket: 37247 Between: Richard Alan Suter Appellant and Her Majesty The Queen Respondent Coram: Abella, Moldaver, Karakatsanis, Wagner, Gascon, Côté and Rowe JJ. Reasons for Judgment: (paras. 1 to 104) Moldaver J. (Abella, Karakatsanis, Wagner, Côté and Rowe JJ. concurring) Reasons Dissenting in Part: (paras. 105 to 202) Gascon J. R. v. Suter, 2018 SCC 34, [2018] 2 S.C.R. 496 Richard Alan Suter Appellant v. Her Majesty The Queen Respondent Indexed as: R. v. Suter 2018 SCC 34 File No.: 37247. 2017: October 11; 2018: June 29. Present: Abella, Moldaver, Karakatsanis, Wagner, Gascon, Côté and Rowe JJ. on appeal from the court of appeal for alberta Criminal law — Sentencing — Considerations — Collateral consequences — Mitigating factors — Accused accidentally driving vehicle onto restaurant patio and killing child — Accused pleading guilty to offence of refusing to provide breath sample knowing that he caused accident resulting in death and sentenced to four months of imprisonment and driving prohibition — Court of Appeal increasing sentence to 26 months of imprisonment — Whether lower courts erred in determining appropriate sentence — Criminal Code, R.S.C. 1985, c. C‑46, ss. 255(3.2), 718 to 718.2. S drove his vehicle onto a restaurant patio, killing a two‑year‑old child. The police demanded a breath sample after the accident but S refused, on the advice of a state‑provided lawyer to whom he spoke after his arrest. He was charged with refusing to provide a breath sample after causing an accident resulting in a death, under s. 255(3.2) of the Criminal Code, and with impaired driving causing death and impaired driving causing bodily harm. Sometime after being charged, S was abducted by vigilantes who cut off his thumb with pruning shears for his role in the child’s death. S eventually pleaded guilty to the s. 255(3.2) offence and the other charges were withdrawn. The sentencing judge imposed a 4‑month sentence of imprisonment on S, coupled with a 30‑month driving prohibition. He found that the accident was caused by a non‑impaired driving error, S having hit the gas pedal instead of the brake pedal. He further found that S’s refusal to provide a breath sample was the result of bad legal advice and was a mistake of law, which fundamentally changed S’s moral culpability. In addition to that and other mitigating factors, the sentencing judge took into account the violent vigilante actions against S. The Court of Appeal allowed a Crown appeal from that sentence and increased the custodial portion of it to 26 months. It found that the deficient legal advice did not constitute a mistake of law and it could not be used to mitigate S’s sentence. It also found that the sentencing judge failed to consider, as an aggravating factor, that S chose to drive while distracted in the context of his health and pre‑existing alcohol problems, and that the sentencing judge erred by taking the vigilante violence into account. Held (Gascon J. dissenting in part): The appeal should be allowed in part. The sentence of 26 months’ imprisonment imposed by the Court of Appeal should be set aside and replaced with one of time served. The 30‑month driving prohibition should be upheld. Per Abella, Moldaver, Karakatsanis, Wagner, Côté and Rowe JJ.: The sentencing range for the s. 255(3.2) offence is the same as for impaired driving causing death and driving “over 80” causing death — low penitentiary sentences of 2 or 3 years to more substantial penitentiary sentences of 8 to 10 years. The sentencing range is broad because these offences cover a broad spectrum of offenders and circumstances. In unique cases, mitigating factors, collateral consequences, or other attenuating circumstances relating to the offence or offender may warrant the imposition of a sentence that falls below this broad range, or aggravating factors may warrant the imposition of a sentence that exceeds this broad range. As long as the objectives and principles of sentencing codified in ss. 718 to 718.2 of the Criminal Code are met and respected, the sentence will be fit. Facts that are irrelevant to the gravity of an offence and to the level of the offender’s moral blameworthiness with respect to that offence cannot be relied on as aggravating in the sentencing analysis. To consider such facts is an error in principle, which may cause a court to punish the offender for an offence for which he or she was neither tried nor convicted, and result in the imposition of an unfit sentence. Tailoring sentences to the circumstances of the offence and the offender may require the sentencing judge to examine collateral consequences. A collateral consequence includes any consequence arising from the commission of an offence, the conviction for an offence, or the sentence imposed for an offence, that impacts the offender. Collateral consequences do not need to be foreseeable, nor must they flow naturally from the conviction, sentence, or commission of the offence, but they must relate to the offence and the circumstances of the offender. There is no rigid formula for taking collateral consequences into account, and there is no requirement that collateral consequences emanate from state misconduct in order to be considered a factor at sentencing. However, the fundamental principle of proportionality must prevail in every case — collateral consequences cannot be used to reduce a sentence to a point where it becomes disproportionate to the gravity of the offence or the moral blameworthiness of the offender. Violent actions against an offender for his or her role in the commission of an offence necessarily form part of the personal circumstances of that offender, and should therefore be taken into account when determining an appropriate sentence. However, vigilante violence should only be considered to a limited extent, as giving it too much weight at sentencing allows this kind of criminal conduct to gain undue legitimacy in the judicial process. Although it is not a defence to a criminal charge, mistake of law can be used as a mitigating factor in sentencing, because offenders who honestly but mistakenly believe in the lawfulness of their actions are less morally blameworthy than offenders who are unsure about the lawfulness of their actions, or know that their actions are unlawful. A mistake of law is a legal concept with rigorous requirements, which occurs only where a person has an honest but mistaken belief in the legality of his or her actions. Confusion or uncertainty as to the lawfulness of one’s actions does not meet the legal requirements for mistake of law, however, such confusion may still be relevant to the sentencing analysis depending on the facts of the particular case. Its mitigating effect, if any, will necessarily be less than in a situation where there is a true mistake of law. A finding of non‑impairment is a relevant mitigating factor when sentencing an offender for refusing to provide a breath sample, but the mitigating effect of such a finding must be limited for several reasons. First, sentencing hearings for refusal offences could be transformed into de facto impaired driving trials, adding to the complexity and length of the proceedings and depleting scarce judicial resources. Second, since refusal offences are in essence an evidence gathering tool to obtain the most reliable evidence of impairment, the seriousness of the offence and the moral blameworthiness of the offender stem primarily from the refusal itself, and not from the offender’s level of impairment. Third, it could create an incentive for individuals not to provide a breath sample, be convicted of the refusal offence, and then subsequently argue at the sentencing hearing that they were not impaired to benefit from a reduced sentence. The extent to which the mitigating effect must be limited is a fact‑driven exercise that depends on the specific circumstances in any given case, and the onus is on the offender to establish, on a balance of probability, that he or she was not impaired at the time the offence was committed. In the instant case, both the sentencing judge and the Court of Appeal committed errors in principle in arriving at the sentences they imposed and these errors resulted in the imposition of unfit sentences. The Court of Appeal erred when it recast the circumstances of the accident and effectively sentenced S for the uncharged offence of careless driving or dangerous driving causing death, and when it held that the vigilante violence inflicted on S could not be considered when crafting an appropriate sentence. The sentencing judge erred in finding that S was acting under a mistake of law when he refused to provide a breath sample and that this factor fundamentally changed his moral culpability, and in giving undue weight to S’s non‑impairment as a mitigating factor. In S’s unique case, the following factors operate to remove his sentence from the normal range for a s. 255(3.2) offence: he was not impaired at the time of the accident, he refused to provide a breath sample because of ill‑informed and incorrect legal advice, and he was attacked by vigilantes. However, they do not justify the sentence imposed by the sentencing judge, which does not properly account for the gravity of the offence. A sentence of 15 to 18 months’ imprisonment would have been a fit sentence at the time of sentencing. However, S has already served just over 10 and a half months of his custodial sentence and has spent almost 9 months awaiting the Court’s decision. It would not be in the interests of justice to re‑incarcerate S at this time — it would cause him undue hardship and serve no useful purpose. Per Gascon J. (dissenting in part): There is agreement with the majority that the Court of Appeal’s sentence of 26 months in prison was unfit. However, there is disagreement with the sentence of 15 to 18 months in prison prescribed by the majority. The four‑month carceral sentence imposed by the sentencing judge should be restored. It cannot be revisited on appeal since it does not implicate a flawed process (material errors in reasoning) — such as an error in principle or an error in weighing a relevant factor unreasonably — or a flawed outcome (demonstrable unfitness). The sentencing judge made no error in principle in his analysis of S’s mistake of law. A mistake of law does not require an offender to be certain as to the lawfulness of their conduct. Such a narrow construction is antithetical to the contextual and individualized nature of sentencing. Mistake of law is a flexible concept broad enough to include some confusion or uncertainty about the law. Accordingly, thinking conduct is likely legal, but being uncertain, is sufficient to constitute a mistake of law. Mistake of law should not be dealt with as a binary, where only a person being completely confident that their conduct is legal fundamentally alters culpability. Trial judges should be trusted to take a contextual approach — one which considers the source, nature and reasonableness of a mistake, along with any degrees of uncertainty — when allocating mitigating weight to a mistake of law for the purpose of sentencing. In any event, here, S was certain as to the legality of refusing to provide a breath sample, and therefore made a mistake of law, even on the majority’s test. The sentencing judge did not give excessive weight to S’s sobriety. An appellate court can intervene on sentence when a sentencing judge weighs a particular factor unreasonably, but not when the appellate court would have simply weighed the relevant factor differently. When reviewing trial reasons, appellate courts must read the reasons as a whole, and should not isolate single passages from trial reasons to find errors in reasoning. In this case, isolating a passage from the sentencing judge’s decision to conclude that he gave excessive weight to S’s sobriety when sentencing him mischaracterizes the sentencing judge’s reasons. The sentencing judge looked at the combined effect of sobriety, bad legal advice, vigilante violence and the many other mitigating factors in this case. Accordingly, his weighing of sobriety provides no basis for appellate intervention. Even if a trial judge makes no errors in the sentencing process, a court of appeal can interfere if the ultimate sentence is demonstrably unfit, meaning that the sentence is clearly unreasonable. Demonstrable unfitness is not an unchecked subjective inquiry. An appellate court must demonstrate the unfitness of a sentence with reference to the Criminal Code ’s sentencing principles, including proportionality ( s. 718.1 ), the sentencing objectives ( s. 718 ), individualization ( s. 718.2 (a)) and parity ( s. 718.2 (b)). A sentence of four months of imprisonment in the circumstances of this case is not demonstrably unfit based on this established approach to sentencing appeals. A fit sentence must be proportionate to the gravity of the offence (how serious the offence is) and the degree of responsibility of the offender (their moral blameworthiness). In the instant case, the gravity of the offence — refusing to provide a breath sample after a fatal car accident — is very high. However, S’s moral blameworthiness — as a sober driver who was in a genuine car accident caused by a non‑impaired driving error and who refused to provide a breath sample only because he was expressly instructed to do so by his lawyer — could hardly be lower. The tensions that inevitably arise when balancing these conflicting considerations underlie the particularly delicate task of ascertaining proportionality. However, appellate courts are not in a better position than the sentencing judge to reconcile these conflicting forces; rather, given the latter’s proximity to the facts of the case and his institutional expertise, appellate courts are in a worse position. Courts of appeal must defer to the sentencing judge with respect to this complex balancing exercise. Other than the sentencing judge’s expertise and S’s unique circumstances, two additional factors reinforce the proportionality of a significantly reduced sentence. First, S was not only mistaken in law, but reasonably mistaken. He relied on a state‑provided lawyer’s advice when refusing to provide a breath sample, reasonably so. His moral blameworthiness is therefore infinitesimal. Second, the constitutional significance of the right to counsel also weighs in favour of a significantly reduced sentence. People must be able to rely on legal advice given when exercising their constitutional right to counsel. The proportionality of the sentence is reinforced by the applicable sentencing objectives. The objective of deterring the offender and other persons from committing offences must be emphasized in the impaired driving context. However, the facts of the case at bar cannot be disregarded, as they are critical to a proportionate sentence for S. S was not tempted to commit a crime, nor did he commit a crime because of insufficient deterrence. He refused to provide a breath sample because he was told to do so by his lawyer. A stiff sentence in this case will not deter others from refusing to provide a breath sample; rather, it will deter others from following their lawyer’s advice. Similarly, the objective of denouncing unlawful conduct and the harm done to victims or to the community that is caused by unlawful conduct must be emphasized, but without disregarding the unique facts of this case. Mere commission of an administrative offence, when advised to do so by a state‑provided lawyer, does not warrant strict denunciation, as it lacks the required moral blameworthiness. A sentence should be reduced to account for any mitigating circumstances relating to the offence. In the instant case, S’s reliance on bad legal advice, his sobriety and the vigilante attacks he suffered are not the only mitigating factors. S’s sentence must also be reduced because of his guilty plea, his extreme remorse, his lack of a criminal record, his strong community support and the fact that he has been a productive member of society. These mitigating factors operate collectively in determining a fit sentence. Viewed together, they are remarkably mitigating. The majority’s willingness to ultimately impose a 10‑and‑a‑half‑month sentence, despite signalling that S’s conduct warrants a sentence as high as 18 months, demonstrates that such a harsh sentence would be disproportionate in S’s circumstances. Precisely how these mitigating factors are reflected in a particular sentence is most empirically determined through the sentencing principle of parity, which provides that a sentence should be similar to sentences imposed on similar offenders for similar offences committed in similar circumstances. The facts in S’s case are entirely unique. There are no similar offenders in similar circumstances against which his sentence can be reasonably measured. The case which is closest in terms of moral blameworthiness imposed the same carceral sentence as was imposed here (four months). Further, the jurisprudence establishes a range of up to a year when sentencing this offence only in terms of obstruction of justice, rather than as a proxy for impaired driving. The flexibility found in the jurisprudence demonstrates that Parliament intended the s. 255(3.2) offence to operate harshly enough to deter any incentive for refusal, but flexibly enough to recognize that refusal is not coterminous with impaired driving in all situations. Accordingly, based on the unique facts in this case and the jurisprudence, there is no basis to claim that S’s four‑month sentence was manifestly unfit. Cases Cited By Moldaver J. Applied: R. v. Lacasse, 2015 SCC 64, [2015] 3 S.C.R. 1089; R. v. Mian, 2014 SCC 54, [2014] 2 S.C.R. 689; referred to: R. v. Brydges, 1990 123 (SCC), [1990] 1 S.C.R. 190; R. v. Shropshire, 1995 47 (SCC), [1995] 4 S.C.R. 227; R. v. L.F.W., 2000 SCC 6, [2000] 1 S.C.R. 132; R. v. L.M., 2008 SCC 31, [2008] 2 S.C.R. 163; R. v. Nasogaluak, 2010 SCC 6, [2010] 1 S.C.R. 206; R. v. Junkert, 2010 ONCA 549, 103 O.R. (3d) 284; R. v. Kummer, 2011 ONCA 39, 103 O.R. (3d) 641; R. v. Angelillo, 2006 SCC 55, [2006] 2 S.C.R. 728; R. v. Larche, 2006 SCC 56, [2006] 2 S.C.R. 762; R. v. Proulx, 2000 SCC 5, [2000] 1 S.C.R. 61; R. v. Ipeelee, 2012 SCC 13, [2012] 1 S.C.R. 433; R. v. Pham, 2013 SCC 15, [2013] 1 S.C.R. 739; R. v. Bunn (1997), 1997 22728 (MB CA), 118 Man. R. (2d) 300; R. v. Bunn, 2000 SCC 9, [2000] 1 S.C.R. 183; Tran v. Canada (Public Safety and Emergency Preparedness), 2017 SCC 50, [2017] 2 S.C.R. 289; R. v. MacFarlane, 2012 ONCA 82, 288 O.A.C. 114; R. v. Folino, 2005 40543 (ON CA), 2005 ONCA 258, 77 O.R. (3d) 641; R. v. Anderson, 2014 ONSC 3646; R. v. Mamarika, 1982 ABCA 281, [1982] FCA 94, 42 A.L.R. 94; R. v. McDonald, 2016 NUCA 4; R. v. Stanberry, 2015 QCCQ 1097, 18 C.R. (7th) 87; R. v. Bell, 2013 MBQB 80, 290 Man. R. (2d) 79; R. v. Heatherington, 2005 ABCA 393, 380 A.R. 395; R. v. Owens (2002), 2002 41948 (ON CA), 161 O.A.C. 229; R. v. Abouabdellah (1996), 1996 6502 (QC CA), 109 C.C.C. (3d) 477; R. v. Carroll (1995), 1995 1123 (BC CA), 56 B.C.A.C. 138; R. v. Forster, 1992 118 (SCC), [1992] 1 S.C.R. 339; R. v. Pontes, 1995 61 (SCC), [1995] 3 S.C.R. 44; R. v. MacDonald, 2014 SCC 3, [2014] 1 S.C.R. 37. By Gascon J. (dissenting in part) R. v. Kresko, 2013 ONSC 1631, 42 M.V.R. (6th) 224; R. v. Lacasse, 2015 SCC 64, [2015] 3 S.C.R. 1089; R. v. Forster, 1992 118 (SCC), [1992] 1 S.C.R. 339; R. v. Pontes, 1995 61 (SCC), [1995] 3 S.C.R. 44; R. v. MacDonald, 2014 SCC 3, [2014] 1 S.C.R. 37; R. v. Oickle, 2000 SCC 38, [2000] 2 S.C.R. 3; Rothman v. The Queen, 1981 23 (SCC), [1981] 1 S.C.R. 640; R. v. Collins, 1987 84 (SCC), [1987] 1 S.C.R. 265; R. v. Sinclair, 2010 SCC 35, [2010] 2 S.C.R. 310; R. v. Nasogaluak, 2010 SCC 6, [2010] 1 S.C.R. 206; R. v. Rhyason, 2007 SCC 39, [2007] 3 S.C.R. 108; R. v. Rezaie (1996), 1996 1241 (ON CA), 31 O.R. (3d) 713; R. v. Shropshire, 1995 47 (SCC), [1995] 4 S.C.R. 227; R. v. Proulx, 2000 SCC 5, [2000] 1 S.C.R. 61; R. v. Ramage, 2010 ONCA 488, 257 C.C.C. (3d) 261; R. v. L.M., 2008 SCC 31, [2008] 2 S.C.R. 163; R. v. Bartle, 1994 64 (SCC), [1994] 3 S.C.R. 173; R. v. Brydges, 1990 123 (SCC), [1990] 1 S.C.R. 190; R. v. Hebert, 1990 118 (SCC), [1990] 2 S.C.R. 151; Clarkson v. The Queen, 1986 61 (SCC), [1986] 1 S.C.R. 383; R. v. Singh, 2007 SCC 48, [2007] 3 S.C.R. 405; R. v. Smith, 2017 MBPC 16, 10 M.V.R. (7th) 152; R. v. Ipeelee, 2012 SCC 13, [2012] 1 S.C.R. 433; R. v. Holliday, 2009 ONCJ 323, 87 M.V.R. (5th) 148; R. v. Wallace, 2012 MBCA 54, 280 Man. R. (2d) 209; R. v. M. (C.A.), 1996 230 (SCC), [1996] 1 S.C.R. 500. Statutes and Regulations Cited Canadian Charter of Rights and Freedoms, s. 10( b ). Criminal Code, R.S.C. 1985, c. C‑46, ss. 19, 249(4), 254(5), 255(1) (a)(i), (2), (2.2) [ad. 2008, c. 6, s. 21(3) ], (3), (3.1), (3.2) [ idem ], (3.3), 258(3), 718 to 718.2, 718, 718(a), (b), 718.1, 718.2(a), (b), (e), 718.3(1), 725(1)(c), (2)(b). Traffic Safety Act, R.S.A. 2000, c. T‑6, s. 115. Authors Cited Ashworth, Andrew. Sentencing and Criminal Justice, 5th ed. Cambridge: Cambridge University Press, 2010. Foy, James. “Proportionality in Sentence Appeals: Towards a Guiding Principle of Appellate Review” (2018), 23 Can. Crim. L.R. 77. Kenny’s Outlines of Criminal Law, 19th ed. by J. W. Cecil Turner. Cambridge: University Press, 1966. Manson, Allan. The Law of Sentencing. Toronto: Irwin Law, 2001. Ruby, Clayton C., Gerald J. Chan and Nader R. Hasan. Sentencing, 8th ed. Markham, Ont.: LexisNexis, 2012. APPEAL from a judgment of the Alberta Court of Appeal (Watson, Bielby and Schutz JJ.A.), 2016 ABCA 235, 341 C.C.C. (3d) 21, 41 Alta. L.R. (6th) 268, 100 M.V.R. (6th) 177, [2016] A.J. No. 785 (QL), 2016 CarswellAlta 1461 (WL Can.), varying the sentence imposed by Anderson A.C.J. Prov. Ct., 2015 ABPC 269, 94 M.V.R. (6th) 91, [2015] A.J. No. 1407 (QL), 2015 CarswellAlta 2333 (WL Can.). Appeal allowed in part, Gascon J. dissenting in part. Dino Bottos, Will Van Engen, Fady Mansour and Peter Sankoff, for the appellant. Joanne Dartana and David A. Labrenz, Q.C., for the respondent. The judgment of Abella, Moldaver, Karakatsanis, Wagner, Côté and Rowe JJ. was delivered by Moldaver J. — I. Overview [ 1 ] The circumstances of this case are tragic. Two-year-old Geo Mounsef was killed when the appellant, Richard Suter, drove his vehicle onto a restaurant patio where the Mounsef family was eating dinner. In a matter of seconds, George Mounsef and Sage Morin lost a son, and Quentin Mounsef lost a brother. These devastating consequences speak to the enormity of the tragedy. [ 2 ] Mr. Suter was initially charged with three offences arising out of this incident, including impaired driving causing death and impaired driving causing bodily harm. The impaired driving charges were later withdrawn by the Crown when Mr. Suter pleaded guilty to one count of refusing to provide a breath sample knowing that he caused an accident resulting in a death, an offence set out in s. 255(3.2) of the Criminal Code, R.S.C. 1985, c. C-46. Refusing to provide the police with a breath sample is always serious, but especially so when a death occurs. The maximum penalty for this offence — like for impaired driving causing death and driving “over 80” causing death — is life imprisonment. And lest there be any doubt, for sentencing purposes, these three offences will typically be treated alike. [ 3 ] That said, the circumstances of this case are unique. As we shall see, the fatal accident was caused by a non-impaired driving error, and Mr. Suter refused to provide the police with a breath sample because he received bad legal advice. The lawyer he called from the police station expressly told him not to provide a breath sample, and Mr. Suter demurred. Added to this, sometime after the accident, Mr. Suter was attacked by a group of vigilantes who used a set of pruning shears to cut off his thumb. His wife was also attacked in a separate incident. [ 4 ] Sentencing is a highly individualized process. A delicate balancing of the various sentencing principles and objectives is called for, in line with the overriding principle that a “sentence must be proportionate to the gravity of the offence and the degree of responsibility of the offender” ( s. 718.1 of the Criminal Code ). Accordingly, there will be cases where the particular circumstances of the offence and/or the offender call for a sentence that falls outside of the normal sentencing range. This is one such case. [ 5 ] The sentencing judge imposed a 4-month sentence of imprisonment on Mr. Suter, coupled with a 30-month driving prohibition. The Court of Appeal of Alberta allowed a Crown appeal from that sentence and increased the custodial portion of it to 26 months. For reasons that follow, I am respectfully of the view that both the sentencing judge and the Court of Appeal committed errors in principle in arriving at the sentences they imposed and these errors resulted in the imposition of unfit sentences. Accordingly, this Court may conduct its own analysis to determine a fit sentence. [ 6 ] The Court of Appeal erred when it recast the circumstances of the accident and effectively sentenced Mr. Suter for the uncharged offence of careless driving or perhaps dangerous driving causing death. It also erred when it held that the vigilante violence inflicted on Mr. Suter could not be considered when crafting an appropriate sentence. The sentencing judge erred in finding that Mr. Suter was acting under a mistake of law when he refused to provide the police with a breath sample and that this factor fundamentally changed Mr. Suter’s moral culpability. He also erred in giving undue weight to Mr. Suter’s non-impairment as a mitigating factor. [ 7 ] The errors committed by both the Court of Appeal and the sentencing judge materially contributed to the respective sentences they imposed. In the circumstances, I would allow Mr. Suter’s appeal from the 26-month custodial sentence ordered by the Court of Appeal and, for reasons that will become apparent, I would reduce it to one of time served — just over 10 and a half months. Like the Court of Appeal, I would not interfere with the driving prohibition. II. Facts [ 8 ] The facts in this case derive from Mr. Suter’s sentencing hearing. What follows is a summary of uncontested facts and pertinent findings made by the sentencing judge. [ 9 ] On May 19, 2013, Mr. Suter and his wife went to dinner at Chili’s restaurant. Each consumed one alcoholic drink. When the food arrived, Mr. Suter’s meal was cold. He became upset and insisted on going elsewhere for dinner. Mrs. Suter was displeased but agreed to leave. An argument ensued as the couple drove to a nearby restaurant known as Ric’s Grill. Upon arriving at Ric’s Grill, Mr. Suter pulled into a parking space adjacent to the outside patio of the restaurant. The vehicle stopped a few yards back from the glass partition that separated the patio from the sidewalk, however, Mr. Suter did not put the vehicle in park as he realized that he had mistakenly pulled into a “by permit only” space. [ 10 ] While the vehicle was stopped in that space, Mrs. Suter turned to her husband and exclaimed “Maybe we should just get a divorce”. At about the same moment, she realized that the vehicle was inching forward, and she yelled at her husband to stop. Unfortunately, Mr. Suter’s foot had come off the brake pedal and instead of hitting the brake, he pressed down on the gas pedal. The vehicle accelerated through the glass partition and within a second or two, it slammed into the restaurant wall. [ 11 ] George Mounsef, his wife Sage Morin, and their two young children Geo and Quentin were having dinner on the patio when Mr. Suter’s vehicle came crashing through the glass partition. They were struck by the vehicle, and Geo Mounsef remained pinned by it against the wall of the restaurant for about 30 seconds. Amidst the screaming, someone told Mr. Suter that there was a child under his vehicle and he backed up slowly. At that point, Mr. Suter was pulled from the driver’s seat, thrown to the ground, and beaten by witnesses at the scene. When the police arrived, they found Mr. Suter lying in a fetal position on the parking lot pavement. Mr. Suter was arrested, he was taken to the police station, and a breath demand was made. [ 12 ] At the station, Mr. Suter tried unsuccessfully to phone a lawyer with whom he was familiar. The police suggested that he call a lawyer on contract with Legal Aid (also known as a “ Brydges lawyer”: see R. v. Brydges, 1990 123 (SCC), [1990] 1 S.C.R. 190) and Mr. Suter complied. During the conversation, the Brydges lawyer confused Mr. Suter with legal jargon. At no point did he inquire of Mr. Suter as to how much alcohol, if any, he had consumed that day. In the end, the Brydges lawyer expressly told Mr. Suter not to provide the police with a breath sample. In line with this advice, when asked to provide a breath sample, Mr. Suter refused, despite being told by the officer that refusing to provide a breath sample was an offence. [ 13 ] Mr. Suter was charged with three offences: refusing to provide a breath sample after causing an accident resulting in a death (under s. 255(3.2) of the Criminal Code ), impaired driving causing death (under s. 255(3) of the Criminal Code ), and impaired driving causing bodily harm (under s. 255(2) of the Criminal Code ). [ 14 ] Sometime after being charged, Mr. Suter was abducted by vigilantes. Three hooded men took him from his home in the middle of the night, handcuffed him, and placed a canvas bag over his head. His attackers then drove him to a secluded area, cut off his thumb with pruning shears, and left him unconscious in the snow. Mrs. Suter was also attacked by vigilantes in a shopping mall parking lot. Both incidents were linked to Mr. and Mrs. Suter’s role in Geo Mounsef’s death. [ 15 ] On June 5, 2015, Mr. Suter entered a plea of guilty to the s. 255(3.2) offence. As indicated, the impaired driving charges were withdrawn by the Crown. III. Decisions Below A. The Sentencing Decision (Anderson A.C.J.), 2015 ABPC 269, 94 M.V.R. (6th) 91 [ 16 ] At the sentencing hearing, defence counsel sought a non-custodial sentence — either a fine or a fine and probation. Crown counsel, on the other hand, sought the imposition of a three-year custodial sentence. As indicated, the sentencing judge imposed a sentence of 4 months’ imprisonment coupled with a 30-month driving prohibition. [ 17 ] In his reasons, the sentencing judge emphasized that this case was unique. As tragic as the consequences were, he characterized the accident as one “caused by a non-impaired driving error” (para. 76). He also found that Mr. Suter’s refusal to provide a breath sample was the result of “hopefully rare, ill-informed and bad legal advice” ( ibid. ). According to the sentencing judge, this fact could not absolve Mr. Suter, “as a mistake of law is not a defence”, but it nevertheless “fundamentally change[d] Mr. Suter’s moral culpability” ( ibid. ). [ 18 ] The sentencing judge noted a number of other mitigating factors in this case, including that: Mr. Suter entered a guilty plea; he was “remorseful far beyond what is reflected in the plea itself”; he had no criminal record and strong community support; and he had been employed virtually all of his adult life (para. 79). The sentencing judge also took into account — although “to a more limited extent” — the “extreme vitriol, public scorn and threats” Mr. Suter had endured, as well as the “violent vigilante actions against both Mr. Suter and Mrs. Suter” (para. 81). According to the sentencing judge, all of these factors operated to “significantly reduce the sentence from what would otherwise be fit” (para. 82). He concluded that a sentence of 4 months’ imprisonment, coupled with a 30-month driving prohibition, was appropriate in the circumstances. Both Mr. Suter and the Crown appealed from that sentence. B. The Court of Appeal Decision (Watson, Bielby and Schutz JJ.A.), 2016 ABCA 235, 41 Alta. L.R. (6th) 268 [ 19 ] The Court of Appeal of Alberta found that the sentencing judge made several errors in his decision, and that these errors in combination resulted in an unfit sentence. [ 20 ] First, the court found that the sentencing judge erred in principle in concluding that Mr. Suter was acting under a mistake of law when he refused to provide the police with a breath sample. In its view, the deficient legal advice did not constitute a mistake of law and it could not be used to mitigate Mr. Suter’s sentence. Second, the court found that the sentencing judge failed to consider as a relevant aggravating factor the fact that Mr. Suter “cho[se] to drive while distracted in the context of his health and pre-existing alcohol problems” (para. 100). Third, the court found that the sentencing judge erred by taking the vigilante violence into account when determining an appropriate sentence. Such violence, it maintained, did not “emanate from state misconduct” and therefore could not change “what would otherwise be a proportional sentence” (para. 106). [ 21 ] The Court of Appeal set aside the 4-month custodial sentence, and imposed a sentence of 26 months of imprisonment. It did not interfere with the 30-month driving prohibition. [ 22 ] Mr. Suter now appeals to this Court from that sentence. IV. Analysis [ 23 ] It is well established that appellate courts cannot interfere with sentencing decisions lightly: see R. v. Shropshire, 1995 47 (SCC), [1995] 4 S.C.R. 227, at para. 48; R. v. L.F.W., 2000 SCC 6, [2000] 1 S.C.R. 132, para. 25; R. v. L.M., 2008 SCC 31, [2008] 2 S.C.R. 163, at para. 14; R. v. Nasogaluak, 2010 SCC 6, [2010] 1 S.C.R. 206, at para. 46; R. v. Lacasse, 2015 SCC 64, [2015] 3 S.C.R. 1089, at para. 39. This is because trial judges have “broad discretion to impose the sentence they consider appropriate within the limits established by law” ( Lacasse, at para. 39 ). [ 24 ] In Lacasse, a majority of this Court held that an appellate court could only interfere with a sentence in one of two situations: (1) where the sentence imposed by the sentencing judge is “demonstrably unfit” (para. 41); or (2) where the sentencing judge commits an error in principle, fails to consider a relevant factor, or erroneously considers an aggravating or mitigating factor, and such an error has an impact on the sentence imposed (para. 44). In both situations, the appellate court may set aside the sentence and conduct its own analysis to determine a fit sentence in the circumstances. [ 25 ] A sentence that falls outside of a certain sentencing range is not necessarily unfit: see Lacasse, at para. 58; Nasogaluak, at para. 44. Sentencing ranges are merely guidelines, and are just “one tool among others that are intended to aid trial judges in their work” ( Lacasse, at para. 69 ). It follows that deviation from a sentencing range does not automatically justify appellate intervention ( ibid., at para. 67). [ 26 ] Both the sentencing judge and the Court of Appeal correctly held that the sentencing range for the s. 255(3.2) offence is the same as for impaired driving causing death. In my view, this range also includes the offence of driving “over 80” causing death (under s. 255(3.1) of the Criminal Code ). All three of these offences carry a maximum penalty of life imprisonment — an indication that Parliament intended that they be treated as equally serious. Moreover, they all have the same overarching objective: to deter drunk driving. [ 27 ] The sentencing range for these offences has been quite broad — low penitentiary sentences of 2 or 3 years to more substantial penitentiary sentences of 8 to 10 years — because courts have recognized that they cover a broad spectrum of offenders and circumstances: see R. v. Junkert, 2010 ONCA 549, 103 O.R. (3d) 284, at para. 40; R. v. Kummer, 2011 ONCA 39, 103 O.R. (3d) 641, at para. 21; Lacasse, at para. 66. An offender’s level of moral blameworthiness will vary significantly depending on the aggravating and mitigating factors in any given case. In unique cases, mitigating factors, collateral consequences, or other attenuating circumstances relating to the offence or offender may warrant a sentence that falls below this broad range. By the same token, the aggravating features in a particular case may warrant the imposition of a sentence that exceeds this broad range. As long as the sentence meets the sentencing principles and objectives codified in ss. 718 to 718.2 of the Criminal Code, and is proportionate to the gravity of the offence and the level of moral blameworthiness of the offender, it will be a fit sentence. [ 28 ] As I will explain, I am respectfully of the view that the sentencing judge and the Court of Appeal committed errors in principle, and that these errors led both courts to impose unfit sentences in the circumstances of this case. A. The Court of Appeal of Alberta’s Decision (1) The Court of Appeal Did Not Err in Raising New Issues [ 29 ] Before turning to the errors committed by the Court of Appeal in its sentencing analysis, I will briefly address Mr. Suter’s argument that the court erred in improperly raising new issues, contrary to this Court’s decision in R. v. Mian, 2014 SCC 54, [2014] 2 S.C.R. 689. [ 30 ] In accordance with Mian, an issue is new if it is “legally and factually distinct from the grounds of appeal raised by the parties” and “cannot reasonably be said to stem from the issues as framed by the parties” (paras. 30 and 35). It may only be raised if failing to do so would risk an injustice — for instance, if the court of appeal has “good reason to believe that the result would realistically have differed had the error not been made” (para. 45). An issue will be properly raised if the parties are given notice and an opportunity to respond (para. 54). Proper notice requires that the court of appeal “make the parties aware that it has discerned a potential issue and ensure that they are sufficiently informed so they may prepare and respond” ( ibid. ). An opportunity to respond includes filing written arguments, addressing the issue orally, or both (para. 59). [ 31 ] Mr. Suter claims that the Court of Appeal improperly raised two issues: (1) whether the vigilante violence he suffered should have been considered as a mitigating factor, and (2) whether his manner of driving should have been treated as an aggravating factor. [ 32 ] I begin with the first issue: the effect of the vigilante violence on Mr. Suter’s sentence. During oral arguments before the Court of Appeal, both parties addressed the weight that should be given to the vigilante violence at sentencing. It follows, in my view, that this was not a new issue as contemplated by Mian — it was not legally and factually distinct from the issues raised by the parties. Accordingly, it was open to the Court of Appeal to address it. [ 33 ] Turning to the second issue — the aggravating effect of Mr. Suter’s manner of driving — I accept that this was a new issue. Nonetheless, I am satisfied that it was properly raised. First, it was open to the Court of Appeal to conclude that the sentence imposed by the sentencing judge would have differed had he considered Mr. Suter’s manner of driving. Second, the Court of Appeal gave the parties adequate notice and provided them with an opportunity to respond. During the oral hearing, the court informed the parties that it had identified a potential issue and invited submissions. This Court in Mian explicitly rejects an approach that would require strict procedural standards to be followed, as such a formalistic approach would “fail to recognize that the issue may arise in different circumstances in different cases” (para. 55). [ 34 ] In my view, the Court of Appeal met the requirements in Mian. (2) The Court of Appeal Erred by Effectively Sentencing Mr. Suter for the Uncharged Offence of Careless Driving or Dangerous Driving Causing Death [ 35 ] As a general rule, courts cannot sentence an offender in respect of a crime for which he or she has not been convicted: see R. v. Angelillo, 2006 SCC 55, [2006] 2 S.C.R. 728, at paras. 23 and 32; R. v. Larche, 2006 SCC 56, [2006] 2 S.C.R. 762, at para. 1. [1] To do so would run counter to the presumption of innocence. With respect, the Court of Appeal did just that in this case. In arriving at a custodial sentence of 26 months, it effectively sentenced Mr. Suter for the uncharged offence of careless driving or dangerous driving causing death (under s. 115 of the Traffic Safety Act, R.S.A. 2000, c. T-6, and s. 249(4) of the Criminal Code, respectively). This error contributed to the imposition of a sentence that was unfit in the circumstances. [ 36 ] The decisions of the sentencing judge and the Court of Appeal paint two very different pictures of the events of May 19, 2013. The sentencing judge, while sensitive to the devastating consequences of Mr. Suter’s driving, found that it was an “accident caused by a non-impaired driving error” (para. 76). In his view, when Mr. Suter was momentarily distracted by a conversation with his wife in which she suggested that the two should perhaps get a divorce, he accidentally pressed on the gas pedal instead of the brake pedal. As a result, the vehicle accelerated onto the restaurant patio where the Mounsef family was seated. This driving error — which lasted only a few seconds — cost two-year-old Geo Mounsef his life. [ 37 ] The Court of Appeal engaged in its own interpretation of the evidence and concluded that what occurred in this case was more than just a momentary driving error. Although the court accepted that Mr. Suter was not impaired by alcohol, it nevertheless held that Mr. Suter’s ability to drive was “knowingly impaired by health and other factors” (para. 1). Specifically, the Court of Appeal held that Mr. Suter’s ability to drive was “impaired by the distraction offered by his argument with his wife, in the context of [his] health and drinking problems” (para. 92). [ 38 ] With respect, the concept of “impaired by distraction” is both novel and confusing, and I would not endorse it. In the present case, I see it primarily as a way of circumventing the sentencing judge’s finding that this accident was simply the result of a “non-impaired driving error” (para. 76). [ 39 ] In describing the circumstances of the accident, the Court of Appeal focused on the fact that Mr. Suter chose to drive in a busy parking lot, while angry and distracted, and in the context of pre-existing marital, health, and alcohol problems: Mindful that the sentencing judge found that Mr. Suter was not impaired at this time, his admissions demonstrate that he nonetheless operated his vehicle while seriously distracted. He drove after he had been drinking, in the context of being a man with a drinking problem whose health and well-being had been disintegrating in the weeks prior to the collision. His marriage was under stress. He chose to drive while angry and upset, and while having an argument with his wife after upsetting her with his conduct at Chili’s restaurant. Mr. Suter had experienced problems with alcohol in the relatively recent past. Those problems resulted in the police having to be called due to his hallucinations, with a resulting hospitalization. He had sustained a head injury two days prior to the accident. He had recently returned to drinking, and had been drinking (although not to the extent of impairment) at the time of the collision. He was angry. He was engaging in a serious, animated argument with his wife, over issues that had the potential to permanently impact his marriage. He was distracted by these events. He nonetheless continued to drive into a busy parking lot, adjacent to a number of outdoor diners. [paras. 89-90] [ 40 ] Many of the facts that the Court of Appeal relied on as aggravating are, in my view, irrelevant to the gravity of the s. 255(3.2) offence and the level of Mr. Suter’s moral blameworthiness with respect to that offence. These facts include that: (1) Mr. Suter had alcohol problems in the past; (2) he had experienced an episode of hallucinations two weeks prior to the incident; (3) he had hit his head after playing golf two days before the accident; and (4) his marriage had been under stress. [ 41 ] The sentencing judge found that Mr. Suter was not impaired by alcohol at the time of the accident. He accepted Mr. and Mrs. Suter’s evidence about the amount of alcohol Mr. Suter had consumed that day. According to an expert report that was admitted on consent, with this amount of alcohol, Mr. Suter would not have blown “over 80” had he provided the police with a breath sample. In light of this, it escapes me how alcohol consumption patterns months prior to the accident could have any bearing on the seriousness of the s. 255(3.2) offence, or on Mr. Suter’s level of moral culpability for that offence. [ 42 ] Similarly, absent any evidence that the hallucinations or the head injury would have affected, or did affect, Mr. Suter’s ability to drive, I fail to see how these facts are relevant to the sentencing analysis. The same holds true of the fact that Mr. Suter’s marriage had been under stress for some period of time preceding the accident. [ 43 ] The Court of Appeal also portrayed Mr. Suter as being distracted, angry, engaged in a “serious, animated argument”, while driving in a “busy parking lot” at the time of the accident (para. 90). At para. 100 of its reasons, the court observed that Mr. Suter caused a child’s death “not intentionally, but carelessly and even recklessly ” (emphasis added). Characterizing the circumstances of the accident in this way, the court gave undue weight to facts that may — and I stress the word “may” — have formed the basis of a separate, uncharged offence of careless driving or perhaps dangerous driving causing death. And it was largely on this basis that the Court of Appeal substantially increased Mr. Suter’s sentence. [ 44 ] In my view, the Court of Appeal improperly recast the accident as one caused by health and alcohol problems, anger, and distraction. It reweighed the evidence and looked to external factors that had no bearing on the gravity of the offence for which Mr. Suter was charged, nor on Mr. Suter’s level of moral blameworthiness. In doing so, the court effectively punished Mr. Suter for a careless driving or dangerous driving causing death offence for which he was neither tried nor convicted. This was an error in principle that, as I will explain, resulted in the imposition of an unfit sentence. (3) The Court of Appeal Erred in Finding That Vigilante Violence Cannot Be Considered at Sentencing [ 45 ] The sentencing judge found, correctly in my view, that the vigilante violence experienced by Mr. Suter could be considered — to a limited extent — when crafting an appropriate sentence. With respect, the Court of Appeal erred in concluding otherwise. This error also contributed to the 26-month custodial sentence it imposed. [ 46 ] As I have observed, sentencing is a highly individualized process: see Lacasse, at para. 54; R. v. Proulx, 2000 SCC 5, [2000] 1 S.C.R. 61, at para. 82; Nasogaluak, at para. 43. In R. v. Ipeelee, 2012 SCC 13, [2012] 1 S.C.R. 433, this Court stated that a sentencing judge must have “sufficient manoeuvrability to tailor sentences to the circumstances of the particular offence and the particular offender” (para. 38). Tailoring sentences to the circumstances of the offence and the offender may require the sentencing judge to look at collateral consequences. Examining collateral consequences enables a sentencing judge to craft a proportionate sentence in a given case by taking into account all the relevant circumstances related to the offence and the offender. [ 47 ] There is no rigid formula for taking collateral consequences into account. They may flow from the length of sentence, or from the conviction itself: see R. v. Pham, 2013 SCC 15, [2013] 1 S.C.R. 739, at para. 11; R. v. Bunn (1997), 1997 22728 (MB CA), 118 Man. R. (2d) 300 (C.A.), at para. 23; R. v. Bunn, 2000 SCC 9, [2000] 1 S.C.R. 183 (“ Bunn (SCC)”), at para. 23; Tran v. Canada (Public Safety and Emergency Preparedness), 2017 SCC 50, [2017] 2 S.C.R. 289. In his text The Law of Sentencing (2001), Professor Allan Manson notes that they may also flow from the very act of committing the offence: As a result of the commission of an offence, the offender may suffer physical, emotional, social, or financial consequences. While not punishment in the true sense of pains or burdens imposed by the state after a finding of guilt, they are often considered in mitigation. [Emphasis added; p. 136.] I agree with Professor Manson’s observation, much as it constitutes an incremental extension of this Court’s characterization of collateral consequences in Pham. In my view, a collateral consequence includes any consequence arising from the commission of an offence, the conviction for an offence, or the sentence imposed for an offence, that impacts the offender. [ 48 ] Though collateral consequences are not necessarily “aggravating” or “mitigating” factors under s. 718.2 (a) of the Criminal Code — as they do not relate to the gravity of the offence or the level of responsibility of the offender — they nevertheless speak to the “personal circumstances of the offender” ( Pham, at para. 11 ). The relevance of collateral consequences stems, in part, from the application of the sentencing principles of individualization and parity: ibid.; s. 718.2(b) of the Criminal Code. [2] The question is not whether collateral consequences diminish the offender’s moral blameworthiness or render the offence itself less serious, but whether the effect of those consequences means that a particular sentence would have a more significant impact on the offender because of his or her circumstances. Like offenders should be treated alike, and collateral consequences may mean that an offender is no longer “like” the others, rendering a given sentence unfit. [3] [ 49 ] Collateral consequences do not need to be foreseeable, nor must they flow naturally from the conviction, sentence, or commission of the offence. In fact, “[w]here the consequence is so directly linked to the nature of an offence as to be almost inevitable, its role as a mitigating factor is greatly diminished” (Manson, at p. 137). Nevertheless, in order to be considered at sentencing, collateral consequences must relate to the offence and the circumstances of the offender. [ 50 ] Professor Manson writes: “When an offender suffers physical injury as a result of an offence, this may be relevant for sentencing purposes especially if there will be long-lasting effects” (p. 136). Though Professor Manson had in mind a scenario where an offender is injured while committing a driving offence, in my view, his comments apply to any offender who suffers injury as a result of an offence. In fact, the attenuating effect of an injury on the sentence imposed will likely be lessened where the injury is so directly linked to the offence as to be almost inevitable (see para. 49). For instance, an injury resulting from an impaired driving offence (a foreseeable consequence of driving while impaired) may have less of an attenuating impact on the sentence imposed than if that same injury resulted from an unforeseeable event arising out of the offence. [ 51 ] Our courts have held that where an offender is attacked by fellow inmates in a prison and the attack is related to the offence for which the offender is in custody, such violence may be considered as a factor at sentencing: see R. v. MacFarlane, 2012 ONCA 82, 288 O.A.C. 114, at para. 3; R. v. Folino, 2005 40543 (ON CA), 2005 ONCA 258, 77 O.R. (3d) 641, at para. 29; R. v. Anderson, 2014 ONSC 3646, at paras. 14 and 18. Although being assaulted by a fellow inmate is not the same thing as being abducted and attacked by vigilantes, the rationale for taking these collateral consequences into account when sentencing an offender remains. In both scenarios, attacks relating to the commission of the offence form part of the personal circumstances of the offender. To ensure that the principles of individualization and parity are respected, these attacks are considered at sentencing. [ 52 ] Australian jurisprudence has recognized violent retribution by members of the public as a relevant collateral consequence for sentencing. In R. v. Mamarika, 1982 ABCA 281, [1982] FCA 94, 42 A.L.R. 94, the Federal Court of Australia accepted that the violence inflicted upon Mr. Mamarika by members of his community as a result of his role in killing the deceased could be taken into account at sentencing. Specifically, it noted:... by reason of his action, the appellant brought on himself the anger of members of the community and..., as a result, he received severe injuries from which he fortunately made a good recovery. So seen, it is a matter properly to be taken into account in determining an appropriate sentence, without giving any sanction to what occurred. [p. 97] [ 53 ] I agree with this approach. As indicated, violent actions against an offender for his or her role in the commission of an offence — whether by a fellow inmate, or by a vigilante group — necessarily form part of the personal circumstances of that offender, and should therefore be taken into account when determining an appropriate sentence. [ 54 ] In this case, the vigilante violence flowed from the public’s perception of the events of May 19, 2013, and the tragic consequences of Mr. Suter’s actions. Although this violence did not flow directly from the commission of the s. 255(3.2) offence (nor did it flow from the length of the sentence or the conviction itself), it is nevertheless a collateral consequence as it is inextricably linked to the circumstances of the offence. [ 55 ] The Court of Appeal acknowledged that Mr. Suter was “brutally attacked, beaten violently, and had a digit savagely removed with shears in a form of vigilante violence” (para. 106). This attack was retaliation for Mr. Suter’s role in Geo Mounsef’s death. The court held, however, that because these facts did not “emanate from state misconduct”, they should not change “what would otherwise be a proportional sentence” ( ibid. ). [ 56 ] I agree with the Court of Appeal that the fundamental principle of proportionality must prevail in every case — collateral consequences cannot be used to reduce a sentence to a point where the sentence becomes disproportionate to the gravity of the offence or the moral blameworthiness of the offender. There is, however, no requirement that collateral consequences emanate from state misconduct in order to be considered as a factor at sentencing: see Bunn (SCC), at para. 23; R. v. McDonald, 2016 NUCA 4, at paras. 41-44; R. v. Stanberry, 2015 QCCQ 1097, 18 C.R. (7th) 87, at paras. 18-20; R. v. Bell, 2013 MBQB 80, 290 Man. R. (2d) 79, at para. 87; Folino, at para. 29; R. v. Heatherington, 2005 ABCA 393, 380 A.R. 395, at paras. 5-6; R. v. Owens (2002), 2002 41948 (ON CA), 161 O.A.C. 229, at paras. 10-11; R. v. Abouabdellah (1996), 1996 6502 (QC CA), 109 C.C.C. (3d) 477 (Que. C.A.), at p. 480; R. v. Carroll (1995), 1995 1123 (BC CA), 56 B.C.A.C. 138, at paras. 11-12. [ 57 ] As such, the violence suffered by Mr. Suter at the hands of non-state vigilante actors can be considered when determining an appropriate sentence. The violent attack was related to Mr. Suter’s role in Geo Mounsef’s death, and both the permanent physical injury and psychological trauma resulting from this attack necessarily form part of Mr. Suter’s personal circumstances. In light of the sentencing principles of individualization and parity, the vigilante attack against Mr. Suter was a relevant collateral consequence to consider at sentencing. [ 58 ] That said, this particular collateral consequence should only be considered to a limited extent. Giving too much weight to vigilante violence at sentencing allows this kind of criminal conduct to gain undue legitimacy in the judicial process. This should be avoided. Vigilantism undermines the rule of law and interferes with the administration of justice. It takes justice out of the hands of the police and the courts, and puts it into the hands of criminals. As a general rule, those who engage in it should expect to be treated severely. [ 59 ] In sum, the sentencing judge was entitled to consider, to a limited extent, the vigilante violence suffered by Mr. Suter for his role in Geo Mounsef’s death. As such, the Court of Appeal erred when it refused to give any effect to it. (4) Conclusion [ 60 ] The errors committed by the Court of Appeal — sentencing Mr. Suter for an uncharged offence of careless driving or dangerous driving causing death, and failing to consider the vigilante violence suffered by Mr. Suter — contributed to the 26-month custodial sentence and, in my respectful view, rendered it unfit. [ 61 ] I turn now to the errors committed by the sentencing judge. B. The Sentencing Judge’s Decision (1) The Sentencing Judge Erred in Finding a Mistake of Law in This Case [ 62 ] The sentencing judge erred in concluding that Mr. Suter was acting under a mistake of law when he refused to provide the police with a sample of his breath. The sentencing judge made no express finding as to whether Mr. Suter honestly but mistakenly believed that his refusal was lawful — an essential element of mistake of law. Without such a finding, it was not open to the sentencing judge to conclude that there was a mistake of law in this case. With respect, this error had a material impact on the sentencing judge’s assessment of Mr. Suter’s moral culpability, and it contributed in no small measure to the manifestly inadequate sentence of four months’ imprisonment imposed by him. [ 63 ] A person’s knowledge of the illegality of his or her actions may be a relevant factor to consider at sentencing. However, as with all factors considered at sentencing, the significance (and potential mitigating effect) of this knowledge will vary depending on the circumstances. [ 64 ] A mistake of law is a legal concept with rigorous requirements. In my view, it occurs only where a person has an honest but mistaken belief in the legality of his or her actions. Although it is not a defence to a criminal charge ( s. 19 of the Criminal Code; R. v. Forster, 1992 118 (SCC), [1992] 1 S.C.R. 339, at p. 346), mistake of law can nevertheless be used as a mitigating factor in sentencing (see R. v. Pontes, 1995 61 (SCC), [1995] 3 S.C.R. 44, at para. 87; R. v. MacDonald, 2014 SCC 3, [2014] 1 S.C.R. 37, at para. 61; Kenny’s Outlines of Criminal Law (19th ed. 1966), by J. W. Cecil Turner, at pp. 61-62). This is because offenders who honestly but mistakenly believe in the lawfulness of their actions are less morally blameworthy than offenders who — in committing the same offence — are unsure about the lawfulness of their actions, or know that their actions are unlawful. [ 65 ] Confusion or uncertainty as to the lawfulness of one’s actions does not, in my view, meet the legal requirements for mistake of law. However, such confusion may still be relevant to the sentencing analysis depending on the facts of the particular case. Its mitigating effect, if any, will necessarily be less than in a situation where there is a true mistake of law. [ 66 ] In this case, the sentencing judge made the following findings of fact: (1) Mr. Suter was initially confused by the lawyer’s legal advice; (2) in the end, the lawyer expressly told Mr. Suter not to provide the police with a breath sample; and (3) Mr. Suter refused to provide the police with a breath sample because of the lawyer’s ill-informed legal advice. [ 67 ] However — and this is critical — the sentencing judge made no express finding as to whether or not Mr. Suter honestly but mistakenly believed that, in refusing to provide the police with a breath sample, he was not committing a criminal offence. At para. 76 of his reasons, the sentencing judge stated: In this case, however, the Court has accepted the testimony of Mr. Suter as to what the lawyer said, and finds that the refusal was based on the lawyer expressly telling him not to provide a sample. This does not absolve Mr. Suter, as a mistake of law is not a defence but it fundamentally changes Mr. Suter’s moral culpability. In other words, the sentencing judge accepted Mr. Suter’s testimony as to what the Brydges lawyer said (do not provide a breath sample). The sentencing judge also accepted that Mr. Suter’s refusal was based on the Brydges lawyer’s express instruction. The sentencing judge then concluded that these findings constitute a mistake of law and therefore fundamentally changed Mr. Suter’s moral culpability. [ 68 ] With respect, this is where the sentencing judge erred. These two findings do not, on their own, meet the requirements for mistake of law. As indicated, an additional finding that Mr. Suter honestly but mistakenly believed that, in refusing to provide a breath sample, he was not committing a criminal offence was required. In this regard it must be remembered that an accused who raises mistake of law as a mitigating factor on sentence has the onus of establishing, on a balance of probability, that the requisite elements of a mistake of law have been made out. [ 69 ] Based on the record before us, this Court is not in a position to infer that the sentencing judge made an implicit finding that Mr. Suter had an honest but mistaken belief in the lawfulness of his refusal. Though Mr. Suter testified that he thought it was “within [his] rights not to blow”, and did not think that the lawyer would be telling him to do something that was illegal, this evidence must be assessed in light of the other evidence adduced at the sentencing hearing, and equally important, evidence that was not adduced. [ 70 ] Nothing in the record suggests that the lawyer told Mr. Suter it was not an offence to refuse to provide a breath sample to police, nor did the lawyer inform Mr. Suter that he was legally justified in refusing because, for example: (1) the police lacked reasonable grounds to make the breath demand; or (2) Mr. Suter had a medical impediment that would justify his refusal. To be clear, this is not to suggest that a mistake of law can only be met when an accused not only believes his or her conduct is lawful, but also knows the legal basis for its legality. Rather, my point here is that, if there was evidence that such an explanation had been provided by the lawyer, then we could more safely infer that Mr. Suter honestly believed in the lawfulness of his conduct, despite the sentencing judge’s failure to make an express finding in this regard. Indeed, the lawyer testified — and it was not disputed — that his instruction to Mr. Suter was a strategic choice based on a misunderstanding of the current state of the law. Accordingly, there is an alternate available inference that the lawyer’s instruction to refuse to provide a breath sample was articulated in terms of sentencing strategy, which presumes an awareness of the illegality of refusing to provide a sample. [ 71 ] Furthermore, Mr. Suter admitted that the police officer told him that it was an offence to refuse to provide a breath sample. Of course, it is understandable, particularly in the context of an arrest, for a detainee to accept the advice of his or her lawyer, who is aligned in interest, over that of a police officer. I only mention this admission by Mr. Suter to note that he was likely not blindsided by the possibility that refusing to provide a breath sample may be a criminal offence. Regardless, mistake of law turns on the accused’s subjective belief in the legality of his or her conduct. While a police officer’s instruction that certain conduct amounts to an offence is not dispositive of that subjective belief, it may be relevant to the sincerity of the belief. [ 72 ] In light of all this, the only conclusion that this Court can safely infer from the record is that Mr. Suter was confused by the lawyer’s advice — a finding that is consistent with the sentencing judge’s description of Mr. Suter’s call with the Brydges lawyer. [ 73 ] But mere confusion as to the lawfulness of one’s actions is insufficient to ground a mistake of law. Though Mr. Suter’s situation was unfortunate, it does not justify departing from the rigorous requirements of mistake of law. [ 74 ] That said, while Mr. Suter’s state of confusion did not give rise to a mistake of law, this does not end the matter. When considering what mitigating weight — if any — Mr. Suter’s uncertainty as to the lawfulness of his refusal will have, it cannot be ignored that this uncertainty stemmed from ill-informed legal advice. As explained in Clayton C. Ruby, Gerald J. Chan and Nader R. Hasan’s Sentencing (8th ed. 2012): The pointlessness of punishing individuals who have no idea they were breaking the law is quickly apparent.... Still less does it make sense to punish someone who had sought legal advice, or set upon a legal position without that assistance. [Emphasis added; footnotes omitted; §5.319.] As indicated, Mr. Suter sought legal advice, was given incorrect legal advice, and refused to cooperate with police based on it. These facts go to attenuate Mr. Suter’s level of moral blameworthiness. [4] [ 75 ] In saying this, I wish to be clear that Mr. Suter’s case is unique, and the mitigation afforded to him should not be misconstrued as suggesting that imperfect legal advice presumptively mitigates a sentence. Here, the Brydges lawyer testified that he did not know the law on the s. 255(3.2) offence, and that he deliberately tried to steer Mr. Suter away from providing the police with a breath sample. The sentencing judge accepted Mr. Suter’s evidence that the lawyer expressly told him not to provide the police with a breath sample — i.e., the lawyer advised Mr. Suter to break the law. Moreover, the sentencing judge found that the reason Mr. Suter did not comply with the breath demand was because he was following his lawyer’s erroneous advice; there was no finding that the refusal was a strategic choice on Mr. Suter’s part. In the circumstances, these unique facts contributed to Mr. Suter’s uncertainty regarding the lawfulness of his refusal, and therefore served to attenuate his moral blameworthiness. [ 76 ] In sum, the sentencing judge erred when he found that the erroneous legal advice given to Mr. Suter and upon which he acted amounted to a mistake of law, and therefore fundamentally changed Mr. Suter’s moral culpability. This error contributed in no small measure to the manifestly inadequate sentence of four months’ imprisonment imposed by the sentencing judge. (2) The Sentencing Judge Erred by Giving Undue Weight to Non-Impairment as a Mitigating Factor [ 77 ] Although a finding of non-impairment is a relevant mitigating factor when sentencing an offender for a refusal offence, its mitigating effect must be limited. In my view, the sentencing judge erred by giving undue weight to Mr. Suter’s non-impairment as a mitigating factor — an error that also contributed to the four-month custodial sentence. [ 78 ] The sentencing judge noted the following: It would be contrary to the fundamental principle of proportionality to equate the moral blameworthiness of a sober person who follows bad legal advice in refusing to comply with a police demand, on the one hand, with the moral culpability of the person who gets drunk and kills someone, on the other. [para. 66] I agree. In fact, I would go further — the moral blameworthiness of a sober person who fails to provide the police with a breath sample clearly differs from that of a drunk person who fails to provide the police with a breath sample. The question then becomes: To what extent? [ 79 ] Answering this question is, of course, a fact-driven exercise that depends on the specific circumstances in any given case. That said, in my view, there are strong policy reasons for limiting the mitigating effect of non-impairment at a sentencing hearing for a refusal offence under the Criminal Code, be it refusing to provide a breath sample simpliciter under s. 254(5); refusing to provide a breath sample after having caused an accident resulting in bodily harm under s. 255(2.2); or refusing to provide a breath sample after having caused an accident resulting in a death under s. 255(3.2). [ 80 ] First, overemphasizing the mitigating effect of non-impairment risks transforming sentencing hearings for refusal offences into de facto impaired driving trials. This would add to the complexity and length of these proceedings, and deplete scarce judicial resources. [ 81 ] Second, while refusal offences are certainly aimed at deterring drunk driving, s. 255(3.2) of the Criminal Code, like ss. 254(5) and 255(2.2), is also in its essence an evidence gathering tool. By refusing to provide a breath sample, a person is depriving the police, the court, the public at large, and the family of the deceased of the most reliable evidence of impairment — or lack thereof. The act of refusing is the gravamen of the offence. Thus, the seriousness of the offence and the moral blameworthiness of the offender stem primarily from the refusal itself, and not from the offender’s level of impairment. [ 82 ] Third, there is a real risk that relying too heavily on non-impairment as a mitigating factor at sentencing would create an incentive for individuals not to provide the police with a breath sample. Parliament sought to eliminate the incentive for refusal through its 2008 amendments to the Criminal Code: see An Act to amend the Criminal Code and to make consequential amendments to other Acts, S.C. 2008, c. 6, s. 21(3), adding s. 255(2.2) and (3.2). Before 2008, there was no aggravated offence for refusing to provide a breath sample where bodily harm or death occurred — there was only the offence of refusing to provide a breath sample simpliciter, which carried with it a maximum penalty of five years imprisonment. There were, however, aggravated offences for impaired driving and driving “over 80” where bodily harm or death occurred, which carried with them maximum penalties of 10 years and life imprisonment, respectively. [ 83 ] In consequence, the pre-2008 regime created an incentive to refuse to provide a breath sample where bodily harm or death occurred because of the significantly more lenient penalty that attached to such an offence compared to the penalty that attached to the aggravated offences for impaired driving and driving “over 80”. In 2008, Parliament amended the Criminal Code, and created two new offences: refusing to provide a breath sample where bodily harm occurs ( s. 255(2.2) ) and refusing to provide a breath sample where death occurs ( s. 255(3.2) ). Because these new offences carried the same maximum penalties as the aggravated offences for impaired driving and driving “over 80”, Parliament eliminated any incentive to refuse to provide the police with a breath sample. [ 84 ] The mitigating weight given to non-impairment at sentencing cannot operate in a way that undermines Parliament’s intention in this regard. If non-impairment is treated as a significant mitigating factor, this could motivate persons to refuse to provide the police with a breath sample, be convicted of the refusal offence, and then subsequently argue at the sentencing hearing that they were not impaired. If successful, offenders would benefit from a significantly reduced sentence than what they would have received if convicted of impaired driving or driving “over 80”. [ 85 ] To avoid the concerns outlined above, the mitigating effect of non-impairment on the offender’s sentence should be limited. In addition, the onus must be on the offender to establish on balance that he or she was not impaired at the time the offence was committed. This will help protect against a deluge of impaired driving trials at the sentencing stage of refusal offences, and will ensure that Parliament’s intentions and objectives are respected. [ 86 ] In this case, the sentencing judge placed considerable weight on the fact that Mr. Suter was not impaired at the time of the accident. He stated that “[i]f the Court were imposing a sentence for impaired driving causing death in these circumstances, the Court would view the sentence suggested by the Crown [three years] as too low” (para. 75). In other words, had Mr. Suter been convicted of impaired driving causing death, the sentencing judge would have imposed a sentence of more than three years’ imprisonment. But the sentencing judge found that Mr. Suter was not impaired when he caused Geo Mounsef’s death. As indicated, he concluded that what occurred in this case was an “accident caused by a non-impaired driving error” (para. 76 (emphasis added)). The evidence of non-impairment — in combination with other mitigating factors, including the bad legal advice Mr. Suter received — “moved the Court from its starting position” ( ibid. ). [ 87 ] Reducing a sentence from over three years’ imprisonment to four months’ imprisonment is a substantial drop. This drastic shift indicates that Mr. Suter’s non-impairment must have contributed in no small measure to the manifestly inadequate sentence of four months’ imprisonment imposed by the sentencing judge. (3) Conclusion [ 88 ] The errors committed by the sentencing judge — mischaracterizing what occurred in this case as a mistake of law and giving undue weight to Mr. Suter’s non-impairment as a mitigating factor — contributed to his decision to impose a four-month custodial sentence. This sentence, even in light of the unique mitigating factors and collateral consequences in this case, is manifestly inadequate for the s. 255(3.2) offence. Accordingly, appellate intervention is warranted. [ 89 ] Having concluded that both the Court of Appeal and the sentencing judge committed errors in principle that impacted the sentences imposed, I turn now to what would have been an appropriate sentence for Mr. Suter. C. An Appropriate Sentence for Mr. Suter [ 90 ] As indicated, the sentencing range for the s. 255(3.2) offence is the same as for impaired driving causing death and driving “over 80” causing death — low penitentiary sentences of 2 or 3 years to more substantial penitentiary sentences of 8 to 10 years, depending on the circumstances. In unique cases, mitigating factors, collateral consequences, or other attenuating circumstances relating to the offence or offender may warrant the imposition of a sentence that falls below this broad range. By the same token, the aggravating features in a particular case may warrant the imposition of a sentence that exceeds this broad range. As long as the objectives and principles of sentencing codified in ss. 718 to 718.2 of the Criminal Code are met and respected, the sentence will be fit. [ 91 ] I pause to note that there is very little sentencing jurisprudence on s. 255(3.2). My colleague, Justice Gascon, correctly points out that, in the handful of decisions which have been reported, “there are no ‘similar offenders... in similar circumstances’ against which [Mr. Suter’s] sentence can be reasonably measured” (para. 191). In light of this, the s. 255(3.2) jurisprudence offers little guidance. [ 92 ] Mr. Suter’s case is unique. While the consequences of his actions are undoubtedly tragic, and the gravity of the refusal offence is significant, there are several factors in this case that, in combination, operate to reduce Mr. Suter’s sentence. The fact that Mr. Suter was not impaired at the time of the accident; that he refused to provide the police with a breath sample because of ill-informed and incorrect legal advice; and that he was attacked by vigilantes and had his thumb cut off with pruning shears, are all factors that must be taken into account in crafting an appropriate sentence. [ 93 ] But for these attenuating circumstances, I am of the view that a sentence of three to five years in the penitentiary would not have been out of line. Unlawfully refusing to provide the police with a breath sample after having caused an accident resulting in a death is an extremely serious offence. Parliament has made this clear. It carries with it a maximum punishment of life imprisonment — and with good cause. When a person refuses to provide a breath sample in response to a lawful request, this deprives the police, the court, the public at large, and the family of the deceased of the best evidence as to the driver’s blood alcohol level and state of impairment. Moreover, it places a barrier in the way of the ongoing efforts and pressing objective of deterring, denouncing, and putting an end to the scourge of impaired driving. [ 94 ] That said, in my view, the various attenuating factors in this case operate to remove Mr. Suter’s sentence from the normal range for a s. 255(3.2) offence (see para. 90 above), and render the 26-month sentence imposed by the Court of Appeal unfit. By the same token, these factors do not justify the four-month sentence imposed by the sentencing judge. With respect, that sentence does not properly account for the gravity of the offence, rendering it manifestly unfit. [ 95 ] In the end, I am of the view that a sentence of 15 to 18 months’ imprisonment would have been a fit sentence at the time of sentencing. Such a sentence takes into account the several attenuating factors that make this case unique, while not losing sight of the gravity of the s. 255(3.2) offence. [ 96 ] In concluding, as I have, that a sentence of 15 to 18 months would have been a fit sentence in Mr. Suter’s case, I am mindful of the reasons of my colleague who would sustain as fit the 4-month custodial sentence imposed by the sentencing judge. While I respectfully disagree with much of my colleague’s reasoning and his ultimate conclusion, I propose to limit my remarks to two matters. [ 97 ] First, it is important to recognize that the sentencing judge, of his own accord, stated that but for the various mitigating factors available to Mr. Suter, including the improper legal advice he received and his non-impairment, he would have imposed a sentence greater than the 36-month sentence the Crown was seeking. I take from this that he would have imposed a sentence in the range of 42 to 48 months. However, in light of the various mitigating factors, he chose to go well below that range, arriving at a custodial sentence of four months. In other words, he took 38 to 44 months off the sentence he otherwise would have imposed. Respectfully, I am of the view that in doing so, he arrived at a sentence that was demonstrably unfit. [ 98 ] In so concluding, I take no issue with the sentencing judge that the mitigating factors in this case warranted a substantial reduction from the sentencing range he otherwise would have sanctioned: 42 to 48 months. I do however take issue with the size of the reduction. In my view, it unduly minimized the gravity and seriousness of Mr. Suter’s crime. [ 99 ] The reduction which I consider to be appropriate — 42 to 48 months down to 15 to 18 months — is a very substantial reduction. It recognizes the mitigating factors available to Mr. Suter, without losing sight of Parliament’s clear and unmistakeable message that refusing to provide a breath sample in circumstances where an accident resulting in death occurs is a very serious crime punishable by up to life imprisonment. [ 100 ] My colleague, on the other hand, sees no reason to interfere with the four-month sentence imposed at the sentencing hearing, even though he himself would have “weighed the gravity of the offence more heavily than the sentencing judge” (para. 170). I take from this that my colleague would have imposed a sentence of more than four months’ imprisonment. This brings me to the second matter I wish to comment on. [ 101 ] In my view, my colleague’s sentencing analysis is flawed. While he acknowledges that the “gravity of the offence — refusing to provide a breath sample after a fatal car accident — is very high” (para. 169), he nevertheless maintains that Mr. Suter’s degree of moral blameworthiness “could hardly be lower” ( ibid. ); that the principles of general deterrence and denunciation have virtually no role to play in this case (paras. 178-84); and that, because Mr. Suter “would have faced no criminal charge if he had received adequate legal advice and provided a breath sample”, his crime was simply “following his lawyer’s negligent advice” (para. 108). [ 102 ] With respect, these findings all but mitigate the crime out of existence. Indeed, my colleague goes so far as to state that Mr. Suter’s only fault was “following his lawyer’s negligent legal advice” — clearly not a recognized crime in the Criminal Code (para. 108). And yet he would have imposed a sentence of more than four months’ imprisonment. His logic, with respect, escapes me. On his analysis, even the four months imposed by the sentencing judge would appear to be untoward. [ 103 ] Having determined that a fit sentence at the time of sentencing would have been one of 15 to 18 months, there are additional factors at this stage that warrant consideration. Mr. Suter has already served just over 10 and a half months of his custodial sentence. He has spent almost nine months awaiting this Court’s decision. In my view, to now impose on Mr. Suter what would have been a fit disposition at the time he was sentenced would cause him undue hardship, and serve no useful purpose. In short, it would not be in the interests of justice to reincarcerate Mr. Suter at this time. V. Conclusion [ 104 ] Accordingly, I would allow Mr. Suter’s appeal in part. I would set aside the sentence of 26 months’ imprisonment imposed by the Court of Appeal and replace it with one of time served. I would uphold the 30-month driving prohibition. The following are the reasons delivered by Gascon J. — I. Overview [ 105 ] I have reviewed the majority’s reasons and agree with their criticism of the Court of Appeal’s decision ( 2016 ABCA 235, 41 Alta. L.R. (6th) 268). Specifically, I agree that the Court of Appeal, by reevaluating the trial record, “effectively punished Mr. Suter for a careless driving or dangerous driving causing death offence for which he was neither tried nor convicted” (para. 44). I also agree that, as sentencing is an “individualized process” (para. 46) not amenable to “rigid formula[s]” (para. 47), the Court of Appeal erred when it held that non-state collateral consequences — like vigilante violence — cannot inform the appropriateness of a sentence. Lastly, I agree that, given the exceptional mitigating factors present in this case, the Court of Appeal’s sentence of 26 months in prison was unfit. However, that is where my agreement with the majority ends. [ 106 ] This case is undeniably tragic. Indeed, few tragedies are worse than the death of a child. But the tragic circumstances of this case should not displace the proper framework that this Court, and Parliament, both require sentencing judges to apply in imposing an appropriate sentence. In my respectful view, the majority have abandoned that framework by prescribing a sentence of 15 to 18 months in prison with an incomplete analysis of proportionality, individualization, parity and the sentencing objectives enumerated in the Criminal Code, R.S.C. 1985, c. C-46. [ 107 ] To be clear, my position is not based in leniency for administrative offences. Administrative offences — especially those pertaining to impaired driving — must be strictly enforced. They are critical to the orderly administration of justice. And in the context of impaired driving, they are central to denouncing and deterring that scourge. But this strict enforcement cannot overlook the individual nature of sentencing or disregard circumstances in which unique facts not only warrant, but require, a reduced sentence. [ 108 ] This case presents such unique facts. Mr. Suter was convicted of refusing to provide a breath sample following a fatal accident, an offence directed at two forms of mischief: (1) directly, obstruction of justice; and (2) indirectly, impaired driving. But two key findings made by the sentencing judge show that Mr. Suter’s conduct engaged neither form of mischief. First, Mr. Suter was expressly instructed to refuse to provide the sample by his state-provided lawyer. As a result, he was not seeking to obstruct justice, but to comply with his lawyer’s advice. Second, he was sober. The sentencing judge did not simply find that the Crown failed to prove Mr. Suter’s impairment beyond a reasonable doubt (a potential consequence of refusal (see e.g. R. v. Kresko, 2013 ONSC 1631, 42 M.V.R. (6th) 224, at para. 42 )). Rather, the sentencing judge found that he was sober, despite the fact that for impaired driving offences the Criminal Code specifically permits an adverse inference to be drawn against accused persons who refused to provide a breath sample (s. 258(3)). Indeed, the only third-party witnesses who assessed Mr. Suter’s impairment before the accident — and thus without the confounding variables of the shock and mob beating that followed it — confirmed his sobriety based on their “considerable experience” in the service industry (A.R., vol. II, at pp. 129-31). Further, an uncontradicted expert report stated that Mr. Suter would not have even blown “over 80” based on the evidence accepted by the sentencing judge. Together, these facts found by the sentencing judge, which neither party contests, demonstrate the apparent injustice in this case: Mr. Suter would have faced no criminal charge if he had received adequate legal advice and provided a breath sample. Simply put, his crime was following his lawyer’s negligent advice. [ 109 ] Based on the parties’ submissions and my review of the relevant case law, the two key findings in this case — bad legal advice and sobriety — have never arisen together in the history of Canadian reported legal decisions regarding refusal offences. Indeed, the sentencing judge described these facts as “unique” ( 2015 ABPC 269, 94 M.V.R. (6th) 91, at para. 48 ). And on top of these two findings, the sentencing judge found that Mr. Suter and his wife suffered multiple violent vigilante attacks in response to this tragedy. Vigilantes kidnapped and mutilated Mr. Suter and left him unconscious in the snow. They also assaulted Mrs. Suter, breaking her nose and teeth. Any one of these three findings is exceptional. That they would occur together is beyond imagination. These are the extraordinary facts that led the sentencing judge to impose a sentence of four months in prison. While I would have weighed the gravity of the offence differently than the sentencing judge, I find that there is no legal basis for this Court to interfere with the initial sentence imposed. [ 110 ] In this regard, I disagree with the majority on four points. First, I dispute that a mistake of law requires an offender to be certain as to the lawfulness of their conduct. I cannot find any authority that supports this position. Further, this narrow construction is, in my view, antithetical to the contextual and individualized nature of sentencing. Second, I disagree that the sentencing judge gave excessive weight to Mr. Suter’s sobriety in this case. In my view, the majority identify this error by misreading the sentencing decision. Third, I question the majority’s approach, which, in effect, stays Mr. Suter’s sentence based solely on the time spent awaiting a decision on his appeal, despite the pressing objectives of deterrence and denunciation they purport to rely on when describing a fit sentence in these circumstances. In my respectful view, this can be motivated only by lingering doubt about the proportionality of a 15- to 18-month sentence in Mr. Suter’s particular circumstances, not by concerns about the burden of reincarceration. Fourth, I cannot conclude, on a principled basis, that a sentence of four months in these circumstances is demonstrably unfit. Neither the unique facts before us, nor the relevant jurisprudence, support such a characterization. [ 111 ] For these reasons, I would have also allowed the appeal in part, but would have restored the initial sentence of four months. II. Background [ 112 ] I do not take issue with the majority’s summary of the relevant facts or decisions below in this case. [ 113 ] On the facts, the majority correctly note the many mitigating factors in this case: (1) the fatal accident was not caused by impairment, and the Crown did not allege or seek to prove that it was caused by distraction; (2) Mr. Suter refused a breath sample only because he was expressly instructed to do so by his state-provided lawyer; (3) Mr. Suter and his wife suffered multiple vigilante attacks; and (4) there were many other mitigating factors, including Mr. Suter’s guilty plea, extreme remorse, lack of criminal record and strong community support. [ 114 ] The majority also correctly summarize the decisions below. They note that the sentencing judge sentenced Mr. Suter to four months’ imprisonment because of the “unique” facts described above, including how his reliance on his lawyer’s advice “fundamentally change[d]” his moral culpability (para. 17). The majority also explain how the Court of Appeal held that the trial sentence was unfit and substituted a sentence of 26 months in prison because Mr. Suter made no mistake of law, should not have had his sentence reduced based on vigilante violence and was criminally distracted. III. Analysis [ 115 ] An initial sentence can be revisited on appeal if it implicates a flawed process (material errors in reasoning) or flawed outcome (demonstrable unfitness) ( R. v. Lacasse, 2015 SCC 64, [2015] 3 S.C.R. 1089, at para. 11 ). As I will explain, neither type of flaw is present in the sentencing judge’s decision. On this basis, I would have restored the initial sentence. A. Flawed Process: Errors in Reasoning [ 116 ] The majority find two errors in the sentencing judge’s reasoning warranting appellate intervention: (1) his legal misunderstanding of mistakes of law (an “error in principle” per Lacasse, at para. 44 ); and (2) the excessive weight he gave to Mr. Suter’s sobriety (weighing a relevant factor “unreasonably” per Lacasse, at para. 49 ). [ 117 ] In my view, the sentencing judge made neither error. Accordingly, this Court cannot interfere with Mr. Suter’s sentence on the basis of an error in reasoning. (1) Mistake of Law [ 118 ] In my assessment, the sentencing judge made no errors in his analysis of Mr. Suter’s mistake of law. The majority’s findings to the contrary rely on an overly restrictive conception of mistakes of law and inappropriately discredit Mr. Suter’s unequivocal testimony, which was uniformly found credible by the sentencing judge. (a) The Legal Test [ 119 ] The majority hold that “mistake of law is a legal concept with rigorous requirements” (para. 64). Specifically, they hold that a mistake of law occurs only when an offender “has an honest but mistaken belief in the legality of his or her actions” ( ibid. ). Critically, they also characterize being “unsure” ( ibid. ) or “uncertai[n]” (para. 65) as fatal to a claim of mistake of law. I find this characterization too narrow. [ 120 ] At the outset, it is critical to clarify the different degrees of awareness an offender may have in respect of the criminality of their conduct. For the sake of discussion, four broad categories of awareness can be defined: (1) “right certainty”: where an offender (correctly) is certain their conduct is illegal; (2) “right probability”: where an offender (correctly) thinks it is likely their conduct is illegal, but is uncertain; (3) “wrong probability”: where an offender (erroneously) thinks it is likely their conduct is legal, but is uncertain; and (4) “wrong certainty”: where an offender (erroneously) is certain their conduct is legal. [ 121 ] The majority hold that “[c]onfusion or uncertainty” negates a mistake of law (para. 65). It would appear, therefore, that only the fourth category above — wrong certainty — would be sufficient to constitute a mistake of law. For example, if an offender is completely confident that it is lawful to refuse a breath sample because their lawyer tells them to do so (wrong certainty), but then gets “confused” by a police officer saying the opposite (now, wrong probability), on the majority’s test, it would appear that such an offender would be too “unsure” to make a mistake of law (see para. 64). [ 122 ] I dispute this overly rigid articulation. To be clear, my position is not that any iota of uncertainty will mitigate a sentence, even if the offender was still confident that their conduct was likely illegal (i.e., right probability). Some offenders may harbour minor doubts about the illegality of their conduct, and will thus lack certainty as to the criminality of that conduct, without any mitigation being warranted for that lack of certainty. Rather, my point is that thinking conduct is likely legal, but being uncertain (i.e., wrong probability), is still sufficient to constitute a mistake of law. [ 123 ] I also dispute that mistake of law should be dealt with as a binary, where only wrong certainty “fundamentally” alters culpability (majority reasons, at para. 67). To the contrary, we should trust trial judges to take a contextual approach — one which considers the source, nature and reasonableness of a mistake, along with any degrees of uncertainty — when allocating mitigating weight to a mistake of law. Indeed, we leave trial judges with similar discretion (and typically abstain from relying on “rigid formula[s]” ( ibid., at para. 47)) in respect of myriad other mitigating factors in sentencing. [ 124 ] I now turn to the authorities on mistake of law. [ 125 ] First, I note that the majority cite no authority for their strict “view” on mistake of law. The majority properly cite R. v. Forster, 1992 118 (SCC), [1992] 1 S.C.R. 339, for the proposition that mistake of law is not a defence to a criminal charge. They also correctly cite R. v. Pontes, 1995 61 (SCC), [1995] 3 S.C.R. 44, and R. v. MacDonald, 2014 SCC 3, [2014] 1 S.C.R. 37, for the proposition that, despite not being a defence, mistake of law can mitigate a sentence. But they provide no authority for their core thesis: that an offender’s “uncertainty” about the legality of their conduct negates a mistake of law. None of the majority’s authorities supports this novel characterization, which they present as established doctrine. [ 126 ] Second, I find that, if anything, the authorities suggest the contrary view: that mistake of law is a flexible concept broad enough to include some confusion or uncertainty about the law. [ 127 ] Section 19 of the Criminal Code provides only that “ [i]gnorance of the law by a person who commits an offence is not an excuse for committing that offence.” It is silent regarding the relationship between ignorance of the law and sentencing. To the extent that the scope of s. 19 informs the scope of mistake of law for sentencing purposes (see e.g. MacDonald, at paras. 58-59 and 61 ), its reliance on a negative statement of awareness (“ignorance”) makes it broad enough to include offenders who are uncertain about the lawfulness of their conduct (i.e., wrong probability). This scope is reflected in the jurisprudence, which is similarly broad enough to encompass partial uncertainty when mitigating a sentence on the basis of a mistake of law (see e.g. Forster, at p. 346; MacDonald, at para. 59 ). [ 128 ] In my view, the true breadth of these authorities is in keeping with the individualized nature of sentencing ( Criminal Code, s. 718.2(a); Lacasse, at para. 54 ). The strict rule in s. 19 of the Criminal Code in terms of guilt warrants flexibility in the interpretation of mistakes of law in terms of sentencing. As Professor Ashworth notes: “English law confines [the mistake of law defence] narrowly, and does so partly because effect can be given at the sentencing stage to variations in culpability” ( Sentencing and Criminal Justice (5th ed. 2010), at p. 149 (footnote omitted)). Indeed, the majority ostensibly concede that the individualized nature of sentencing requires flexibility when measuring degrees of culpability based on an offender’s level of awareness of the criminality of their conduct. To quote para. 64 of the majority reasons, if honestly mistaken offenders (wrong certainty) are “ less morally blameworthy” (emphasis in original) than offenders who are “unsure about the lawfulness of their actions” (wrong probability), it surely follows that offenders with such uncertainty are also less morally blameworthy than offenders who “know that their actions are unlawful” (right certainty). Criminality presumably carries greater moral blameworthiness when it is deliberate rather than inadvertent. (b) Application to the Instant Case [ 129 ] As I have explained, I would hold that an offender who is slightly confused about the legality of their conduct — but still thinks that conduct is likely legal — is mistaken in law. The majority concede that Mr. Suter was confused as to the legality of refusing to provide a breath sample. The majority do not specify how confused Mr. Suter was (i.e., between wrong probability and right probability). But as I explain below, that “confusion” (if any) would still leave Mr. Suter in the category of wrong probability. Accordingly, he made a mistake of law, the sentencing judge did not err in this regard, and his sentence cannot be revisited on this basis. [ 130 ] I would add that, in any event, the strongest inference from the sentencing judge’s reasons and the record is that Mr. Suter was erroneously certain that his conduct was lawful (i.e., wrong certainty). It follows that, even on the majority’s narrow test, Mr. Suter made a mistake of law. [ 131 ] The sentencing judge distinguished the mitigating effect of two types of legal advice (para. 76): (1) poor strategic advice (“a mis-guided presentation of legal options”, i.e., refuse to provide a breath sample because that will carry a smaller penalty than blowing over the legal limit); and (2) poor legal advice (“the lawyer expressly telling him not to provide a sample”, i.e., you must refuse to provide the breath sample). He reasoned that the latter is more mitigating. This is because, upon being expressly instructed by a Brydges lawyer to take one course of action (refuse to provide the breath sample), an individual would assume that single course of action was lawful. Indeed, that is precisely what Mr. Suter described in his testimony, which the sentencing judge uniformly found credible: I just didn’t think the lawyer would be telling me to... do something that would be illegal. Like I thought it was within my rights not to blow....... Why would [the Brydges lawyer] tell me to do something that was illegal? (R.R., at pp. 40 and 85) [ 132 ] I dispute that these unequivocal statements — read in the context of Mr. Suter’s unwavering credibility — leave any doubt as to the honesty or degree of Mr. Suter’s belief that his conduct was lawful. While the sentencing judge did not do a line-by-line review of Mr. Suter’s examination transcript and identify which phrases he found credible, he clearly accepted Mr. Suter’s “side of the story” and never expressed any concerns with his credibility. [ 133 ] With respect, none of the majority’s reasons for discrediting these accepted statements persuade. [ 134 ] First, the majority discredit Mr. Suter on appeal because the Brydges lawyer’s advice was “strategic” (para. 70), which supposedly contradicts Mr. Suter’s claim that he thought his refusal was lawful. But this relies on evidence from the Brydges lawyer which the sentencing judge rejected and ignores evidence from Mr. Suter, who expressly denied that the Brydges lawyer’s advice was strategic (A.R., vol. II, at pp. 42-43). I consider it inappropriate for this Court to discredit a witness found credible at first instance with the contradictory testimony of another witness found non-credible at first instance. By speculating about whether “the lawyer’s instruction to refuse to provide a breath sample was articulated in terms of sentencing strategy” (para. 70), the majority improperly reevaluate the record in a manner similar to what they criticize the Court of Appeal for doing. In any event, the majority concede that “there was no finding that the refusal was a strategic choice on Mr. Suter’s part” (para. 75). I therefore fail to see how the lawyer’s reason for providing advice proves Mr. Suter’s understanding of the lawfulness of his conduct based on that advice. [ 135 ] Second, the majority discredit Mr. Suter on appeal because he “admitted that the police officer told him that it was an offence to refuse to provide a breath sample”, which is “relevant to the sincerity of the belief” he had as to the lawfulness of his conduct (para. 71). But, as the majority concede, “it is understandable, particularly in the context of an arrest, for a detainee to accept the advice of his or her lawyer, who is aligned in interest, over that of a police officer” ( ibid. ). This is especially important given the broad latitude this Court has granted the police to use “forms of deceit” — including confronting detainees with “exaggerated” or even “fake” evidence that purports to be “absolutely overwhelming” proof of guilt — when extracting prejudicial admissions from detainees ( R. v. Oickle, 2000 SCC 38, [2000] 2 S.C.R. 3, at paras. 66, 91, 95 and 99-100; Rothman v. The Queen, 1981 23 (SCC), [1981] 1 S.C.R. 640, at p. 697; R. v. Collins, 1987 84 (SCC), [1987] 1 S.C.R. 265, at pp. 286-87; R. v. Sinclair, 2010 SCC 35, [2010] 2 S.C.R. 310, at paras. 60, 116 and 142). Regardless, it is for the sentencing judge to assess the impact of this admission on Mr. Suter’s credibility, not this Court. Mr. Suter expressly and repeatedly stated that he did not trust the police when they told him he would be charged for refusing to provide a breath sample and that, in his understanding, he was supposed to prefer his lawyer’s advice over statements from the police (A.R., vol. II, at pp. 13-14 and 38-40; R.R., at p. 40). Again, I find it inappropriate for this Court to discredit these unequivocal statements by parsing the record when the sentencing judge found Mr. Suter’s testimony credible on that same record. [ 136 ] Third, the majority discredit Mr. Suter’s claim that he honestly believed his conduct was lawful because the sentencing judge found that he was confused during his call with the Brydges lawyer. But Mr. Suter’s confusion during the call is irrelevant given that, by the end of the call, he “clearly got the intended message” and “understood the lawyer’s advice”, i.e., refuse to provide the breath sample (sentencing decision, at paras. 38 and 41). [ 137 ] Lastly, the majority discredit Mr. Suter on appeal because “[n]othing in the record suggests that the lawyer told Mr. Suter... that he was legally justified in refusing” (para. 70). In my view, this conflates Mr. Suter’s belief with the foundation for that belief. Mr. Suter’s belief does not depend on the Brydges lawyer’s explanation or his own legal acumen. Rather, it depends on his sincerity. In his credible and uncontradicted testimony, Mr. Suter stated that he “didn’t think the lawyer would be telling [him] to... do something that would be illegal” and that he “thought it was within [his] rights not to blow” (R.R., at p. 40). Whether or not Mr. Suter knew why he could refuse to blow, he certainly thought it was a lawful decision. The whole point of having a lawyer in the first place — in the detention context or otherwise — is so that they can tell you what to do based on what they know. The exercise of legal rights should not be conditioned on a comprehensive understanding of their constitutional underpinnings. Otherwise, unsophisticated detainees will necessarily have diluted constitutional rights, despite needing those rights the most. [ 138 ] Accordingly, even on the majority’s test, I find that Mr. Suter made a mistake of law. I agree that we could “more safely infer” Mr. Suter’s honest belief in the lawfulness of his conduct with even more evidence (majority reasons, at para. 70). But the sentencing judge’s reasons, coupled with Mr. Suter’s unequivocal and credible statements affirming that honest belief, are more than sufficient in the circumstances. (2) Sobriety [ 139 ] In my view, the sentencing judge also made no errors in his weighing of Mr. Suter’s sobriety. With respect, the majority’s finding to the contrary mischaracterizes the sentencing judge’s reasons. [ 140 ] To begin, I note that an appellate court can intervene on sentence when a sentencing judge weighs a particular factor “unreasonably”, but not when the appellate court would have simply “weighed the relevant facto[r] differently ” ( Lacasse, at para. 49 (emphasis added)). Indeed, to intervene on sentence based merely on the view that “the trial judge gave too much weight to one relevant factor... is to abandon deference altogether” ( Lacasse, at para. 49, citing R. v. Nasogaluak, 2010 scc 6, [2010] 1 s.c.r. 206, at para. 46). [ 141 ] The majority hold that the sentencing judge “placed considerable weight” on Mr. Suter’s sobriety (para. 86) and that sobriety “must have contributed in no small measure” to Mr. Suter’s four-month sentence (para. 87). But they are, in my view, inconsistent when describing that weight. At first, the majority appear to suggest that the sentencing judge allocated at least 32 months of mitigation to Mr. Suter’s sobriety alone. Specifically, they claim that if “Mr. Suter [had] been convicted of impaired driving causing death, the sentencing judge would have imposed a sentence of more than three years imprisonment” (para. 86). However, they later concede that the sentence reduction was based on Mr. Suter’s sobriety “in combination with other mitigating factors, including the bad legal advice” ( ibid. ). In my view, the former interpretation (i.e., that sobriety alone reduced the sentence to four months) is a misreading of the sentencing decision, while the latter interpretation (i.e., that the many mitigating factors reduced the sentence together) fails to demonstrate an unreasonable independent weighing of sobriety. [ 142 ] The former interpretation links the sentence reduction to sobriety alone. When reviewing trial reasons, appellate courts must read the reasons “as a whole” ( R. v. Rhyason, 2007 SCC 39, [2007] 3 S.C.R. 108, at para. 10 ). It follows that appellate courts should not isolate single passages from trial reasons to find errors in reasoning. Yet that is what this interpretation does. It isolates the following passage from the sentencing judge’s decision (para. 75) to conclude that he gave excessive weight to Mr. Suter’s sobriety when sentencing him: “If the court were imposing a sentence for impaired driving causing death in these circumstances, the court would view the sentence suggested by the Crown [three years] as too low.” [ 143 ] At first instance, the Crown sought a sentence of three years in prison. Accordingly, the former interpretation reads para. 75 of the sentencing decision as follows: if Mr. Suter had been impaired (“[i]f the court were imposing a sentence for impaired driving”), but all other facts were the same (“in these circumstances”), his sentence would have been more than three years. By corollary, according to the former interpretation, the sentencing judge relied on sobriety alone to lower a sentence of at least three years to four months. I agree that this would give significant weight to sobriety as a mitigating factor. However, this interpretation is defeated when the passage is read in the immediate context of the paragraphs following para. 75 and in the broader context of the reasons as a whole. [ 144 ] Paragraph 75 and the immediate context following it are as follows: If the court were imposing a sentence for impaired driving causing death in these circumstances, the court would view the sentence suggested by the Crown as too low. However, the evidence has moved the court from its starting position. The court finds, on balance, that as tragic as the consequences have been, this collision was an accident caused by a non-impaired driving error. As earlier outlined, the court also finds that Mr. Suter’s refusal to the lawful demand was the result of, hopefully rare, ill-informed and bad legal advice. If the advice had stopped with a mis-guided presentation of legal options, even if aimed at steering the suspect away from blowing, the mitigating effect of the advice would be significantly less. In this case, however, the court has accepted the testimony of Mr. Suter as to what the lawyer said, and finds that the refusal was based on the lawyer expressly telling him not to provide a sample. This does not absolve Mr. Suter, as a mistake of law, is not a defence but it fundamentally changes Mr. Suter’s moral culpability. People must be able to rely on legal advice given when exercising their constitutional right to counsel. [Emphasis added; paras. 75-77] [ 145 ] If the sentencing judge’s intent were to say that he would have sentenced Mr. Suter to three years in prison but for his sobriety, then the text following the word “[h]owever” at the start of para. 76 would discuss only that sobriety. It does not. Rather, it discusses Mr. Suter’s sobriety and, critically, also discusses the bad legal advice he received. Likewise, the last paragraph above is devoted solely to the implications of failing to see moral culpability as being diminished when someone is instructed to commit an offence by their lawyer, and makes no reference whatsoever to Mr. Suter’s sobriety. Clearly, the sentencing judge viewed both Mr. Suter’s sobriety and the bad legal advice he received as acting together to significantly reduce his sentence. [ 146 ] Consequently, the “circumstances” referred to in para. 75 must be other circumstances not mentioned in the paragraphs below it (paras. 76-82), e.g., the death of a toddler. Indeed, this is consistent with para. 78, where the sentencing judge adds mitigating factors to “the circumstances of the event”, indicating that his references to “circumstances” in paras. 75 and 78 are non-exhaustive and exclude certain mitigating circumstances in this case. [ 147 ] An interpretation that links the sentence reduction to sobriety alone therefore mischaracterizes the sentencing judge’s reasons. If anything, it was the bad legal advice — which the sentencing judge viewed as “fundamentally chang[ing] Mr. Suter’s moral culpability” (para. 76) — that contributed most to the sentence reduction. [ 148 ] The broader context of the reasons as a whole reinforces this conclusion. Throughout his reasons, the sentencing judge is clear that it is the unique combination of mitigating factors that culminates in a four-month sentence and that no one factor is dispositive. He writes: The sentence imposed should be similar to sentences imposed on similar offenders for similar offences committed in similar circumstances. Applying this principle in this case is a particular challenge because this offence is new and the combination of factors is very unique. This is a serious offence by any measure. The gravity is reflected in part by the maximum penalty of life imprisonment. On the other hand, there are many factors in this case that reduce Mr. Suter’s degree of responsibility from what may, with time, become a normal sentencing range for this offence. The constellation of factors in this case is, to say the least, very unusual. They are so unlikely in combination that it may be unique. All of [the eight mitigating] factors [listed] operate to significantly reduce the sentence from what would otherwise be fit. [Emphasis added; paras. 48, 74 and 82.] [ 149 ] Accordingly, the suggestion that sobriety alone, not the unique mix of mitigating factors, substantially reduced Mr. Suter’s sentence in this case, is belied by the reasons when read in their entirety. While the sentencing judge’s phrasing at para. 75 (“in these circumstances”) is not ideal, it is entirely understandable when read in context and viewed through the proper, deferential lens. [ 150 ] The latter interpretation (i.e., that “[t]he evidence of non-impairment — in combination with other mitigating factors, including the bad legal advice Mr. Suter received — ‘moved the court from its starting position’” (majority reasons, at para. 86)), I agree with. But, in my view, it fails to provide a basis for appellate intervention. Simply put, since the sentencing judge looked at the combined effect of sobriety, bad legal advice, vigilante violence and the many other mitigating factors in this case — all of which he considered in concert — there is nothing to substantiate the majority’s view of the independent weight assigned to sobriety or, in turn, of the unreasonableness of that weight. (3) Conclusion on Flawed Process [ 151 ] Based on the foregoing, I find that the sentencing judge made neither process error found by the majority. It follows that, in my view, this Court cannot interfere with Mr. Suter’s initial sentence on the basis of an error in principle (i.e., the mistake of law issue) or an error in weighing a relevant factor unreasonably (i.e., the sobriety issue). B. Flawed Outcome: Demonstrable Unfitness [ 152 ] Still, one potential basis for intervention remains in this case: demonstrable unfitness ( Lacasse, at para. 52 ). Even if a trial judge makes no errors in the sentencing process, a court of appeal can interfere if the outcome is “demonstrably unfit”, a concept best articulated by Laskin J.A. as follows: The Supreme Court has used a variety of phrases to convey this point: an appellate court cannot interfere unless the sentence is “clearly unreasonable”, “clearly or manifestly excessive”, “clearly excessive or inadequate”, “demonstrably unfit” or a “substantial and marked departure”. Whatever the phrase, the bottom line is the same: appellate courts should defer to sentences imposed by trial judges unless the sentence is outside an acceptable range. ( R. v. Rezaie (1996), 1996 1241 (ON CA), 31 O.R. (3d) 713 (C.A.), at p. 720) This “very high threshold” ( Lacasse, at para. 52 ) is, however, not met in this case. [ 153 ] Trial judges have “broad discretion” to impose fit sentences ( Lacasse, at para. 39 ). Yet this discretion is not unbridled. A carefully drafted scheme of Criminal Code provisions outlines the analytical framework for sentencing. These provisions were enacted as “a step towards more standardized sentencing, ensuring uniformity of approach” (C. C. Ruby, G. J. Chan and N. R. Hasan, Sentencing (8th ed. 2012), at §1.59). Following this approach, I am compelled to restore the sentencing judge’s four-month sentence. Even though I would have weighed the gravity of the offence differently at first instance, this is beside the point. It suffices to note that a four-month sentence in this case is not “clearly unreasonable” ( Lacasse, at para. 52, citing Rezaie, at p. 720). (1) Mr. Suter’s Initial Sentence [ 154 ] To begin, I note that Mr. Suter’s sentence at first instance was not limited to four months’ imprisonment. Rather, it included (1) 5 days of pre-sentence custody; [5] (2) a victim fine surcharge of $200; (3) a 30-month driving prohibition commencing on his release from prison; and (4) a pre-conviction driving suspension, which, in conjunction with the driving prohibition, bars him from operating a motor vehicle for 5 years. [ 155 ] This complete punishment must be in our minds when assessing the fitness of Mr. Suter’s sentence. (2) The Majority’s Approach [ 156 ] I find the majority’s approach to reviewing Mr. Suter’s sentence incomplete. [ 157 ] First, the majority’s approach to choosing a fit sentence is deficient. They note that the sentencing range for impaired driving causing death is 2 to 10 years of imprisonment, make a passing reference to the statutory scheme governing sentencing, weigh the exact same factors as the sentencing judge (sobriety, bad legal advice and vigilante violence), assert that a 4-month prison sentence is “manifestly unfit” with minimal explanation (it “does not properly account for the gravity of the offence” (para. 94)), and conclude that a sentence as high as 18 months of imprisonment would have been fit in the circumstances — again, with minimal explanation (it balances “the several attenuating factors” with “the gravity” of the offence (para. 95)). As I explain below, this approach hardly amounts to the proper sentencing analysis required by the Criminal Code. The Court’s decisions should provide guidance to courts of appeal on how to rigorously assess the fitness of a sentence. As one commentator notes: “... a uniform approach to sentencing review is required to maintain confidence in the justice system. In the absence of a principled basis for determining when to intervene on appeal, a uniform approach is impossible” (J. Foy, “Proportionality in Sentence Appeals: Towards a Guiding Principle of Appellate Review” (2018), 23 Can. Crim. L.R. 77, at p. 78). Yet the majority fail to cite a single case involving a sentence under s. 255(3.2) outside the typical sentencing range to provide support for their statements regarding the fitness of the sentence in these unique circumstances. [ 158 ] Second, the majority’s approach to imposing a fit sentence is also deficient. Despite signalling the need for a sentence as high as 18 months, the majority pivot and ultimately impose a sentence of 10 and a half months, resulting from a 4-month sentence deemed to have been served in two thirds of its time (i.e., 80 days) combined with 6 and a half additional months served before being released on bail. [6] They do so because Mr. Suter has spent almost nine months awaiting this Court’s decision. While this is not an “absolute” rule, courts certainly can consider the intervening time between an initial sentence being imposed or served and an appeal being brought when fashioning appropriate sentences on appeal (Ruby, Chan and Hasan, at §§4.24 to 4.29). However, in a context such as the one in the instant case — i.e., “where further appeals to higher courts take place” — intervening time will generally constitute a mitigating factor, not a basis to stay the execution of the sentence; unlike, for example, Crown delay in seeking an appeal, which more appropriately warrants such a stay ( ibid., at §4.28). Accordingly, though it should come as no surprise that I am favourable to Mr. Suter’s release given his minimal moral blameworthiness, the majority’s decision to nearly halve his sentence seems, with respect, artificial to me. They insist on a sentence as high as 18 months for the purpose of “deterring” and “denouncing” Mr. Suter’s conduct (para. 93), and then effectively stay the remaining sentence as it now serves “no useful purpose” (para. 103). [ 159 ] I see only one explanation for this apparently irreconcilable tension: the majority want to signal their opposition to impaired driving through a stiff sentence, but hesitate in imposing that sentence on Mr. Suter because it would be disproportionate to his circumstances. In my view, this approach is flawed for two reasons. First, denouncing the offence of refusing to provide a breath sample is not irreconcilable with recognizing unique facts that engage neither the direct mischief (obstruction of justice) nor the indirect mischief (impaired driving) at which the offence is directed. Second, this approach, while it enhances the proportionality of Mr. Suter’s sentence, undermines the proportionality of future sentences, since accused with similarly sympathetic circumstances will have their sentences measured against what I consider to be the excessive benchmark of 15 to 18 months of imprisonment endorsed by the majority. (3) The Established Approach [ 160 ] In my respectful view, and in the unique circumstances of this case, I cannot conclude that a four-month prison sentence is demonstrably unfit based on the established approach to sentencing appeals. I will explain this by first discussing the deferential process in sentencing appeals and then turning to the relevant sentencing principles applicable here. (a) Deference in Sentencing Appeals [ 161 ] Sentencing is a “delicate” process ( Lacasse, at para. 1 ) that is “profoundly subjective” ( ibid., at para. 40, citing R. v. Shropshire, 1995 47 (SCC), [1995] 4 S.C.R. 227, at para. 46 ). For this reason, we give “wide latitude to sentencing judges” because they, unlike appellate courts, have the advantage of hearing and seeing the witnesses and hearing the parties’ sentencing submissions, and are familiar with the circumstances in their districts ( Lacasse, at paras. 11 and 48 ). They are therefore “in the best position to determine [a] just and appropriate” sentence ( ibid., at para. 102). This is why Parliament grants trial judges “discretion” when negotiating amongst “different degrees or kinds of punishment in respect of an offence” ( Criminal Code, s. 718.3(1)). [ 162 ] This case is no exception. Its extreme complexity only reinforces the need for deference. To quote the sentencing judge: This case is unique in several respects. The sheer volume of evidence and submissions presented in this sentencing is exceptional. Rarely do sentencing proceedings take more than a day, particularly when following a guilty plea with an Agreed Statement of Facts. In this case, a week of evidence and argument was presented. In addition to receiving a lengthy Agreed Statement of Facts with exhibits attached including a transcript of the preliminary inquiry, the court viewed video evidence, received expert reports, heard from experts, heard testimony from Mr. Suter and others and over the course of one of the days, heard very passionate and articulate victim impact statements from approximately 25 persons impacted by this tragedy. Five volumes of legal authorities were presented to the court in addition to extensive written and oral argument. The list of authorities considered is appended as Appendix “A”. Although only one of the authorities is specifically cited in this judgment, they have all been considered and absorbed into the court’s reasoning. [para. 6] [ 163 ] As trial judges are the “experts” in sentence fitness — both in fact and by Parliamentary mandate — mere deviation from a jurisprudential sentencing range does not justify appellate intervention ( Lacasse, at para. 11 ). Rather, the ultimate sentence alone, assuming no errors in reasoning, warrants appellate intervention only when it is “demonstrably unfit” ( ibid. ). This reflects the “considerable deference” that appellate courts must pay to trial sentences ( ibid., at para. 41, citing R. v. Proulx, 2000 SCC 5, [2000] 1 S.C.R. 61, at para. 123 ). [ 164 ] A deferential approach to sentence appeals respects the distinct institutional roles between trial and appellate courts. Each court operates in its realm of expertise. Trial judges — who see a higher volume of diverse cases and are immersed in the unique (or common) facts of the many cases that come before our courts — develop an expertise in the complex balancing exercise of sentencing ( Lacasse, at paras. 48-49 ). In contrast, appellate courts — which carefully scrutinize relatively fewer cases, primarily in terms of legal principles — develop an expertise in reviewing decisions for error correction ( ibid., at para. 11). As Doherty J.A. noted in R. v. Ramage, 2010 ONCA 488, 257 C.C.C. (3d) 261, para. 70: Appellate repetition of the exercise of judicial discretion by the trial judge, without any reason to think that the second effort will improve upon the results of the first, is a misuse of judicial resources. The exercise also delays the final resolution of the criminal process, without any countervailing benefit to the process. [ 165 ] Given the above, this Court has consistently cautioned against unrestrained interference with sentencing decisions on the claimed basis of demonstrable unfitness, noting that such interference should not happen “lightly” ( Lacasse, at para. 39 ), that it cannot happen merely because an appellate court “feels that a different order ought to have been made” ( Shropshire, at para. 46 ) or “second-guess[es]” the sentencing judge ( Proulx, at para. 125 ), and that appellate courts can interfere only when a sentence is “clearly unreasonable” ( R. v. L.M., 2008 SCC 31, [2008] 2 S.C.R. 163, at para. 35 ). [ 166 ] As such, demonstrable unfitness is not an unchecked subjective inquiry. In other words, an appellate court cannot simply assert that a sentence is demonstrably unfit without meaningful explanation. Rather, it must demonstrate that unfitness with reference to the Criminal Code ’s sentencing principles, including proportionality ( s. 718.1 ), the sentencing objectives ( s. 718 ), individualization ( s. 718.2 (a)) and parity ( s. 718.2 (b)). Indeed, as the Court noted in Lacasse, “[i]ndividualization and parity of sentences must be reconciled for a sentence to be proportionate” (para. 53). [ 167 ] I now turn to these concepts. (b) Sentencing Principles (i) Proportionality (Section 718.1) [ 168 ] A fit sentence “must be proportionate to the gravity of the offence and the degree of responsibility of the offender” ( Criminal Code, s. 718.1). In other words, the ultimate sentence must correspond to the degree of “gravity of the offence” (i.e., how serious the offence is) and the degree of “responsibility of the offender” (i.e., their moral blameworthiness) ( Lacasse, at para. 12 ). [ 169 ] This case is difficult precisely because these two factors pull us in opposite directions. The gravity of the offence — refusing to provide a breath sample after a fatal car accident — is very high. Indeed, that gravity was the basis for the sentencing judge’s desire to impose a carceral sentence. But Mr. Suter’s moral blameworthiness — as a sober driver who was in a genuine car “accident” (caused by a “non-impaired driving error” (sentencing decision, at para. 76)) and who refused to provide a breath sample only because he was expressly instructed to do so by his lawyer — could hardly be lower. The tensions that inevitably arise when balancing these conflicting considerations underlie the particularly “delicate” task of ascertaining proportionality ( Lacasse, at para. 12 ). For this reason, “what objectives should be pursued and the best way to do so” is up to the trial judge ( ibid., at para. 41, citing Proulx, at para. 125 ). Simply put, we are not in a better position than the sentencing judge to reconcile these conflicting forces. Rather, given his proximity to the facts of this case and his institutional expertise, we are in a worse position. [ 170 ] The majority concede that Mr. Suter’s low moral blameworthiness brings him below the established sentencing range for this offence, i.e., below the range of 2 to 10 years in prison. This is precisely the conclusion reached by the sentencing judge. Accordingly, both the majority and the sentencing judge agree that Mr. Suter’s circumstances are unique and bring him below the typical sentencing range for this offence. The question then becomes how far below that sentencing range Mr. Suter’s sentence should lie. Put differently, the only remaining question is: Where in the technically available range from zero to two years should Mr. Suter’s unique circumstances place his sentence? The sentencing judge said 4 months; the majority says 15 to 18. While I personally would have weighed the gravity of the offence more heavily than the sentencing judge, this is precisely the complex balancing exercise to which we have repeatedly instructed courts of appeal to defer. [ 171 ] Other than the sentencing judge’s expertise and Mr. Suter’s unique circumstances, two additional factors reinforce the proportionality of a significantly reduced sentence in these circumstances. [ 172 ] First, Mr. Suter was not only mistaken in law, but reasonably mistaken. The sentencing judge found that Mr. Suter relied on the Brydges lawyer’s advice when refusing to provide a breath sample. This reliance was reasonable. Providing a breath sample upon detention was an unfamiliar administrative procedure. Faced with this uncertainty, Mr. Suter did not seek to obstruct justice or remain willfully blind to his obligations. Rather, he diligently considered his legal position by consulting a Brydges lawyer, the very individual the police directed him to consult. Ultimately, the only reason Mr. Suter refused to blow was his reliance on that lawyer’s advice. As the sentencing judge noted, this “fundamentally change[d] Mr. Suter’s moral culpability” (para. 76). [ 173 ] The fact that Mr. Suter’s advice came from a state-provided lawyer takes on particular significance here. The police directed Mr. Suter to a Brydges lawyer; that lawyer told Mr. Suter to refuse the breath sample; and Mr. Suter complied. This is the full extent of his criminal conduct. His moral blameworthiness is infinitesimal. [ 174 ] Second, the constitutional significance of the right to counsel also weighs in favour of a significantly reduced sentence in this case. As the sentencing judge noted: “People must be able to rely on legal advice given when exercising their constitutional right to counsel” (para. 77). [ 175 ] Section 10( b ) of the Canadian Charter of Rights and Freedoms provides that “[e]veryone has the right on arrest or detention... to retain and instruct counsel without delay”. This is because, when detained, an offender is “put in a position of disadvantage relative to the state” ( R. v. Bartle, 1994 64 (SCC), [1994] 3 S.C.R. 173, at p. 191) and has a corresponding need “to be informed of his rights and obligations under the law” ( Sinclair, at para. 26; see also R. v. Brydges, 1990 123 (SCC), [1990] 1 S.C.R. 190, at pp. 209 and 214) and to receive “appropriate advice with respect to the choice he faces” ( R. v. Hebert, 1990 118 (SCC), [1990] 2 S.C.R. 151, at pp. 176-77), which is chiefly “whether to cooperate with the police or not” ( Sinclair, at para. 24 ). Fulfilling these needs serves the philosophy underlying s. 10 ( b ): to ensure that people who are “detained are treated fairly in the criminal process” ( Bartle, at p. 191; see also Brydges, at p. 203; Clarkson v. The Queen, 1986 61 (SCC), [1986] 1 S.C.R. 383, at p. 394). [ 176 ] Mr. Suter’s experience shows the importance of these constitutional obligations. He merely followed the advice he received from his Brydges lawyer. For this, he has been thrust through 3 layers of court, 10 and a half months of prison deemed served and years of uncertainty. For my part, I cannot accept that a majority of this Court would countenance such a severe prison term when the failure on these facts was not that of Mr. Suter’s moral conscience, but of his province’s duty counsel system (a point conceded by the majority at para. 75, fn. 4), which is “part of [a province’s] constitutional responsibility for the administration of justice” ( Bartle, at p. 196). [ 177 ] Further, a specific constitutional purpose underlying the right to counsel reinforces Mr. Suter’s low moral blameworthiness in this case and the reasonableness of his mistake. A fundamental component of the right to counsel is preserving a detainee’s right against self-incrimination ( R. v. Singh, 2007 SCC 48, [2007] 3 S.C.R. 405, at para. 21 ). Yet as the Court explained in Bartle, “[w]hat is singular about the refusal offence in the impaired driving context is that it punishes a person who refuses to incriminate” themselves (p. 215). The consistency between the general right to abstain from self-incrimination and refusal to provide a breath sample speaks to the particular importance of clear and proper legal advice in this anomalous circumstance. [ 178 ] Based on the foregoing considerations, I cannot conclude that the sentence of four months was demonstrably unfit. Further, I find that the proportionality of the sentence is reinforced by the applicable sentencing objectives. Specifically, neither deterrence nor denunciation — the sentencing objectives on which the majority rely — establishes the demonstrable unfitness of a four-month sentence here. (ii) Deterrence (Section 718(b)) [ 179 ] One objective in sentencing is “to deter the offender and other persons from committing offences” ( Criminal Code, s. 718(b)). In the impaired driving context, this objective “must be emphasized” ( Lacasse, at para. 5 ). Indeed, the “pressing objective of deterring... the scourge of impaired driving” is one basis for the majority’s imposition of a 15- to 18-month sentence in this case (para. 93). But such reasoning disregards the facts before us, which are critical to a proportionate sentence for Mr. Suter. Mr. Suter was not “tempted” to commit a crime, nor did he commit a crime because of insufficient deterrence. He simply refused to provide a breath sample because he was told to do so by his lawyer. As a result, a stiff sentence in this case will not deter others from refusing to provide a breath sample; rather, it will deter others from following their lawyer’s advice, despite the constitutional right to seek out that advice ( Singh, at para. 33 ). [ 180 ] The significant weight given to general deterrence in the context of crimes related to impaired driving is often predicated on the fact that such crimes are inordinately committed by otherwise law-abiding citizens ( Ramage, at para. 75 ). But on these exceptional facts, where Mr. Suter was not impaired and his only offence was refusing to blow because he was instructed to do so, general deterrence plays a minor role. We must recall that sentencing is an individualized process, not a sledgehammer for conveying the Court’s distaste for impaired driving regardless of the circumstances. [ 181 ] I also note that the low sentence given in this case will not encourage detainees to seek out poor legal advice with a view to mitigating their ultimate sentence. The only reason Mr. Suter was charged with an offence in the first place was because of the poor legal advice he received. It would therefore be absurd to claim that this legal advice ultimately helped him more than it harmed him. In any event, concerns about defence counsel instructing individuals to unknowingly commit administrative offences (surely a rare event) should be dealt with through disciplinary or other proceedings against defence lawyers, not harsh prison terms for the victims of their negligent advice. [ 182 ] Accordingly, the objective of deterrence does not justify interfering with the four-month sentence on the unique facts of this case. (iii) Denunciation (Section 718(a)) [ 183 ] Another objective in sentencing is “to denounce unlawful conduct and the harm done to victims or to the community that is caused by unlawful conduct” ( Criminal Code, s. 718(a)). In the impaired driving context, this objective, too, “must be emphasized” ( Lacasse, at para. 5 ). Indeed, the “pressing objective of... denouncing... the scourge of impaired driving” is the other basis for the majority’s more severe sentence (para. 93). But for similar reasons, this approach equally disregards the unique facts in this case. Mr. Suter did not truly “choose” to refuse to blow; he simply followed instructions. Mere commission of an administrative offence, when advised to do so by a state-provided lawyer, does not warrant strict denunciation. In such a circumstance, the conduct — while criminal — originates from faulty advice reasonably relied upon, thereby lacking the moral blameworthiness needed to warrant strict denunciation. If anyone’s conduct should be denounced in this case, it is that of the Brydges lawyer who counselled Mr. Suter’s offence (see e.g. Bartle, at pp. 215-16), without whose negligent advice none of this would have ever happened. As noted, our criminal justice system is predicated on the belief that the constitutional right to counsel will effectively preserve the legal rights of unsophisticated offenders. It is perverse to harshly denounce the conduct of an accused who has done nothing but exercise that constitutional right and whose only “offence [was] the act of following legal advice”, as the sentencing judge put it during counsel’s submissions on sentence (A.R., vol. II, at p. 109). [ 184 ] In consequence, I conclude that the objective of denunciation also fails to justify interfering with the four-month sentence on the unique facts of this case. (iv) Individualization (Section 718.2(a)) [ 185 ] A sentence should be “reduced” to account for any “mitigating circumstances relating to the offence” ( Criminal Code, s. 718.2(a)). [ 186 ] These reasons focus on three such factors: primarily, Mr. Suter’s reliance on bad legal advice and his sobriety; and secondarily, the gruesome vigilante attacks he and his wife suffered. However, these are not the only mitigating factors in this case. Rather, as the sentencing judge noted, Mr. Suter’s sentence must also be reduced because of: (1) his guilty plea; (2) his extreme remorse, “beyond what is reflected in the plea itself” (para. 79); (3) his lack of a criminal record; (4) his strong community support; and (5) the fact that he has been a “productive member of society, having been employed virtually all of his adult life before retirement” (para. 80). [ 187 ] All eight of these mitigating factors operate collectively in determining a fit sentence in the circumstances. Viewed together, these facts are remarkably mitigating. Indeed, it is difficult to conceive of more sympathetic facts than those present here in terms of Mr. Suter’s moral blameworthiness. One might even argue that this case approaches what could be notionally considered the “mandatory minimum” for s. 255(3.2) of the Criminal Code. [7] [ 188 ] Precisely how these factors are reflected in a particular sentence is most empirically determined through the sentencing principle of parity, which I end with below. (v) Parity (Section 718.2(b)) [ 189 ] The Criminal Code provides that “a sentence should be similar to sentences imposed on similar offenders for similar offences committed in similar circumstances” ( s. 718.2 (b)). [ 190 ] Here, the sentencing judge held that “[a]pplying [the parity] principle in this case is a particular challenge because this offence is new and the combination of factors is very unique” (para. 48). I agree. Based on my review of the jurisprudence, Mr. Suter’s facts are, indeed, entirely unique. Typically, criminal cases fall somewhere between various previous cases, leaving trial judges with the complex task of fitting the facts before them — and a proportionate sentence — into the mix of the jurisprudence. But this case differs. There are no reported decisions where someone convicted of refusing a breath sample after a fatal accident was found factually sober and was expressly instructed by a state-provided lawyer to refuse to provide that sample. I have, of course, also not found any such case where the individual convicted, and his wife, both suffered appalling vigilante attacks. [ 191 ] The extent to which Mr. Suter’s circumstances are markedly unique in itself supports deference to the initial sentence. In other words, there are no “similar offenders... in similar circumstances” against which his sentence can be reasonably measured ( Criminal Code, s. 718.2(b)). As such, it is hard to dispute that the sentencing judge was in the best position to attach a proportionate sentence to the unique facts of this case. [ 192 ] Moreover, despite the absence of similarly sympathetic facts in the s. 255(3.2) jurisprudence, a brief review of that jurisprudence reinforces my conclusion that a four-month sentence is not demonstrably unfit in these circumstances. [ 193 ] Based on my review, the case with facts suggestive of moral blameworthiness most similar to Mr. Suter’s is R. v. Smith, 2017 MBPC 16, 10 M.V.R. (7th) 152. I say this cognizant of the fact that the circumstances in the two cases are actually quite different; as I explained above, this warrants deference to the sentencing judge’s initial sentence, not intervention. [ 194 ] In Smith, the accused had consumed multiple drinks and marijuana in the hours leading up to the fatal accident (para. 4), which in itself favoured a higher sentence than the one that should be imposed on Mr. Suter, who was sober. But the accident in Smith appeared to have been an “accident” just like here (the key analogous fact in Smith and Mr. Suter’s case). The victim was a drunk person wearing dark clothing who was standing in the middle of the road at night and who appeared only when a car in front of the accused’s vehicle swerved at the last second (paras. 4-5 and 7). On these unique facts, the sentencing judge found that it was unclear that the accused was even driving dangerously (para. 7) and, in turn, the judge was unsure as to whether the accused had even caused the accident (para. 15). Further, the accused’s background was highly sympathetic, like Mr. Suter’s. He had “relevant factors relating to his Indigenous heritage” (para. 23; see also paras. 19 and 27), had a limited prior record from his youth (paras. 24 and 32), financially supported his family (paras. 28 and 30) and was being recommended for community supervision (paras. 29-30). For “withholding crucial evidence” from the state (i.e., for refusing to provide a breath sample following a fatal accident), Mr. Smith received a four-month carceral sentence to be served on weekends (paras. 31-34). And because the “circumstances of this offence involve[d] [Mr. Smith] using illegal drugs (marijuana) and consuming alcohol throughout the evening and well into the early hours of the morning” (again, unlike Mr. Suter), Mr. Smith was placed on supervised probation for 18 months to “ensure that he remains at home and does not embark on the sort of partying” that culminated in the fatal accident in that case (para. 36). [ 195 ] Smith is the closest reported case in terms of moral culpability to Mr. Suter’s and, coincidentally, the accused was given the same carceral sentence imposed by the sentencing judge in the instant case. Of course, Indigeneity can play a significant role in sentencing ( Criminal Code, s. 718.2(e); R. v. Ipeelee, 2012 SCC 13, [2012] 1 S.C.R. 433, at para. 59 ). But the accused in Smith was not factually sober, nor was he instructed to refuse a breath sample by his lawyer — the two most significant mitigating factors in Mr. Suter’s case. [ 196 ] Other cases suggest a trend in the jurisprudence to punish an offence under s. 255(3.2) in one of two ways: as a pure obstruction offence (when it is paired with a substantive driving conviction); or as a proxy for a substantive driving offence (when it is not paired with a substantive driving conviction). In the case of a pure obstruction offence, the sentences range from one month to a year of imprisonment (see e.g. R. v. Holliday, 2009 ONCJ 323, 87 M.V.R. (5th) 148, at para. 24, and R. v. Wallace, 2012 MBCA 54, 280 Man. R. (2d) 209, at paras. 11 and 63 ). In contrast, when the offence serves as a proxy for a substantive driving offence, the sentences, predictably, extend as high as five years of imprisonment (see e.g. Kresko, at paras. 4, 42 and 87; I note, though, that there were two victims in Kresko, not one). My point is not to doctrinally endorse this obstruction/proxy divide in the jurisprudence. Rather, it is to show that, when trial judges artificially isolate the obstructive component of s. 255(3.2), the sentences tend to hover around one year or less. In this case — where Mr. Suter was sober and involved in an accident unrelated to impairment — only the obstructive component of s. 255(3.2) is implicated. It follows that a sentence of less than a year in prison would appear to be sufficiently fit in light of the jurisprudence, which reflects the wisdom of trial judges who regularly determine the appropriate sentences for these types of offences. [ 197 ] The flexibility described above demonstrates, in my view, precisely how Parliament intended the offence to operate: harshly enough to deter any incentive for refusal, but flexibly enough to recognize that refusal is not coterminous with impaired driving in all situations. As the sentencing judge noted below: The court cannot accept that a sentence for this offence must always be the same as it would be for impaired driving causing death, regardless of the facts. The disincentive for one to refuse comes from the exposure to the same penalty, not some kind of pre-ordained range triggered by a refusal regardless of context. If Parliament intended that there should be a fixed penalty or fixed range, it would have been easy to say that. [Emphasis in original; para. 65.] [ 198 ] Indeed, the mandatory minimum sentence for refusing to provide a breath sample following a fatal accident is a non-custodial one, namely a $1,000 fine (see Criminal Code, s. 255(1)(a)(i) and (3.3) ). (4) Conclusion on Flawed Outcome [ 199 ] Based on the unique facts in this case and the jurisprudence described above, I find that the majority has no basis to claim that Mr. Suter’s four-month sentence was “manifestly unfit” (at para. 94). In reality, their criticism is not that his sentence was a “substantial and marked departure from the sentences customarily imposed for similar offenders committing similar crimes” ( R. v. M. (C.A.), 1996 230 (SCC), [1996] 1 S.C.R. 500, at para. 92; see also Lacasse, at para. 52 ), but that it was not one they would have imposed. [ 200 ] I end with this. The majority admonish the Court of Appeal for sentencing Mr. Suter as if he had been convicted of dangerous driving causing death (the 26-month sentence). But they also criticize the sentencing judge for giving excessive weight to Mr. Suter’s confusion and sobriety (the four-month sentence). It appears to me that the majority have far more significant concerns about the Court of Appeal’s sentencing methodology than about that of the sentencing judge. Indeed, the majority, in the end, essentially weigh all the same factors that were weighed by the sentencing judge and simply come up with a different number. With that in mind, it seems odd to me that the majority — setting aside their generous views on the mitigating effect of awaiting a decision on an appeal — would favour a sentence closer to that imposed by the Court of Appeal than to that imposed by the sentencing judge. IV. Conclusion [ 201 ] This case is about a fatal car accident which was just that: an accident. It was neither caused by impairment nor proven to be caused by distraction. Mr. Suter clearly committed a criminal offence by refusing to provide a breath sample. But he refused only because his state-provided lawyer — to whose advice he was constitutionally entitled — expressly and repeatedly instructed him to do so. Mr. Suter did what, I expect, most Canadians would have done. In a complex and unfamiliar administrative context, he followed his lawyer’s unequivocal advice (in keeping with his constitutional right to seek out that advice ( Singh, at para. 33 )). The majority hold that — but for their notional stay on appeal — Mr. Suter should go to prison for as many as 18 months. Respectfully, I cannot agree. [ 202 ] I would therefore allow Mr. Suter’s appeal in part. But unlike the majority, as I find there is no legal basis to justify appellate intervention with the initial sentence here, I would set aside the 26-month sentence of imprisonment imposed by the Court of Appeal and restore the 4-month sentence imposed by the sentencing judge. I would uphold the 30-month driving prohibition. Appeal allowed in part, Gascon J. dissenting in part. Solicitors for the appellant: Bottos Law Group, Edmonton. Solicitor for the respondent: Attorney General of Alberta, Edmonton. [1] This general rule is subject to two exceptions, neither of which applies in this case. The first exception is codified in s. 725(1) (c) of the Criminal Code. Section 725(1) (c) provides that a court may — when determining an appropriate sentence for an offender — consider any facts forming part of the circumstances of the offence that could form the basis of a separate charge. Where this provision is invoked, the court must note these facts on the information or indictment pursuant to s. 725(2) (b) of the Criminal Code. This was not the case here. As such, the s. 725(1) (c) exception does not apply. The second exception is where the facts that may form the basis of a separate charge are relevant to the offender’s character or reputation (see Angelillo, at paras. 29-31 ). In such a case, courts may consider these facts when sentencing an offender. Again, this was not the case here. This second exception therefore does not apply. [2] Collateral consequences may also be relevant to the objectives of general and specific deterrence and denunciation. However, such considerations have no application on the facts of this case. [3] To be clear, though relevant collateral consequences must be taken into account in a sentencing analysis, the attenuating effect of such consequences on the sentence imposed will differ depending on the circumstances. Indeed, in some cases it may be that the collateral consequence will have no impact on the sentence imposed. [4] Accepting, as we must for present purposes, the sentencing judge’s findings against the Brydges lawyer, the lawyer’s advice was not only incorrect, it represented a serious failure to keep abreast of the governing law and a troublesome breakdown in the provincial program carrying out the Brydges mandate — a program designed to ensure that all detained persons have access to legal advice under s. 10( b ) of the Canadian Charter of Rights and Freedoms (see Brydges, at pp. 214-15). [5] The duration of the pre-sentence custody was not stated in the sentencing decision, but has been confirmed with the Chief Sentence Administrator of the Edmonton Remand Centre and the parties on appeal. [6] The length of time served by Mr. Suter was not stated in the sentencing decision, but has been confirmed with the Chief Sentence Administrator of the Edmonton Remand Centre and the parties on appeal. [7] Though, since the majority finds that Mr. Suter was uncertain, they would presumably endorse a lower sentence in a future case where a trial judge expressly finds that a detainee felt certain that refusal was lawful, assuming all other facts are equal.